本期机器人顶刊精选

Switchable elements are central to both technological devices and biological machines because they enable controlled and reversible transitions between distinct functional states. Here, we present a DNA origami–based, mechanically bistable snap-through mechanism that can be electrically controlled. This nanoscale switching mechanism exhibits long-term stability in both states in the absence of external stimuli while achieving millisecond-scale switching times upon application of an electric field. Individual devices sustain hundreds of thousands of switching cycles over several hours and remain functional for actuation over several days, offering a powerful platform for systematically studying the endurance and failure mechanisms of biomolecular nanoswitches. As a nanoscale electromechanical interface, our device enables applications in molecular information processing, optical nanodevices, and the dynamic control of chemical reactions. We demonstrate that functionalization with gold nanorods facilitates polarization-dependent optical modulation, establishing direct application in plasmonics. We further show that controlling the accessibility of a molecular binding site allows electrical regulation of reaction kinetics, thereby directly coupling mechanical switching to biochemical function.

Visibility underwater is challenging and degrades as the distance between the subject and the camera increases. That is why forward‐looking underwater computer vision tasks are difficult. We have collected underwater forward‐looking stereovision and visual–inertial image sets using two underwater imaging platforms, a stereo camera rig, and an ROV in the Mediterranean and Red Seas. To our knowledge, there are no other public data sets in the underwater environment with this forward‐looking camera‐sensor orientation that have published ground‐truth depth maps as well as pose. These data sets are critical for the development of several underwater applications, including autonomous obstacle avoidance, visual odometry, 3D tracking, Simultaneous Localization and Mapping and depth estimation through deep learning. The stereo data sets contain synchronized stereo images, and the visual–inertial data sets include monocular images and inertial measurement unit (IMU) measurements with millisecond‐level timestamp alignment. All data was collected in dynamic underwater environments with objects of known size. Both sensor configurations allow for scale estimation, with the calibrated baseline in the stereo setup and the IMU in the visual–inertial setup. Ground‐truth depth maps were created offline for both data set types using a commercial photogrammetry software (Agisoft Metashape). The ground truth is validated with multiple known measurements placed throughout the imaged environment. There are four stereo and 12 visual–inertial data sets in total, each containing thousands of images, with a range of different underwater visibility and ambient light conditions, natural and man‐made structures, and dynamic camera motions. The forward‐looking orientation of the camera plus the corresponding ground truth makes these data sets unique and ideal for testing underwater obstacle‐avoidance algorithms and for navigation close to the seafloor in dynamic environments. We show results from an experiment with a monocular depth estimation algorithm to demonstrate the applicability of the data sets. With our data sets, we hope to encourage the advancement of autonomous functionality for underwater vehicles in dynamic and/or shallow‐water environments.

The mechanical complexity of soft robots creates significant challenges for their model-based control. Specifically, linear data-driven models have struggled to control soft robots on complex, spatially extended paths that explore regions with significant nonlinear behavior. To account for these nonlinearities, we develop here a model-predictive control strategy based on the recent theory of adiabatic spectral submanifolds (aSSMs). This theory is applicable because the internal vibrations of heavily overdamped robots decay at a speed that is much faster than the desired speed of the robot along its intended path. In that case, low-dimensional attracting invariant manifolds (aSSMs) emanate from the path and carry the dominant dynamics of the robot. Aided by this recent theory, we devise an aSSM-based model-predictive control scheme purely from data. We demonstrate the effectiveness of our data-driven model in tracking dynamic trajectories across diverse tasks. We validate on high-fidelity, high-dimensional finite-element models of a soft trunk robot and Cosserat-rod-based elastic soft arms, with additional experiments confirming robust performance even in the presence of experimental noise. Notably, we find that five- or six-dimensional aSSM-reduced models outperform the tracking performance of other data-driven modeling methods by a factor up to 10 across all closed-loop control tasks.

Conventional field phenotyping methods are typically manual, time‐consuming, and destructive, creating a bottleneck for breeding progress. To address this challenge, robotics and automation technologies offer efficient sensing tools to monitor field evolution and crop development throughout the season. This study presents an end‐to‐end automated pipeline for terrestrial laser scanning (TLS) in plant breeding trials, built around a Husky ground robot equipped with a high‐resolution survey‐grade FARO 3D LiDAR scanner. Unlike prior TLS phenotyping approaches relying on manual scan placement or heuristic site selection, our system integrates a novel analytical 3D ray‐casting method for optimized TLS site planning with an offline route optimization algorithm that accounts for crop growth stages and field accessibility constraints. This enables efficient planning in complex breeding environments, reduces manual labor, and improves data collection efficiency, addressing scalability challenges in large breeding trials. Leveraging Real Time Kinematic‐Global Navigation Satellite System (RTK‐GNSS) and sensor fusion, the system achieved average errors below 0.6 cm for position and for heading, enabling point cloud registration with mean errors around 2 cm, comparable to traditional manual methods that require artificial targets. The platform was successfully deployed and evaluated in two distinct cotton breeding field layouts, demonstrating the platform's capability to autonomously collect accurate TLS data for quantitative plant phenotyping across varying plot configurations. The proposed autonomous phenotyping system advances scalable, efficient phenotyping workflows to support breeding programs for crop improvement, highlighting the potential for broader deployment in field phenomics.

Accurate and robust localization is a fundamental requirement for service and inspection robots, particularly in feature-sparse indoor environments where traditional systems struggle due to a lack of distinct landmarks. While prior maps can enhance robustness, precise and compact maps capturing real-world details are often unavailable for new or frequently changing environments. This paper presents BIM-Loc, a novel discrepancy-aware LiDAR-based localization method that directly integrates Building Information Models (BIM) from the design phase. BIM-Loc simultaneously estimates trajectories aligned with the BIM coordinate system and identifies discrepancies between real-world observations and the as-designed BIM in an online fashion. Our core contributions include: (1) a novel multi-hit ray casting strategy for efficient BIM-point data association and projection of 3D observations into 2D texture space; (2) a pose graph optimization framework with BIM-integrated factors that enforces consistency among odometry, sequential scans, and BIM structures; and (3) a hierarchical Bayesian inference module that incrementally updates a continuous 2D surface representation for discrepancy detection, propagating updates from the pixel to the structure level. Extensive evaluations in both simulation and real-world applications demonstrate that BIM-Loc significantly outperforms state-of-the-art map-based methods in localization accuracy and robustness. More experimental results are available at our project website: https://bim-loc.github.io/bim-loc .

This article studies the time-optimal path planning problem for a convexified Reeds-Shepp (CRS) vehicle on a unit sphere, capable of both forward and backward motion, with speed bounded in magnitude by 1 and turning rate bounded in magnitude by a given constant. For the case in which the turning-rate bound is at least 1, using Pontryagin's Maximum Principle and a phase-portrait analysis, we show that the optimal path connecting a given initial configuration to a desired terminal configuration consists of at most six segments drawn from three motion primitives: tight turns, great circular arcs, and turn-in-place motions. A complete classification yields a finite sufficient list of 23 optimal path types with closed-form segment angles derived. The complementary case in which the turning-rate bound is less than 1 is addressed via an equivalent reformulation. The proposed formulation is applicable to underactuated satellite attitude control, spherical rolling robots, and mobile robots operating on spherical or gently curved surfaces. The source code for solving the time-optimal path problem and visualization is publicly available at https://github.com/sixuli97/Optimal-Spherical-Convexified-Reeds-Shepp-Paths.

This paper presents a framework to prompt multi-robot teams with high-level tasks using natural language expressions. Our objective is to use the reasoning capabilities of language models in understanding and decomposing multi-robot collaboration and decision-making tasks, but in settings where such models cannot be called at deployment time. However, it is hard to specify the behavior of an individual robot from a team instruction, and have it continuously adapt to actions from other robots. This necessitates a framework with the representational capacity required by the logic and semantics of a task, and yet supports decentralized, real-time operation. We solve this dilemma by recognizing that a task can be represented as a deterministic finite automaton, and that recurrent neural networks (RNNs) can encode numerous automata. This allows us to distill the logic and sequential decompositions of sub-tasks obtained from a language model into an RNN, and align its internal states with the semantics of a given task. This leads to a tiny model that encapsulates the reasoning of the language model and can be implemented onboard. To interpret the internal state of the RNN for a decentralized execution, we train a graph neural network control policy conditioned on the hidden states of the RNN and the language embeddings. We present evaluations on simulated and real-world multi-robot tasks that require sequential and collaborative behavior by the team, demonstrating scalable, robust, real-time performance -- sites.google.com/view/prompting-teams.

honey bees serve as inspiration for long-range autonomous flight in resource-constrained aerial robot navigation.

Learning long-horizon robotic manipulation requires jointly achieving expressive behavior modeling, real-time inference, and stable execution, which remains challenging for existing generative policies. Diffusion-based approaches offer strong modeling capacity but incur high inference latency, while flow matching enables fast, near-single-step generation yet often suffers from unstable execution when operating directly in the raw action space. We propose Continuous Latent Action Flow Policy (CoLA-Flow Policy), a trajectory-level imitation learning framework that performs flow matching in a continuous latent action space. By encoding action sequences into temporally coherent latent trajectories and learning an explicit latent-space flow, CoLA-Flow Policy decouples global motion structure from low-level control noise, enabling smooth and reliable long-horizon execution. The framework further integrates geometry-aware point cloud conditioning and execution-time multimodal modulation, using visual cues as a representative modality to enhance real-world robustness. Experiments in simulation and on real robots show that CoLA-Flow Policy achieves near-single-step inference, improves trajectory smoothness by up to 93.7% and task success by up to 25 percentage points over raw action-space flow baselines, while remaining significantly faster than diffusion-based policies.

Trunk impairment from neurological conditions such as spinal cord injury, cerebral palsy, and stroke limits postural coordination and functional independence. Sense of agency (SoA), the perceived control over one’s movements, influences motor learning and rehabilitation engagement, yet has not been examined in seated postural training. This study investigates how robotic pelvic guidance and visual feedback (VF) designs influence motor coordination and SoA during a seated reaching task in virtual reality (VR), validated here with able-bodied subjects. Thirty-two healthy adults were randomly assigned to four groups ( n = 8) in a 2 × 2 between-subjects design: pelvic guidance (guided vs unguided) and VF type (control VF, CVF: a pursuit-tracking moving target, vs error-based VF, EVF: displaying only the hand–target error). In the guided condition, the pelvic Wheelchair Robot for Active Postural Support (pWRAPS) actively guided pelvic orientation during a four-direction hand-tracking game. In the unguided condition, the robot operated in transparent mode. Participants receiving pelvic guidance showed significantly greater improvements in task performance and pelvis–trunk–hand coordination than unguided participants, regardless of VF type. EVF influenced trunk–pelvis coordination independently of guidance, with CVF yielding greater trunk–pelvis improvement. Trunk–pelvis coordination emerged as the strongest predictor of SoA, and SoA increased significantly at post-test, suggesting that perceived agency reflects coordination quality rather than the presence or absence of robotic assistance. Critically, guidance improved coordination without diminishing SoA, indicating compatible biomechanical and psychological rehabilitation goals. These findings support integrating robotic pelvic guidance with VR-based feedback for seated postural rehabilitation.

Active perception is a fundamental problem in autonomous robotics in which the robot must decide where to move and what to sense in order to obtain the most informative observations for accomplishing its mission. Existing approaches either solve a computationally expensive traveling salesman problem over heuristically selected informative nodes, or adopt a more efficient but overly constrained shortest path tree formulation. To address these limitations, we explore beam search algorithms as scalable alternatives. While the standard beam search provides scalability by preserving the top- B paths at each depth level, it is prone to local optima and exhibits parameter sensitivity. Our first contribution is a node-wise beam search (NBS) algorithm, which maintains top- B candidates per node to enable more effective exploration of the solution space. Systematic benchmarking on graphs shows that NBS consistently outperforms other baselines and maintains strong performance even at low beam widths. As a second contribution, we integrate the concept of frontiers into the path selection criterion, introducing the expected gain metric, which better balances exploration and exploitation compared to existing alternatives. Our third contribution proposes the rapidly-exploring random annulus graph (RRAG), a novel graph construction method that preserves full orientation sampling and ensures connectivity in cluttered environments through a fallback local sampling-based planner. Extensive experiments demonstrate that NBS combined with RRAG achieves the highest performance across all three representative active perception tasks, outperforming state-of-the-art algorithms by at least 20% in one or more tasks. We further validate the approach on real robotic platforms in different scenarios. Project page: https://efficient-beam-search.github.io/

Many hazardous tasks, such as those related to search and rescue scenarios, involve interactions with granular materials. In such examples, tactile sensing can be invaluable, since objects of interest are often not visible but may still be haptically salient. When a tool, such as a robot finger, approaches a buried object, the displaced granular material ahead of the tool begins to compact against the buried object. The resulting granular media jamming causes an increase in contact force felt on the robot finger without the finger making direct contact with the buried object. In this work, we present a framework for tactile‐based proximity sensing that leverages theoretical models of soil failure during soil‐tool interactions and empirical models of granular material failure zones to haptically detect the onset of granular media jamming and estimate the distance from a sensorized robot fingertip to an object buried in granular material. We investigated the effects of granular material type on proximity sensing accuracy by studying glass beads, poppy seeds, and playground sand. For various fingertip and object depths, we estimated the distance from a sensorized finger to a buried object during haptic exploration in granular materials on a real robot. We developed particle filters to estimate the force state on the robot fingertip in real‐time, and showed how occupancy maps can be continuously updated during haptic exploration within granular materials with a real robot. Our modeling framework can also be leveraged to simulate haptic interactions with granular materials prior to conducting real robot experiments.

With the increasing demand for inspection, maintenance, and operation in vertical and complex‐geometry environments, wall‐climbing robots have become essential equipment for high‐rise infrastructure maintenance, large‐scale industrial inspection, and emergency response. The codesign of adsorption and locomotion is fundamental to ensuring reliable attachment, stable mobility, and robust environmental adaptability. This paper provides a comprehensive review of major adsorption mechanisms—negative pressure, magnetic adsorption, electrostatic adsorption, bioinspired dry adsorption, and mechanical interlocking—and examines their underlying dynamics. Typical locomotion architectures, including wheeled, tracked, legged, peristaltic, and flipping mechanisms, are compared in terms of mobility characteristics and application boundaries across different surface conditions. Key engineering advances are summarized, including adsorption–locomotion coupling, curvature‐conforming structures, tunable adsorption, lightweight design, multimodal integration, and intelligent control. Considering emerging application scenarios, such as offshore wind turbines and nuclear power facilities—where resistance to wind loads, radiation, corrosion, and cross‐medium operation are required—future research directions are identified, such as multimodal adsorption fusion, compliant structural materials, enhanced environmental perception, and adaptive motion planning. This review aims to provide coherent technical guidance for the development of next‐generation high‐reliability, high‐adaptability wall‐climbing robotic systems

Deployment of legged robots for navigating challenging terrains (e.g., stairs, slopes, and unstructured environments) has gained increasing preference over wheel-based platforms. In such scenarios, accurate odometry estimation is a preliminary requirement for stable locomotion, localization, and mapping. Traditional proprioceptive approaches, which rely on leg kinematics sensor modalities and inertial sensing, suffer from irrepressible vertical drift caused by frequent contact impacts, foot slippage, and vibrations, particularly affected by inaccurate roll and pitch estimation. Existing methods incorporate exteroceptive sensors such as light detection and ranging (LiDAR) or cameras. Further enhancement has been introduced by leveraging gravity vector estimation to add additional observations on roll and pitch, thereby increasing the accuracy of vertical pose estimation. However, these approaches tend to degrade in feature-sparse or repetitive scenes and are prone to errors from double-integrated IMU acceleration. To address these challenges, we propose GaRLILEO , a novel gravity-aligned continuous-time radar-leg-inertial odometry framework. GaRLILEO decouples velocity from the IMU by building a continuous-time ego-velocity spline from SoC radar Doppler and leg kinematics information, enabling seamless sensor fusion which mitigates odometry distortion. In addition, GaRLILEO can reliably capture accurate gravity vectors leveraging a novel soft S 2 -constrained gravity factor, improving vertical pose accuracy without relying on LiDAR or cameras. Evaluated on a self-collected real-world dataset with diverse indoor-outdoor trajectories, GaRLILEO demonstrates state-of-the-art accuracy, particularly in vertical odometry estimation on stairs and slopes. We open-source both our dataset and algorithm to foster further research in legged robot odometry and SLAM. https://garlileo.github.io/GaRLILEO/ .

The problem of exploration in unknown environments is still a great challenge for autonomous mobile robots due to the lack of a priori knowledge. Active Simultaneous Localization and Mapping (SLAM) is an effective method to realize obstacle avoidance and autonomous navigation. Traditional Active SLAM is usually complex to model and difficult to adapt automatically to new operating areas. This paper presents a novel hybrid technique for Active SLAM algorithm based on Deep Reinforcement Learning (DRL). The Relational Proximal Policy Optimization (RPPO) model with deep separable convolution and data batch processing is used to predict the action strategy and generate the action plan through the acquired environment RGB images, so as to realize the autonomous collision free exploration of the environment. Meanwhile, Gmapping is applied to locate and map the environment. The data show that the new RPPO model has higher training efficiency due to the 30 percent reduction of training time per round, and the exploration efficiency is significantly improved under the premise of effective obstacle avoidance. Then, on the basis of the Transfer Learning, the Active SLAM algorithm is further trained in a more complex environment to optimize the model parameters to adapt to the new complex environment. Finally, we conduct several experiments to demonstrate the feasibility of the Active scanning algorithm in a gradually complex environment for the task of effective exploration and mapping that needs to be completed.

In this paper, we propose a new modeling approach and a fast algorithm for 3D motion planning, applicable for fixed-wing unmanned aerial vehicles. The goal is to construct the shortest path connecting given initial and final configurations subject to motion constraints. Our work differs from existing literature in two ways. First, we consider full vehicle orientation using a body-attached frame, which includes roll, pitch, and yaw angles. However, existing work uses only pitch and/or heading angle, which is insufficient to uniquely determine orientation. Second, we use two control inputs to represent bounded pitch and yaw rates, reflecting control by two separate actuators. In contrast, most previous methods rely on a single input, such as path curvature, which is insufficient for accurately modeling the vehicle's kinematics in 3D. We use a rotation minimizing frame to describe the vehicle's configuration and its evolution, and construct paths by concatenating optimal Dubins paths on spherical, cylindrical, or planar surfaces. Numerical simulations show our approach generates feasible paths within 10 seconds on average and yields shorter paths than existing methods in most cases.

Human-level contact-rich manipulation relies on the distinct roles of two key modalities: vision provides spatially rich but temporally slow global context, while force sensing captures rapid, high-frequency local contact dynamics. Integrating these signals is challenging due to their fundamental frequency and informational disparities. In this work, we propose ImplicitRDP, a unified end-to-end visual-force diffusion policy that integrates visual planning and reactive force control within a single network. We introduce Structural Slow-Fast Learning, a mechanism utilizing causal attention to simultaneously process asynchronous visual and force tokens, allowing the policy to perform closed-loop adjustments at the force frequency while maintaining the temporal coherence of action chunks. Furthermore, to mitigate modality collapse where end-to-end models fail to adjust the weights across different modalities, we propose Virtual-target-based Representation Regularization. This auxiliary objective maps force feedback into the same space as the action, providing a stronger, physics-grounded learning signal than raw force prediction. Extensive experiments on contact-rich tasks demonstrate that ImplicitRDP significantly outperforms both vision-only and hierarchical baselines, achieving superior reactivity and success rates with a streamlined training pipeline. Code and videos will be publicly available at https://implicit-rdp.github.io.

Bundle adjustment (BA) is a critical technique in various robotic applications such as simultaneous localization and mapping (SLAM), augmented reality (AR), and photogrammetry. BA optimizes parameters such as camera poses and 3D landmarks to align them with observations. With the growing importance of deep learning in perception systems, there is an increasing need to integrate BA with deep learning frameworks for enhanced reliability and performance. However, widely-used C++-based BA libraries, such as GTSAM, g$^2$o, and Ceres Solver, lack native integration with modern deep learning libraries like PyTorch. This limitation affects their flexibility, ease of debugging, and overall implementation efficiency. To address this gap, we introduce an eager-mode BA library seamlessly integrated with PyTorch with high efficiency. Our approach includes a sparsity-aware auto-differentiation design and GPU-accelerated sparse operations designed for 2nd-order optimization. Our eager-mode BA on GPU demonstrates substantial runtime efficiency, achieving an average speedup of 18.5$\times$, 22$\times$, and 23$\times$ across all benchmarks compared to GTSAM, g$^2$o, and Ceres, respectively.

Ensuring robot safety can be challenging; user-defined constraints can miss edge cases, policies can become unsafe even when trained from safe data, and safety can be subjective. Thus, we learn about robot safety by showing policy trajectories to a human who flags unsafe behavior. From this binary feedback, we use the statistical method of conformal prediction to identify a region of states, potentially in learned latent space, guaranteed to contain a user-specified fraction of future policy errors. Our method is sample-efficient, as it builds on nearest neighbor classification and avoids withholding data as is common with conformal prediction. By alerting if the robot reaches the suspected unsafe region, we obtain a warning system that mimics the human's safety preferences with guaranteed miss rate. From video labeling, our system can detect when a quadcopter visuomotor policy will fail to steer through a designated gate. We present an approach for policy improvement by avoiding the suspected unsafe region. With it we improve a model predictive controller's safety, as shown in experimental testing with 30 quadcopter flights across 6 navigation tasks. Code and videos are provided.

In recent years, multimodal locomotion capabilities have enabled robots to maneuver in both terrestrial and aerial domains. However, most of these robots are designed only for locomotion, and few possess the manipulation capabilities required for practical tasks. By adding a manipulator, ground robots can perform manipulation, and some drones with robotic arms have demonstrated aerial manipulation. Nonetheless, such multirotors cannot be directly used for manipulation on the ground, and this configuration itself is unsuitable for air-ground hybrid locomotion. This is because their thruster-centralized structure makes it difficult to achieve both sufficient degrees of freedom (DoF) for manipulation and stable motion with contact and transformation. Therefore, in this work, we develop a new multilink multirotor with thrusters on each link and capable of contact with the environments. This robot can perform terrestrial rolling locomotion, aerial flight locomotion, and manipulation in multiple environments using joint actuation. First, we introduce a minimal configuration design of the proposed robot. We also describe a kinematic model and propose a design for each component based on this model. Second, we propose a real-time control method based on nonlinear optimization that considers contact and joint motion, which can be applied to various multirotors. Third, we propose motion strategies that include contact constraints specific to air-ground hybrid multilink multirotors, and analyze the limitations of manipulation capabilities based on multi-contact model. Finally, we demonstrate a variety of motions in both domains using the implemented prototype. To the best of our knowledge, this is the first demonstration of air-ground hybrid locomotion and manipulation by a multilink multirotor.

An electrically controlled DNA origami snap-through switch provides robust, programmable logic for molecular robotics.

The Infinite Sadness of Small Appliances imagines the multiple ways domestic home robots can violate the privacy of a family.

Researchers have increasingly adopted Transformer-based models for inertial odometry. While Transformers excel at modeling long-range dependencies, their limited sensitivity to local, fine-grained motion variations and lack of inherent inductive biases often hinder localization accuracy and generalization. Recent studies have shown that incorporating large-kernel convolutions and Transformer-inspired architectural designs into CNN can effectively expand the receptive field, thereby improving global motion perception. Motivated by these insights, we propose a novel CNN-based module called the Dual-wing Adaptive Dynamic Mixer (DADM), which adaptively captures both global motion patterns and local, fine-grained motion features from dynamic inputs. This module dynamically generates selective weights based on the input, enabling efficient multi-scale feature aggregation. To further improve temporal modeling, we introduce the Spatio-Temporal Gating Unit (STGU), which selectively extracts representative and task-relevant motion features in the temporal domain. This unit addresses the limitations of temporal modeling observed in existing CNN approaches. Built upon DADM and STGU, we present a new CNN-based inertial odometry backbone, named Next Era of Inertial Odometry (IONext). Extensive experiments on six public datasets demonstrate that IONext consistently outperforms state-of-the-art (SOTA) Transformer- and CNN-based methods. For instance, on the RNIN dataset, IONext reduces the average ATE by 10% and the average RTE by 12% compared to the representative model iMOT.

Imitation learning has shown promise in learning from large-scale real-world datasets. However, pretrained policies usually perform poorly without sufficient in-domain data. Besides, human-collected demonstrations entail substantial labour and tend to encompass mixed-quality data and redundant information. As a workaround, human-in-the-loop systems gather domain-specific data for policy post-training, and exploit closed-loop policy feedback to offer informative guidance, but usually require full-time human surveillance during policy rollout. In this work, we devise ARMADA, a multi-robot deployment and adaptation system with human-in-the-loop shared control, featuring an autonomous online failure detection method named FLOAT. Thanks to FLOAT, ARMADA enables paralleled policy rollout and requests human intervention only when necessary, significantly reducing reliance on human supervision. Hence, ARMADA enables efficient acquisition of in-domain data, and leads to more scalable deployment and faster adaptation to new scenarios. We evaluate the performance of ARMADA on four real-world tasks. FLOAT achieves nearly 95% accuracy on average, surpassing prior state-of-the-art failure detection approaches by over 20%. Besides, ARMADA manifests more than 4$\times$ increase in success rate and greater than 2$\times$ reduction in human intervention rate over multiple rounds of policy rollout and post-training, compared to previous human-in-the-loop learning methods.

Enabling robots to autonomously discover high-level spatial concepts (e.g., rooms and walls) from primitive geometric observations (e.g., planar surfaces) within 3D Scene Graphs is essential for robust indoor navigation and mapping. These graphs provide a hierarchical metric-semantic representation in which such concepts are organized. To further enhance graph-SLAM performance, Factorized 3D Scene Graphs incorporate these concepts as optimization factors that constrain relative geometry and enforce global consistency. However, both stages of this process remain largely manual: concepts are typically derived using hand-crafted, concept-specific heuristics, while factors and their covariances are likewise manually designed. This reliance on manual specification limits generalization across diverse environments and scalability to new concept classes. This paper presents a novel learning-based method that infers spatial concepts online from observed vertical planes and introduces them as optimizable factors within a SLAM backend, eliminating the need to handcraft concept generation, factor design, and covariance specification. We evaluate our approach in simulated environments with complex layouts, improving room detection by 20.7% and trajectory estimation by 19.2%. Validated on real construction sites, room detection improves by 5.3% and map matching accuracy by 3.8%.

Consistent maps are key for most autonomous mobile robots, and they often use SLAM approaches to build such maps. Loop closures via place recognition help to maintain accurate pose estimates by mitigating global drift, and are thus key for realizing an effective SLAM system. This paper presents a robust loop closure detection pipeline for outdoor SLAM with LiDAR-equipped robots. Our method handles various LiDAR sensors with different scanning patterns, fields of view, and resolutions. It generates local maps from LiDAR scans and aligns them using a ground alignment module to handle both planar and non-planar motion of the LiDAR, ensuring applicability across platforms. The method uses density-preserving bird’s-eye-view projections of these local maps and extracts ORB feature descriptors for place recognition. It stores the feature descriptors in a binary search tree for efficient retrieval, and self-similarity pruning addresses perceptual aliasing in repetitive environments. Extensive experiments on public and self-recorded datasets demonstrate accurate loop closure detection, long-term localization, and cross-platform multi-map alignment, agnostic to the LiDAR scanning patterns, fields of view, and motion profiles. We provide the code for our pipeline as open-source software at https://github.com/PRBonn/MapClosures .

Quadrotor unmanned aerial vehicles (UAVs) are increasingly deployed in complex missions that demand reliable autonomous navigation and robust obstacle avoidance. However, traditional modular pipelines often incur cumulative latency, whereas purely reinforcement learning (RL) approaches typically provide limited formal safety guarantees. To bridge this gap, we propose an end-to-end RL framework augmented with model-based safety mechanisms. We incorporate physical priors in both training and deployment. During training, we design a physics-informed reward structure that provides global navigational guidance. During deployment, we integrate a real-time safety filter that projects the policy outputs onto a provably safe set to enforce strict collision-avoidance constraints. This hybrid architecture reconciles high-speed flight with robust safety assurances. Benchmark evaluations demonstrate that our method outperforms both traditional planners and recent end-to-end obstacle avoidance approaches based on differentiable physics. Extensive experiments demonstrate strong generalization, enabling reliable high-speed navigation in dense clutter and challenging outdoor forest environments at velocities up to$7.5 \rm{ m/s}$.

Autonomous robots are projected to significantly augment the manual workforce, especially in repetitive and hazardous tasks. For a successful deployment of such robots in human environments, it is crucial to guarantee human safety. State-of-the-art approaches to ensure human safety are either too conservative to permit a natural human-robot collaboration or make strong assumptions that do not hold for autonomous robots, e.g., knowledge of a pre-defined trajectory. Therefore, we propose the shield for Safe Autonomous human-robot collaboration through Reachability Analysis (SARA shield). This novel power and force limiting framework provides formal safety guarantees for manipulation in human environments while realizing fast robot speeds. As unconstrained contacts allow for significantly higher contact forces than constrained contacts (also known as clamping), we use reachability analysis to classify potential contacts by their type in a formally correct way. For each contact type, we formally verify that the kinetic energy of the robot is below pain and injury thresholds for the respective human body part in contact. Our experiments show that SARA shield satisfies the contact safety constraints while significantly improving the robot performance in comparison to state-of-the-art approaches.

Traditional rigid submersibles often lack the acoustic stealth and agility required for close‐range deep‐sea exploration. While soft robotics have emerged as a promising approach for navigating complex marine environments, these systems typically face a trade‐off between pressure tolerance and practical maneuverability. We present DeepTrack, an untethered soft robotic fish driven by a compact electromagnetic propulsion system. Integrating a pressure‐independent actuation mechanism into a compliant silicone body keeps the robot's mass at just 475 g and eliminates the need for bulky pressure vessels. A two‐fin configuration decouples forward propulsion from depth regulation, enabling agile 3D maneuvering. An onboard visual servoing system ensures precise closed‐loop target tracking. Experimental validations demonstrated a maximum swimming speed of 5.78 cm/s (0.27 body length (BL)) and a yaw rate of 0.26 rad/s. During dynamic trajectory tracking, the system maintained a root mean square error (RMSE) below 0.37 BL. Open‐water field trials further validated the robot's robustness in autonomously tracking dynamic targets amid ambient currents and unstructured visual clutter. Furthermore, hyperbaric chamber tests confirm that the robot preserves its structural integrity and tracking performance at hydrostatic pressures up to 30 MPa. These results demonstrate the potential of DeepTrack as a mechatronic platform for future visually guided operations in deep sea environments.

Self-reconfigurable wheeled mobile robots (SRWMRs) are capable of achieving multiple motion modes and subsequent switching between them through the coordinated sequential movements of multiple rocker–bogie joints, leading to overcoming dynamic obstacles posed by different terrains. However, real-time coordination of internal joints for executing reconfiguration actions while maintaining strong adhesion between the wheels and terrain remains challenging, particularly during action transitions. To enhance multimode motion capability by using a unified model, this work develops an inverse kinematics control (IKC) method, including 3D kinematic modeling of an SRWMR with actively and passively articulated suspensions, and additional motion-constraint inequalities for multi-joints in the wheel-suspension system. Specifically, the 3D model is built to achieve horizontal movements and vertical lifting of the robot chassis and its wheels. To further stabilize body posture and reduce wheel slippage during multimode motion, motion constraints are proposed to regulate the relative velocities among multiple joints and the displacement of the robot’s center of mass. According to the results of physical experiments with the HIT-MRII robot, Wheel Rolling, Wheel Crabbing, Wheel Lifting, Robot Chassis Lifting, Robot Creeping modes, and parts of their hybrid modes are achieved steadily and safely by the developed IKC method. The maximum motion performance of each mode is achieved by the proposed motion constraints. The enhanced mobility of the robot is demonstrated by comparing various traversal modes on soil terrain and presenting the corresponding control strategies.

Partially observable Markov decision processes ( pomdp s) are a general and principled framework for motion planning under uncertainty. Despite tremendous improvement in the scalability of pomdp solvers, long-horizon pomdp s remain difficult to solve. To alleviate the difficulty, this paper proposes a new approximate online pomdp solver, called reference-based online pomdp planning via rapid state space sampling ( rop-ras3 ). rop-ras3 uses novel extremely fast sampling-based motion planning techniques to sample the state space and generate a diverse set of macro-actions online, which are then used to bias belief-space sampling and infer high-quality policies without requiring exhaustive enumeration of the action space—a fundamental constraint for modern online pomdp solvers. rop-ras3 converges to a near-optimal reference-based solution at a rate that depends on the number of sampled actions, rather than the size of the action space. rop-ras3 is evaluated on various long-horizon pomdp s with up to 3000 lookahead steps and 35-dimensional state spaces, where the state, action and observation spaces can be continuous, discrete, or a hybrid of discrete and continuous. Although the reference-based optimal solution may not be the same as the optimal pomdp solution, empirical results indicate that in all of these problems, in terms of success rate, rop-ras3 outperforms other state-of-the-art methods by up to multiple folds . We also demonstrate the capability of our approach on a physical robot demonstration. This work extends the theory and empirical results of our ISRR24 paper. Code can be found at https://github.com/RDLLab/ROPRAS3 .

This paper proposes an online lifelong dynamic learning-based outer-loop velocity compensation control scheme for$n$-degree-of-freedom robotic manipulators operating in continuous multi-task environments. A radial basis function neural network (RBF NN) incorporating a neuron dynamic-growing strategy is employed in the actual controller, enabling real-time adjustment of neuron compact sizes according to NN inputs and facilitating online identification of unknown system dynamics. Furthermore, an online weight feedback mechanism is integrated into the neural network learning law to preserve previously learned weight parameters during task execution. By introducing an S-shaped filtering function, significant synaptic weights are assigned higher feedback gains, whereas less important weights are gradually suppressed toward zero, effectively mitigating catastrophic forgetting. In contrast to conventional dynamic learning control approaches, the proposed scheme enables the retrieval of historical knowledge when online revisiting prior tasks, thereby ensuring sustained control accuracy over time. Rigorous theoretical analysis demonstrates that all closed-loop signals remain uniformly bounded, and both weight estimation errors and system identification errors converge exponentially to a small residual neighborhood around zero. Finally, experiments on a UR5 robotic manipulator validate the effectiveness of the proposed method.

The endurance limitations of onboard batteries and operational interruptions caused by static charging remain major barriers to scalable unmanned aerial vehicle (UAV) deployment in emerging low‐altitude economy. Dynamic wireless power transfer (DWPT) has recently emerged as a key enabling technology for persistent aerial mobility by allowing energy replenishment during flight without landing. This paper presents a comprehensive survey and introduces a unified field‐motion system framework that links DWPT mechanisms with UAV mobility states and operational requirements. Existing DWPT approaches are systematically classified along two orthogonal dimensions: physical field characteristics (near‐field inductive coupling and far‐field radiative transmission) and motion dynamics (quasi‐static hovering to fully dynamic flight). The survey reveals that near‐field ground‐to‐air configurations achieve high efficiency and robustness through Parity‐Time (PT)‐symmetric compensation and three‐dimensional coil structures, while far‐field air‐to‐air energy transfer enabled by beam‐shaped laser and microwave arrays supports long‐range energy relay under dynamic alignment uncertainty. Building upon these insights, the paper proposes a layered heterogeneous energy architecture integrating high‐efficiency local charging nodes with flexible long‐distance energy relays, forming a multi‐tier aerial energy replenishment network analogous to aerial refuelling systems. By explicitly connecting energy transfer modalities with UAV operational dynamics, infrastructure deployment, and persistent mission planning, this review establishes a system‐level perspective for energy‐aware UAV transportation networks. The presented framework provides design guidelines and research directions toward scalable, continuously operating aerial logistics, inspection, and sensing services in future low‐altitude intelligent transportation systems.

Reliable autonomous docking of AUV under ice is critical for operational safety, as maintaining vehicle integrity and achieving full‐lifecycle autonomous operations are unique challenges in ice hole recovery. This study presents a vision‐guided navigation framework specifically designed for ice hole docking, which enables accurate identification and localization of ice hole light beacons array. The framework integrates an adaptive selection algorithm that ensures robust light source detection across varying lighting conditions and operational distances, along with a dual‐quadrilateral geometry‐based recognition method capable of accurately discerning beacon positions in different visibility conditions during docking process. A monocular pose estimation algorithm provides real‐time and precise relative position and heading information, while an ice hole alignment strategy guides the AUV for safe and accurate docking. Images of six beacons arranged around the ice hole were captured by an upward‐facing camera mounted on the AUV and processed through the proposed pipeline. Extensive simulations and real‐world experiments demonstrate that the method significantly improves positioning accuracy, robustness, and overall navigation reliability, offering a practical and effective solution for underwater optical vision‐guided docking under ice.

After a stroke, individuals often experience mobility impairments because of weakness and loss of independent joint control in the lower limbs. As a result, gait recovery becomes a primary goal of physical rehabilitation, traditionally achieved through high-intensity therapist-led training. However, conventional therapist-led approaches involving manual assistance or resistance can be physically demanding and limit interaction at multiple joints simultaneously. Robotic exoskeletons have emerged as a promising solution, enabling multijoint support, reducing therapist strain, and offering objective performance feedback. However, typical exoskeleton control strategies limit the physical therapist’s involvement and adaptability to the patient’s needs, which may hinder clinical adoption and outcomes. In this study, we introduce a gait rehabilitation paradigm based on physical human-robot-human interaction that we call therapist-exoskeleton-patient interaction (TEPI), in which a therapist and a patient with stroke are each equipped with a lower-limb exoskeleton virtually connected at the hips and knees via spring-damper elements. This connection enables bidirectional physical interaction, allowing the therapist to guide the patient’s movement while receiving real-time haptic feedback. We evaluated this approach with eight patients with chronic stroke using a within-subject design, comparing TEPI training with conventional therapist-guided mobilization during treadmill walking. Results showed that, compared with conventional therapy, TEPI led to greater joint range of motion, increased step length and height, similar muscle activation, and high self-reported motivation and enjoyment. These findings suggest that TEPI can integrate robotic precision with therapist intuition, offering a framework for enhancing gait rehabilitation outcomes in populations recovering from stroke.

The Moving Target Traveling Salesman Problem (MT-TSP) seeks a trajectory that intercepts several moving targets, within a particular time window for each target. When generic nonlinear target trajectories or kinematic constraints on the agent are present, no prior algorithm guarantees convergence to an optimal MT-TSP solution. Therefore, we introduce the Iterated Random Generalized (IRG) TSP framework. The idea behind IRG is to alternate between randomly sampling a set of agent configuration-time points, corresponding to interceptions of targets, and finding a sequence of interception points by solving a generalized TSP (GTSP). This alternation asymptotically converges to the optimum. We introduce two parallel algorithms within the IRG framework. The first algorithm, IRG-PGLNS, solves GTSPs using PGLNS, our parallelized extension of state-of-the-art solver GLNS. The second algorithm, Parallel Communicating GTSPs (PCG), solves GTSPs for several sets of points simultaneously. We present numerical results for three MT-TSP variants: one where intercepting a target only requires coming within a particular distance, another where the agent is a variable-speed Dubins car, and a third where the agent is a robot arm. We show that IRG-PGLNS and PCG converge faster than a baseline based on prior work. We further validate our framework with physical robot experiments.

Trust forms the bedrock of successful human interactions, and its integration into human–robot collaboration remains a critical challenge. Contemporary research predominantly explores human trust in robotic systems, focusing on refining the appearance and behavior of artificial agents to foster their acceptability in social settings. However, this systematic review centers on the less-explored dimension of trust mechanisms within autonomous robotic systems. Our aim is to survey what we define as Computational Trust models, which encompass a robot’s capability to both assess the trustworthiness of other agents (“Artificial Trust”) and to predict their levels of trust towards itself (“Natural Trust”). To achieve this objective, an initial set of 1916 papers, ranging from 2013 to 2023, was collected from IEEE Xplore, Scopus, and ISI Web of Science. Eligibility criteria were then applied to this set to select works that designed a Computational Trust model for a robotics application, which was validated through an experiment. These criteria were agreed upon by all authors to ensure unanimous decisions on whether to retain or remove results. At the end of this process, 101 key papers were identified. Following the selection process, we conducted thorough analyses to cluster these works based on the type of Computational Trust model used, the application domain, the robotic platforms employed in the validation, the experimental design, and the evaluation metrics. Finally, we identify common trends in this emerging branch of Human–Robot Interaction and provide guidelines for scholars wishing to contribute to this field.

In this letter, we present a hierarchical control framework that enables wheeled bipedal robots to perform planar object sliding tasks with their wheeled legs. The proposed approach formulates a nonlinear model predictive controller (NMPC) based on a reduced-order three rigid bodies (TRB) dynamical model that explicitly accounts for the hip roll degree of freedom and multiple wheel-environment contact modes, which is essential for lateral stepping and pedipulation tasks. Within this framework, the NMPC simultaneously regulates robot locomotion and interaction forces, allowing the robot to stably execute both rolling and object manipulation behaviors. A trajectory-optimization-based robot-object motion planner is developed to generate reference motions that incorporate stick-slip transitions in ground-object contact. Two representative pedipulation motions, namely scooting and lateral sliding, are validated through real-world hardware experiments, in which the robot successfully retrieves a 1 kg object from under a desk and slides a 4 kg object over a distance of 0.228 m via scooting.

We present CHAI (Compliant Human-centered Adaptive Interaction), a novel language-driven framework for real-time modulation of a robot's kinematics and mechanical compliance in real-world environments. By interpreting natural language instructions and visual context through pre-trained vision-language encoders, CHAI combines a transformer-based geometry encoder with a conditional diffusion model to iteratively refine a nominal kinematic trajectory and its associated compliance profile. CHAI introduces on-the-fly language-driven impedance (compliance) modulation along both translational and rotational directions—including motion-aligned and radial axes—executed through a passivity-aware and therefore stable Cartesian impedance controller. This capability is key for supporting compliant interaction, improving adaptability and reducing hazardous contact in physical interaction tasks. Comprehensive experiments demonstrate significant gains in scalability, adaptability, trajectory accuracy, and interactive behaviour over prior methods. Ablation studies further validate the contributions of stiffness control and multi-modal conditioning.

For underwater tasks that require an efficient approach over long distances and stable adhesion for close‐range crawling, this paper proposes a propeller–leg hybrid underwater adhesion robot with integrated swimming and crawling capabilities. First, we present the overall design of a robot that integrates a bionic leg–foot mechanism and a propeller propulsion system. Second, we design a hydrogel biomimetic sucker with reversible adhesion and detachment, accompanied by an adhesion performance test. A dual‐criterion system for static stability margin and adhesion force reliability was proposed during the robot crawling stage. On the basis of computational fluid dynamics simulation results, a critical instability model was developed to quantify the effects of flow velocity, slope, and load on system stability. We have developed a crawling–swimming collaborative control strategy to address the mode‐switching issue during locomotion. This strategy employs a hierarchical control architecture guided by stability criteria, in which central pattern generator–based rhythmic control dynamically integrates with sliding‐mode propulsion compensation via switching signals. The experimental results show that the proposed robot can achieve stable attachment motion through propeller–leg coordination under special working conditions (inverted surfaces and slopes). The proposed robot provides a multimodal solution for underwater operations that require a rapid target approach in open water, followed by stable attachment and adhesive crawling on structural surfaces.

Left Atrial Appendage Closure (LAAC) is an established therapy for reducing stroke risk in patients with atrial fibrillation who are contraindicated for long-term anticoagulation. Current LAAC procedures rely on manual catheter manipulation under fluoroscopic image guidance to seal the LAA by means of a self-expanding occluder. Assessment of occluder deployment typically involves qualitative methods such as tactile feedback and x-ray imaging. This work presents an autonomous robotic approach for LAA occluder implantation based on electromagnetic (EM) tracking. Additionally, to overcome the limitations of qualitative assessment of occluder-tissue engagement during deployment, a novel fiber Bragg grating (FBG)-based force sensor is integrated at the tip of the occluder deployment system. A Naive Bayes classifier is trained to predict the occluder deployment outcome based on intra-operatively acquired force-displacement curves. The system's capability is demonstrated in a phantom model of the left atrial anatomy, achieving mean positional errors of 0.84 mm during autonomous navigation and 1.32 mm during autonomous deployment across five runs. All runs culminated in successful occluder placement and a force-sensing based deployment quality classification accuracy of 83.3%.

Reliable fault diagnosis is essential for the safe operation of autonomous underwater vehicles (AUVs) in uncertain and dynamic environments. However, conventional closed‐set diagnosis methods are unable to handle previously unseen fault conditions, while existing open‐set techniques often struggle with low‐frequency, highly coupled multivariate telemetry. This paper proposes ProtoNet‐MDAG, a prototype‐based open‐set diagnosis framework for AUV fault recognition. Specifically, a dilated multi‐scale encoder is developed to extract discriminative temporal features from low‐frequency sensor streams, and a variance‐penalized prototypical learning strategy is introduced to enforce compact and well‐structured known‐class manifolds in the latent space. Based on this representation, an adaptive Mahalanobis distance gating mechanism is constructed to perform statistically calibrated open‐set rejection. Experimental results on a Haizhe AUV and a BlueROV‐class underwater robotic platform show that the proposed method achieves 97.19% and 89.12% accuracy, respectively. The results demonstrate strong unknown‐state rejection on the Haizhe AUV and competitive known‐class recognition on the BlueROV‐class platform, while also revealing that platform‐dependent distribution shift can weaken unknown‐state rejection in more heterogeneous operating conditions. Robustness tests on the Haizhe dataset further show that the framework remains stable under representative telemetry perturbations. These results demonstrate that the proposed framework provides an effective and reliable solution for open‐set fault diagnosis in underwater robotic systems.

Visual navigation models often struggle in real-world dynamic environments due to limited robustness to the sim-to-real gap and the difficulty of training policies tailored to target deployment environments (e.g., households, restaurants, and factories). Although real-to-sim navigation simulation using 3D Gaussian Splatting (GS) can mitigate these challenges, prior GS-based works have considered only static scenes or non-photorealistic human obstacles built from simulator assets, despite the importance of safe navigation in dynamic environments. To address these issues, we propose ReaDy-Go, a novel real-to-sim simulation pipeline that synthesizes photorealistic dynamic scenarios in target environments by augmenting a reconstructed static GS scene with dynamic human GS obstacles, and trains navigation policies using the generated datasets. The pipeline provides three key contributions: (1) a dynamic GS simulator that integrates static scene GS with a human animation module, enabling the insertion of animatable human GS avatars and the synthesis of plausible human motions from 2D trajectories, (2) a navigation dataset generation framework that leverages the simulator along with a robot expert planner designed for dynamic GS representations and a human planner, and (3) robust navigation policies to both the sim-to-real gap and moving obstacles. The proposed simulator generates thousands of photorealistic navigation scenarios with animatable human GS avatars from arbitrary viewpoints. ReaDy-Go outperforms baselines across target environments in both simulation and real-world experiments, demonstrating improved navigation performance even after sim-to-real transfer and in the presence of moving obstacles. Moreover, zero-shot sim-to-real deployment in an unseen environment indicates its generalization potential. Project page: https://syeon-yoo.github.io/ready-go-site/.

Unmanned underwater vehicles (UUVs) have been deployed for fish net-pen visual inspection (FNVI) in offshore aquaculture. Limited energy capacity of onboard power supplies constrains the UUV’s working range and operating time. To minimize the energy consumption by the UUV during the FNVI of the Blue Endeavour Project (an offshore salmon farm of the New Zealand King Salmon Company), an energy-optimal linear quadratic tracking (EO-LQT) control scheme is proposed in this paper. For EO-LQTs implementation, a new Linear-Parameter-Varying (LPV) system that approximates the nonlinear UUV dynamics model with an accuracy of approximately 99% regardless of the operating points in real-time, with the modified versions of Bhāskara I’s sine approximation and Shirali’s cosine approximation, is developed. The use of the Lagrangian under the Principle of Least Action with the UUV’s kinetic energy and the non-quadratic thruster power function in the EO-LQT performance index (PI) is demonstrated. The steps to solve the Hamilton-Jacobi-Bellman (HJB) equation with the non-quadratic Hamiltonian H are detailed to derive the new analytical EO-LQT optimal control form. Five EO-LQT controllers with different PIs are tested against the conventional LQT (CO-LQT) controller in both high-fidelity simulations under simulated disturbance speed up to 0.9 m/s and pool experiments, reducing energy consumption up to 37.1%. As key comparison metrics for the pose tracking and energy consumption, the mean-absolute-error (MAE) and T200 thruster power function are used to validate the effectiveness of the proposed EO-LQT controllers, compared to the CO-LQT controller.

HRI should report methods clearly and in full and treat cross-context differences as data, not as replication failures.

Physiological and qualitative data reveal insights into human perceived safety of mobile robot encounters.

The advancement of agricultural robotics offers effective solutions to the complexities of harvesting fragile fruits. This paper provides a comprehensive review of research and development within the domain of fragile fruit harvesting, with a specific focus on analyzing core technologies across the perception–reasoning–execution pipeline. This analysis encompasses fruit recognition and localization, decision‐making, trajectory planning, and end‐effector design. By examining representative robotic systems from both academia and industry, critical technical parameters are systematically summarized. The analysis reveals that the deployment of current agricultural robots remains constrained by high costs and low efficiency. Moreover, harvesting fragile fruits faces significant challenges, including severe environmental occlusion, pronounced localization errors, and elevated fruit damage rates. To address these issues, the standardization of cultivation models, the integration of multimodal sensing, and the application of embodied intelligent algorithms are identified as primary strategies. Future efforts should integrate relevant technologies from general‐purpose robotics into agricultural production. By leveraging multidisciplinary synergy, precise, high‐efficiency, and nondestructive fruit harvesting can be facilitated.

Underwater area surveillance plays a crucial role in upholding national security, safeguarding strategic interests, and mitigating potential threats. Despite its significance, a notable gap persists in the availability of comprehensive models for assessing its effectiveness. This article introduces a methodology for evaluating the operational efficiency of underwater area‐surveillance missions, with a focus on quantifying their effectiveness. By combining the availability vector, credibility matrix, and capability matrix, this approach yields evaluations of the operational efficiency of multi‐Unmanned Underwater Vehicle (multi‐UUV) systems during underwater area‐surveillance missions. Moreover, the article includes simulations of underwater area‐surveillance scenarios involving submarines, UUVs, and other targets. These simulations have pinpointed shortcomings in surveillance strategies, offering valuable insights to drive the subsequent execution, refinement, and advancement of surveillance strategies for multi‐UUV systems.

This paper presents the design, cooperative implementation, and experimental validation of an autonomous underwater vehicle swarm system–the Wukong1000 AUVs. The primary contributions of this work include: (i) The AUV's overall design work was conducted to meet pressure resistance requirements and seabed surveying mission needs, including hull form and general arrangement design, structural design and strength verification, and hydrodynamic performance validation. (ii) For multi‐AUV formation cooperative mission requirements, a modular, network‐centric responsive software and hardware framework was designed to facilitate the scalability of the AUV swarm. (iii) A swarm system consisting of three AUVs (one leader and two followers) was established, featuring an integrated acoustic communication and ranging system for cooperative formation navigation and control. Through comprehensive lake trials, the multi‐mission execution capabilities of the swarm system were fully validated: a 3‐AUV formation bathymetric mapping mission achieved high‐resolution coverage of within 5000 s under a 96.23% communication success rate; autonomous homing of AUVs using only low‐cost range information under dual‐motion conditions was successfully demonstrated; and high‐precision optical‐guided autonomous recovery tests were successfully conducted in an ice‐hole environment. These trials collectively validated the system's reliability and multi‐mission adaptability.

This paper introduces 2Fast-2Lamaa, a lidar-inertial state estimation framework for odometry, mapping, and localization. Its first key component is the optimization-based undistortion of lidar scans, which uses continuous IMU preintegration to model the system’s pose at every lidar point timestamp. The continuous trajectory over 100–200 ms is parameterized only by the initial scan conditions (linear velocity and gravity orientation) and IMU biases, yielding eleven state variables. These are estimated by minimizing point-to-line and point-to-plane distances between lidar-extracted features without relying on previous estimates, resulting in a prior-less motion-distortion correction strategy. Because the method performs local state estimation, it directly provides scan-to-scan odometry. To maintain geometric consistency over longer periods, undistorted scans are used for scan-to-map registration. The map representation employs Gaussian Processes to form a continuous distance field, enabling point-to-surface distance queries anywhere in space. Poses of the undistorted scans are refined by minimizing these distances through non-linear least-squares optimization. For odometry and mapping, the map is built incrementally in real time; for pure localization, existing maps are reused. The incremental map construction also includes mechanisms for removing dynamic objects. We benchmark 2Fast-2Lamaa on over 750 km of public and self-collected datasets from both automotive and handheld systems. The framework achieves state-of-the-art performance across diverse and challenging scenarios, reaching odometry and localization errors as low as 0.22% and 0.06 m, respectively. The real-time implementation is publicly available at https://github.com/clegenti/2fast2lamaa .

Loop closing is crucial for correcting drift in ego‐localization and mapping. Current approaches face a critical precision–recall trade‐off. To ensure precision and accurate loop pose estimation, traditional methods that impose strict geometric verification inevitably suffer from low recall. Moreover, existing semantic methods, while addressing perceptual aliasing, have yet to effectively utilize semantic information to enhance the recall of geometrically valid loop candidates. In response, we propose a novel loop‐closing method that integrates geometric and semantic verification to enhance loop recall while strictly maintaining precision under the same geometric verification. To effectively utilize semantic information, we utilize a semantic topological graph to organize semantic details. To measure the similarity between semantic topological graphs, we propose semantic object associations after long intervals. This association leverages geometric constraints, appearance similarity, and coarse‐grained object similarity, effectively formulating object associations as a linear matching problem. Finally, we implement an object‐level Bundle Adjustment method that accurately computes geometric transformations between matching keyframes, to improve loop recall and trajectory estimation accuracy. Experimental results demonstrate that the proposed object associations, even after long intervals, can handle dense, occluded, and small objects. Moreover, our loop closing significantly improves loop recall rates and trajectory estimation accuracy, while maintaining strict geometric consistency, as validated on the KITTI and KITTI‐360 data sets.

Task execution quality significantly impacts multi‐robot missions, yet existing task allocation frameworks rarely consider quality of service as a decision variable, despite its importance in applications like robotic disinfection and cleaning. We introduce the multi‐robot, multi‐objective, and multi‐mode routing and scheduling (M 3 RS) problem, designed for time‐constrained missions. In M 3 RS, each task offers multiple execution modes with varying resource needs, durations, and quality levels, allowing trade‐offs across mission objectives. M 3 RS is modeled as a mixed‐integer linear programming (MIP) problem that optimizes task sequencing and execution modes for each agent. We apply M 3 RS to multi‐robot disinfection in healthcare and public spaces, optimizing disinfection quality and task completion rates. Through synthetic case studies, M 3 RS demonstrates 3%–46% performance improvements over the standard task allocation method across various metrics. Further, to improve compute time, we propose a clustering‐based column generation algorithm that achieves solutions comparable to or better than the baseline MIP solver while reducing computation time by 60%. We also conduct case studies with simulated and real robots. Experimental videos are available on the project page ( https://sites.google.com/view/g-robot/m3rs/ ).

Marine operations traditionally rely on human intervention, a costly and disruptive method. Autonomous surface vehicles (ASVs) offer a powerful alternative, capable of operating autonomously, for extended periods, and with various sensors for various missions. However, the use of a single ASV face limitations, such as lacking the flexibility and fault tolerance needed for complex tasks, particularly in applications requiring rapid exploration of large areas. Therefore, recent research has highlighted the growing interest in swarms of ASVs for rapid and robust exploration of large ocean workspaces. These offer advantages in terms of data measurement accuracy, precision, and consistency. However, conventional ASVs often face limitations due to their size and turning radius, particularly when operating in confined environments. This paper addresses these challenges by presenting the design, construction, linear physical modeling, identification, and control of a novel small, low‐cost, and modular ASV suitable for swarming applications. This omnidirectional vehicle offers a near‐zero turning radius, enabling efficient maneuvering within confined spaces. Additionally, its modular design facilitates the seamless integration of various sensors, allowing for adaptation to a wide range of missions. Under specific assumptions and through experimental validation, a linear time‐invariant mathematical model can effectively capture the vehicle's dynamics. This is in contrast with the usual practice of deriving physical‐based nonlinear vehicle models, that are overcomplicated, difficult to tune appropriately and unnecessary for the planned operative regimes of the vehicle and for control design purposes. The simple linear model here considered is particularly beneficial for reducing the computational burden associated with developing predictive control strategies.

This communication presents a dataset collected in July 2023 using an autonomous underwater vehicle (AUV) testbed deployed in the intricate and unstructured environment of the flooded karstic network at Fontaine de Nîmes . This study is part of the Robots for Karstic Exploration (RKE) project. The AUV, named Ulysse, is equipped with a suite of advanced sensors, including two mechanically scanned imaging sonars (MSIS) for mapping horizontal and vertical cave surfaces, a Doppler Velocity Log (DVL), an Inertial Measurement Unit (IMU), a depth sensor, and a forward‐looking echo‐sounder. The operational capabilities of Ulysse in ROV mode were first demonstrated in the Gourneyras cave. Subsequently, the sensor suite was adapted for use on a diver‐operated platform, named NavScoot 2, to acquire a broader dataset representing this highly complex and intricate environment. To facilitate usability, the dataset includes the original raw data in human‐readable text files, supplemented with basic MATLAB scripts for preliminary data fusion and analysis.

6D pose estimation is a key task in computer vision and embodied AI, widely used in robotic manipulation, augmented reality, etc. Existing methods directly regress in a high-dimensional continuous space, facing two key challenges in category-level pose estimation: limited accuracy due to noise and local optima, and inefficient search over an infinite space that hinders real-time performance. This paper proposes Flow6D, a hierarchical flow matching framework with a two-stage discrete latent space localization-continuous pose regression strategy. Rotation and translation parameters are first discretized into bins, with a discrete flow matching model locking the latent space around the true pose to reduce search complexity. Then, by sampling in the latent space, a continuous flow matching model predicts local pose residuals to optimize the estimate and regress to an accurate pose. The framework also naturally extends to articulated objects, outperforming state-of-the-art methods on synthetic and real datasets with real-time inference at 70 FPS. Project website: https://flow6d.github.io/.

Learning transferable visuomotor imitation policies that generalize across diverse manipulation tasks and adapt rapidly to new tasks from only a handful of demonstrations remains challenging. Most modern policies are trained end-to-end to map observations directly to low-level actions, offering little explicit structure for reusing and recombining behaviors across tasks and making transfer data-inefficient under limited supervision. We propose SkillPlug, a plug-in framework that augments an existing visuomotor policy with a skill-conditioning module and mines a shared, transferable skill library from raw multi-task demonstrations. SkillPlug learns skills via self-supervised objectives that promote compact, reusable, and non-redundant behavior-level primitives, forming a task-shared prior for compositional control. After skill mining, we keep the learned skills fixed and specialize to unseen tasks by fine-tuning only lightweight router and action head, enabling efficient adaptation without full end-to-end retraining. We evaluate SkillPlug on two simulation benchmarks and on a real robot, and observe that the mined transferable skills consistently improve both multi-task performance and few-shot adaptation. Overall, SkillPlug offers a scalable way to mine reusable skills that improve data-efficient generalization in robotic manipulation.

Autonomous bicycles offer a promising agile solution for urban mobility and last-mile logistics. However, conventional control strategies often struggle with underactuated nonlinear dynamics, suffering from sensitivity to model mismatches and limited adaptability to real-world uncertainties. To address this, we develop CycleRL, a comprehensive sim-to-real framework for robust autonomous bicycle control. Our approach establishes a direct perception-to-action mapping within the high-fidelity NVIDIA Isaac Sim environment, leveraging Proximal Policy Optimization (PPO) to optimize the control policy. The framework features a composite reward function tailored for concurrent balance maintenance, velocity tracking, and steering control. Crucially, systematic domain randomization is employed to reduce the reliance on precise system modeling, bridge the simulation-to-reality gap and facilitate direct transfer. In simulation, CycleRL achieves promising performance, including a 99.90% balance success rate, a heading tracking error of 1.15°, and a velocity tracking error of 0.18 m/s. These quantitative results, coupled with successful hardware deployment, validate DRL as an effective paradigm for autonomous bicycle control, offering superior adaptability over traditional methods. Video demonstrations are available at https://cpnt-lab.github.io/CycleRL/.

Learning robust manipulation policies typically requires large and diverse datasets, the collection of which is time-consuming, labor-intensive, and often impractical for dynamic environments. In this work, we introduce DynaMimicGen (D-MG), a scalable dataset generation framework that enables policy training from minimal human supervision while uniquely supporting dynamic task settings. Given only a few human demonstrations, D-MG first segments the demonstrations into meaningful sub-tasks, then leverages Dynamic Movement Primitives (DMPs) to adapt and generalize the demonstrated behaviors to novel and dynamically changing environments. Improving prior methods that rely on static assumptions or simplistic trajectory interpolation, D-MG produces smooth, realistic, and task-consistent Cartesian trajectories that adapt in real time to changes in object poses, robot states, or scene geometry during task execution. Our method supports different scenarios - including scene layouts, object instances, and robot configurations - making it suitable for both static and highly dynamic manipulation tasks. We show that robot agents trained via imitation learning on D-MG-generated data achieve strong performance across long-horizon and contact-rich benchmarks, including tasks like cube stacking and placing mugs in drawers, even under unpredictable environment changes. By eliminating the need for extensive human demonstrations and enabling generalization in dynamic settings, D-MG offers a powerful and efficient alternative to manual data collection, paving the way toward scalable, autonomous robot learning.

We present SAGA, a versatile and adaptive framework for visuomotor control that can generalize across various environments, task objectives, and user specifications. To efficiently learn such capability, our key idea is to disentangle high-level semantic intent from low-level visuomotor control by explicitly grounding task objectives in the observed environment. Using an affordance-based task representation, we express diverse and complex behaviors in a unified, structured form. By leveraging multimodal foundation models, SAGA grounds the proposed task representation to the robot's visual observation as 3D affordance heatmaps, highlighting task-relevant entities while abstracting away spurious appearance variations that would hinder generalization. These grounded affordances enable us to effectively train a conditional policy on multi-task demonstration data for whole-body control. In a unified framework, SAGA can solve tasks specified in different forms, including language instructions, selected points, and example demonstrations, enabling both zero-shot execution and few-shot adaptation. We instantiate SAGA on a quadrupedal manipulator and conduct extensive experiments across eleven real-world tasks. SAGA consistently outperforms end-to-end and modular baselines by substantial margins. Together, these results demonstrate that structured affordance grounding offers a scalable and effective pathway toward generalist mobile manipulation.

Recent advancements in legged robot perceptive locomotion have shown promising progress. However, terrain-aware humanoid locomotion remains largely constrained to two paradigms: depth image-based end-to-end learning and elevation map-based methods. The former suffers from limited training efficiency and a significant sim-to-real gap in depth perception, while the latter depends heavily on multiple vision sensors and localization systems, resulting in latency and reduced robustness. To overcome these challenges, we propose a novel framework that tightly integrates three key components: (1) Terrain-Aware Locomotion Policy with a Blind Backbone, which leverages pre-trained elevation map-based perception to guide reinforcement learning with minimal visual input; (2) Multi-Modality Cross-Attention Transformer, which reconstructs structured terrain representations from noisy depth images; (3) Realistic Depth Images Synthetic Method, which employs self-occlusion-aware ray casting and noise-aware modeling to synthesize realistic depth observations, achieving over 30% reduction in terrain reconstruction error. This combination enables efficient policy training with limited data and hardware resources, while preserving critical terrain features essential for generalization. We validate our framework on a full-sized humanoid robot, demonstrating agile and adaptive locomotion across diverse and challenging terrains.

To catch a thrown object, a robot must be able to perceive the object's motion and generate control actions in a timely manner. Rather than explicitly estimating the object's 3D position, this work focuses on a novel approach that recognizes object motion using pixel-level visual information extracted from consecutive RGB frames. Such visual cues capture changes in the object's position and scale, allowing the policy to reason about the object's motion. Furthermore, to achieve stable learning in a high-DoF system composed of a robot arm equipped with a multi-fingered hand, we design a heterogeneous multi-agent reinforcement learning framework that defines the arm and hand as independent agents with distinct roles. Each agent is trained cooperatively using role-specific observations and rewards, and the learned policies are successfully transferred from simulation to the real world. Project page:https://seongdrgn.github.io/pixel2catch/

We study the problem of collision-free humanoid traversal in cluttered indoor scenes, such as hurdling over objects scattered on the floor, crouching under low-hanging obstacles, or squeezing through narrow passages. To achieve this goal, the humanoid needs to map its perception of surrounding obstacles with diverse spatial layouts and geometries to the corresponding traversal skills. However, the lack of an effective representation that captures humanoid–obstacle relationships during collision avoidance makes directly learning such mappings difficult. We therefore propose Humanoid Potential Field (HumanoidPF), which encodes these relationships as collision-free motion directions, significantly facilitating RL-based traversal skill learning. We also find that HumanoidPF exhibits a surprisingly negligible sim-to-real gap as a perceptual representation. To further enable generalizable traversal skills through diverse and challenging cluttered indoor scenes, we further propose a hybrid scene generation method, incorporating crops of realistic 3D indoor scenes and procedurally synthesized obstacles. We successfully transfer our policy to the real world and develop a teleoperation system where users could command the humanoid to traverse in cluttered indoor scenes with just a single click. Extensive experiments are conducted in both simulation and the real world to validate the effectiveness of our method.

Crawler robots represent a vital subclass of mobile robots, widely deployed in unstructured field environments. On complex, uneven terrain, track slippage (TS) is almost unavoidable. In addition, signal time delay (STD) is common in sensing and actuation processes, further increasing control complexity. As a result, the coupling of TS and STD poses significant challenges to the accuracy and smoothness of path tracking control (PTC) in crawler robots. Recognizing the strengths of pure pursuit (PP), notably its robustness and straightforward structure, we set out to address the above challenges by improving the pure pursuit method. We propose a PTC method that incorporates a look‐ahead heading error compensation (LHEC) algorithm and a PP controller, achieving real‐time adjustment of the control inputs by calculating the heading deviation between the look‐ahead point and the crawler robot and feeding it back to the control loop as a dynamic compensation signal. This method provides a robust solution to the challenges posed by TS and STD without requiring exhaustive systemic modeling, effectively leveraging the inherent ability of these factors to mitigate oscillations under specific conditions, as we found, thereby enhancing both tracking accuracy and smoothness simultaneously. According to the real‐world experiment results, our control method has high accuracy, with the maximum absolute displacement error of 0.0762 m across all experiments. The proposed method can reduce the maximum absolute displacement error by at least 41.34% compared to state‐of‐the‐art yaw‐rate‐compensated methods, including pure pursuit, nonlinear model predictive control, and Stanley control. Moreover, the proposed method also exhibits superior smoothness. The average yaw jerk did not exceed 13.95 rad/s 3 . Compared with state‐of‐the‐art yaw rate compensation methods based on pure pursuit or Stanley control, the proposed method can reduce the average absolute yaw jerk by at least 19.55%. Furthermore, 15 sets of repeated trials on continuous curve paths in plowed dry land demonstrate that the controller maintains high consistency. By the way, this study clarifies the inherent limitations of look‐ahead distance adjustment and yaw rate compensation strategies under the coupled influence of TS and STD, providing new insights for the development of robust field‐robotic control.

To address the heavy reliance on manual labor, low operational efficiency, and high safety risks in distribution network maintenance, this paper proposes an aerial continuum manipulator (ACM) system for real‐world deployment. The system integrates a quadrotor UAV with a wave spring continuum manipulator (WSCM), thereby combining aerial mobility with compliant manipulation capability to accommodate complex operating environments. To cope with compound uncertainties arising during flight operations, including rotor downwash, platform vibration, and attitude disturbances, a disturbance rejection zeroing neurodynamic model‐free control method with dual input fuzzy regulation (DFRZND) is developed. By introducing the tracking error and its rate of change as fuzzy inputs, the proposed method enables adaptive adjustment of the convergence parameter. In conjunction with a smooth composite activation function, it further enhances disturbance rejection and trajectory tracking accuracy. Simulation results demonstrate that the proposed method maintains high precision motion control performance under various challenging conditions, including persistent disturbances, oscillatory disturbances, and abrupt disturbances. Furthermore, a physical prototype was developed and outdoor experiments were conducted to verify that the system can reliably perform representative distribution network maintenance tasks, including insulating cover installation on parallel groove clamps and hook engagement of bypass cable lifters. The results indicate that the proposed ACM system and its control strategy exhibit strong potential for practical application in live power maintenance operations.

This study presents a novel and unified framework for modeling hybrid terrestrial mobile robots with flippers, tracks, and wheels, explicitly addressing chassis pose control. Using a differential kinematic approach, the unified model combines a generic chassis and locomotion mechanisms models, and is specifically tailored for an actively actuated tracked vehicle. The introduction of two motion groups and four distinct controllers enables precise manipulation of roll, pitch, and ground clearance. The proposed models' algorithms, implemented in the Robot Operating System, are simple, efficient, and easily embedded in a robot designed for industrial services. Experiments conducted with the prototype in laboratory, open‐field, and industrial environments evaluate the proposed methodology. The results show that the differential models properly coordinate the locomotion mechanisms, allowing the chassis to achieve the desired input velocities.

Current Visual Simultaneous Localization and Mapping (VSLAM) systems often struggle to create maps that are both semantically rich and easy to interpret. While incorporating semantic scene knowledge helps build richer maps with contextual associations among mapped objects, representing them in structured formats such as scene graphs has not been widely addressed, leading to complex map comprehension and limited scalability. This paper introduces vS-Graphs, a novel real-time VSLAM framework that integrates vision-based scene understanding with map reconstruction and comprehensible graphbased representation. The framework infers structural elements (i.e., rooms and floors) from detected building components (i.e., walls and ground surfaces) and incorporates them into optimizable 3D scene graphs. This solution enhances the reconstructed map's semantic richness, comprehensibility, and localization accuracy. Extensive experiments on standard benchmarks and real-world datasets demonstrate that vS-Graphs achieves an average of 15.22% accuracy gain across all tested datasets compared to state-of-the-art VSLAM methods. Furthermore, the proposed framework achieves environment-driven semantic entity detection accuracy comparable to that of precise LiDARbased frameworks, using only visual features. The code is publicly available athttps://github.com/snt-arg/visual sgraphs and is actively being improved.

This paper presents a novel data-driven Reference Governor with Model Predictive Control, integrating local motion replanning and path following for collision avoidance. Employing a model-free Reference Governor, the proposed framework utilises system knowledge through Bayesian Optimisation to augment predetermined evasive trajectories, minimising pathfollowing errors and simultaneously ensuring obstacle safety margins. A single-track vehicle model in combination with a nonlinear tyre model is used to capture the vehicle's dynamics. The optimised control action is the vehicle steering angle, whilst the Reference Governor optimises parameters of a sigmoid reference signal to minimise the tracking error and guarantee safety with respect to obstacles in emergency manoeuvres. The proposed approach is evaluated on a single lane change using a highfidelity simulation environment, and its performance is compared to a baseline controller integrating path following and obstacle avoidance. The results demonstrate a 14% reduction in safety critical overshoot, maximising obstacle safety distance and a four times lower controller cycle time compared to the baseline. Furthermore, through a robustness analysis, it is demonstrated that the proposed approach is more robust towards model mismatches and perception-based errors, as seen by average 30% and 40% reductions in near-miss and collision rates.

Path tracking control is a critical component of autonomous driving technology. However, it faces significant challenges due to the inherent strong nonlinearity and dynamic uncertainties of autonomous vehicles (AVs), which remains a severe challenge in the field. To address the problem, this paper proposes a novel robust data-driven control framework designed to balance modeling accuracy with disturbance rejection performance. First, a control-oriented Deep Koopman model is constructed to map the nonlinear AV path-tracking error dynamics into a high-dimensional lifted space. Second, to address inevitable modeling residuals and external disturbances, a high-order super-twisting observer (HOSTO) is designed within the lifted space, which guarantees finite-time convergence of disturbance estimation, providing strong robust compensation. Finally, an approximate lifted-space MPC with Jacobian-based cost mapping is formulated as a standard Quadratic Program (QP), avoiding repeated nonlinear decoding during online optimization. The effectiveness of the proposed method is validated through comprehensive simulations and experiments, showing superior control accuracy and robustness against disturbances compared to existing algorithms.

Curriculum learning has demonstrated substantial effectiveness in robot learning. However, it still faces limitations when scaling to complex, wide-ranging task spaces. Such task spaces often lack a well-defined difficulty structure, making the difficulty ordering required by previous methods challenging to define. We propose a Learning Progress-based Automatic Curriculum Reinforcement Learning (LP-ACRL) framework, which estimates the agent's learning progress online and adaptively adjusts the task-sampling distribution, thereby enabling automatic curriculum generation without prior knowledge of the difficulty distribution over the task space. Policies trained with LP-ACRL enable the ANYmal D quadruped to achieve and maintain stable, high-speed locomotion at 2.5 m/s linear velocity and 3.0,/s angular velocity across diverse terrains, including stairs, slopes, gravel, and low-friction flat surfaces–whereas previous methods have generally been limited to high speeds on flat terrain or low speeds on complex terrain. Experimental results demonstrate that LP-ACRL exhibits strong scalability and real-world applicability, providing a robust baseline for future research on curriculum generation in complex, wide-ranging robotic learning task spaces.

Inverse kinematics (IK) of continuum robots is inherently set-valued: a single task-space target corresponds to a family of feasible configurations. However, in practice, this structure is only observable through finite samples and tolerance-based approximations, making it dependent on sampling density, tolerance, and estimator design. In this work, IK is studied as a conditional solution-set problem by approximating inverse neighborhoods from dense configuration sampling and analyzing their structure using graph-based connectivity and intrinsic dimension (ID) estimators. Experiments on constant-curvature robots with increasing redundancy ($n=3,6,9$), under both position-only and full-pose ($SE(3)$) formulations, show that inverse neighborhoods exhibit stable low-dimensional structure despite increasing actuation dimension. In contrast, connectivity estimates are highly scale-dependent: fixed-radius graphs produce strong fragmentation, whereas robust graph constructions reveal neighborhoods dominated by a single connected component. Increasing the sampling density further shows that the intrinsic dimension and the dominant-component structure remain stable, while apparent separation decreases. To assess functional implications, a trajectory-level path-lifting experiment is introduced and validated on CBHA data, demonstrating continuous inverse solutions with low tracking error and no branch switching. These results suggest that the observed fragmentation is largely a finite-sample and estimator-dependent effect. This motivates modeling IK as a conditional distribution$p(q \mid y)$rather than a deterministic function, providing empirical support for generative approaches such as conditional normalizing flows.

This work presents a mapless navigation approach for outdoor applications. It combines the exploratory capacity of conditional variational autoencoders (CVAEs) to generate trajectories and the semantic segmentation capabilities of a lightweight visual language model (VLM) to select the trajectory to execute. Open-vocabulary segmentation is used to score and select the generated trajectories based on natural language, and a state-of-the-art local planner executes velocity commands. One of the key features of the proposed approach is its ability to generate a large variability of trajectories and select them to navigate in real-time. In real-world outdoor experiments, Sem-NaVAE achieves a 90% success rate across routes of 120–240m in unseen environments, outperforming the nearest baseline by 10% while remaining within 7% of a map-based upper bound. A video showing an experimental run of the system can be found inhttps://youtu.be/i3R5ey5O2yk.

Inverse kinematics (IK) for 6-DoF manipulators is inherently set-valued, since multiple distinct joint configurations can realize the same end-effector pose. Nevertheless, learning-based IK is frequently posed as single-output regression, which can induce mode collapse and unstable predictions near branch boundaries. In this work, IK is cast as a deterministic set-prediction problem, in which an unordered set of feasible joint candidates is produced in a single forward pass. A multi-head network is trained with permutation-invariant supervision via optimal bipartite matching, eliminating the need for a fixed global ordering of IK branches. High-fidelity multi-solution supervision on a UR3 platform is obtained through a hybrid analytic–numerical procedure, in which branch seeds are enumerated and refined using Levenberg–Marquardt optimization, followed by feasibility filtering. At deployment, the predicted set is mapped to an executable joint-command sequence through a continuity-aware selection rule, thereby promoting temporally consistent branch choices. On real UR3 trajectories, improved set fidelity and branch coverage are observed relative to single-solution and fixed-assignment baselines, while accurate joint-space and task-space tracking is maintained on held-out motions. Runtime benchmarks further indicate that multi-branch inference is faster than repeated numerical IK when multiple solutions are required.

Continuous-time SLAM provides a principled framework for fusing heterogeneous sensors while estimating smooth trajectories, and is particularly well-suited for handling heterogeneous, asynchronous sensor streams with non-uniform readout patterns, such as rolling shutter cameras, LiDAR scanners, radar sweeps, or event-based sensors. In this work, we introduce G-solver, a fully Gaussian and distributed framework that combines Gaussian Belief Propagation (GBP) with Gaussian Process (GP) motion priors for continuous-time trajectory estimation. Our GP model provides a probabilistic representation of the trajectory, enabling consistent interpolation and the use of data-driven hyperparameters, while GBP offers a scalable message-passing formulation well-suited for decentralized settings. The resulting solver naturally extends to multi-camera scenarios without specialized synchronization or engineering effort. We evaluate the approach on synthetic and real data, including rolling shutter and distributed multi-camera optimization, demonstrating accurate and stable estimation with runtimes comparable to existing continuous-time methods. An open-source implementation is released athttps://github.com/rvp-group/gsolver.

Data-driven control methods such as data-enabled predictive control (DeePC) have shown strong potential for efficient control of soft robots without explicit parametric models. However, in object manipulation tasks, unknown external payloads and disturbances can significantly alter the system dynamics and behavior, leading to offset errors and degraded control performance. In this paper, we present a novel velocity-form DeePC framework that achieves robust and optimal control of soft robots under unknown payloads. The proposed framework leverages input/output data in an incremental representation to mitigate performance degradation induced by unknown payloads, eliminating the need for weighted datasets or disturbance estimators. We validate the method experimentally on a planar soft robot and demonstrate its improved performance compared to standard DeePC in scenarios involving unknown payloads.

Learning structured task representations from human demonstrations is essential for bimanual manipulation, where action ordering, object involvement, and interaction geometry vary significantly across executions. A key challenge lies in jointly capturing the discrete semantic task structure and the temporal evolution of object-centric geometric relations in a form that supports reasoning over task progression. We introduce a semantic–geometric graph-based task representation that jointly encodes object identities, inter-object semantic relations, and per-object motion histories, via a Message Passing Neural Network (MPNN) encoder and a Transformer-based decoder. The encoder operates solely on the temporal scene graph, producing structured representations decoupled from action labels. The decoder then conditions on action-context to forecast future actions, associated objects, and object motions. This decoupling learns task-agnostic representations, enabling encoder reuse across embodiments through decoder-only finetuning on a small robot dataset. Across eleven bimanual tasks from two datasets, we find that the benefit of structured semantic–geometric representations over simpler sequence-based models grows with task variability in action ordering and object involvement. At deployment, a planner couples the action and motion predictions with learned Probabilistic Movement Primitives, achieving full task success on two real-robot bimanual tasks and outperforming graph ablations, Transformer, decoder-only, and finetuned vision-language model baselines. Website:https://frherbert.github.io/bimanual-task-graphs

Blind locomotion policies trained in simulation for unstructured terrains often suffer from significant performance degradation when deployed on real bipedal robots. A typical manifestation is the emergence of abnormal behaviors (e.g., sudden high-stepping motions) even on simple terrains like flat ground. This issue arises from environmental noise interference and the inability to reliably obtain certain critical state information (such as terrain heightmaps and linear velocities) through proprioception. Existing methods commonly attempt to address this problem by injecting large amounts of noise and learning explicit or implicit representations of the missing information. Nevertheless, excessive noise can severely restrict the potential performance of the policy. The more fundamental challenge lies in the fact that both explicit and implicit representations constructed solely from historical proprioceptive data are unable to accurately reconstruct key unobservable state variables. This inevitable estimation bias ultimately leads to abnormal behaviors. To address these challenges, we propose a bipedal training framework based on a Smoothed and Denoised World Model (SDWM), which explicitly mitigates the effects of observation noise, incomplete state information, and inaccurate reconstruction. We validate the effectiveness of SDWMthrough comparative tests conducted in simulation as well as indoor and outdoor realworld scenarios.

Origami robotic modules offer compactness, compliance, and reconfigurability, but their system-level capabilities are often limited by the lack of reliable self-perception and closed-loop 3D motion control at the module level. This work presents a compact and lightweight shape memory alloy (SMA)-driven origami robotic module that achieves self-sensing and closed-loop three-dimensional motion control. The module integrates a foldable origami skeleton with four symmetrically arranged SMA spring actuators for omnidirectional bending and axial contraction. Embedded Hall-effect sensors enable intrinsic three-dimensional motion perception by measuring the module deformation state. A geometric kinematic model is established to map the skeleton deformation to the central-axis ending position of the module for reliable feedback control. Experimental results show mean calibration errors of 0.21 mm (x), 0.15 mm (y), 0.33 mm (z), and 4.78° for orientation estimation. Benefiting from module-level perception and control, multiple modules can be coordinated to perform crawling, confined-space climbing, walking, and object manipulation tasks. These results demonstrate the importance of closed-loop module-level motion control for reconfigurable modular robotic systems.

Multi-agent autonomous exploration is essential for applications such as environmental monitoring, search and rescue, and industrial-scale surveillance. However, effective coordination under communication constraints remains a significant challenge. Frontier exploration algorithms analyze the boundary between the known and unknown regions to determine the next-best view that maximizes exploratory gain. This article proposes an enhancement to existing frontier-based exploration algorithms by introducing a probabilistic approach to frontier prioritization. By leveraging Dirichlet process Gaussian mixture model (DP-GMM) and a probabilistic formulation of information gain, the method improves the quality of frontier prioritization. The proposed enhancement, integrated into two state-of-the-art multi-agent exploration algorithms, consistently improves performance across environments of varying clutter, communication constraints, and team sizes. Simulations showcase an average exploration time improvement of 10% and 14% for the two algorithms across all combinations. Successful deployment in real-world experiments with a dual-drone system further corroborates these findings.

Manual multirotor piloting requires continuous visual monitoring and dual-stick manipulation, increasing cognitive workload and limiting usability for novice users and attention-constrained scenarios. This paper presents a bidirectional wearable human–drone interface that integrates an inertial measurement unit (IMU)-based gesture controller with an abdominal vibrotactile device to enable non-visual egocentric 3D relative-position awareness. The proposed GRU–ECOC gesture pipeline achieves 97.58% accuracy (macro F1 = 0.9773). A user-in-the-loop waypoint-following study with 12 novice users shows learnability, reducing completion time from 263 s to 186 s over five trials. For spatial feedback, we implement a 3-by-12 vibrotactile device that conveys egocentric 3D position and achieves 82.75% complete 3D recognition accuracy in an indoor study with 12 participants and 1,200 trials. At the integrated level, a real-drone point-to-point task compares Visual-only and Tactile-only operation under the same gesture-control condition. Tactile-only operation remains feasible but slower than the Visual-only baseline, with 8/9 successes and a mean completion time of 45.1 s compared with 9/9 successes and 31.4 s for Visual only operation. A separate sensory-blocked forward-position holding task further demonstrates the feasibility of non-visual closed-loop recovery, with successful recovery in all 8 boundary exit events. These results support the feasibility of bidirectional wearable human–drone interaction while clarifying the cost of tactile-only state feedback relative to visual operation.

Capturing 4D spatiotemporal scene structure is crucial for the safe and reliable operation of robots in dynamic environments. However, existing approaches typically address only part of the problem: they either provide coarse geometric tracking via bounding boxes or detailed 3D occupancy estimates that lack explicit temporal association and instance-level reasoning. In this work, we present Latent Gaussian Splatting (LaGS) for 4D Panoptic Occupancy Tracking (4D-POT). We revisit the underlying representation and model 3D features as a sparse set of feature-bearing Gaussians. These act as dynamic, volume-oriented keypoints that enable spatially continuous, distance-weighted aggregation of multi-view features before being splatted into a voxel grid for decoding. This point-centric formulation enables flexible, data-dependent receptive fields and long-range spatial interactions that are difficult to capture with local and dense voxel-based operators. A hierarchical Gaussian representation further enables multi-scale reasoning by combining global context from coarse super-points with fine-grained detail from higher-resolution streams. Extensive experiments on Occ3D nuScenes and Waymo demonstrate state-of-the-art performance for 4D-POT. We provide code and models at https://lags.cs.uni-freiburg.de/.

Autonomous target search is crucial for deploying Micro Aerial Vehicles (MAVs) in emergency response and rescue missions. Existing approaches either focus on 2D semantic navigation in structured environments – which is less effective in complex 3D settings, or on robotic exploration in cluttered spaces – which often lacks the semantic reasoning needed for efficient target search. This paper overcomes these limitations by proposing a novel framework that utilizes a semantically-guided viewpoint planner to minimize target search and exploration time in unstructured 3D environments using an MAV. Specifically, we develop a combinatorial planner that generates efficient semantic exploration plans by prioritizing viewpoints that likely lead to the target. To guide the planner towards the target, an active perception pipeline is developed that propagates semantic priorities of observed objects into neighboring frontier voxels for computing semantic information gains of frontier viewpoints. In addition, we demonstrate how LLM-based similarity scores can be leveraged as semantic priority input to our pipeline. Evaluations in two distinct simulation environments show that the proposed method consistently outperforms baselines by quickly finding the target while maintaining reasonable exploration times. Real-world experiments with an MAV further demonstrate the method’s ability to handle practical constraints like limited battery life, small sensor range, and semantic uncertainty.

This paper presents a comprehensive review of robotics research in search and rescue (SAR) operations conducted in caverns, underground environments, disaster zones, and other areas where Global Navigation Satellite System (GNSS) signals are unavailable. The majority of applications for Simultaneous Localisation and Mapping (SLAM), despite its maturity, are still restricted to structured indoor settings or outdoor environments under normal weather conditions. Standard SLAM frameworks often experience degradation and malfunction or even fail when deployed in increasingly complex and unstructured scenarios. This review identifies three major challenges that robots face in SAR environments: (i) increasingly complex terrain, (ii) changing environments and visibility, and (iii) autonomous exploration requirements, along with corresponding technological evolutions in robot mobility, sensor technologies, and SLAM algorithms. A comprehensive and quantitative evaluation of existing approaches is provided, focusing on SLAM on uneven terrain, multisensor fusion, and active SLAM. Additionally, this paper outlines ongoing challenges for guiding future development toward more robust and reliable deployment‐oriented SLAM solutions for SAR applications. These include: (i) short‐term dynamics and structural changes that undermine data association and loop closure, (ii) observability loss and degeneracy in confined and cluttered spaces, and (iii) multirobot consistency under constrained communication. Two cross‐cutting constraints, which are sensor non‐stationarity and safety‐critical autonomy, are highlighted as key factors that turn deployable SAR SLAM into a system‐level reliability problem. Finally, potential research directions and a practical research roadmap toward robust, real‐time, and evaluable SAR SLAM are outlined.

To enhance adaptability and obstacle‐crossing performance in unstructured environments, this study proposes a large‐expansion‐ratio deformable mobile platform, referred to as the “Wheel–Claw Climber,” which is based on spatial folding and linkage deformation mechanisms. The platform incorporates a radially expandable deformable wheel with a maximum expansion ratio of 2.72, substantially exceeding the existing benchmark of 2.4. The “Wheel–Claw Climber” operates in three distinct configurations—wheel, claw, and intermediate—enabling adaptation to diverse terrains. An obstacle‐crossing model is established, and the maximum obstacle‐crossing height is determined to be 4.5 R , representing a 27.7% improvement over current designs and demonstrating superior obstacle‐surmounting capability. Furthermore, a cross‐slope traversal strategy is proposed, whereby deformation and expansion of the low‐side wheel relative to the high‐side wheel increase passability and stability by 31.1% and 21.65%, respectively, on the same slope. Experimental validation confirms that the Wheel–Claw Climber can seamlessly transition between wheel and claw modes. At a wheel rotation speed of 1 rad/s, the platform achieves a maximum obstacle‐crossing height of 550 mm, with a 100% success rate for obstacles below 450 mm. Overall, the deformable wheel architecture and experimental findings provide new insights for the design and development of next‐generation mobile platforms.

Inspired by the human hand, this work presents a hybrid soft-rigid architecture for a robotic palm to enhance dexterity and robustness. While soft-robotic hands often lack the structural integrity to handle heavy objects, classical rigid designs are sensitive to impacts and frequently neglect the palm's functional role. We bridge that gap by combining soft compliance with skeletal elements for robust force transmission. In our design, rigid metacarpal bones house local actuation and transmission units. These units receive torque from forearm-based motors via flexible shafts to drive tendon actuation. We embed these elements in a 3D-printed silicone matrix, which eliminates the design restrictions of traditional moulding. Ball joints for the thumb and little finger enable the formation of three palmar arches. In experiments, the palm achieved active and passive deformations of 74% and 67%, respectively, relative to its neutral width. This shape adaptability offers practical benefits in confined spaces, where internal palm movements enable new grasping and manipulation techniques previously difficult to achieve with traditional designs.

A deformable mobile robot with single-degree-of freedom (single-DOF) can autonomously navigate and avoid obstacles in complex environments by altering its three dimensional dimensions through deformation. This capability allows it to traverse passages that fixed-size robots cannot, such as height-restricted or narrow channels, thereby improving task efficiency. This paper proposes a height-aware navigation framework that projects the varying 3D geometric constraints into a grid map and costmap, enabling coupled motion deformation planning for single-DOF robots. To achieve this, the height restrictions in complex three-dimensional environments are projected as grayscale information into a two-dimensional grid map. Using these grayscale values, we design a costmap layer to reflect the robot's footprint dimensions in different deformation states. An A* algorithm with height encoding then obtains a global path containing the robot's height constraints. Finally, an enhanced Dynamic Window Approach (DWA) algorithm is proposed to calculate the robot's movement and deformation speeds, enabling it to utilize its single-DOF deformation capability to traverse height-restricted or narrow obstacle passages. To validate the effectiveness of each core module and highlight the advantages over traditional methods, simulations and experiments were conducted using a Miura-ori based single-DOF deployable robot.

Driven by the demand for efficient robotic propulsion in complex underwater environments, this work presents a biomimetic jellyfish robot inspired by the locomotory characteristics of biological jellyfish. Establishing the biomimetic propulsion foundation based on the kinematic features of three representative jellyfish species spanning large, medium, and small scales, this study designed a double‐crescent biomimetic tentacle by integrating the Fin Ray effect and the crescent‐shaped caudal fin morphology of tuna. A computational fluid dynamics analysis was conducted to investigate the effects of swing frequency, swing angle, and geometric parameters on the hydrodynamic characteristics of the tentacle structure during in situ propulsion. To investigate the propulsion performance of the biomimetic tentacle and to optimize its structural and dynamic parameters, a dedicated propulsion testbed was fabricated, and both orthogonal and vortex‐visualization experiments were conducted. On the basis of the optimized tentacle structure and dynamic parameters, a biomimetic jellyfish robot prototype was fabricated and subjected to underwater propulsion evaluation experiments. The developed prototype exhibited an average thrust of 3.66 N and an average propulsion velocity of 98.6 mm/s. This study provides a systematic methodology for the design and optimization of tentacle structures and dynamic parameters in biomimetic jellyfish robots.

Legged animals can self-right and stand up from arbitrary postures across diverse environments. Empowering robots with animal-like recovery capabilities expands their real-world applications. However, existing recovery controllers that rely on predefined trajectories are limited to flat terrain, whereas current learning-based policies often produce overly aggressive or sluggish control outputs, leading to poor hardware performance. To bridge the gap, this paper proposes a novel real-to-sim-to-real learning framework that enables quadruped robots to acquire both agile and robust recovery skills. Unlike previous learning approaches that train from scratch, we employ adversarial imitation learning with hardware-collected demonstrations to promote natural motions and enhance learning efficiency. For robust recovery on complex terrains, a terrain imagination module is integrated to predict key terrain properties via onboard sensing. To ensure reliable sim-to-real transfer, the recovery policy is trained with domain randomization, terrain curriculum learning, and geometry decomposition techniques. The trained policy can be generalized to various quadruped robots and is directly deployed on the Unitree Go2 Hardware. To the best of our knowledge, this is the first time that a small-sized quadruped demonstrates such advanced recovery skills across diverse environments, outperforming prior studies with more powerful hardware. The project page is available at:https://rsr-recovery.github.io/.

We presentPROBE(PRobabilisticOccupancyBEVEncoding), a learning-free LiDAR place recognition descriptor that models each BEV cell's occupancy as a Bernoulli random variable. Rather than relying on discrete point-cloud perturbations, PROBE analytically marginalizes over continuous Cartesian translations via the polar Jacobian, yielding a distance-adaptive angular uncertainty$\sigma \_\theta = \sigma \_{t} / r$in$\mathcal {O}(R{\cdot }S)$time. The primary parameter$\sigma \_{t}$represents the expected translational uncertainty in meters, a sensor-independent physical quantity that enhances cross-sensor generalization while reducing the need for extensive per-dataset tuning. Pairwise similarity combines aBernoulli-KL Jaccardwith exponential uncertainty gating and FFT-based height cosine similarity for rotation alignment. Evaluated on four datasets spanning four diverse LiDAR types, PROBE achieves the highest accuracy among handcrafted descriptors in multi-session evaluation and competitive single-session performance relative to both handcrafted and supervised baselines. The source code and supplementary materials are available athttps://sites.google.com/view/probe-pr.

Vision-Language-Action (VLA) models pretrained on large-scale multimodal datasets have emerged as powerful foundations for robotic perception and control. However, their massive scale, often billions of parameters, poses significant challenges for real-time deployment, as inference becomes computationally expensive and latency-sensitive in dynamic environments. To address this, we propose Token Expand-and-Merge-VLA (TEAM-VLA), a training-free token compression framework that accelerates VLA inference while preserving task performance. TEAM-VLA introduces a dynamic token expansion mechanism that identifies and samples additional informative tokens in the spatial vicinity of attention-highlighted regions, enhancing contextual completeness. These expanded tokens are then selectively merged in deeper layers under action-aware guidance, effectively reducing redundancy while maintaining semantic coherence. By coupling expansion and merging within a single feed-forward pass, TEAM-VLA achieves a balanced trade-off between efficiency and effectiveness, without any retraining or parameter updates. Extensive experiments on LIBERO benchmark demonstrate that TEAM-VLA consistently improves inference speed while maintaining or even surpassing the task success rate of full VLA models.

Learning soft continuum robot (SCR) dynamics from video offers flexibility but existing methods lack interpretability or rely on prior assumptions. Model-based approaches require prior knowledge and manual design. We bridge this gap by introducing: (1) The Attention Broadcast Decoder (ABCD), a plug-and-play module for autoencoder-based latent dynamics learning that generates pixel-accurate attention maps localizing each latent dimension's contribution while filtering static backgrounds, enabling visual interpretability via spatially grounded latents and on-image overlays. (2) Visual Oscillator Networks (VONs), a 2D latent oscillator network coupled to ABCD attention maps for on-image visualization of learned masses, coupling stiffness, and forces, thereby enabling mechanical interpretability. We validate our approach on single- and double-segment SCRs, demonstrating that ABCD-based models significantly improve multi-step prediction accuracy with 5.8$times$error reduction for Koopman operators and 3.5$times$for oscillator networks on a two-segment robot. VONs autonomously discover a chain structure of oscillators. This fully data-driven approach yields compact, mechanically interpretable models with potential relevance for future control applications.

Continuum robots are flexible, slender manipulators well suited for confined surgical environments. In these settings, unknown interaction forces and model uncertainty significantly affect robot shape, motivating state estimation from external observations. Existing estimation methods either neglect actuation modeling or rely on simplified deterministic actuation models. In contrast, we jointly estimate robot shape, external loads, and actuation inputs using mechanically principled actuation priors. To achieve this, we present a discrete Cosserat rod formulation with piecewise-linear strain integration that provides high numerical accuracy while inducing a sparse factor graph structure for efficient nonlinear optimization. We extend the framework to tendon-driven and parallel robots in simulation and validate it experimentally on a surgical concentric tube robot. Overall, our approach enables principled real-time estimation across multiple robot architectures while providing direct access to manipulator Jacobians through the linearized factor graph.

Stereo matching is crucial for 3D perception in autonomous driving and generalist robotics. Recent advancements transitioned from cost volume method to iterative refinement approach, offering superior details and spatial consistency. However, iterative methods remain constrained by expensive matching encoding and numerous refinement iterations, limiting latency-sensitive practical application. To unleash the accuracy superiority of iterative paradigms in real-time systems, we present Compact Iterative Refinement Network (CIRNet). Specifically, we design Compact Correlation Encoding (CCE) to allocate essential matching evidence by functional attributes, constructing global geometric context at deep level and fine-grained local similarity at shallow level to reduce redundant computation and provide reliable coarse initialization. Meanwhile, we introduce Staged Iterative Refinement (SIR) using early coarse correction for large residuals and a lightweight shared operator for fine details, improving the accuracy-latency trade-off and reaching near-saturated accuracy at lower latency. Evaluations on SceneFlow and KITTI show CIRNet sets new state-of-the-art in balancing accuracy and efficiency. Furthermore, CIRNet exhibits robust cross-domain generalization on multiple real-world datasets and practical deployability on embedded edge platform.

Miniaturization of climbing robots is of interest for accessing narrow and complex structures that are inaccessible to conventional robots. This paper presents the design, fabrication, and experimental validation of an insect-scale climbing robot with magnetic wheels. The robot integrates two key technologies: a micro ultrasonic motor, which is one of the smallest electricity-driven motors, and a micro planetary gear train with a gear ratio of 64, fabricated using a micro-stereolithography 3D printer. The combination of these technologies enables the miniaturization of climbing robots. With a differential steering mechanism driven by two geared motors, the prototype robot measures 13 mm×13 mm and weighs 0.77 g. To obtain a suitable magnetic adhesion force, the magnetic wheels are designed and optimized to ensure reliable locomotion on vertical surfaces. Experiments demonstrate that the robot can climb vertically while carrying a payload exceeding five times its own weight. The steerability tests confirm the robot's ability to perform turns along a desired trajectory, although some errors remain. These experiments have demonstrated the feasibility of the smallest magnet-wheel-based climbing robot, providing insights for future studies on micro-scale robotic exploration in constrained environments.

Reliable relative localization using only onboard sensing is challenging for multi-robot systems operating in GPS-denied environments, particularly when robots exhibit limited mutual motion, under which the available measurements cannot provide persistently excited regression signals. To address this challenge, this letter proposes an adaptive online relative localization method that enables accurate relative-position estimation using only onboard measurements under a mild interval excitation (IE) condition. By introducing two dynamic auxiliary variables, the relative localization problem is reformulated as an online parameter estimation framework, which guarantees globally exponential convergence without persistent excitation (PE). Furthermore, an explicit lower bound on the estimation gain is provided in terms of the strength and duration of the interval excitation induced by the robots' relative motion. Finally, simulations and physical experiments demonstrate that the proposed method achieves reliable localization performance under both interval and persistently excited motions.

LiDAR place recognition plays a crucial role in SLAM, robot navigation, and autonomous driving. However, existing LiDAR place recognition methods often struggle to adapt to new environments without forgetting previously learned knowledge, a challenge widely known as catastrophic forgetting. To address this issue, we proposeKDF+, a novel continual learning framework for LiDAR place recognition that extends the KDF paradigm with a loss-aware sampling strategy and a rehearsal enhancement mechanism. The proposed sampling strategy estimates the learning difficulty of each sample via its loss value and selects samples for replay according to their estimated difficulty. Harder samples, which tend to encode more discriminative information, are sampled with higher probability while maintaining distributional coverage across the dataset. In addition, the rehearsal enhancement mechanism encourages memory samples to be further refined during new-task training by slightly reducing their loss relative to previous tasks, thereby reinforcing long-term knowledge retention. Extensive experiments across multiple benchmarks demonstrate that KDF+ consistently outperforms existing continual learning methods and can be seamlessly integrated into state-of-the-art continual learning for LiDAR place recognition frameworks to yield significant and stable performance gains. The code is available athttps://github.com/Thunder-Volcano/KDF-plus.

Vision-language-action (VLA) models are gaining attention in robotics, yet their robustness to adversarial attacks remains largely unexplored. Existing work shows that adversarial patches can mislead VLA-based robots but assumes full access to the entire execution trajectory, an unrealistic requirement in practice. We address this limitation by formulating a partially observable threat model, where the adversary can exploit only a short prefix of the trajectory to generate a fixed patch applied to all subsequent frames. Under this setting, we propose a twophase framework. First, we localize the patch using the model's attention maps to identify visually critical regions that correspond to the full instruction. Then, we optimize the patch to disrupt the semantic grounding of target objects and increase the curvature of action trajectories, thereby compounding failures in both perception and control. Extensive experiments in simulation and real-world robotic environments show that our method sustains adversarial effects under partial observability, inducing longhorizon disruptions and significantly reducing task success rates.

This paper studies dynamic multi-robot cooperative task allocation in multi-stage scout-rescue missions with unknown and evolving task demands. We propose IAOM, an Intention-Aware Optimization Model that encodes each robot's decision state–target task, role, and planner-derived ETA–as an intention vector and organizes robots and tasks into an execution-aware intention graph capturing residual demand and active commitments. A GAT-PPO policy is learned on this graph and coupled to a hybrid A* based motion layer, so that allocation, coalition formation, and congestion handling are optimized within a single framework. We evaluate IAOM on grid based urban rescue benchmarks with multiple map sizes and resource density regimes against optimization-, heuristic-, and learning-based baselines. Under abundant and normal resources, IAOM consistently reduces mission makespan and average completion time relative to the best baseline, achieving up to about$20\%$shorter makespan and about$50\%$lower completion time with comparable load balance. Under tight resources where robots barely meet total demand, gains are smaller and all methods are largely bounded by resource scarcity, but IAOM remains competitive.

In [1], the corresponding author information was incorrectly published. Yixun Niu was incorrectly listed as the corresponding author. The correct corresponding author should be Xin Cheng. This correction applies only to the corresponding author information; all other contents of the article remain unchanged.

Picking robots often encounter significant challenges when navigating unstructured agricultural environments due to obstacles such as dense branches, immature crops, and other obstacles. This paper presents a Sampling Step Guiding Rapidly‐exploring Random Tree (SSG‐RRT) path planning algorithm for wheeled picking robots. The proposed algorithm addresses key issues, including excessive redundancy in sampling points, low tree expansion efficiency, poor convergence guidance, and abrupt path turns by constructing a Sampling Step Rapidly‐exploring Random Tree (SS‐RRT) algorithm that combines a greedy biased sampling strategy and adaptive step size, and further integrating the Artificial Potential Field (APF) algorithm to achieve convergence‐oriented optimization. Additionally, a reconnection optimization strategy is employed to eliminate unnecessary path nodes that do not account for obstacles, and cubic B‐spline curve smoothing is applied to refine the generated path. To further improve local obstacle avoidance, the Dynamic Window Approach (DWA) is integrated with SSG‐RRT. The DWA algorithm tracks the globally planned path generated by SSG‐RRT while dynamically adjusting the local path based on velocity constraints to avoid obstacles in real time. Compared with the SS‐RRT algorithm, simulation results demonstrate that the SSG‐RRT algorithm reduces path length by 21.6%, sampling time by 87.5%, and overall planning time by 84.1%. The proposed approach is successfully applied to real‐time obstacle avoidance for both static and dynamic obstacles, effectively addressing challenges such as poor convergence, excessive path inflection points, and weak dynamic obstacle avoidance capabilities in complex and dynamic picking environments.

Neural Radiance Fields and Gaussian Splatting have emerged as powerful representations for reconstructing complex dynamic urban scenes from multi-view driving logs. By producing photorealistic and geometrically consistent renderings, these methods offer a foundation for closed-loop simulation as well as scalable data augmentation engine, enabling the synthesis of diverse viewpoints and conditions without the need for costly additional data collection in the real world. However, existing approaches are almost exclusively tailored to well-lit, daytime environments and struggle with the challenges of nighttime settings, where noise, strong flares, and moving light sources dominate the visual signal. To address these limitations, we propose NightSplat, a Gaussian Splatting framework specifically designed for high-fidelity reconstruction and novel view synthesis of nighttime driving scenes. Through dedicated components, our method explicitly models sensor noise, lens flare effects, and introduces a lightweight module to represent dynamic vehicle lights. Extensive evaluations on nuScenes and Waymo Open Dataset demonstrate that NightSplat significantly outperforms prior state-of-the-art baselines, both quantitatively and qualitatively, thereby extending simulation and testing to nighttime scenes. Project page: night-splat.github.io

Autonomous navigation in orchards is essential for enhancing operational efficiency and ensuring safe agricultural operations. However, autonomous navigation in orchard environments presents significant challenges due to uneven surfaces and limited visual information in natural environments. To address these issues, this study proposed a shortest‐path planning method for autonomous orchard navigation based on 3D LiDAR SLAM. First, a global 3D map was constructed using the LIO‐SAM algorithm. Ground points were then separated using the Cloth Simulation Filter (CSF), and terrain roughness information was extracted from the ground point cloud to identify rugged areas that might compromise robot stability. In parallel, an improved Random Forest model was used to segment fruit‐tree points, after which DBSCAN was applied to extract individual tree centers and the Kernel Density Estimation (KDE) method was used to estimate tree‐row direction. Finally, a cost map integrating fruit‐tree distribution and terrain roughness information was constructed, and an improved A* algorithm was employed to generate efficient and terrain‐adaptive paths. The proposed method was evaluated in both a simulation and a real pear orchard. The results showed that the proposed approach reduced traversal over rugged terrain by more than 50% and lowered estimated energy consumption by nearly 48%, while maintaining comparable path lengths and high computational efficiency. Field experiments further demonstrated reliable path‐following performance, with average lateral and longitudinal deviations within 0.18 meters and heading deviation below 3.1°. These findings highlight the practical value of incorporating terrain roughness into path planning for robust and efficient orchard navigation.

In reinforcement learning-based control, reliable zero-shot transfer from simulation to real hardware remains a major open challenge. The offline training phase typically produces multiple models, each corresponding to a different policy, and choosing which model to deploy on physical hardware is a critical decision. However, there is no established methodology for this selection process. A common practice is to deploy the policy that achieves the highest reward in simulation, without evaluating its robustness to uncertainties. As a result, due to the sim-to-real gap, the selected policy may perform poorly when transferred to a physical robotic system, potentially leading to unsafe or undesirable behavior. This work proposes to exploit the notion of “closed-loop sensitivity” for applying it, for the first time, to the analysis and comparison of different reinforcement learning models, i.e., different policies, and to estimate their expected performance on real hardware. The proposed metric captures how the closed-loop system reacts to modeling errors, providing a quantitative measure of policy robustness. Both simulation results and real world experimental campaigns demonstrate that the proposed approach enables reliable zero-shot transfer to a real robot, while effectively handling uncertainties and reducing the risk of unsafe behaviors.

Reliable underwater calibration pipelines require a robust front-end that can localize checkerboard corners at sub-pixel accuracy and assign consistent row–column indices across views. Underwater-acquired calibration images often suffer from spatially varying illumination, motion blur, specular highlights, dome-port distortion, and large viewpoint changes, which weaken corner evidence and reduce the fraction of calibration-usable views. We present an illumination-adaptive framework for ordered checkerboard corner detection in underwater-acquired calibration images. The method combines illumination-adaptive feature enhancement, structured prediction of corner confidence, sub-pixel offsets, and row/column indices, and Hungarian assignment for topology-consistent decoding. On an independent synthetic degradation benchmark, the proposed method achieves 75.4% view-aware correct corner rate (VCCR) at a 0.5-pixel threshold on the composite degradation setting, outperforming the best baseline with Multi-Scale Retinex with Color Restoration preprocessing by 9.9 percentage points. On real underwater data, under a shared OpenCV calibration backend, it achieves the lowest reprojection error among compared ordered-output methods and remains best on a fixed common-view subset. In downstream stereo reconstruction on a second real-world underwater stereo dataset, our calibration increases the common valid stereo area ratio from 67.3% to 88.8% and reduces checkerboard plane/edge root mean square error (RMSE) from 9.35/3.04 mm to 6.02/1.46 mm.

This letter concerns decentralized multi-agent systems in which individual agents operate under partial observability and uncertainty, but with the option of invoking inter-agent communication. When information exchange is infeasible or prohibitively expensive —owing to bandwidth limits, energy constraints, or the demands of clandestine operation— one is prevented from continually maintaining a joint belief over the underlying system state. In such cases, the decision of when to request information from others is key. In revisiting the prior work addressing this decision head-on, we see conservative strategies: agents are constrained to act solely on information that is accessible to all, with useful local observations being ignored whenever communication is too costly. Such an approach, though apposite to high-risk settings when consistent decision-making must be guaranteed, can degrade overall performance significantly. We explore how agents may integrate their local information, even when it has not been globally shared, and then communicate only when prudent—we term this anask-typeparadigm. This re-formulation, in which the agents are not compelled to synchronize, leads to a novel method that enables the decentralized execution of a centralized policy by reasoning, at execution time, about whether the cost of exchanging information will be offset by the information's value. We empirically evaluate our approach across several benchmarks and a case study, assessing in each case its effectiveness in trading between tolerating some loss of coordination and reducing communication costs.

The Robot Operating System (ROS) is an open-source software framework often utilised in robotics, providing standardised libraries for building complex robotic systems. However, ROS's distributed architecture and messaging system create barriers for understanding robot status and diagnosing system anomalies, particularly for developers with limited ROS expertise. We proposeROS Help Desk(RosHD), an agentic AI system for autonomous anomaly detection and debugging of ROS-based robotic systems, with expertise-adaptive assistance. RosHD continuously monitors system logs and multimodal sensor streams (lidar, RGB cameras) to detect anomalies, and provides developers with customised debugging support via specialised sensor analysis tools, diagnostic tools, ROS diagnostic utilities, code analysis capabilities, and a unified chat interface. We present the first empirical evaluation of expertise-adaptive LLM-powered debugging for ROS through a controlled user study with novice through advanced developers ($N=28$). Participants using RosHD achieved$92.85\%$task success compared to$17.86\%$with baseline assistance, reducing debugging time by up to$72\%$across experience levels.

Grasping moving objects with minimal response time while limiting jerk is essential for many industrial and service robotics applications, and is considerably more challenging than grasping static targets. Existing approaches, such as sample-based methods, are not well-suited to this task because the stochastic nature of their solutions often produces jerky or oscillatory motions. This work introduces a method for grasping objects in motion on a conveyor belt by formulating a lightweight quadratic programming problem that explicitly controls the robot's arrival time at the target pose while constraining jerk. The method further supports real-time replanning, enabling the system to handle unpredictable events and environmental uncertainties, such as conveyor starts and stops or fluctuating velocities, without prior knowledge of object speed and without requiring synchronization between the robot and the conveyor. We demonstrate the capabilities of the proposed dynamic grasp-planning framework through multiple experiments on a physical robot at very high speeds up to 16m/minute (i.e., 26.66cm/s). A comparison with a state-of-the-art baseline highlights the superior performance of the proposed approach.

To address the limitations of existing wind testing equipment and enable realistic airflow simulation for unmanned aerial vehicle (UAV) flights, a novel rotor‐based wind wall system was developed. This system accurately replicates complex aerodynamic environments, allowing for precise, controlled indoor testing of UAV performance and aerodynamic behavior. It provides a reliable and versatile platform for UAV research under realistic yet reproducible conditions. The rotor array design was validated using computational fluid dynamics (CFD) simulations in ANSYS Fluent, employing the finite volume method under representative operating conditions. Results showed that beyond 0.4 m from the wall, the wind field achieved high uniformity, with deviations in the velocity streamlines normal to the plane below 0.01 m. Both uniform and non‐uniform wind profiles were attainable by adjusting the distance from the wall, without compromising adjacent airflow stability, confirming the design's effectiveness in generating controllable and stable airflow. Wind speed accuracy was evaluated by comparing simulation outputs with actual data from UAV flights. The system demonstrated a maximum error of less than 0.5 m/s and a mean error of 0.147 m/s, indicating strong agreement with real‐world conditions. These results validate the device's high accuracy and reliability in simulating realistic wind fields. The rotor‐based wind wall represents a significant advancement in indoor aerodynamic testing, enabling the detailed investigation of UAV performance across diverse wind conditions. Its precise control, stability, and fidelity make it a valuable tool for advancing UAV design, control algorithms, and flight testing methodologies in a safe and repeatable environment.

Uncrewed underwater vehicles (UUVs) have transformed oceanographic research through autonomous data gathering. Similarly, lake and other aquatic research can potentially be automated and transformed. However, UUVs would need to be smaller, lighter, less complex, and cheaper than currently available to make them more practical and user‐friendly lake research tools. Conventionally powered micro‐UUVs are entering the market, but micro‐glider UUV (which uniquely provides long‐duration mobile monitoring capability) development has largely been the preserve of “hobbyists.” Here, we present innovative design guidelines for a micro‐glider UUV that features the novel use of a single actuator to achieve control in three dimensions. This minimizes UUV size, weight, and complexity, and optimizes maneuverability to promote suitability for use in lakes. Presented is a comprehensive and dynamic modeling technique that uses three external and three internal principal glider dimensions to explore the complete micro‐glider UUV design space. This model can be used to predict transitory and steady‐state glider performance and ultimately optimize design parameters. The prototype proof‐of‐concept testing has demonstrated practical validation of the design and modeling work. This work is significant as it explores an innovative micro‐glider UUV design for lake and reservoir applications; perhaps of greater importance, the modeling approach presented can form the base for micro‐glider UUV design. This will allow the development of enhanced automated monitoring capabilities across a broad range of aquatic systems and promote understanding of the freshwater and marine worlds.

Water tanks are vital in supplying and storing water for both domestic and industrial needs. Improper maintenance of water tanks leads to water contamination. Contaminated water is one of the main contributors to skin and hair diseases. It is crucial to ensure that water tanks are kept clean and free of contaminants, especially in residential and commercial settings. Currently, there are hardly any method for dirt or stain detection and cleanliness evaluation after cleaning the tanks. In addition, manual methods are still widely used to clean these tanks, and the process is tiresome and time‐consuming. Autonomous water tank cleaning robots have been suggested to address these challenges, performing the cleaning process with minimal human intervention, thereby reducing time, effort, and risks. Although research has been conducted in this field, there has been limited progress in integrating artificial intelligence (AI)‐based cleanliness evaluation modes into robotic systems. This research proposes an autonomous robot capable of cleaning both the floor and walls of concrete water tank structures. The robot incorporates a custom‐trained modified U‐Net model to detect and clean residual dirt patches missed during the initial cleaning. The models were trained and validated on a custom dataset of 7300 images. Among all the trained segmentation models, the modified U‐Net model achieved a validation accuracy of 96%.

This paper presents a novel Embroidery Actuator, a fabric-integrated pneumatic actuator that enables diverse and controllable deformations through embroidery pattern design. Unlike conventional fabric actuators that rely on fiber- or thread-shaped actuators, the proposed actuator is fabricated by directly stitching an inflatable tube onto the fabric using a cord-embroidery technique. The embroidered thread and the fabric jointly form a sleeve that constrains the expansion of the inflatable tube, converting internal pressure into targeted bending or stretching deformations. By varying the embroidery pattern, such as zigzag or cross configurations, different geometric constraints can be realized, allowing for flexible control of deformation direction and magnitude. Analytical deformation models based on theNeo-Hookean modelandLagrange's equationswere developed to predict the relationship between pneumatic pressure and bending angle. And then,experiments demonstrated that the actuator achieved 47 degrees of flexion on the fabric surface side and 165 degrees on the reverse side by altering the embroidery pattern. Additionally, the created model expressed deformation with an error margin of several degrees.

Inspired by the biological mechanism of fish caudal peduncles, which are key structures connecting musculature to the caudal fin and modulating stiffness for controlled energy transfer to enable high maneuverability across diverse swimming scenarios, this letter proposed a variable-stiffness caudal peduncle for a robotic fish that integrates thermoplastic polymer polycaprolactone, using temperature control to modulate the molten state of the material and thereby adjust stiffness. This structure enables the formation of an optimal body profile across a wide frequency range, enhancing propulsion performance. The Pseudo-Rigid-Body Model and Lagrangian method were used to model the flexible caudal peduncle and dynamic behavior of the robotic fish, respectively. Thrust results show that the caudal peduncle, in its molten state, exhibits superior performance at low frequencies, while in its solid state, it performs better at high frequencies. Simulations and experiments revealed an optimal stiffness for maximum thrust, with a peak average thrust of 0.72 N. Untethered swimming tests confirmed that temperature-based stiffness regulation of the PCL molten state enables a maximum speed of 0.47 m/s (0.88 body lengths per second) and a minimum cost of transport of 68.9 J/m/kg.

This paper discusses lap time optimization, focusing on a single lap without considering opponents in autonomous racing. The paper presents a control and optimization architecture composed of a model‐based low level controller and a higher level iterative learning algorithm with the goal of obtaining the fastest qualifying lap in autonomous racing competitions. First principles models are extremely expensive to calibrate near the handling limit, to solve this issue our algorithm learns the position varying acceleration limits of the vehicle over multiple laps. The proposed algorithm brings together the robustness and generalization capability of model‐based approaches with the performance of data‐driven methods. To validate the approach and its computational efficiency, we implement the solution on a high performance small scale vehicle and test it against a human driver on a racing track with speed up to 50 km/h and lateral accelerations of 1.2 g. The proposed approach beats a national level champion in terms of qualifying lap for small scale vehicles, on the considered test track.

Existing skiing robots are severely limited in speed and terrain adaptability, because their controllers neglect to explicitly model the complex ski‐snow interaction. This paper presents DynaSki, a complete locomotion framework that overcomes this limitation through a novel model predictive controller (SKIMPC) centered on a Line‐Ski Contact (LSC) model. The LSC model, formulated as a specialized Contact Wrench Cone (CWC), enables the controller to optimize the contact wrench at the ski edge. This core controller is synergistically integrated within the complete DynaSki framework, which also includes: an analytical kinematic solver for the robot's unique parallel mechanism to ensure precise wrench mapping; a dynamics‐aware trajectory planner that converts user commands into feasible trajectory references; and a dedicated landing controller to robustly manage airborne phases. The effectiveness of DynaSki is validated in real world experiments. The robot achieved a top speed exceeding 18 m/s on a steep trail (max inclination 22.8°), traversed wavy slopes with undulations, and executed a stable landing after a 0.3 s airborne phase for the first time in this field.

Heavy unmanned underwater vehicles (UUVs) with high‐capacity batteries and various payloads demand advanced energy replenishment and data retrieval technologies. Conventional suspended docking stations are unable to effectively counteract the impact forces of heavy UUVs, making seabed docking a more feasible solution. This study presents a seabed docking system for heavy UUVs that utilizes innovative differentiated optical‐guidance beacons, significantly enhancing the effective docking range to 35 m. Through the study of underwater optical characteristics, an effective‐workspace index for underwater optical guidance was proposed, guiding the design of differential beacons. Upon successful docking, the system enables high‐power magnetic‐coupling wireless power transfer (MC‐WPT) and high‐speed underwater wireless optical communication (UWOC). The MC‐WPT system operates at a power level of 6.6 kW, representing a high‐power level for underwater wireless charging and achieving an efficiency of 88.6%. A maximum data rate of 27 Mbps was attained by the UWOC system by employing a “wide‐angle transmission and wide‐field‐of‐view reception” technique. Validation tests conducted at a depth of 60 m in Qiandao Lake, using a UUV with a diameter of 533 mm, demonstrated a 100% successful docking rate. The results validate the feasibility and robustness of the proposed docking system in real‐world underwater conditions.

Autonomous navigation for inspection robots in cattle barns critically depends on localization and path‐planning algorithms. To address the issues of low navigation accuracy, long planning time, and insufficient trajectory smoothness in barn environments, this study proposes a navigation framework that combines LiDAR–inertial odometry and a rapidly‐exploring random tree (LIO‐RRTNav). For relocalization, this study develops a Fast_LIO2 with relocalization and pose optimization method (RP‐Fast_LIO2). It exploits the geometric structure of the barn to associate the current frame point cloud with historical point clouds to suppress drift. Meanwhile, stable features are extracted from dynamic scenes and registered to the global map using the Iterative Closest Point (ICP) algorithm; the registration result is used as the initial state for an iterated extended Kalman filter (IEKF) to refine pose estimation. For path planning, this study proposes a highly efficient, robust, and smooth rapidly exploring random tree method (HRS‐RRT). By incorporating a goal‐oriented random sampling strategy and a safety‐distance constraint, it enables safe and efficient planning, and the resulting path is further optimized through path pruning and cubic B‐spline smoothing based on a spring potential energy model. Simulation test results showed that the HRS‐RRT algorithm reduced 22.17%, 75.00%, and 83.09% in terms of path length, planning time and number of iterations, respectively, when compared with the traditional RRT algorithm. The experimental results revealed that RP‐Fast_LIO2 algorithm reduced the mean and root mean square error of the absolute position error by 83.21% and 79.89%, when comparing with the traditional Fast_LIO2 algorithm. In navigation experiments conducted in two cattle yards, when the robot operated at speeds of 0.3, 0.5, and 1.0 m/s, the maximum lateral and longitudinal deviations did not exceed 0.13 and 0.08 m, respectively, and the maximum heading error did not exceed 7.24°. The results acquired verified the adaptability of the LIO‐RRTNav algorithm in the cattle yard environment, meeting the requirements of cattle yard inspection robot operation.

This study focuses on the perception of traversable areas and ground materials in complex field environments, driven by the application demands of autonomous ground intelligent agents in critical tasks such as battlefield support and disaster relief. Compared to existing studies on fine‐grained segmentation in structured environments (e.g., urban roads), perception in unstructured field environments is far more challenging due to irregular terrain and complex conditions, while research in this area remains limited. To address this, this paper proposes a cross‐modal synergistic optimization multi‐task segmentation network (CSOM‐Net) for ground intelligent agents in field environments, based on an in‐depth analysis of the technical requirements in complex field settings. This paper proposes a conflict‐consistency guided feature fusion (CCGF) method to resolve the challenge of feature conflicts in multimodal data that hinder effective fusion in complex field environments. A cross‐modal bidirectional mutual adaptation (CBMA) learning strategy is proposed to address the feature interference caused by significant differences in optimization objectives in cross‐modal tasks under field environments. Extensive comparative and ablation experiments were conducted on both real‐world and simulated complex field datasets to evaluate the proposed CSOM‐Net model. Experimental results show that CSOM‐Net outperforms conventional models, achieving approximately 5% and 2% mIoU improvements in point cloud traversable area and image ground material segmentation tasks, respectively. The model proves effective for joint segmentation tasks of point cloud traversable areas and image ground materials in complex field environments, providing a robust solution to enhance environmental perception capabilities for autonomous ground agents operating under challenging field conditions.

To maintain ecological balance (EB), eco‐robots are equipped with situational Intelligence (SI). Yet, the robot itself causes pollution through emissions. None of the existing studies focused on designing carbon‐neutral eco‐robots (CNER) for air quality monitoring (AQM). Thus, this paper proposes an enhanced EB model using CNERs with SI for AQM. The process begins with deploying the lightweight optimal eco‐robot design with a minimum carbon footprint. Next, by using the air quality along with the pollution assessment dataset, an AI‐assisted eco‐robots‐based AQM framework is trained. Here, the dataset is pre‐processed; then, features are extracted. Next, to predict the air quality as good, moderate, poor, and hazardous, the proposed probabilistic Matusita neural soft clipish network (PMNSCN) is employed. In real‐time, the environmental data are sensed via the multi‐factor sensors. Afterward, to avoid unnecessary delays, the noisy data are refined. Then, the air quality is predicted, followed by on‐site active interventions (OAI). Risk‐based prioritization is done to offer long‐term measures. According to the risk zones, the authorities undertake eco‐robot‐based clean‐up efforts. The proposed approach ultimately promotes a sustainable environment with 98.9773% accuracy.