本期机器人顶刊精选

LiDAR point cloud registration is fundamental to robotic perception and navigation. However, in geometrically degenerate or narrow environments (e.g., corridors), registration problems become ill-conditioned: certain motion directions are weakly constrained, leading to unstable solutions and degraded accuracy. Existing methods adopt a detect-then-mitigate paradigm but fail to reliably detect, physically interpret, and stabilize such ill-conditioning without corrupting the optimization problem. In this study, we introduce DCReg ( Decoupled Characterization for Ill-conditioned Registration ), which establishes a principled detect–characterize–mitigate paradigm that systematically addresses ill-conditioned registration through three integrated innovations. First, DCReg achieves reliable ill-conditioning detection by employing a Schur complement decomposition to the hessian matrix. This structure-aware technique decouples the 6-DoF registration problem into 3-DoF clean rotational and translational subspaces, eliminating coupling effects that mask degeneracy patterns in conventional full-Hessian analyses. Second, within these decoupled subspaces, we develop physically interpretable characterization techniques that resolve eigenbasis ambiguities through principled basis alignment. This establishes stable and quantitative mappings between mathematical eigenspaces and physical motion directions, providing actionable insights about which specific motions lack constraints and to what extent. Third, leveraging this directional spectral information, we design a targeted mitigation strategy via a structured preconditioner. While guided by MAP regularization principles for spectral shaping, our approach implements eigenvalue clamping exclusively within the preconditioner rather than modifying the origin problem. This preserves the original least-squares objective and its minimizer while enabling efficient and stable optimization via Preconditioned Conjugate Gradient with a single physically interpretable parameter. Extensive experiments demonstrate DCReg achieves at least 20%–50% improvement in long-duration localization accuracy and typical speedups of 5 − 30× (up to 116× in certain scenarios) over comparable degeneracy-aware baseline methods across diverse environments. Our implementation will be available at https://github.com/JokerJohn/DCReg .

Robotic manipulation policies are commonly initialized through imitation learning, but their performance is limited by the scarcity and narrow coverage of expert data. Reinforcement learning can refine polices to alleviate this limitation, yet real-robot training is costly and unsafe, while training in simulators suffers from the sim-to-real gap. Recent advances in generative models have demonstrated remarkable capabilities in real-world simulation, with diffusion models in particular excelling at generation. This raises the question of how diffusion model-based world models can be combined to enhance pre-trained policies in robotic manipulation. In this work, we propose World4RL, a framework that employs diffusion-based world models as high-fidelity simulators to refine pre-trained policies entirely in imagined environments for robotic manipulation. Unlike prior works that primarily employ world models for planning, our framework enables direct end-to-end policy optimization. World4RL is designed around two principles: pre-training a diffusion world model that captures diverse dynamics on multi-task datasets and refining policies entirely within a frozen world model to avoid online real-world interactions. We further design a two-hot action encoding scheme tailored for robotic manipulation and adopt diffusion backbones to improve modeling fidelity. Extensive simulation and real-world experiments demonstrate that World4RL provides high-fidelity environment modeling and enables consistent policy refinement, yielding significantly higher success rates compared to imitation learning and other baselines.

The humanlike form of humanoid robots uniquely positions them to achieve the agility and versatility in motor skills that humans have. Learning from human demonstrations offers a scalable approach to acquiring these capabilities. However, prior works either produced unnatural motions or relied on motion-specific tuning to achieve satisfactory naturalness. Furthermore, these methods are often motion or goal specific, lacking the versatility to compose diverse skills, especially when solving unseen tasks. We present BeyondMimic, a framework that scales to diverse motions and carries the versatility to compose them seamlessly in tackling unseen downstream tasks. A compact motion tracking formulation enables mastery of a wide range of highly agile behaviors, including aerial cartwheels, spin kicks, flip kicks, and sprinting, with a single setup and shared hyperparameters, all while achieving humanlike performance. Moving beyond the mere imitation of existing motions, we propose a unified latent diffusion model that empowers versatile goal specification, seamless task switching, and dynamic composition of these agile behaviors. Leveraging classifier guidance, a diffusion-specific technique for test-time optimization toward unseen objectives, our model extended its capability to solve downstream tasks never encountered during training, including motion inpainting, joystick teleoperation, and obstacle avoidance, and transferred these skills zero-shot to real hardware. Together, these components enable scalable acquisition of humanlike motor skills from human motion and motion synthesis that generalizes and adapts beyond the training setup.

Wildfires and deforestation pose significant threats to forest ecosystems, motivating scalable methods for reforestation beyond labor‐intensive cultivation of native forest seedlings in nurseries. To enable large‐scale direct seeding of native species, this paper presents the design, analysis, and field evaluation of a UAV‐based seed‐dropping system. The approach integrates a lightweight carriage mechanism for gravity‐based seeding systems with an open‐source multirotor UAV to enable precision seeding operations. Precise seed placement enables adherence to ecologically motivated spacing, for example, 1.8 m for framework‐species plots, which broadcast seeding cannot readily achieve. Additionally, the system uses 30 mm biochar‐coated seed balls, which facilitate the handling of multiple seed varieties and enhance overall germination success in framework‐species approaches. The autonomy stack combines LiDAR‐based terrain following with deterministic actuation and on‐board seed‐drop confirmation via an outlet limit switch, providing a geotagged record of each seed drop for post hoc analysis of the reforestation efforts. Field evaluations at an active reforestation site in Northern Thailand have demonstrated the feasibility of the proposed solution. An altitude sensitivity study at 5, 10, and 15 m height revealed wind‐aligned drift and terrain interactions; operations at 3–5 m height were found to provide practical operating windows for precision deployments, with 95% of drops landing within 0.66 m of the target for a controlled test at 5 m height. A representative 27‐waypoint autonomous mission demonstrated a 90% success rate, with 24 drops confirmed by the on‐board system and matched ground observations. Unlike conventional aerial seeding systems, the proposed precision seed‐dropping approach significantly reduces seed wastage, making the solution scalable for large‐scale ecological restoration and biodiversity recovery initiatives.

The design of robotic binocular camera systems has been inspired by human vision, as has their use in humanoid robots. One aspect of this inspiration has yet to play a major role, namely, that humans determine depth using a convergent binocular imaging geometry with both eyes pointing at the same location. Although some robot heads have the functionality to use vergence and version movements and thus alter binocular imaging geometry, a method to exploit it for depth computation has not been explored. Even though to an observer, their motions and appearance might resemble that of humans, in detail, the resemblance to the motions required for actual convergent depth computation is not seen. To bridge this gap, we present convergent binocular stereo (CBS), a stereo algorithm designed to provide a foundation for purposeful binocular computations under the humanoid constraint, intended for active binocular robots. CBS computes horizontal and vertical disparities using a coarse-to-fine refinement strategy with Gabor-filtered responses. We also introduce the Convergent Binocular Stereo–BenchMark (CBS-BM), a convergent, natural image dataset containing 49 scenes with ground-truth horizontal disparity collected on a four-degrees-of-freedom robotic system. Our evaluation, a quantitative comparison between parallel and convergent stereo systems, shows that CBS is broadly competitive with state-of-the-art parallel methods, even outperforming them in scenes with repeated patterns and in mean horizontal disparity and depth error over all scenes. Although not intended to replace parallel stereo where humanlike behavior is unnecessary, CBS enables functionally realistic depth computation for humanoid robotic heads.

Learning sensorimotor contingencies—that is, the link between one’s actions and their sensory effects—is fundamental to developing body knowledge, understanding causality, and developing a sense of agency. In developmental psychology, this process is classically studied using the mobile paradigm, where infants learn that movement of a limb causes motion of a connected mobile. To expand our understanding of how infants learn this, we tested an embodied computational model that learns through two biologically inspired mechanisms: prediction and curiosity. Implemented on the child-sized iCub humanoid robot interacting with a mobile, the model detected sensorimotor contingencies across several experimental conditions using a variety of movement strategies. Our findings suggest that contingency learning cannot be captured by a single behavioral metric, such as the amount of movement, but instead emerges through a spectrum of exploratory behaviors. Analysis of the robot’s internal activity reveals that these behaviors emerge from the dynamic trade-off between prediction and curiosity—between exploitation and exploration. Our work provides a biologically motivated, physically embodied model of sensorimotor interaction that connects theories of infant learning with robotic implementations. The results allow us to generate testable hypotheses for developmental research and to inform the design of autonomous learning systems.

This special issue examines what it takes to move humanoids beyond task-specific demonstrations toward dependable generality.

Despite the rise of billion-parameter foundation models trained across thousands of graphical processing units (GPUs), similar scaling gains have not been shown for humanoid control. Current neural controllers for humanoids remain modest in size, target a limited set of behaviors, and are trained on a handful of GPUs. We show that scaling model capacity, data, and compute yields a generalist humanoid controller capable of natural, robust whole-body movements. We position motion tracking as a scalable task for humanoid control, leveraging dense supervision from diverse motion-capture data to acquire human motion priors without manual reward engineering. We build a foundation model for motion tracking by scaling along three axes: network size (1.2 to 42 million parameters), dataset volume (more than 100 million frames from 700 hours of motion capture), and compute (21,000 GPU hours). Beyond demonstrating the benefits of scale, we further show downstream utility through a real-time kinematic planner that bridges motion tracking to tasks such as navigation, enabling natural and interactive control, as well as a unified token space that supports virtual reality (VR) teleoperation and vision-language-action (VLA) models with a single policy. Through this interface, we demonstrate autonomous VLA-driven whole-body locomanipulation requiring coordinated hand and foot placement. Scaling motion tracking exhibits favorable properties: Performance improves steadily with compute and data diversity, and learned policies generalize to unseen motions, establishing motion tracking at scale as a practical foundation for humanoid control.

Achieving robust, humanlike whole-body control on humanoid robots for agile, contact-rich behaviors remains a central challenge, demanding heavy per-skill engineering and a brittle process of tuning controllers. We introduce ZEST (zero-shot embodied skill transfer), a streamlined motion-imitation framework that trains policies via reinforcement learning from diverse sources—high-fidelity motion capture, noisy monocular video, and non–physics-constrained animation—and deploys them to hardware zero-shot. ZEST generalizes across behaviors and platforms without relying on contact labels, reference or observation windows, state estimators, or extensive reward shaping. Its training pipeline combines adaptive sampling, which focuses training on difficult motion segments, and an automatic curriculum using a model-based assistive wrench, together enabling dynamic, long-horizon maneuvers. We further provide a procedure for selecting joint-level gains from approximate analytical armature values for closed-chain actuators, along with a refined model of actuators. Trained entirely in simulation with moderate domain randomization, ZEST demonstrated broad generality. On Boston Dynamics’ Atlas humanoid, ZEST learned dynamic, multicontact skills (army crawl and breakdancing) from motion capture. It transferred expressive dance and scene-interaction skills, such as box climbing, directly from videos to Atlas and the Unitree G1. Furthermore, it extended across morphologies to the Spot quadruped, enabling acrobatics, such as a continuous backflip, through animation. Together, these results demonstrate robust zero-shot deployment across heterogeneous data sources and embodiments, establishing ZEST as a scalable interface between biological movements and their robotic counterparts.

Multimodal large language models (MLLMs) have advanced zero-shot end-to-end Vision-Language Navigation (VLN), yet robust navigation requires not only semantic understanding but also predictive modeling of environment dynamics and spatial structure. We propose PROSPECT, a unified streaming navigation agent that couples a streaming Vision-Language-Action (VLA) policy with latent predictive representation learning. PROSPECT uses CUT3R as a streaming 3D foundation spatial encoder to produce long-context, absolute-scale spatial features, and fuses them with SigLIP semantic features via cross-attention. During training, we introduce learnable stream query tokens that query the streaming context and predict next-step 2D and 3D latent features (rather than pixels or explicit modalities), supervised in the latent spaces of frozen SigLIP and CUT3R teachers. The predictive branch shapes internal representations without inference overhead. Experiments on VLN-CE benchmarks and real-robot deployment demonstrate state-of-the-art performance and improved long-horizon robustness under diverse lighting. We will release code for the community soon.

Many aerial tasks involving quadrotors demand both instant reactivity and long-horizon planning for obstacle avoidance, energy efficiency, or trajectory tracking. High-fidelity models enable accurate control but are too slow for long horizons. Low-fidelity planners scale but cannot directly control the system, necessitating cascaded architectures. Prevailing hierarchical approaches plan with a simplified model and use a high-fidelity controller for tracking, yet this decomposition is inherently suboptimal. The controller is limited by the coarse plan, and conventional MPC alternatives shorten the horizon to stay real-time feasible. We present UNIQUE, an MPC architecture that replaces this hierarchical stacking with temporal cascading. The planning problem is formulated as the second-tail horizon of a single multi-phase MPC, rather than being solved separately. We align costs across horizons, derive feasibility constraints for the point-mass planning model, and introduce transition constraints that convert high-fidelity states into meaningful low-fidelity states. Parallel point-mass and mixed-integer solvers address nonconvexities while incorporating progressive 3D obstacle smoothing over the planning horizon. In simulations and real flights, under equal computational budgets, UNIQUE improves closed-loop tracking by up to 75% compared with standard MPC and hierarchical baselines. Ablations and Pareto analyses confirm performance gains across variations in horizon, constraint approximations, and smoothing schedules.

Humanoid soccer poses a representative challenge for embodied intelligence, requiring robots to coordinate agile locomotion with unreliable visual perception in dynamic environments. However, existing systems typically rely on modular pipelines that separate perception from control or assume ideal sensing, making it difficult to achieve coherent and reactive behavior under real-world perceptual limitations. In this work, we present a unified reinforcement learning–based controller that enables humanoid robots to learn vision-driven reactive soccer skills by directly coupling visual perception with locomotion control. The robot is trained in simulation to acquire soccer behaviors, and adversarial motion priors guide policy learning toward natural motion patterns. To support robust performance under imperfect sensing, we introduce an encoder-decoder architecture together with a virtual perception system that models key characteristics of onboard vision, exposing the policy to perceptual noise and detection failures during training. This design encourages the policy to internalize perceptual uncertainty and continuously adapt its motion in a closed loop. The resulting controller produces coordinated soccer behaviors using only onboard vision, including ball searching, chasing, and multidirectional kicking. It reduces ball position estimation error by 46% and shortens time-to-kick by up to 64% compared with a rule-based baseline, achieving around 90% kicking success in frontfield positions. Experiments across diverse environments and dynamic scenarios, including real RoboCup competitions, further demonstrate the robust performance of the controller. These results highlight the practical effectiveness of integrating perceptual uncertainty directly into policy learning for achieving reliable vision-driven behaviors in humanoid robots operating under real-world conditions.

Humanoid robots stand at the forefront of robotics, aiming to capture the agility, robustness, and expressivity of human movement in an anthropomorphic form. The locomotion control of humanoids has evolved from classical model–based methods to reinforcement learning powered by large-scale simulation and now to generative models that produce adaptive, whole-body behaviors, propelling humanoids toward operation in real-world environments. This survey positions humanoid control at a turning point, converging toward a unified paradigm of physics-guided generative intelligence that integrates optimization, learning, and predictive reasoning. We identify three core principles linking these paradigms: physics-based modeling, constrained decision-making, and adaptation to uncertainty. Building on these connections, we provide recommendations for researchers and outline open challenges in safety, accessibility, and human-level capability. These directions represent a transformation from engineered stability to intelligent autonomy, laying the groundwork for humanoid generalists capable of operating safely, collaborating naturally, and extending human capability in the open world.

We propose a sensor-guided adaptive contrastive learning framework for visual representation learning in PointGoal navigation. During training, privileged LiDAR sensing guides the contrastive objective through a geometry-aware similarity metric and adaptive temperature scaling, encouraging visual embeddings to capture navigation-relevant structure rather than scene-specific appearance. The resulting encoder is pretrained independently, frozen, and used as the perceptual backbone for reinforcement learning, decoupling representation learning from policy optimization. We further introduce a cross-stage domain mismatch between representation pretraining and policy learning to suppress environment-specific shortcuts and promote reliance on task-relevant features. Extensive experiments in high-fidelity simulation demonstrate that our approach significantly improves policy-level scene transfer across diverse indoor and outdoor environments. At deployment, the agent relies only on monocular RGB observations together with standard task-related inputs such as goal position and proprioceptive signals, without access to LiDAR or other privileged sensors. Our method outperforms large pretrained vision models and standard contrastive baselines under severe appearance and semantic shifts. We also release a multimodal dataset to support future research on privileged-guided visual representation learning for navigation. The code is available at:

Monolithic vision-action models represent an emerging paradigm in autonomous driving. However, this architecture produces token sequences that quickly exceed real-time computational budgets when encoding extended temporal context for complex interactions. While approaches like linear transformers and external memory try to make the context lightweight, token compression is most compatible with the architecture as it requires no backbone modifications. Yet existing compression adopts rule-based heuristics like temporal decay, decoupled from planning, risking loss of decision-critical information. We propose COMPACT-VA, a planning-aligned working memory framework built on conditional VQ-VAE, compressing extended context into bounded representations. Compression is conditioned on both historical trajectory and a learned planning intent that the posterior encoder distills from future trajectories during training, while the prior encoder learns to predict it from compressed observations. The compressed memory, concatenated with the predicted latent, feeds the policy for end-to-end optimization, planning with retained decision-critical information. We evaluate on high-signal dynamic scenarios where historical context is most critical for behavior correctness (e.g., stop, yield, or proceed), and accordingly design behavioral metrics. Under comparable token budgets, we achieve $>$6% improvement (68.3%) on success rates with consistent gains across metrics. Ablations validate planning-aligned coupling effectiveness. Closed-loop evaluation confirms that COMPACT-VA maintained general driving performance with 3.3* speedup and 2.7* memory reduction over uncompressed processing.

Image-goal navigation using lightweight topological maps is a practical paradigm for indoor robot deployment: the map requires only geotagged images, and localization relies on visual matching rather than precise pose estimation. However, learned waypoint predictors can produce targets that violate geometric constraints or deviate from the global path. Executing these waypoints safely further requires a local planner capable of collision avoidance, yet existing systems either lack one or rely on fixed parameters that cannot adapt to confined spaces. To address these limitations while retaining the navigational intuition of the learned predictor, we present EgoNav, a hierarchical system that implements this idea by generating candidates from semantically segmented traversable regions and scoring them alongside the learned waypoint for geometric safety, directional coherence, and fidelity to the learned prior. An adaptive local path planner then executes the refined waypoint with parameters modulated based on the refinement outcome. Experiments in Habitat-sim and on a physical humanoid robot show that EgoNav consistently outperforms contemporary baselines in both success rate and path efficiency.

This paper introduces Embodied Scene Rearrangement Planning (ESRP), a novel task requiring embodied agents to rearrange furniture in 3D scenes to match a target configuration using only egocentric observations and a top-down target layout. Unlike prior rearrangement tasks, ESRP precludes global state access and introduces mutual object occlusions, reflecting the practical constraints of real-world robotic deployment. These factors make aligning partial egocentric observations with the global target layout particularly challenging for long-horizon planning. To facilitate research, we present ESRP-Bench, a comprehensive benchmark built on OmniGibson featuring over 5,400 scene pairs and 8,200 objects. We define three multi-level metrics to evaluate rearrangement quality and provide four baselines: a hierarchical task-and-motion planning method, a vision-language-model-based method, and two learning-based approaches (IL and RL). Experimental results demonstrate that current methods struggle to complete the task efficiently, highlighting ESRP as a challenging frontier for embodied agents in scene understanding and long-horizon task planning. This work serves as a stepping stone toward deploying intelligent agents in real-world scenarios. Project page: https://pie-lab.cn/ESRP/.

Navigation in the natural world is a feat of adaptive inference, where biological organisms maintain goal-directed behaviour despite noisy and incomplete sensory streams. Central to this ability is the Free Energy Principle (FEP), which posits that perception is a generative process where the brain minimises Variational Free Energy (VFE) to maintain accurate internal models of the world. While Deep Neural Networks (DNNs) have served as powerful analogues for biological brains, they typically lack the real-time plasticity required to handle abrupt sensory shifts. We introduce FEP-Nav, a biologically inspired framework for real-time perceptual adaptation in robust visual navigation. Motivated by the decomposition of VFE into prediction error and Bayesian surprise, FEP-Nav combines a Top-down Decoder, which provides an internal expectation of uncorrupted sensory input, with Adaptive Normalisation, which adjusts shifted feature distributions toward prior statistics. We interpret reconstruction and normalisation as approximate mechanisms for reducing the corresponding VFE-related terms during inference without gradient-based updates. Experiments across simulated and real-world visual corruptions show that FEP-Nav restores performance lost under visual corruption, outperforming non-adaptive baselines and strong adaptive methods. These results suggest that variational principles can provide a useful design perspective for robust autonomous behaviour under degraded sensory conditions.

Unmanned aerial vehicle (UAV) swarm agents operating in radio-frequency (RF)-degraded environments require coordination mechanisms that connect directly observable signals to executable responses. This work presents an embodied visual-communication approach in which bio-inspired motion–LED glyphs are represented by a reduced six-parameter semantic chart embedded within a full 24-dimensional hybrid execution manifold. A learned translator large language model (LLM) maps a perceived glyph to a response that is instantiated as an executable trajectory and propagated through closed-loop quadrotor dynamics. The system is evaluated in 200 three-UAV search-and-rescue trials with receiver-specific degradation from sensing range, field of view, occlusion, and relative motion. Under clean observations, the quantized small model and rule-based translator both achieve $$100\%$$ 100 % semantic correctness, but under single- and two-parameter corruption, the quantized small model achieves $$64.7\%$$ 64.7 % and $$64.1\%$$ 64.1 % , compared with $$44.8\%$$ 44.8 % and $$35.2\%$$ 35.2 % for rule-based, while reducing mean multi-agent trajectory error from $$2.902~\textrm{m}$$ 2.902 m to $$0.993~\textrm{m}$$ 0.993 m . All methods maintain $$100\%$$ 100 % rotor-allocation and finite-horizon feasibility. The quantized small model also has comparable overall semantic correctness with the large model ( $$83.3\%$$ 83.3 % vs. $$86.3\%$$ 86.3 % ) while dropping latency from $$5.115~\textrm{s}$$ 5.115 s to $$2.789~\textrm{s}$$ 2.789 s and is validated onboard a Jetson–Pixhawk UAV platform, where airborne inference and bounded command execution produce measurable physical motion. These results establish a unified pathway from degraded visual observation to semantically meaningful, dynamically grounded UAV coordination.

Flapping-wing micro aerial vehicles offer quieter and safer operation than rotary-wing drones, yet achieving precise autonomous control of bird-scale ornithopters remains challenging: lift, airspeed, and turning authority are tightly coupled and governed by only a few control inputs. Conventional cascaded controllers treat altitude, speed, and heading independently, producing persistent tracking errors during complex maneuvers, while time-parameterized trajectory tracking requires predefined speed profiles that existing methods cannot robustly produce for these coupled dynamics. We address both limitations simultaneously with a Model Predictive Contouring Control (MPCC) approach that tracks arc-length-parameterized trajectories while optimizing progress online, eliminating the need for predefined timing. However, MPCC requires a dynamical model that captures the coupled aerodynamics without exceeding the computational budget of real-time nonlinear optimization. Here, we propose a compact, continuously differentiable model that captures the dominant couplings of bird-scale ornithopters, enabling real-time predictive control. We validated the method with the XFly ornithopter flying along circular and three-dimensional racing trajectories and achieved a mean deviation from the reference trajectory between 6.5 and 9 cm at speeds up to 3 m/s, which represents an 8.5x improvement over prior ornithopter control methods.

Robust stereo visual-inertial odometry (VIO) remains challenging in low-texture scenes and under abrupt illumination changes, where point features become sparse and unstable, leading to ambiguous association and under-constrained estimation. Line structures offer complementary geometric cues, yet many efficient point-line systems still rely on point-guided line association, which can break down when point support is weak and may lead to biased constraints. We present a stereo point-line VIO system in which line segments are equipped with dedicated deep descriptors and matched using an entropy-regularized optimal transport formulation, which performs global mass-transport assignment and supports unmatched observations under ambiguity, outliers, and partial observations. The proposed descriptor is training-free and is computed by sampling and pooling network feature maps. To improve estimation stability, we analyze the impact of line measurement noise and introduce reliability-adaptive weighting to regulate the influence of line constraints during optimization. Experiments on EuRoC and UMA-VI, together with real-world deployments in low-texture and illumination-challenging environments, demonstrate improved accuracy and robustness over representative baselines while maintaining real-time performance.

We present Riemannian Informed Trees (RIT*), a planning framework that replaces Euclidean primitives in batch-informed search with their Riemannian counterparts. RIT* constructs a tighter, cost-consistent informed set, performs a nearest-neighbour search under an anisotropic distance metric, and evaluates edge costs efficiently via a cascading scheme. We further introduce a Collision-Adaptive Metric Refinement (CARM), which learns an obstacle-proximity cost field online from collision feedback, reducing the reliance on prior metric design in practical settings. Experiments across environments from 2-D to 14-D show that RIT* is competitive in low-dimensional and spatially constant-metric settings and produces substantially lower-cost solutions when the metric varies spatially in high-dimensional configuration spaces. Performance gains scale with anisotropy and dimension, reaching up to 13.0% improvement in median initial cost over BIT* in the 3-D anisotropic benchmark, up to 9.0% in median final cost over BIT* in 6-DOF manipulation, and 24.8-63.5% in a 14-DOF bimanual planning problem, where Euclidean-informed baselines degrade. Videos and code can be found here: https://muhayyuddin.github.io/ritstar/

Skill abstraction---the process of learning reusable and temporally extended behaviors---has emerged as a key paradigm for improving sample efficiency and generalization in robot learning. For efficient skill transfer to real robots, learned skills must generalize across tasks and domains, remain robust to visual and dynamic perturbations, and be efficient enough for practical deployment. However, existing methods typically satisfy only a subset of these properties, as they capture either high-level semantic intent (what) or low-level motion dynamics (how). This incomplete skill transfer yields weak priors for policy learning, thereby demanding substantial in-domain data for downstream adaptation. To address these challenges, we introduce BooST, a two-stage framework that explicitly bridges semantics and motions to satisfy all three desiderata. BooST first leverages a cross-modal VQ-VAE to capture both semantic intent and motion dynamics, yielding a unified skill representation. It then distills this representation into a lightweight policy for efficient downstream adaptation to new tasks. Extensive experiments across simulation and real-robot settings demonstrate that BooST achieves superior few-shot adaptation, cross-domain skill transfer, and robustness to dynamic visual distractors, while maintaining a lightweight yet expressive design suitable for real-world deployment.

From early Movement Primitive (MP) techniques to modern Vision-Language Models (VLMs), autonomous manipulation has remained a pivotal topic in robotics. As two extremes, VLM-based methods emphasize zero-shot and adaptive manipulation but struggle with fine-grained planning. In contrast, MP-based approaches excel in precise trajectory generalization but lack decision-making ability. To leverage the strengths of the two frameworks, we propose VL-MP, which integrates VLM with Kernelized Movement Primitives (KMP) via a low-distortion decision information transfer bridge, enabling fine-grained robotic manipulation under ambiguous situations. One key of VL-MP is the accurate representation of task decision parameters through semantic keypoints constraints, leading to more precise task parameter generation. Additionally, we introduce a local trajectory feature-enhanced KMP to support VL-MP, thereby achieving shape preservation for complex trajectories. Extensive experiments conducted in complex real-world environments validate the effectiveness of VL-MP for adaptive and fine-grained manipulation.

The Will Smith 2004 blockbuster I, Robot predicts pervasive humanoid robots in 2035; investors agree, roboticists disagree.

In the automated sorting and processing of agricultural products, the nonlinear mechanical degradation induced by lipid oxidation severely restricts the grasping reliability of high‐fat fragile targets. To address the resulting damage and slip issues, this study proposes a damage‐less grasping strategy driven by an oxidation prior, integrating variable impedance control and quadratic programming‐based control barrier functions (QP‐CBF). By fusing multimodal optical and tactile information, the proposed strategy utilizes hyperspectral imaging to predict the degree of oxidative deterioration, mapping this prior into a dynamic safety threshold for the normal gripping force. Furthermore, fractional‐order calculus is introduced to reconstruct the viscoelastic mechanical response of the targets, effectively suppressing the normal force overshoot during the initial contact phase. To mitigate the spatial slippage risks exacerbated by strict underlying force constraints, the framework employs a QP‐CBF module to dynamically constrain hazardous spatial accelerations, thereby achieving active anti‐slip intervention. Experiments conducted on multi‐gradient oxidized samples demonstrate that under severe oxidation‐induced embrittlement, the proposed intervention mechanism effectively compensates for the friction deficit. Compared to conventional feedforward perception strategies that frequently fail due to insufficient normal gripping force, our approach improves the intact handling success rate by 37%, achieving an overall success rate exceeding 93%. These results indicate that the integrated control framework successfully resolves the physical contradiction between preventing breakage and avoiding slippage, providing a highly robust algorithmic reference for the compliant robotic manipulation of fragile objects.

Multi-modal trajectory generation is essential for safe autonomous driving, yet existing diffusion-based planners suffer from high inference latency due to iterative neural function evaluations. This paper presents MISTY (Mixer-based Inference for Single-step Trajectory-drifting Yield), a high-throughput generative motion planner that achieves state-of-the-art closed-loop performance with pure single-step inference. MISTY integrates a vectorized Sub-Graph encoder to capture environment context, a Variational Autoencoder to structure expert trajectories into a compact 32-dimensional latent manifold, and an ultra-lightweight MLP-Mixer decoder to eliminate quadratic attention complexity. Importantly, we introduce a latent-space drifting loss that shifts the complex distribution evolution entirely to the training phase. By formulating explicit attractive and repulsive forces, this mechanism empowers the model to synthesize novel, proactive maneuvers, such as active overtaking, that are virtually absent from the raw expert demonstrations. Extensive evaluations on the nuPlan benchmark demonstrate that MISTY achieves state-of-the-art results on the challenging Test14-hard split, with comprehensive scores of 80.32 and 82.21 in non-reactive and reactive settings, respectively. Operating at over 99 FPS with an end-to-end latency of 10.1 ms, MISTY offers an order-of-magnitude speedup over iterative diffusion planners while while achieving significantly robust generation.

Musculoskeletal robots require an internal body schema that remains consistent under a wide range of physical state changes, including abnormalities such as muscle rupture and actuator jamming. Conventional approaches based on autoencoders or variational autoencoders learn average behaviors by projecting sensor and actuator signals into a low-dimensional latent space; however, exploration within the latent space alone has limited capability to handle out-of-distribution or abnormal states that are not included in the training data. To address this limitation, this study proposes a diffusion-based framework for body schema learning in musculoskeletal robots. Unlike generative models that operate through low-dimensional latent spaces, diffusion models can directly and iteratively estimate physically consistent sensor and actuator values in the high-dimensional space through a denoising process, even under partial observations and constraints, without requiring retraining. By formulating body schema adaptation as a gradient-guided denoising process, the proposed method enables adaptive estimation of appropriate muscle lengths and muscle tensions even under abnormal conditions such as muscle rupture and actuator jamming. The validity of the proposed framework is verified through simulation experiments using a musculoskeletal robot model.

Kinodynamic planning for car-like vehicles on uneven terrain requires simultaneously optimizing competing objectives such as path efficiency and pose stability. This work presents an adaptive epsilon-constraint method integrated into a Model Predictive Control (MPC) framework, where the epsilon bounds are dynamically adjusted based on terrain descriptors to explore the Pareto front in real time. To capture vehicle-terrain dynamics, we develop a semi-parametric model combining analytical vehicle dynamics with a Sparse Gaussian Process (SGP) trained on the same terrain descriptors. The proposed epsilon-MPC is evaluated against MPPI and GAKD baselines, achieving a 94% navigation success rate while reducing maximum orientation deviation by 24% and improving multi-objective trade-off quality by 23%.

In structured multi-agent transportation systems, agents often must follow predefined routes, making spatial rerouting undesirable or impossible. This paper addresses route-constrained multi-agent coordination by optimizing waypoint passage times while preserving each agent's assigned waypoint order and nominal route assignment. A differentiable surrogate trajectory model maps waypoint timings to smooth position profiles and captures first-order tracking lag, enabling pairwise safety to be encoded through distance-based penalties evaluated on a dense temporal grid spanning the mission horizon. The resulting nonlinear and nonconvex velocity-scheduling problem is solved using an inexact-projection Alternating Direction Method of Multipliers (ADMM) algorithm that combines structured timing updates with gradient-based collision-correction steps and avoids explicit integer sequencing variables. Numerical experiments on random-crossing, bottleneck, and graph-based network scenarios show that the proposed method computes feasible and time-efficient schedules across a range of congestion levels and yields shorter mission completion times than a representative hierarchical baseline in the tested bottleneck cases.

Many fish species are capable of leaping out of water, and several robotic fish have been developed to mimic this behavior. However, enhancing jumping performance and achieving low‐altitude gliding—as seen in flying fish—remain challenging. This study presents a bionic flying‐fish robot capable of efficient high‐speed swimming and water‐exiting motion. Its morphology is reconstructed from a 3D scan of a flying fish, a representative cross‐medium organism. Unlike conventional jumping robotic fish, the proposed design features an optimized body shape and an improved tail‐drive mechanism, which reduce kinetic energy loss during high‐frequency oscillation. Experimental tests and analyses of the Strouhal number (St) and Reynolds number (Re) confirm efficient propulsion and swimming characteristics comparable to real fish, with a maximum speed of 2.54 m/s. The simplified pectoral‐fin structure further enables agile steering and pitching control. During high‐speed swimming, the Froude number (Fr) exceeds 1, demonstrating the robot's ability to fully exit the water. In the test, the robotic fish was able to leap out of the water to a height of 0.5–0.55 m at a sprint angle of to . These results highlight the robot's exceptional locomotion performance and provide insights for the design of cross‐medium robotic systems.

Electroaerodynamic propulsion is compelling for use in micro air vehicles due to its silent and solid-state nature, but its limited efficiency has thus far precluded a path towards power-autonomous flight. Recent work has shown that thrust density and efficiency for small-scale atmospheric ion thrusters can be vastly increased when operating close to a ground plane. Here, we explore the design space of centimeter-scale hovercraft, which can leverage this ground effect for low-altitude flight. We first perform an empirical investigation, characterizing the performance benefits and trade-offs for different geometries and configurations of passive hovercraft skirts, then use the results to fabricate a viable point design. We demonstrate a palm-sized hovercraft that, while tethered to an external power source, can fly for extended periods, withstand dozens of takeoff and landing cycles, passively stabilize to reject significant mechanical disturbances, and generate practically zero audible noise signature. The measured thrust efficiency of 16 mN/W and additional payload capacity of almost 1.5 grams above the vehicle's self mass of about 1.6 grams exceeds any similarly sized electroaerodynamically propelled robot by an order of magnitude. This is the first time an ion-propelled micro hovercraft has been shown in the open literature, and our work points the way towards an entirely new class of robot.

Autonomous 3D reconstruction with depth sensors is strongly affected by reflective surfaces, which cause missing or unreliable measurements and reduce the effectiveness of conventional Next-Best-View (NBV) strategies. This limitation is particularly critical in industrial applications involving reflective components and low-cost, low-resolution depth sensing, where robustness to sensing failures is essential. This paper proposes a Fast Reflectivity-Aware Next-Best-View (FRA-NBV) strategy that explicitly addresses reflection-induced depth loss without relying on prior object models or assumptions on material reflectance, making it suitable for a wide range of industrial configurations. Reflective regions are identified from the spatial distribution of missing depth measurements and localized in three-dimensional space using an online ellipsoid-based representation of the object estimate. A recovery strategy then selects additional poses that modify the sensor's angle of incidence to improve the likelihood of reconstructing the affected regions. Experiments on objects with different geometric and reflective complexity demonstrate that the approach significantly improves reconstruction coverage under realistic industrial conditions.

Clothes manipulation, such as folding or hanging, is a critical capability for home service robots. Despite recent advances, most existing methods remain limited to specific clothes types and tasks due to the complex, high-dimensional geometry of clothes. This paper presents CLothes mAnipulation with Semantic keyPoints (CLASP). CLASP aims at general-purpose clothes manipulation across diverse clothes types, such as T-shirts, shorts, skirts, long dresses, … and across different tasks, such as folding, flattening, hanging, …. The key idea of CLASP is semantic keypoints —for example, “left sleeve” and “right shoulder”—a sparse spatial-semantic representation, salient for both perception and action. Semantic keypoints of clothes can be reliably extracted from RGB-D images and provide an effective representation for a wide range of clothes manipulation policies. CLASP uses semantic keypoints as an intermediate representation to connect high-level task planning and low-level action execution. At the high level, it exploits a vision-language model (VLM) to predict task plans over the semantic keypoints. At the low level, it executes the plans with the help of a set of pre-built manipulation skills conditioned on the keypoints. Extensive simulation experiments show that CLASP outperforms state-of-the-art baseline methods on multiple tasks across diverse clothes types, demonstrating strong performance and generalization. Further experiments with a Franka dual-arm system on four distinct tasks—folding, flattening, hanging, and placing—confirm CLASP’s performance on real-life clothes manipulation.

Bimanual manipulation in cluttered scenes requires policies that remain stable under occlusions, viewpoint changes and scene variations. Existing vision-language-action models often lack such robustness because (i) multi-view features are fused via view-agnostic token concatenation, yielding limited cross-view spatial representations, and (ii) language is injected as global conditioning, resulting in coarse instruction grounding. In this paper, we introduce PEAfowl, a perception-enhanced multi-view VLA policy for bimanual manipulation. For spatial perception, PEAfowl predicts per-token depth distributions, performs differentiable 3D lifting, and aggregates local cross-view neighbors to form geometrically grounded, cross-view aligned representations. For language utilization, we propose to replace global conditioning with a Perceiver-style text-aware readout over frozen CLIP visual features, enabling iterative evidence accumulation. To better exploit commodity RGB-D sensing despite noisy and incomplete depth, PEAfowl's depth-distribution lifting naturally supports training-only depth distillation, where a pretrained depth teacher supervises the depth-distribution head to inject refined geometric priors without adding inference overhead. On RoboTwin 2.0 under domain-randomized setting, PEAfowl improves the strongest baseline by 23.0 pp in success rate, and physical experiments further demonstrate improved performance on the evaluated real-robot tasks. Project website: https://peafowlvla.github.io/.

Visual traversability estimation is central to autonomous navigation, yet most approaches either rely on prompt-driven Vision-Language Model (VLM) or decouple traversability from trajectory planning, requiring separate planners with heavy mapping, manual tuning, and extended deployment time. We propose EmbodiedDiffusion, a diffusion-based framework that simultaneously predicts traversability maps and generates feasible trajectories from RGB images using planner-free synthetic supervision and embodiment conditioning for cross-platform transfer. The framework distills category-level traversability semantics from a VLM teacher into a lightweight student model during training, enabling prompt-free, real-time inference at deployment. A modular FiLM-based conditioning mechanism isolates embodiment-specific reasoning into a compact trainable subset of the network, allowing rapid adaptation to new robot platforms without retraining the visual backbone or the trajectory diffusion model. Across indoor environments with quadruped and aerial robots, EmbodiedDiffusion achieves 80-100% navigation success in the full-data regime with real-time inference (90 ms) and adapts to new platforms using only 10 min of visual data collection, demonstrating scalable, unified traversability reasoning and trajectory generation for heterogeneous robots.

Multi-agent path finding (MAPF) is the problem of planning conflict-free paths from the designated start locations to goal positions for multiple agents. It underlies a variety of real-world tasks, including multi-robot coordination, robot-assisted logistics, and social navigation. Recent decentralized learnable solvers have shown great promise for large-scale MAPF, especially when leveraging foundation models and large datasets. However, these agents are reactive policy models and exhibit limited modeling of environmental temporal dynamics and inter-agent dependencies, resulting in performance degradation in complex, long-term planning scenarios. To address these limitations, we propose MAPF-World, an autoregressive action world model for MAPF that unifies situation understanding and action generation, guiding decisions beyond immediate local observations. It improves situational awareness by explicitly modeling environmental dynamics, including spatial features and temporal dependencies, through future state and actions prediction. By incorporating these predicted futures, MAPF-World enables more informed, coordinated, and far-sighted decision-making, especially in complex multi-agent settings. Furthermore, we augment MAPF benchmarks by introducing an automatic map generator grounded in real-world scenarios, capturing practical map layouts for training and evaluating MAPF solvers. Extensive experiments demonstrate that MAPF-World outperforms state-of-the-art learnable solvers, showcasing superior zero-shot generalization to out-of-distribution cases. Notably, MAPF-World is trained with a 96.5% smaller model size and 92% reduced data.

Designing large soft robots capable of generating high forces for physical human-robot interaction remains a significant challenge in soft robotics. Prior work in large soft robots has focused on proof-of-concept prototypes, and no systematic framework exists for determining the suitability of a design paradigm for a desired task. This manuscript introduces a method for optimizing the geometry of a soft robot limb, maximizing its blocking force subject to an anti-bucking constraint under its own gravitational loading. We demonstrate that an explicit solution exists to the proposed optimization problem under certain assumptions. Experiments with three geometries of a large, soft, pneumatically-actuated manipulator demonstrate that the method correctly predicts which designs meet constraints and which produces the largest end-effector forces. This method, with its closed-form solution, can allow designers to determine a-priori if an intended class of soft manipulators is an appropriate choice for physical interaction at large size scales.

Egocentric human videos provide a scalable source of manipulation demonstrations; however, deploying them on robots requires active viewpoint control to maintain task-critical visibility, which human viewpoint imitation often fails to provide due to human-specific priors. We propose EgoAVFlow, which learns manipulation and active vision from egocentric videos through a shared 3D flow representation that supports geometric visibility reasoning and transfers without robot demonstrations. EgoAVFlow uses diffusion models to predict robot actions, future 3D flow, and camera trajectories, and refines viewpoints at test time with reward-maximizing denoising under a visibility-aware reward computed from predicted motion and scene geometry. Real-world experiments under actively changing viewpoints show that EgoAVFlow consistently outperforms prior human-demo-based baselines, demonstrating effective visibility maintenance and robust manipulation without robot demonstrations.

Generalist robots should be able to understand and follow user instructions. Despite providing a powerful architecture for mapping open-vocabulary language instructions to robot actions, current vision-language-action (VLA) models struggle to follow fine-grained commands. One cause for this is a lack of semantic diversity and language grounding in existing robot datasets and, specifically, a lack of fine-grained task diversity for similar observations. To address this, we present a novel method to augment existing robot datasets by leveraging vision-language models to create counterfactual labels. By augmenting existing datasets with these labels, we increase the diversity and granularity of language grounding for robot datasets, ultimately improving the language-following capabilities of VLAs. We evaluate the resulting model's ability to follow language instructions, ranging from simple object-centric commands to complex referential tasks, by conducting vision-language navigation experiments in 3 different indoor and outdoor environments. Our experiments show that counterfactual relabeling (without additional data collection) significantly improves instruction-following in VLA policies, outperforming state-of-the-art methods and doubling the success rate compared to VLAs trained on unaugmented data. We also evaluate our method for manipulation VLAs and find a similar gain in performance on tasks with distractors.

The LiDAR 3D object detector that balances accuracy and speed is crucial for achieving real-time perception in autonomous driving. However, many existing LiDAR detection models rely on complex feature transformations, leading to poor real-time performance and high resource consumption, which limits their practical effectiveness. In this work, we propose a Faster LiDAR 3D object detection framework that Adaptively aligns Sparse voxels to enable efficient heterogeneous knowledge Distillation, called FASD. We aim to distill the Transformer's sequence modeling capability into Mamba models, significantly boosting accuracy through knowledge transfer. Specifically, we first design a cross-model knowledge distillation architecture to convey the global contextual understanding capabilities of the Transformer to Mamba. The Transformer-based teacher model employs a scale-adaptive attention mechanism to enhance multi-scale fusion. In contrast, the Mamba-based student model leverages feature alignment through spatial alignment adapters, supervised with latent-space features and span-head logit distributions, leading to improved performance and efficiency. We evaluated FASD on the Waymo and nuScenes datasets, achieving up to a 2x reduction in FLOPs and a 4x reduction in memory consumption, while improving baseline performance by 1-2 percentage points and maintaining high deployment efficiency.

The prosthesis grasping force feedback can be conveyed to users via sensory substitution (vibrotactile encoding) or modality-matched stimulation (physical force exertion). To date, the two approaches are rarely compared and, typically, they are designed to convey only the normal grasping force component. This letter developed a lab based closed-loop myoelectric prosthetic control system using a collaborative robotic arm to render high-fidelity 3D grasping force from the prosthetic end-effector on the user. Experiments with 11 subjects (10 able-bodied, 1 amputee) demonstrated that: (1) Tangential force feedback facilitated accurate perception of slippage tendencies under visual deprivation. (2) In stiffness identification, visual + force feedback significantly outperformed both visual only and visual + vibrotactile feedback in accuracy (87.1% vs. 43.1% and 80.2%) and decision time (9.8 s vs. 11.6 s and 10.7 s). (3) In force tracking, exclusive force feedback (rather than exclusive vibrotactile feedback) significantly minimized tracking errors compared to visual feedback, with visual + force feedback achieving the lowest RMSE (3.70 N) compared to the visual feedback (4.89 N) and no feedback (5.34 N). The results indicate that visual feedback alone is inadequate for force control, and the addition of a haptic channel significantly improves performance. Overall, the proposed 3D force feedback system offers broad applicability, implying that its socket-integrated miniaturization could facilitate dynamic grasping and manipulation.

The robustness of event cameras to high dynamic range and motion blur holds the potential to improve visual odometry systems in challenging environments. Although their high temporal resolution does not require synchronous processing, most event-based odometry methods still run at fixed rates, which simplifies system design but restricts latency and throughput. In this work, we present AERO-VIS, a stereo event-inertial SLAM system with an integrated, data-driven, robust, and performance-optimized keypoint detector. By processing the event stream asynchronously, the system dynamically adapts to downstream runtime demands, ensuring low-latency and real-time performance. When deploying AERO-VIS on a UAV, we achieve unprecedented accuracy in onboard event-based SLAM. These unique characteristics enable us to present the first purely event-based inertial SLAM system that demonstrates closed-loop UAV control and large-scale state estimation while relying solely on onboard compute. A video of the experiments and the source code are available at ethz-mrl.github.io/AERO-VIS.

Teleoperating multiple robots simultaneously enables additional views and coordinated control. Yet, it poses fundamental challenges: the system must present sensor data cohesively and allow operators to manage multiple robot bases, arms, and cameras while maintaining low latency. Current multi-robot teleoperation systems require multiple operators, rely on autonomy, or restrict operators to high-level commands. We present GHOST: an open-source VR teleoperation system that enables single operator control of two mobile manipulators via direct lowlevel commands using only onboard sensing. GHOST creates an exocentric 3D workspace by aligning real-time point clouds from the robots' RGB-D cameras, where scene coverage is improved through learning-based completion to aid operator spatial awareness. For control, the operator uses a mode-switching architecture to command either robot individually or both robots simultaneously. Experiments with 15 novice participants demonstrate 1.6-4x the success rate of an off-the-shelf tablet interface. For experts across nine challenging dual-robot tasks, our system enabled completion of two tasks that were infeasible with the tablet, and was 1.47x faster on average than the tablet. Website and code: https://h2r.github.io/GHOST/.

Various drawbacks exist in traditional visual SLAM methods for wind turbine blade polishing scenarios. Insufficient feature extraction, unstable matching, and reduced positioning accuracy are easily produced. These issues are triggered by weak textures, strong reflections, and illumination fluctuations. A point‐line fused visual‐inertial positioning method named EPL‐VINS is therefore proposed. FAST feature points and EDLines feature lines are adopted in the visual front end of this method. An adaptive threshold feature detection strategy and a texture complexity‐based feature distribution optimization mechanism are put forward, so that the capability of effective feature extraction is strengthened and the spatial distribution of features is optimized. At the back end, a tightly coupled optimization framework based on sliding window is constructed. Marginal prior information, IMU pre‐integration residuals, visual point‐line feature residuals, and loop closure constraints are all integrated into the optimized objective function to realize the joint estimation of visual and inertial information. Meanwhile, the LM optimization algorithm is improved by the dynamic update strategy of damping factors, and the convergence speed and numerical stability of back‐end nonlinear optimization are enhanced. On the public EuRoC data set, an average positioning accuracy improvement of 22.3% is achieved by EPL‐VINS compared with VINS‐Mono. Higher positioning accuracy and environmental adaptability are also exhibited by EPL‐VINS in actual factory scenarios for wind turbine blade polishing. High‐precision robot positioning is realized by the proposed method in complex industrial environments.

Path planning for orchard robots is one of the most fundamental technologies of smart agriculture. Introducing reinforcement learning (RL) into path planning has become a promising approach at the frontier of robotics research, addressing the problem that existing methods primarily focus on the optimal path at the immediate moment, lack a holistic perspective, and fail to achieve optimal overall performance. However, the sim‐to‐real issue presents a critical challenge for implementing RL in orchard robots. To solve this issue, an RL‐enhanced path‐planning method is proposed for orchard robots. The proposed method avoids directly using RL to generate a path, thereby eliminating the primary source of the sim‐to‐real problem in RL. In the proposed method, multiple candidate paths are computed using a Bézier curve and are physically navigable for robots, leveraging LiDAR to dynamically update the environmental map. Meanwhile, a comprehensive reward function is designed for RL to achieve optimal path planning. Subsequently, RL is employed to select the optimal path among multiple candidate paths, thereby avoiding the sim‐to‐real problem in complex orchard environments. Finally, validation tests are carried out in both simulation and field environments. The experimental results show that the proposed method effectively addresses the sim‐to‐real challenges of RL and achieves optimal path planning.

LiDAR Inertial Odometry (LIO) is a critical component for many mobile robots that need to navigate without relying on external positioning (e.g., GPS). Platforms that operate autonomously in different environments and with heterogeneous LiDAR sensors require a LIO approach that can adapt to these different scenarios without human intervention. Existing LIO approaches can typically provide reliable and accurate odometry in scenarios with similar environments and sensors when suitably tuned. However, many approaches struggle to retain robust odometry across heterogeneous environments and sensors while using a consistent configuration. This paper presents EllipseLIO, a real-time LIO approach that generalises between scenarios by using methods for LiDAR scan filtering and registration that adapt to the sensor capabilities and environment without requiring scenario-specific tuning. Experiments with EllipseLIO and state-of-the-art LIO approaches on five datasets with diverse and challenging scenarios demonstrate that EllipseLIO is the best performing approach overall. It achieves a 35% lower odometry error on average than the second-best approach and is the only approach that does not diverge in any experiment. An open-source version of EllipseLIO is available at https://github.com/v4rl-ucy/ellipselio.

Autonomous robots operating in natural karstic caves face perception and navigation challenges that are qualitatively distinct from those encountered in mines or tunnels: irregular geometry, reflective wet surfaces, near-zero ambient light, and complex branching passages. Yet publicly available datasets targeting this environment remain scarce and offer limited sensing modalities and environmental diversity. We present CAVERS, a multimodal dataset acquired in two structurally distinct rooms of Cueva de la Victoria, Málaga, Spain, containing 24 sequences with approximately 335 GB of recorded data. The sensor suite combines an Intel RealSense D435i RGB-D-I camera, an Optris PI640i near-IR thermal camera, and a Velodyne VLP-16 LiDAR, operated both handheld and mounted on a wheeled rover under full darkness and artificial illumination. For most of the sequences, mm-accurate 6-DoF ground truth pose and velocity at 120 Hz are provided by an Optitrack motion capture system installed directly inside the cave. We benchmark seven state-of-the-art SLAM and odometry algorithms spanning visual, visual-inertial, thermal-inertial, and LiDAR-based pipelines, as well as a 3D reconstruction pipeline, demonstrating the dataset's usability. The dataset and all supplementary material are publicly available at: https://github.com/spaceuma/cavers.

High-definition (HD) maps are evolving from pre-annotated to real-time construction to better support autonomous driving in diverse scenarios. However, this process is hindered by low-quality input data caused by onboard sensors limited capability and frequent occlusions, leading to incomplete, noisy, or missing data, and thus reduced mapping accuracy and robustness. Recent efforts have introduced satellite images as auxiliary input, offering a stable, wide-area view to complement the limited ego perspective. However, satellite images in Bird's Eye View are often degraded by shadows and occlusions from vegetation and buildings. Prior methods using basic feature extraction and fusion remain ineffective. To address these challenges, we propose SATMapTR, a novel online map construction model that effectively fuses satellite image through two key components: (1) a gated feature refinement module that adaptively filters satellite image features by integrating high-level semantics with low-level structural cues to extract high signal-to-noise ratio map-relevant representations; and (2) a geometry-aware fusion module that consistently fuse satellite and BEV features at a grid-to-grid level, minimizing interference from irrelevant regions and low-quality inputs. Experimental results on the nuScenes dataset show that SATMapTR achieves the highest mean average precision (mAP) of 73.8, outperforming state-of-the-art satellite-enhanced models by up to 14.2 mAP. It also shows lower mAP degradation under adverse weather and sensor failures, and achieves nearly 3 times higher mAP at extended perception ranges.

Robot-assisted minimally invasive surgery improves surgical precision but introduces complexity, making technical error detection essential for ensuring patient safety. Current executional error detection methods using video data often overlook fine-grained contextual descriptions of activities and error types within the hierarchical structure of surgical procedures. They also under-utilize complementary multimodal information. We propose a unified framework for executional error detection that leverages multimodal input, including video, kinematics, and descriptive textual prompts. Through activity prompting, we integrate descriptive language in gesture-level activities, instrument-object interactions, and error definitions. We also introduce activity-aware visual embeddings derived from vision encoders pretrained on surgical activity labels to compare the effectiveness of contrastive language-image embeddings with traditional image-based embeddings for error detection. By seamlessly integrating kinematic data with video and textual modalities, our framework significantly improves error detection performance. Achieving up to 5\% and 16.6\% F1 score improvements over state-of-the-art baselines on the JIGSAWS and SAR-RARP50 datasets, respectively, we demonstrate the value of combining curated textual prompts with multimodal data for accurate error detection.

Natural language provides an intuitive way to express spatial intent in geospatial applications. While existing localization methods often rely on dense point cloud maps or high-resolution imagery, OpenStreetMap (OSM) offers a compact and freely available map representation that encodes rich semantic and structural information, making it well-suited for large-scale localization. However, text-to-OSM (T2O) localization remains largely unexplored. In this paper, we formulate the T2O localization task, which aims to estimate accurate 2D positions in urban environments from textual scene descriptions without relying on geometric observations or GNSS-based initial location. To support the proposed task, we introduce TOL, a large-scale benchmark spanning multiple continents and diverse urban environments. TOL contains approximately 121K textual queries paired with OSM map tiles and covers about 316 km of road trajectories across Boston, Karlsruhe, and Singapore. We further propose TOLoc, a coarse-to-fine localization framework that explicitly models the semantics of surrounding objects and their directional information. In the coarse stage, direction-aware features are extracted from both textual descriptions and OSM tiles to construct global descriptors, which are used to retrieve candidate locations for the query. In the fine stage, the query text and top-1 retrieved tile are jointly processed, where a dedicated alignment module fuses the textual descriptor and local map features to regress the 2-DoF pose. Experimental results demonstrate that TOLoc achieves strong localization performance, outperforming the best existing method by 6.53\%, 9.93\%, and 8.32\% at 5 m, 10 m, and 25 m thresholds, respectively, and shows strong generalization to unseen environments. Dataset, code and models will be publicly available at: https://github.com/WHU-USI3DV/TOL.

Large language models (LLMs) have shown great promise for autonomous driving. However, discretizing numbers into tokens limits precise numerical reasoning, fails to reflect the positional significance of digits in the training objective, and makes it difficult to achieve both decoding efficiency and numerical precision. These limitations affect both the processing of sensor measurements and the generation of precise control commands, creating a fundamental barrier for deploying LLM-based autonomous driving systems. In this paper, we introduce DriveCode, a novel numerical encoding method that represents numbers as dedicated embeddings rather than discrete text tokens. DriveCode employs a number projector to map numbers into the language model's hidden space, enabling seamless integration with visual and textual features in a unified multimodal sequence. Evaluated on OmniDrive, DriveGPT4, and DriveGPT4-V2 datasets, DriveCode demonstrates superior performance in trajectory prediction and control signal generation, confirming its effectiveness for LLM-based autonomous driving systems.

Off-road nighttime autonomous driving suffers from unreliable visible-light perception, making infrared modality crucial for accurate freespace detection. However, progress remains limited due to the scarcity of annotated infrared off-road datasets and the inter-frame inconsistencies inherent to current single-frame methods. To address these gaps, we present the IRON dataset, which, to our knowledge, is the first large-scale infrared dataset for off-road temporal freespace detection under all-day conditions, with strong support for nighttime perception. The dataset comprises 24,314 densely annotated infrared images with synchronized RGB images in diverse scenes and different light conditions. Building upon this dataset, we propose IRONet, a novel flow-free framework for temporal freespace detection that addresses inter-frame inconsistencies by aggregating historical context via a memory-attention mechanism and a carefully designed mask decoder. On our IRON dataset, IRONet achieves state-of-the-art performance, reaching 82.93%(+1.19%) IoU and 90.66%(+0.71%) F1 score at real-time inference. Remarkably, IRONet also exhibits robust generalization to RGB modalities on ORFD and Rellis datasets. Overall, our work establishes a foundation for reliable all-day off-road autonomous driving and future research in infrared temporal perception. The code and IRON dataset are available at https://github.com/wsnbws/IRON.

Real-time 6-DoF object pose tracking is essential for many robotics applications, and several approaches exist. Yet even today's approaches remain unreliable under temporary full occlusions and rapid object motions. Once tracking is lost, most methods struggle to detect the failure and recover automatically, often requiring manual re-initialization. In this paper, we address the problem of robust model-based 6-DoF tracking of unseen objects from RGB-D data, especially in scenarios with occlusion and fast motion. We propose a novel method that combines efficient learning-based keypoint matching with optimization-based alignment and introduces a novel failure detection and recovery module. Our system monitors pose reliability, detects tracking divergence or occlusions, and performs a global re-detection and pose estimation step that robustly verifies recovery candidates before resuming tracking. Our evaluation on standard tracking benchmarks and on a new dataset of occluded and fast-moving scenes shows that our method matches state-of-the-art accuracy on easy tracking sequences, maintains high tracking speed at 57.6 frames per second, and provides the most robust tracking performance under challenging conditions. Thus, we believe that our approach is a relevant step forward in robust 6-DoF object tracking from RGB-D data.

Human-to-robot (H2R) object handover is a fundamental capability for human-robot collaboration, yet progress is hindered by the scarcity of large-scale, human-centric datasets and the significant sim-to-real gap. To address these challenges, we introduce Hand2Bot, an RGB-D video dataset that provides rich contextual information such as body posture and facial expressions, specifically collected for handover scenarios with real-world noise patterns. We further propose PassGen, a generative pipeline that leverages stable video diffusion and an Intention-Aware Temporal Face Encoder to synthesize realistic handover sequences while ensuring hand-object consistency. To bridge the sim-to-real gap, we implement a morphology-based depth editing strategy that replicates realistic sensor noise found in physical depth maps. Experimental evaluations demonstrate that our framework achieves high intention identification accuracy and low false trigger rates in both ablation studies and real-world deployment on a physical robot platform. Our results confirm that training on PassGen allows for robust zero-shot transfer and earlier intention anticipation compared to traditional hand-centric baselines, effectively enabling socially aware robotic behavior in shared workspaces.

Robot Task and Motion Planning (TAMP) algorithms enable autonomous operation by incorporating the specific functions and constraints of end-effector tools, such as grippers or soldering irons, directly into the planning process. In this paper, we explore sampling-based TAMP algorithms specifically designed for a critical subset of devices whose unique properties make traditional planning methods ineffective. Visibility-based instruments, such as exteroceptive sensors, cameras, flashlights and directional antennas, are essential across a vast array of human activities. The unique properties of these devices, and particularly, their field-of-view, render many widely used heuristics and distance metrics less effective. We introduce two new sampling-based algorithms, FOV-PRM and FOV-RRT, designed to tackle visibility-based tasks. FOV-PRM employs a hierarchical decomposition of the environment, leveraging the concept of visibility integrity, to efficiently sample configurations with a clear line-of-sight to the target. A specialized Inverse Kinematics solver enables FOV-RRT to "glance" in the direction of the target at opportune moments, facilitating the rapid discovery of key configurations. We show that FOV-PRM and FOV-RRT achieve a higher success rate and faster runtimes compared to adaptations of RRT, PRM and VIR, through both simulated and physical experiments.

Embodied artificial intelligence (Embodied AI) aims to develop agents that can perceive, reason, and act through continuous interaction with physical world. To achieve this goal, agents must not only understand visual scenes and language instructions, but also transform multimodal information into executable actions. This requirement calls for models that can integrate perception, semantic reasoning, and action generation within a unified decision‐making framework. Recent advances in large language models (LLMs) have provided important technical foundations for this integration, leading to emergence of vision‐language‐action (VLA) models as a central paradigm for embodied systems. Although existing surveys have reviewed Embodied AI from perspectives such as robotic systems, simulation environments and task settings, few have systematically analyzed action‐generation mechanisms of VLA models. This limitation makes it difficult to distinguish VLA architectures. To address this gap, we provide a comprehensive review of VLA models for Embodied AI from an action‐generation perspective. We first formulate VLA models as embodied decision‐making systems that map visual observations and language instructions to action sequences through interaction with dynamic environments. Building on this formulation, we review historical evolution of VLA models. Furthermore, we propose an action‐generation‐centered taxonomy that categorizes VLA models into three paradigms: direct policy learning, generative action modeling, and reasoning‐guided modeling. For each paradigm, we analyze representative methods, action‐generation mechanisms, advantages and limitations, thereby providing a coherent framework for comparing diverse VLA architectures. Beyond model architectures, we summarize key resources and benchmarking settings and discuss recent applications. Finally, we identify major challenges and future directions. By organizing VLA models around the central problem of action generation, we aim to provide a structured technical map for future research toward more general, reliable, and physically grounded Embodied AI.

Multi-robot systems must simultaneously optimize competing objectives while maintaining coordinated behavior. Existing multi-agent reinforcement learning approaches often rely on fixed or centralized coordination, which limits adaptability and violates distributed constraints. This work introduces the Coordination-Informed Multi-Objective Reinforcement Learning (CIMORL) framework, integrating a distributed weight prediction mechanism, a privileged expert training strategy, and theoretical guarantees for Pareto-optimal solutions. We present the base CIMORL method alongside two sampling-based variants, CIMORL-TS (Tree Search) and CIMORL-MPPI (MPPI), which leverage privileged global information during training to enable fully decentralized deployment. Experimental validation in cooperative and adversarial scenarios demonstrates a $21.2\%$ hypervolume improvement and superior policy stability compared to state-of-the-art baselines. Real-world experiments with Crazyflie drones further validate the framework's robustness in resource allocation and multi-attacker multi-defend scenarios under partial observability.

With the goal of effective chlorophyll mapping in space and time using autonomous underwater vehicles, we study algorithms for hotspot sampling. We develop, implement and test a sampling algorithm using the expected improvement in chlorophyll to evaluate potential sampling designs. Working in a depth and lateral mode, the vehicle intelligently chooses the survey path for each sampling stage. The onboard system for design evaluation and data assimilation builds on a Gaussian random field representation of the log chlorophyll variations in north, east, depth, and time coordinates. Using a grid-free approach with spatio-temporal thinning of data, the random field model is quickly updated with in situ measurements made by the autonomous vehicle. This approach further enables real-time calculations for the expected improvement criterion used for sampling decisions. Via simulation studies, we compare the suggested sampling approach with others. Results of trials in the Mausund area, Norway, are used to demonstrate field deployments of our implementation for adaptive sampling. Here, we also discuss and compare the chlorophyll samples with that of zooplankton data gathered at the same locations.

Path planning for construction robots in complex construction environments suffers from insufficient adaptability, poor path quality, and high computational cost. This paper proposes a hierarchical path planning method based on improved‐bidirectional‐rapidly‐exploring random tree‐artificial potential field. The proposed method integrates the following key techniques: (1) a semantically enhanced unified bounding volume hierarchy collision detection mechanism for identifying multiple categories of construction obstacle; (2) goal‐biased sampling and dynamic step size to enhance global search efficiency, combined with a modified potential field to improve obstacle avoidance; turning‐angle constraints and B‐spline smoothing are applied to optimize trajectory feasibility; (3) a weakly coupled mechanism model of the mobile vehicle and the robotic arm to reduce computational complexity. Digital twin and large language model are introduced to further enhance the engineering practicality. Ablation experiments are conducted in multiple simulation scenarios based on a scaled cable‐truss structure, demonstrating the superiority of the method in effectiveness and efficiency.

Autonomous aerial tracking with drones offers vast potential for surveillance, cinematography, and industrial inspection applications. While single-drone tracking systems have been extensively studied, swarm-based target tracking remains underexplored, despite its unique advantages of distributed perception, fault-tolerant redundancy, and multidirectional target coverage. To bridge this gap, we propose a novel decentralized LiDAR-based swarm tracking framework that enables visibility-aware, cooperative target tracking in complex environments, while fully harnessing the unique capabilities of swarm systems. To address visibility, we introduce a novel Spherical Signed Distance Field (SSDF)-based metric for 3-D environmental occlusion representation, coupled with an efficient algorithm that enables real-time onboard SSDF updating. A general Field-of-View (FOV) alignment cost supporting heterogeneous LiDAR configurations is proposed for consistent target observation. Swarm coordination is enhanced through cooperative costs that enforce inter-robot safe clearance, prevent mutual occlusions, and notably facilitate 3-D multidirectional target encirclement via a novel electrostatic-potential-inspired distribution metric. These innovations are integrated into a hierarchical planner, combining a kinodynamic front-end searcher with a spatiotemporal $SE(3)$ back-end optimizer to generate collision-free, visibility-optimized trajectories.Deployed on heterogeneous LiDAR swarms, our fully decentralized implementation features collaborative perception, distributed planning, and dynamic swarm reconfigurability. Validated through rigorous real-world experiments in cluttered outdoor environments, the proposed system demonstrates robust cooperative tracking of agile targets (drones, humans) while achieving superior visibility maintenance.

Traditional Task and Motion Planning (TAMP) systems depend on physics models for motion planning and discrete symbolic models for task planning. Although physics model are often available, symbolic models (consisting of symbolic state interpretation and action models) must be meticulously handcrafted or learned from labeled data. This process is both resource-intensive and constrains the solution to the specific domain, limiting scalability and adaptability. On the other hand, Visual Language Models (VLMs) show desirable zero-shot visual understanding (due to their extensive training on heterogeneous data), but still achieve limited planning capabilities. Therefore, integrating VLMs with classical planning for long-horizon reasoning in TAMP problems offers high potential. Recent works in this direction still lack generality and depend on handcrafted, task-specific solutions, e.g. describing all possible objects in advance, or using symbolic action models. We propose a framework that generalizes well to unseen problem instances. The method requires only lifted predicates describing relations among objects and uses VLMs to ground them from images to obtain the symbolic state. Planning is performed with domain-independent heuristic search using goal-count and width-based heuristics, without need for action models. Symbolic search over VLM-grounded state-space outperforms direct VLM-based planning and performs on par with approaches that use a VLM-derived heuristic. This shows that domain-independent search can effectively solve problems across domains with large combinatorial state spaces. We extensively evaluate on extensively evaluate our method and achieve state-of-the-art results on the ProDG and ViPlan benchmarks.

Long-horizon navigation in unstructured environments demands terrain abstractions that scale to tens of square kilometers while preserving semantic and geometric structure. Grids scale poorly, and quadtrees misalign with terrain boundaries. Neither encodes terrain semantics essential for traversability-aware planning, yielding infeasible or inefficient paths for autonomous ground vehicles operating over more than 10 square kilometers. CLEAR (Connected Landcover Elevation Abstract Representation) is a reusable terrain abstraction framework for large-scale planning that produces convex, semantically aligned regions encoded as a terrain-aware graph. Evaluated on digital terrain maps spanning 9 to 100 square kilometers with physics-based simulation, CLEAR reduces per-query planning by 2x to 19.6x over unabstracted raw-grid A-star after one-time abstraction, with 6.7 percent cost overhead. In physics-based simulation, CLEAR delivers 5 to 8.6 percent shorter executed paths and 100 percent task completion across all maps, compared with 90 to 100 percent for Quadtree. These results hold against AMRA-star, a strong anytime multi-resolution planner, and generalize to a learned elevation-driven cost, demonstrating CLEAR's scalability and cost-model independence as a reusable planning layer.

Collisions of rigid-link robots and rigid environments are often modeled as instantaneous events. Under this idealization, the impact forces become impulsive and the system velocities nonsmooth. In this work, we systematically analyze pre- and post-impact velocities focusing on what we refer to as the nonsmooth impact direction (NSID). We show that it is a characteristic direction of a robotic impact and largely independent of contact properties. The results are directly applicable to large classes of backdrivable robotic systems with rigid links. We address particularities of systems with nonelastic and flexible joints, unconstrained as well as constrained systems. Further, we show that the approach direction w.r.t the NSID sets the direction of the impulsive force in frictional, inelastic impacts. The comprehensive theoretical analysis of this work supported by an experimental validation may serve as a foundation for future planning and control algorithms for various robotic impact applications. These can include humanoid locomotion on a slippery surface or repetitive hammering.

We are entering an era where AI enables multipurpose applications, but the mechanical hardware must be done right.

This study develops a binocular stereo vision system to enhance the operational safety of agricultural unmanned aerial vehicles (UAVs) by detecting and localizing thin cables—one of the most prevalent and hazardous obstacles in farmland environments. The system integrates cost‐effective hardware (dual OV4689 CMOS sensors, 36°/8 mm lenses, 120 mm baseline) with a process‐enhanced SemiGlobal Block Matching algorithm, specifically tailored to address the core challenge of low‐contrast cable detection in complex agricultural settings. The algorithmic innovation combines Hue, Saturation, and Lightness color space segmentation with adaptive dynamic threshold processing, effectively resolving cable feature extraction under variable illumination conditions and intense background interference from crops or terrain. Experimental results demonstrate maximum detection distances of 14.547 and 16.846 m for 20 and 40 mm diameter cables, respectively, with an average ranging error of 0.055 m (0.58% error rate). UAV Flight tests confirm reliable obstacle avoidance within a 2–5 m/s velocity range, achieving an average braking distance of 11.828 m—sufficient to ensure safe maneuvering given agricultural UAVs' inherent dynamic constraints of high moment of inertia and low thrust‐to‐weight ratio. Compared with conventional approaches, the proposed system exhibits superior chromatic separation between cables and backgrounds and enhanced dynamic noise suppression, while maintaining low hardware cost. This renders it a highly cost‐efficient and deployable safety solution for agricultural scenarios with dense cable distributions. The research outcomes provide robust technical support for autonomous obstacle avoidance of agricultural UAVs and offer significant practical value for improving operational safety in precision agriculture.

In robot operation, ensuring the robot can safely contact the target requires accurate contact force information. At present, in most scenarios, robots can accurately measure contact force using force sensors. However, in special environments such as healthcare and nuclear industries, due to the limited installation space for sensors and the harsh working conditions, the use of force sensors faces numerous challenges and is difficult to deploy widely or operate stably. Therefore, this paper proposes a robot end force estimation method without force sensors. This method utilizes the state information of each joint of the robot and establishes the correlation between the contact force and the state information through the Neighbor‐Aware RNN Network we proposed to accurately estimate the contact force at the robot end. This neural network can handle the dynamic temporal characteristics of the information of each joint of the robot, as well as the spatial correlation information of each joint. It learns the relationship between the state information of each joint of the robot and the contact force through an end‐to‐end approach, and ultimately outputs the three‐axis force estimation results. The experimental results show that this method performs exceptionally well in the task of robot contact force estimation, and its various evaluation indicators are significantly superior to those of other force estimation methods.

Maintaining traffic restrictions has become more difficult because growing number of vehicles on the road, particularly in locations with high populations. Efficient traffic monitoring and road safety management have become critical challenges due to the rapid increase in vehicular density in urban environments. This paper proposes a novel IoT‐aided robotic framework, termed MAGCNN‐RTMS‐IoT‐AR, for real‐time traffic sign detection and recognition using the German Traffic Sign Recognition Benchmark (GTSRB) dataset. Initially, the input images are pre‐processed using an Innovation Saturated Koopman Kalman Filter (ISKKF) to enhance data quality and robustness against noise. Subsequently, a Multi‐component Attention Graph Convolutional Neural Network (MAGCNN) is employed to effectively capture spatial and contextual dependencies for accurate classification of traffic signs. To further improve model performance, a Gradient Jellyfish Search Algorithm (GJSA) is integrated for optimal parameter tuning. The proposed approach is evaluated using multiple performance metrics, including accuracy, precision, recall, F 1‐score, error rate, loss, and mean square error (MSE). Experimental results demonstrate that the proposed method; significantly outperforms existing models such as RTMS‐IoT‐LeNet‐5, EATS‐CR‐CNN, and ETSR‐DS‐VGG‐16, achieving notable improvements in 19.36%, 24.82%, and 31.36%, higher accuracy; 21.62%, 24.54%, and 30.54% higher precision. The integration of IoT, robotics, and optimized deep learning makes the proposed system highly suitable for real‐time intelligent transportation applications.

Rice cultivation is a cornerstone of global food security, yet effective weed management remains a critical challenge. Autonomous weeding robots offer a sustainable alternative, but their efficacy hinges on precise navigation, particularly in the complex visual environments of paddy fields. This study introduces a novel instance segmentation‐based navigation line extraction method tailored for paddy‐field robots. Unlike conventional machine vision techniques, which struggle with noise, water reflections, and irregular crop rows, our approach leverages a lightweight instance segmentation model to accurately identify and differentiate rice seedling rows. The model, optimized for deployment on resource‐constrained edge devices like the Jetson Nano, achieves high precision with minimal angular and pixel deviations (0.54° and 6.40 pixels, respectively). It also demonstrates real‐time inference speeds of 16.13 FPS, making it suitable for practical applications. Furthermore, our method eliminates the need for high‐precision data annotation, reducing labeling complexity and enhancing generalizability. Comparative experiments with existing methods, such as YOLOv8‐seg, ESNet, VGG‐UNet, Mobile‐UNet, and Mobile‐DeeplabV3+, demonstrate that our method achieves superior performance in both accuracy and computational efficiency. The proposed navigation line extraction method not only meets the demands of agricultural robots in paddy fields but also lays a foundation for future advancements in automated field navigation, potentially integrating GNSS/INS systems for fully autonomous operations. This research represents a significant step toward sustainable and efficient agricultural practices, addressing both environmental and economic challenges in rice cultivation.

In this study, a bionic fin drive system suitable for deep‐sea environments is proposed. The system heats the phase change material (PCM) through the heat‐conducting medium and converts the expansion generated during the phase change process into hydraulic energy. Finally, the tail fin swing and pectoral fin expansion are realized through a novel motion conversion device, which offers the advantages of simple structure and low cost. In addition, the effects of the heat transfer medium and the thickness of the fish body outside the phase change actuator (PCA) on the output performance of the fin PCA were analyzed. The flexible adjustment of the tail fin's swinging characteristics can be realized by adjusting the voltage and duty cycle applied to the PCA heating sheet. A bionic fish prototype equipped with a phase‐change drive system has been successfully fabricated. The experimental results indicate that the robot can swim straight at a speed of 36 mm/s and complete a quick turn with a radius of 3.2 BL. Through the simulation of deep‐sea environment experiments, the robot can still achieve flexible underwater movement under water pressure of up to 110 MPa without bulky pressure‐resistant devices. These excellent performances demonstrate the reliable movement ability of the bionic robotic fish in the extreme deep‐sea environment.

Acoustic fields are emerging as a promising wireless and tissue‐penetrant power source for micro‐ and nanorobots, with propulsion efficiency, steering precision, and biocompatibility all critically linked to fabrication strategy. To clarify these interdependencies, we conducted a Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA)‐guided systematic review of literature from Scopus, Web of Science, PubMed, and IEEE Xplore (January 1, 2010–May 31 2025). Of 144 records, 63 primary studies and 22 secondary/background sources met predefined criteria and were synthesized. The evidence reveals five dominant actuation modes: boundary‐driven streaming (~40%), bubble oscillation (~25%), acoustic radiation forces (~12%), whole‐body resonance (~13%), and hybrid magneto‐/chemo‐acoustic mechanisms (~9%), each shaped by distinct fabrication approaches. Photolithography produces sharp fins for high‐Re streaming; soft lithography forms gas‐tight cavities for cavitation; two‐photon and micro‐DLP printing achieve nanometer‐scale smoothness for radiation‐pressure traps; and emerging multimaterial printers combine rigid acoustic chambers with magnetic or catalytic components. Three themes recur: (i) geometry–material–control co‐design consistently outperforms sequential prototyping; (ii) scalability and biocompatibility remain major barriers, with ad hoc quality metrics; and (iii) inverse‐design algorithms and closed‐loop acoustic imaging are accelerating optimization. This review maps fabrication techniques to actuation demands and application contexts, offering a consolidated roadmap toward clinically and industrially viable acoustic microrobots.

This paper presents an active disturbance rejection dynamic positioning (DP) control scheme of overactuated autonomous underwater vehicles (AUVs) for inspection missions. The proposed control scheme consists of dynamic control and control allocation. First, the DP process of the slender‐shaped AUV is decomposed into three sequential stages: yaw regulation, path following and position keeping. The tracking error calculation and control allocation are distinctively designed based on the motion performance of the AUV in each stage. Second, the adaptive line‐of‐sight (ALOS) guidance and extended state observer (ESO) are constructed to estimate and compensate for external disturbances to meet the high‐accuracy positioning requirements. A stability analysis is provided to establish the local UUB property of the closed‐loop system. Third, the thruster dead‐zone effect is compensated by shifting the control commands, and smooth control allocation between rudder and thrusters is achieved using a smooth transition function. Finally, the effectiveness and robustness of the proposed DP control scheme are verified through comparative lake trials. Compared to the conventional PID controller, the proposed method achieves statistically significant tracking error reductions () and substantially reduces equivalent control effort during steady‐state station‐keeping by suppressing actuator chattering.

The architecture of apple orchards can be converted to easier operation by implementing the use of a low‐cost, tractor‐mounted, dual manipulator based on a robotic harvesting system with lower degrees of freedom (DoF). To accurately avoid injuries by detecting apples, an efferent, real‐time vision system that includes a new gripping mechanism was developed. In this context, by introducing a new gripping system, spindle wall‐type apple orchard architectures have high potential for easy harvesting using a robotic platform. Therefore, this study was conducted to design and develop a tractor‐mounted, machine vision‐based, 3‐DoF, dual telescopic robotic cartesian system with a novel 2‐DoF gripping mechanism for apple harvesting. The machine vision system was developed using the EfficientDet object detection deep neural network with a mAP@0.5 of 0.775. The tracking algorithm DeepSORT was selected to track the localized apples and transfer the G ‐code to the robotic arm system. The developed novel gripping mechanism consists of three vacuum‐powered, (400 mmHg) spring‐loaded, silicone fingers that provide 0.73 N.m (±0.1) twisting torque to apples in a natural orientation (90°) and a dragging force of 17.65 N (±0.5). The average harvesting cycle time was approximately eleven seconds for both robotic arms, including detection and completion of the harvesting cycle.

Humanoid loco-manipulation requires coordinated task-space motion planning with stable loco-manipulation command tracking under complex robot-environment dynamics and long-horizon tasks. While diffusion policies (DPs) show promise for learning from demonstrations, deploying them on humanoids poses critical challenges: the motion planner trained offline is decoupled from the loco-manipulation controller, leading to poor command tracking, compounding distribution shift, and task failures. The common approach of scaling demonstration data is prohibitively expensive for high-dimensional humanoid systems. To address this challenge, we present REFINE-DP (REinforcement learning FINE-tuning of Diffusion Policy), a hierarchical framework that jointly optimizes a DP motion planner and an RL-based loco-manipulation controller. The DP is fine-tuned via a PPO-based diffusion policy gradient to improve task success rate, while the controller is simultaneously updated to accurately track the planner's evolving command distribution, reducing the distributional mismatch that degrades motion quality. We validate REFINE-DP on a humanoid robot performing loco-manipulation tasks, including door traversal and long-horizon object transport. REFINE-DP achieves an over 90% success rate in simulation, even in out-of-distribution cases not seen in the pre-training data, and enables real-world execution without privileged state information. Our proposed method substantially outperforms pre-trained DP baselines and demonstrates that RL fine-tuning is key to reliable humanoid loco-manipulation. https://refine-dp.github.io/REFINE-DP/

LiDAR place recognition supports loop closure, relocalization, and multi-agent map management. As robotic platforms increasingly combine LiDARs with different fields of view, resolutions, and scanning patterns, existing descriptors degrade because they are tightly coupled to sensor-specific characteristics. We present InLiER, a learning-free pipeline based on an intermediate tokenization step. Height-sliced keypoints from structural elements receive mixed-radix token IDs encoding height, radial distance, local shape, and azimuth from local 3D geometry, in a compact sub-2KB representation. The same vocabulary is reorganized across three retrieval stages: height-ceiling histogram intersection for fast rotation-invariant shortlisting, binary bitmask alignment for yaw estimation and reranking, and token-guided geometric verification for 6-DoF pose estimation. InLiER achieves state-of-the-art performance on the HeLiPR dataset and in real-world field experiments, among modern handcrafted methods and outperforms the learning-based baseline on most cross-sensor configurations.

Navigation within complex pipelines in sectors such as construction, mining, and underwater exploration presents significant demands on robotic adaptability and maneuverability. While traditional soft robots achieve such navigation through large material deformations, they face an inherent conflict between deformation capacity and structural rigidity, which hinders the simultaneous attainment of high traversability and strong load‐bearing capacity. To address this limitation, we propose a novel structural framework to overcome existing constraints of pipeline robots through integrating the characteristics of auxetic metamaterials. Firstly, the proposed robot is monolithically fabricated via 3D printing techniques. The mechanical properties of the printing material are characterized through quasi‐static tensile tests. Subsequently, the deformation behavior of the robotic body is analyzed and compared between numerical and experimental methods. Ultimately, the robot's navigation through complex pipelines is simulated and compared with physical pipe‐passing tests. The results demonstrate that during navigation in complex pipelines, the auxetic structure of the robot deforms radially, which releases circumferential pressures. This pressure release thereby reduces motion resistance, enhancing the robot's overall maneuverability. The findings of this work transcend the performance limitations of rigid materials, thereby offering a viable pathway for robotic navigation within the complex piping networks.

Recent developments in planetary exploration have shown the potential of Unmanned Aerial Vehicles (UAVs), such as the Ingenuity helicopter that provided valuable mapping data. However, limited payload capabilities constrain the flight times and compute available for localization, which restrict their applicability. By providing a tethered connection, issues such as battery and computational constraints are offloaded to the base rover. At the same time, the cable can be exploited for non-drifting localization. This work presents a novel Tether-Inertial Localization approach that uses tether length and angle measurements to estimate the UAV position relative to its base. The method combines a computationally efficient analytical catenary model with a Gaussian Process (GP) residual error compensation. This accounts for systematic sensor inaccuracies and model limitations. Experimental validation across circular, triangular, and figure-eight trajectories with tether lengths up to 4.5 m and a total flight time of 37 minutes demonstrates the effectiveness of the proposed approach. Using only tether-based position estimates for feedback, the analytical catenary model achieves an average RMSE of 7.4 cm, which is further reduced to 5.2 cm through GP-based residual compensation, one order of magnitude better than the state-of-the-art. These results establish Tether-Inertial Localization as a practical alternative to vision- and GNSS-based localization for Tethered Unmanned Aerial Vehicles (TUAVs).

Long-horizon, high-dynamic motion tracking on humanoids remains brittle: retargeted reference motions are typically kinematically plausible but dynamically inconsistent with the robot, so small tracking errors accumulate and eventually destabilize control. We present RobotDancing, a practical single-stage reinforcement learning recipe that tracks retargeted motions by predicting reference-conditioned residual joint targets. By parameterizing actions as residual corrections on top of the reference, the policy can focus on compensating reference--robot dynamics mismatch (e.g., actuation limits, latency, friction, inertia) rather than re-synthesizing the motion. We train one policy per reference sequence while reusing a common training and deployment recipe across motions and platforms. We evaluate RobotDancing on eight LAFAN1 dance motions on Unitree G1 and conduct cross-platform experiments on H1 and H1-2. The resulting G1 policies execute long-horizon, high-energy behaviors on hardware without test-time modifications, and qualitative H1/H1-2 clips further illustrate cross-platform feasibility.

Stable traversal over geometrically complex terrain increasingly requires exteroceptive perception, yet prior perceptive humanoid locomotion methods often remain tied to explicit geometric abstractions, either by mediating control through robot-centric 2.5D terrain representations or by shaping depth learning with auxiliary geometry-related targets. While effective, these approaches introduce additional map-construction procedures or multi-stage skill-transfer processes beyond direct depth-to-control learning. We propose CReF (Cross-modal and Recurrent Fusion), a single-stage depth-conditioned humanoid locomotion framework that learns locomotion-relevant features directly from raw forward-facing depth without explicit geometric intermediates. CReF couples proprioception and depth tokens through proprioception-queried cross-modal attention, fuses the resulting representation with a gated residual fusion block, and performs temporal integration with a Gated Recurrent Unit (GRU) regulated by a highway-style output gate for state-dependent blending of recurrent and feedforward features. To further improve terrain interaction, we introduce a terrain-aware foothold placement reward that extracts supportable foothold candidates from foot-end point-cloud samples and rewards touchdown locations that lie close to the nearest supportable candidate. Experiments in simulation and on a physical humanoid demonstrate robust traversal over diverse terrains and effective zero-shot transfer to real-world scenes containing handrails, hollow pallet assemblies, severe reflective interference, and visually cluttered outdoor surroundings.

Robot demonstration datasets used to train vision-language-action policies can contain a subtle but harmful failure mode: trajectories that are behaviorally correct but paired with the wrong language instruction. We study post-hoc auditing of these Instruction-Trajectory Mismatches (ITMs). Unlike failed rollouts, ITMs often look plausible, and can corrupt the language-behavior mapping learned by the policy. We propose Multimodal Probabilistic Fusion (MMPF), a training-free auditing framework that treats each modality as an expert, estimates a task-label distribution from local neighborhood agreement and global prototype similarity, and then fuses modalities with predictive-entropy weighting in a product of experts. Across LIBERO benchmarks with injected instruction mismatches and noisy real-robot data, MMPF achieves the strongest overall ITM detection and label correction accuracy. We also show that auditing improves most downstream policy learning in settings where language is needed to disambiguate the task. We demonstrate in real robot experiments that our method can achieve improved policy performance and show the trade-off of filtering demonstrations compared to relabeling.

Open-Vocabulary Object Navigation (OVON) requires an embodied agent to locate a language-specified target in unknown environments. Many zero-shot methods rely on frontier-candidate reasoning under incomplete observations, while topology-aware methods reduce candidate redundancy but may still introduce panoramic inspection overhead and repeated reconsideration. We present DRIVE-Nav, a structured framework that organizes exploration around persistent directions rather than raw frontiers. By inspecting encountered directions more completely and restricting subsequent decisions to still-relevant directions within a forward 240-degree view range, DRIVE-Nav reduces redundant revisits and improves path efficiency. The framework extracts and tracks directional candidates from weighted Fast Marching Method (FMM) paths, maintains representative views for semantic inspection, and combines vision-language-guided prompt enrichment with cross-frame verification to improve grounding reliability. Experiments on HM3D-OVON, HM3Dv1, HM3Dv2, and MP3D demonstrate strong overall performance and consistent efficiency gains. On HM3D-OVON, DRIVE-Nav achieves 50.2% SR and 32.6% SPL, improving the previous best method by 1.9% SR and 5.6% SPL. It also delivers the best SPL on HM3Dv1, HM3Dv2, and MP3D and transfers to a physical humanoid robot. Real-world deployment also demonstrates its effectiveness.

Imitation learning from human demonstrations has achieved significant success in robotic control, yet most visuomotor policies still condition on single-step observations or short-context histories, making them struggle with non-Markovian tasks that require long-term memory. Simply enlarging the context window incurs substantial computational and memory costs and encourages overfitting to spurious correlations, leading to catastrophic failures under distribution shift and violating real-time constraints in robotic systems. By contrast, humans can compress important past experiences into long-term memories and exploit them to solve tasks throughout their lifetime. In this paper, we propose VPWEM, a non-Markovian visuomotor policy equipped with working and episodic memories. VPWEM retains a sliding window of recent observation embeddings as short-term working memory, and introduces a Transformer-based contextual memory compressor that recursively converts out-of-window observations into a fixed number of episodic memory embeddings. The compressor uses self-attention over a cache of past summary embeddings and cross-attention over a cache of historical observations, and is trained jointly with the policy. We instantiate VPWEM on diffusion policies to exploit both short-term and episode-wide information for action generation with nearly constant memory and computation per step. Experiments demonstrate that VPWEM outperforms state-of-the-art baselines including diffusion policies and vision-language-action (VLA) models by more than 20\% on the memory-intensive manipulation tasks in MIKASA and achieves an average 5\% improvement on the mobile manipulation benchmark MoMaRT. Code is available at https://github.com/HarryLui98/code_vpwem.

Visual localization in large-scale UAV scenarios is a critical capability for autonomous systems, yet it remains challenging due to geometric complexity and environmental variations. While 3D Gaussian Splatting (3DGS) has emerged as a promising scene representation, existing 3DGS-based visual localization methods struggle with robust pose initialization and sensitivity to rendering artifacts in large-scale settings. To address these limitations, we propose LSGS-Loc, a novel visual localization pipeline tailored for large-scale 3DGS scenes. Specifically, we introduce a scale-aware pose initialization strategy that combines scene-agnostic relative pose estimation with explicit 3DGS scale constraints, enabling geometrically grounded localization without scene-specific training. Furthermore, in the pose refinement, to mitigate the impact of reconstruction artifacts such as blur and floaters, we develop a Laplacian-based reliability masking mechanism that guides photometric refinement toward high-quality regions. Extensive experiments on large-scale UAV benchmarks demonstrate that our method achieves state-of-the-art accuracy and robustness for unordered image queries, significantly outperforming existing 3DGS-based approaches. Code is available at: https://github.com/xzhang-z/LSGS-Loc

Dual‐arm mobile manipulators can transport and manipulate large objects with simple end‐effectors. Interacting with dynamic environments subject to strict safety and compliance requirements, achieving whole‐body motion planning online while meeting various hard constraints for highly redundant mobile manipulators poses a significant challenge. We tackle this challenge by presenting an efficient whole‐body motion‐planning approach based on model‐based predictive control (MPC). We construct a hierarchical MPC framework in which the first MPC optimizes the end‐effectors' collision‐free motion, thereby guiding the second MPC to optimize whole‐body joint trajectories. In the first stage, we employ a Bézier‐curve representation to parameterize the high‐degree‐of‐freedom (high‐DOF) motion in of two collaborating end‐effectors, notably introducing a novel and efficient method to optimize quaternion trajectories. This facilitates fast long‐horizon motion planning of coupled translational and rotational trajectories while accounting for approximated feasibility constraints. In the second stage, this is the first work to novelly incorporate the representation into a whole‐body MPC for online high‐DOF motion generation with predictive admittance control over a relatively short horizon while satisfying whole‐body hard constraints. Compared with the usage of a discretization approach, our whole‐body MPC ensures accurate model‐state transitions with faster, more stable computation speeds, and consistent motion command generation which enhances tracking performance for our hybrid position/velocity‐controlled robot. Both MPCs perform replanning in each control loop to adapt to a changing environment. Simulations and real‐world experiments validate the proposed hierarchical MPC framework, demonstrating that its two novel MPCs achieve efficient and robust performance in scenarios involving static and dynamic obstacle avoidance, compliant interaction with manipulated objects, and external disturbances.

International audience

We present LatentAM, an online 3D Gaussian Splatting (3DGS) mapping framework that builds scalable latent feature maps from streaming RGB-D observations for open-vocabulary robotic perception. Instead of distilling high-dimensional Vision-Language Model (VLM) embeddings using model-specific decoders, LatentAM proposes an online dictionary learning approach that is both model-agnostic and pretraining-free, enabling plug-and-play integration with different VLMs at test time. Specifically, our approach associates each Gaussian primitive with a compact query vector that can be converted into approximate VLM embeddings using an attention mechanism with a learnable dictionary. The dictionary is initialized efficiently from streaming observations and optimized online to adapt to evolving scene semantics under trust-region regularization. To scale to long trajectories and large environments, we further propose an efficient map management strategy based on voxel hashing, where optimization is restricted to an active local map on the GPU, while the global map is stored and indexed on the CPU to maintain bounded GPU memory usage. Experiments on public benchmarks and a large-scale custom dataset demonstrate that LatentAM attains significantly better feature reconstruction fidelity compared to state-of-the-art methods, while achieving near-real-time speed (12-35 FPS) on the evaluated datasets. Our project page, including the code, is available at: https://junwoonlee.github.io/projects/LatentAM

This paper presents a real-time control-constrained Differential Dynamic Programming (DDP) framework for underactuated legged robots. To address the limitation of classical DDP in handling control constraints, we propose an Accelerated Projected Gradient (APG)-based control-constrained DDP (ABC-DDP), which efficiently computes constrained solutions and identifies active sets without repeated Karush-Kuhn-Tucker (KKT) inversions. A virtual constraint is introduced to integrate control constraints within a feasibility-driven multiple-shooting framework, enabling stable optimization even from dynamically infeasible initializations. The proposed method supports real-time model predictive control (MPC) with short horizons under strong underactuation. Simulation results demonstrate static two-leg standing under external disturbances, along with diverse dynamic motions including slow catwalk, upright walking, and high-speed running within a unified MPC framework. To the best of our knowledge, this is the first demonstration of static two-leg standing of a quadruped robot achieved using real-time finite-horizon MPC.

Generating safe corridors is essential for collision-free robotic motion planning, yet most existing methods rely on predefined reference paths, which bias corridor geometry and implicitly limit the homotopy classes that can be explored. We propose a reference-path-free corridor generation framework on graphs of convex sets (GCS) that constructs corridors directly as sequences of convex sets, allowing corridor structure to emerge from the free-space representation rather than from a guiding path. To reason about similarity among corridors, we extend visibility-based deformation from paths to convex-set sequences, enabling the fusion of topologically redundant corridors while preserving distinct alternatives. To overcome the limited adaptability of existing GCS methods based on static global decompositions, we further develop an adaptive multi-scale GCS, in which a sampling-based fine-scale graph supports localized updates and a visibility-based coarse-scale graph enables compact global exploration. The two levels maintain topological consistency, allowing incremental updates without full graph reconstruction under environmental uncertainty. Numerical experiments characterize GCS construction, corridor generation, homotopy-aware exploration, and local updates, showing efficient graph construction, stable trajectory-level performance, and shorter-duration homotopy-aware trajectories than existing baselines. Hardware experiments on ground and aerial robots, including deployment with onboard localization, further validate the framework under translated and previously unknown obstacles.

Instance Image-Goal Navigation (IIN) requires autonomous agents to identify and navigate to a target object or location depicted in a reference image captured from any viewpoint. While recent methods leverage powerful novel view synthesis (NVS) techniques, such as three-dimensional Gaussian splatting (3DGS), they typically rely on randomly sampling multiple viewpoints or trajectories to ensure comprehensive coverage of discriminative visual cues. This approach, however, creates significant redundancy through overlapping image samples and lacks principled view selection, substantially increasing both rendering and comparison overhead. In this paper, we introduce a novel IIN framework with a hierarchical scoring paradigm that estimates optimal viewpoints for target matching. Our approach integrates cross-level semantic scoring, utilizing CLIP-derived relevancy fields to identify regions with high semantic similarity to the target object class, with fine-grained local geometric scoring that performs precise pose estimation within promising regions. Extensive evaluations demonstrate that our method achieves state-of-the-art performance on simulated IIN benchmarks and real-world applicability.

A cooperative‐navigation algorithm based on an extended Kalman filter (EKF) is proposed for Leader–Follower autonomous underwater vehicles (AUVs) to address inconsistencies in inter‐vehicle navigation coordinate systems caused by the absence of geo‐referencing signals in anchor‐free environments. The algorithm integrates inertial compensation and a dynamic ranging threshold to address challenges posed by low update rates (0.05–0.07 Hz), high latency, time‐varying hydroacoustic channels, and multipath interference in hydroacoustic communication, all of which significantly degrade navigation accuracy. To mitigate latency from low‐rate hydroacoustic links, we use time‐tag alignment (UTC‐referenced clock synchronization) to propagate the Leader state to the effective ranging time for inertial latency compensation. In response to the hydroacoustic channel variability and multipath interference, a dynamic ranging threshold based on the AUV's motion capabilities is proposed to eliminate outlier ranging data. To enhance the robustness of cooperative navigation and maintain a coordinate system, a velocity‐bounded constraint is introduced into the EKF framework, limiting the magnitude of individual position corrections. During a field test at Danjiangkou Reservoir, Henan Province, China, AUVs cooperatively mapped an underwater terrain area of at a water depth of 45–55 m. Compared with terrain mapping using a single‐AUV inertial navigation system, the proposed cooperative‐navigation method improved the agreement of the terrain mapping with a Global Navigation Satellite System (GNSS)‐aided surface reference map by 15.89% relative to the inertial navigation baseline, which supports the effectiveness of the proposed cooperative‐navigation method during the mapping mission. Furthermore, the root mean square error (RMSE) of distance measurements between Leader AUV and Follower AUVs during linear transects decreased by 59.13% and 55.21%, respectively, indicating improved Leader–Follower range consistency achieved by the proposed method. In a Weihai open‐sea campaign with a desired Leader–Follower separation of 100 m on straight‐line segments, the resulting separation RMSE was 6.29 and 4.41 m for the two Followers, respectively. Moreover, in a Danjiangkou surface test with continuous GNSS ground truth (GNSS used only as an external reference, not as filter inputs, except for mission‐start initialization where applicable), the Followers' horizontal RMSE decreased from 15.09 m/13.00 m (INS) to 9.38 m/8.71 m (cooperative).

As a critical component for robotic citrus harvesting, the mechanical design of the end‐effector determines harvesting efficiency, success rate, and fruit quality. This paper aims to enhance citrus harvesting efficiency by proposing an integrated gripper‐shear harvesting end‐effector and its control method suitable for harvesting operations in Ehime No. 38 Citrus orchards. First, the fruits were collected and their phenotypic characteristics and mechanical properties were measured. On the basis of mechanical design theory and CAE analysis methods, the flexible clamping mechanism and the sliding shear mechanism were developed. Second, a low‐cost clamping force control method combining finite element analysis and linear regression was proposed. Its validity was verified through transient dynamic analysis experiments conducted in Ansys Workbench, involving 15 clamping experiments (10 for initial testing and 5 for validation). Finally, 50 harvesting trials conducted in citrus orchards demonstrated a 96.0% success rate for stalk severing and 94.0% success rate for harvesting undamaged citrus fruits. The proposed harvesting end‐effector is easy to control, simple to maintain, low in manufacturing cost, minimally damaging to citrus, and highly successful in harvesting.

Odontomas, classified as hamartomas rather than true neoplasms, are the most common type of odontogenic tumors. These lesions are typically asymptomatic, show no gender predilection, and are often detected incidentally through routine radiological examinations. Complete surgical resection is the gold standard for odontoma treatment. Traditional approaches are limited by surgeon‐dependent precision, while digital guide plate‐assisted surgery has fixed procedural constraints. This study reports the first clinical application of a semi‐active oral robotic surgical system for the minimally invasive resection of a compound odontoma embedded in the alveolar bone of the left mandibular body between the mandibular left canine (tooth 33) and first premolar (tooth 34) in a 15‐year‐old female patient. Preoperative planning was completed with high‐resolution CBCT and dedicated robotic navigation software. The odontoma was completely removed without damage to adjacent teeth. The recovery was uneventful, and a 6‐month follow‐up showed excellent bone healing with no residual lesion. The use of oral robots for odontoma extraction enables precise positioning and minimally invasive removal, offering a promising approach for the treatment of these tumors.

This work identifies a passive aerodynamic damping effect induced by co-contraction in antagonistic redundant propulsion. Complementing prior work on aerodynamic promptness, which addressed active wrench-rate authority along constant-wrench fibers, we study the passive side: the local derivative of aerodynamic force with respect to air-relative velocity at a trim. This derivative defines an incremental aerodynamic damping coefficient. We prove that it increases monotonically along constant-force fibers under a mild aerodynamic hardening condition, and derive this property from a first-order Blade Element Theory model exposing the relevant speed-inflow coupling. The resulting mechanism, Variable Aerodynamic Damping Actuation (VADA), is formulated as an antagonistic aerodynamic actuation module and allocation principle, structurally analogous to variable-stiffness actuation at the level of fiber motions and incremental impedance modulation. An impedance-form interpretation clarifies common- and differential-mode roles, while a propeller-data-based assessment using the UIUC Propeller Database shows that the identified damping has practical small-UAV magnitude, is comparable to ordinary low-speed body-drag damping, and depends strongly on low-advance-ratio thrust sensitivity.

End-to-end (E2E) autonomous driving systems offer a promising alternative to traditional modular pipelines by reducing information loss and error accumulation, with significant potential to enhance both mobility and safety. However, most existing E2E approaches directly generate plans based on dense bird's-eye view (BEV) grid features, leading to inefficiency and limited planning awareness. To address these limitations, we propose iterative Proposal-centric autonomous driving (iPad), a novel framework that places proposals - a set of candidate future plans - at the center of feature extraction and auxiliary tasks. Central to iPad is ProFormer, a BEV encoder that iteratively refines proposals and their associated features through proposal-anchored attention, effectively fusing multi-view image data. Additionally, we introduce two lightweight, proposal-centric auxiliary tasks - mapping and prediction - that improve planning quality with minimal computational overhead. Extensive experiments on the NAVSIM and CARLA Bench2Drive benchmarks demonstrate that iPad achieves state-of-the-art performance while being significantly more efficient than prior leading methods.

Real-time reconstruction of deformable surgical scenes is vital for advancing robotic surgery, improving intraoperative guidance, and enabling automation. Recent methods achieve dense reconstructions from da Vinci robotic surgery videos, with Gaussian Splatting (GS) offering real-time performance via graphics acceleration. However, reconstruction quality in occluded regions remains limited, and depth accuracy has not been fully assessed, as benchmarks like EndoNeRF and StereoMIS lack 3D ground truth. We propose Diff2DGS, a two-stage framework for reliable 3D reconstruction of occluded surgical scenes. First, a diffusion-based video module with temporal priors inpaints tissue occluded by instruments with high spatiotemporal consistency. Second, we adapt 2D Gaussian Splatting (2DGS) with a Learnable Deformation Model (LDM) to capture dynamic tissue deformation and anatomical geometry, and introduce adaptive depth weight to improve geometric fidelity. We further extend evaluation beyond image-quality metrics by performing quantitative depth analysis on the SCARED dataset. Diff2DGS outperforms state-of-the-art methods in both appearance and geometry, reaching 38.02 dB PSNR on EndoNeRF and 34.40 dB on StereoMIS. Our experiments also show that optimizing image quality alone does not necessarily ensure accurate 3D geometry.The code is available at https://diff2dgs.github.io/.

The design of a hexapod is complex and requires integration between kinematic models, control systems, and sensing. Existing literature has reviewed these sub‐systems in isolation. Since 2021, there has been no review of the field despite significant advancements in soft soil, lunar traversal, and artificial intelligence. This paper addresses that gap by establishing a novel terrain‐coded locomotion layer scaffold via taxonomy creation. A subset of 58 primary studies were deeply analyzed and synthesized to quantify method choices across hexapod locomotion. Findings show that the DH‐Parameter system and geometric inverse kinematics models are most widely used across all terrain types. Jacobian‐based inverse kinematic models prove to be computationally demanding, yet more accurate. Jacobian approaches also result in multiple end‐effector positions and unnatural poses. Bio‐inspired control systems and proprioceptive methods show promise for smoother gait switching and real‐time adaptability. Approaches related to reinforcement learning, convolutional neural networks and long short‐term memory models were also introduced in recent years. These contribute toward increasing performance, terrain adaptability and path planning. The primary contribution of this paper is a set of data‐driven design patterns linking mathematical models, layered controls, and simulator‐terrain couplings for structured and unstructured environments. These patterns support more defensible deployment‐oriented design choices. Future directions and generalized guidelines based on these design patterns for hexapod locomotion are also highlighted. Thus documenting future‐facing research which shows the hexapod becoming a fully autonomous system capable of harsh terrain traversal (lunar surfaces, ice, soft soil, underwater).

Drone light shows (DLShows) represent a rapidly growing application of swarm robotics, creating captivating aerial displays through the synchronized flight of hundreds or thousands of unmanned aerial vehicles (UAVs) as environmentally friendly and reusable alternatives to traditional pyrotechnics. This domain presents unique challenges in optimally assigning drones to visual waypoints and generating smooth, collision‐free trajectories at a very large scale. This article introduces the Unified Assignment and Trajectory Generation (UATG) framework. The proposed approach concurrently solves two core problems: the optimal assignment of drones to designated goal locations and the generation of dynamically feasible, collision‐free, time‐parameterized trajectories. The UATG framework is specifically designed for DLShows, ensuring minimal transition times between formations and guaranteeing inter‐drone collision avoidance. A key innovation is its exceptional computational efficiency, enabling the coordination of large‐scale in real‐time; for instance, it computes the optimal assignment and trajectories for 1008 drones in approximately 1 s on a standard laptop. Extensive simulations in realistic environments validate the framework's performance, demonstrating its capability to orchestrate complex formations—from alphanumeric characters to intricate 3D shapes—with precision and visual smoothness. This work provides a critical advancement for the DLShow industry, offering a practical and scalable solution for generating complex aerial choreography and establishing a valuable benchmark for ground control station software designed for the efficient coordination of multiple UAVs. A supplemental animated simulation of this work is available at https://youtu.be/-Fjrhw03594 .

During coal mining, coal and gangue with varying shapes, sizes, and masses are transporte together on conveyor belts, posing significant challenges for gangue sorting. Traditional manual sorting is increasingly limited by safety risks, labor costs, and the shortage of skilled workers. To address this problem, this paper presents a push‐based gangue‐sorting robotic system for industrial dynamic conveyor scenarios. The system is organized as a task‐specific closed‐loop robotic sorting framework that couples gangue perception, identity‐consistent tracking, contour‐level pose estimation, time‐aware push planning, coal‐carryover avoidance, and multimodal execution‐outcome verification for dynamic mine‐conveyor operation. In contrast to conventional grasp‐lift‐place sorting systems, the proposed system exploits conveyor support and uses planar pushing as the manipulation primitive, which is better suited to large, heavy, and irregular gangue pieces in constrained mine environments. Long‐term tests show that the proposed system can stably complete gangue separation under isolated deployment conditions without manual intervention. In industrial online tests, the recognition accuracy for dynamic gangue exceeds 92%, the average planning time of a single robotic arm is about 0.4 s, the average execution time is about 2.8 s, and the highest single‐arm gangue removal rate reaches 61.52%. A two‐stage cascade estimate based on the highest single‐arm result suggests the potential for line‐level removal above 80%. The proposed execution‐outcome verification module achieves verification accuracy exceeding 95%.

Efficient material logistics are critical in controlling costs and schedules in the construction industry. However, manual material handling remains prone to inefficiencies, delays, and safety risks. Autonomous forklifts offer a solution to streamline on‐site logistics, reducing reliance on human operators and mitigating labor shortages. This paper presents the development and evaluation of ADAPT (Autonomous Dynamic All‐terrain Pallet Transporter), an autonomous off‐road forklift designed for construction environments. Unlike structured warehouses, construction sites pose significant challenges, including dynamic obstacles, unstructured terrain, and varying weather conditions. To address these challenges, our system integrates AI‐driven perception with traditional approaches for decision making, planning, and control, enabling reliable operation in complex environments. We validate the system through extensive real‐world testing, comparing its performance against an experienced human operator across various weather conditions. Our findings demonstrate that autonomous outdoor forklifts can operate near human‐level performance, offering a viable path toward safer and more efficient construction logistics.

Agricultural robots are increasingly used to improve orchard productivity and reduce labor dependence, yet collaborative orchard mapping remains difficult because dense canopy degrades GNSS reliability, repetitive row geometry induces structural aliasing, and long‐duration operation over large orchard blocks causes mergeable overlap to emerge only after extended traversal, thereby destabilizing generic registration pipelines. In this work, we formulate orchard map merging as an operation‐time progressive submap association and registration problem for a dual‐robot field setting and propose PROMO‐Orchard, an orchard‐aware semantic map‐merging framework. The framework integrates four tightly coupled components: class‐dependent voxelization with persistent orchard‐prior extraction, Orchard‐AKS for operation‐aware submap scheduling, Sem‐OREOS for semantic‐enhanced overlap retrieval and merge triggering, and Sem‐GICP for class‐weighted semantic refinement with orchard structural regularization. Together, these components turn semantic map merging into an operation‐time, orchard‐aware process rather than a generic post‐hoc registration step. Field experiments in three representative orchard settings, namely I‐Tunnel, U‐Tunnel, and Long‐Duration, show that the proposed method consistently suppresses row aliasing and maintains semantically coherent shared maps under in‐row overlap, headland U‐turning, and delayed‐overlap operation. Compared with Segregator and a semantic‐ablation variant, the proposed framework reduces ICP error by 78–94% while maintaining FPMR above 91% across all scenarios, and achieves clear improvements in semantic fidelity, boundary continuity, and structural consistency. These results support the conclusion that reliable orchard map merging requires not only semantic cues, but their structured integration with orchard‐specific priors, thereby providing a practical basis for collaborative orchard mapping and downstream robotic operations such as navigation, canopy management, harvesting, and logistics. While the present study experimentally validates the dual‐robot case, the proposed progressive pairwise merging formulation provides the system‐level basis for future extension to larger multi‐robot orchard teams.

The cover image is based on the article TriRock6W: Autonomous Mobile Robot with Six Wheels, Three Rocker Arms in Complex Environments by shao shiliang et al., https://doi.org/10.1002/rob.70229.