Legged and wheeled robot locomotion.
Yufei Jia, Zhanxiang Cao, Mingrui Yu et al. · arXiv preprint · May 2026
Simulation-based RL for contemporary robot control is increasingly organized around GPU-resident simulation: physics, rollout collection, and learning are placed on a single GPU-centric execution path
Junheng Li, Liang Wu, Sergio A. Esteban et al. · arXiv preprint · Jun 2026
In humanoid motion control, model predictive control (MPC) offers physically grounded prediction and constraint handling, while reinforcement learning (RL) enables robust whole-body skills through large-scale simulation. However, using MPC inside RL often requires time-consuming problem construction or excessive training overhead, making such frameworks difficult to justify in practice. This work studies efficient training-time MPC guidance for humanoid locomotion and manipulation, termed MPC-RL. We introduce a centroidal-dynamics MPC reward formulation that leverages guidance from MPC trajectories in training time. To make this practical in massively parallel RL, we develop π^nMPC, a parallel-in-horizon and construction-free batched GPU MPC solver that operates directly on time-varying dynamics to avoid high memory usage and pre-compilation. Through a variety of comparative studies and hardware validations, we have found that MPC-RL achieves superior performance in locomotion and manipulation skills.
Satoshi Yamamori, Koji Ishihara, Kentaro Minamikawa et al. · arXiv preprint · Jul 2026
Sim-to-real transfer in robot learning is often limited by discrepancies between the ideal actuator dynamics assumed during policy training and the nonlinear, hardware-dependent be...
Kaifeng Zhao, Mathis Petrovich, Haotian Zhang et al. · arXiv preprint · Jul 2026
Generating realistic 3D human motions in real-time within interactive applications is key for animation, simulation, and humanoid robotics. While recent offline motion generation approaches offer precise control via text and kinematic constraints, they lack the inference speed required for intera...
Hongjin Chen, Zijun Xu, Shihao Ma et al. · arXiv preprint · Sep 2026
Humanoid locomotion across complex terrain demands forward-looking exteroception to anticipate obstacles, yet this signal is unreliable in real-world deployment, failing partially and intermittently. Existing perceptive policies often assume that depth observations remain clean and in-distributio...
Siyuan Ma, Boshi Zhang, Yutian Zhang et al. · arXiv preprint · Aug 2026
Mobile manipulation requires a robot to predict how locomotion and arm motion jointly alter future observations and control. Existing world-action models, developed largely for fixed-base platforms, do not explicitly distinguish camera ego-motion from base and arm actions. Here we introduce DECOW...
Alexey Zemtsov, Maxim Bobrin, Alexander Nikulin et al. · ICML 2026 · Jun 2026
Offline goal-conditioned reinforcement learning requires both long-horizon reachability estimates and local action comparisons. Dual goal representations provide value fields that capture global goal reachability, but they do not directly specify which action should be preferred at a given state. We...
Lav R. Varshney · arXiv preprint · Sep 2026
Open-loop multilegged locomotion over rough terrain has been interpreted as matter transport over a noisy channel: leg-ground interactions are discrete basic active contacts, terrain deletes or perturbs those contacts, and spatial redundancy concentrates the resulting thrust and arrival time. Tha...
Haoxuan Han, Chen Chen, Linao Gong et al. · arXiv preprint · Jun 2026
Humanoid robots have achieved strong locomotion capabilities, but reliable navigation on versatile terrains remains challenging because obstacle avoidance must be coordinated with dynamically feasible motion. In this work, we present GuideWalk, a unified end-to-end framework that integrates traversa...
Junnosuke Kamohara, Feiyang Wu, Andy Ningan Zong et al. · arXiv preprint · Sep 2026
Humanoid locomotion on granular terrain remains a significant challenge due to its complex foot-terrain interaction dynamics that are difficult to model. Existing approaches either ignore granular contact dynamics or incorporate simplified normal force models with heuristic tangential components....
Zuxing Lu, Ziang Zheng, Yao Lyu et al. · arXiv preprint · Jun 2026
Building a general-purpose whole-body controller is essential for enabling diverse motion capabilities in humanoid robots across a wide range of downstream tasks, including locomotion and loco-manipulation. Different tasks rely on distinct motion reference modalities: locomotion primarily depends on...
Jian Zhou, Xingyu Zhang, Rui Ma et al. · arXiv preprint · Sep 2026
Muscle-driven locomotion provides a physically grounded approach to generating realistic human movement. However, achieving both physiological plausibility and adaptability to changes in musculoskeletal capacity and external disturbances remains a fundamental challenge. To address this limitation...
Shutong Ding, Zejia Zhong, Zhongyi Wang et al. · ICML 2026 · May 2026
Recent advances in reinforcement learning (RL) have achieved great successes by leveraging the multimodality and exploration capability of diffusion policies. Among these approaches, one representativ
Xiao Ren, Yuhui Yang, Zongbiao Weng et al. · arXiv preprint · Jun 2026
Recent reinforcement learning approaches have shown great promise in improving humanoid motion tracking performance and achieving fall recovery under disturbances. However, most existing works treat motion tracking and fall recovery as different tasks and require multi-stage training with specialize...
Yujie Xiong, Peng Zhai, Taixian Hou et al. · arXiv preprint · Sep 2026
Brachiation enables primates to move across overhead supports when ground paths are blocked, suggesting a complementary locomotion mode for robots operating in cluttered or hazardous environments. Bringing this capabil?ity to high-DoF humanoid robots is difficult because the controller must disco...
Mahmud Hasan Saikot, Sydney Spiegel, Sudheera Akalanka Kariyawasam et al. · arXiv preprint · Aug 2026
Robots that can change their morphologies and behaviors for different tasks and environments hold great promise for adaptable, multifunctional systems. Modular reconfigurable robots (MRRs) can achieve such functionalities by docking and rearranging individual units, but most rely on rigid modules...
Xincheng Tang, Yiji Chen, Youhan Xie et al. · arXiv preprint · Aug 2026
Door opening and traversal is a long-horizon loco-manipulation task that requires precise handle interaction and coordinated base-arm control. We present Video2DoorTraversal, a single-video real-to-sim-to-real framework for wheel-legged mobile manipulators. Given one RGB video of a real door, Doo...
Zhe Li, Zhenzhe Zhang, Yangyang Wei et al. · arXiv preprint · Aug 2026
Humanoid household tasks often require concurrent loco-manipulation, where the robot must move, adjust posture, maintain balance, and manipulate objects as a single coordinated behavior. Yet existing humanoid policies typically decompose locomotion and manipulation, while recent world-action mode...
Haoran Jiang, Jin Chen, Qingwen Bu et al. · arXiv preprint · Dec 2025
Humanoid robots require precise locomotion and dexterous manipulation to perform challenging loco-manipulation tasks. Yet existing approaches, modular or end-to-end, are deficient in manipulation-aware locomotion. This confines the robot to a limited workspace, preventing it from performing large...
Nikita Rudin, Fabian Jenelten, Timon Homberger et al. · ICRA 2024 · May 2024
Unitree Go2 uses cross-modal vision + proprioception transformers trained in Isaac Gym to achieve agile navigation over rough terrain with minimal sim-to-real gap.
Nikita Rudin, David Hoeller, Philipp Reist et al. · RSS 2022 Workshop / arXiv 2109.11978 · Jun 2022
Legged Gym is an open-source training framework built on NVIDIA Isaac Gym that enables massively parallel deep RL for legged robot locomotion, achieving policy training times of minutes rather than days.
Nikita Rudin, David Hoeller, Marco Hutter · Open Source Library · Jun 2022
RSL-RL is a lightweight, high-performance PyTorch library implementing PPO tailored for locomotion. Used as the default RL backend in Legged Gym, it enables training thousands of agents in parallel on GPU.
Read these papers in order to build expertise in Locomotion.
Yufei Jia, Zhanxiang Cao et al. · 2026
Junheng Li, Liang Wu et al. · 2026
Satoshi Yamamori, Koji Ishihara et al. · 2026
Kaifeng Zhao, Mathis Petrovich et al. · 2026
Hongjin Chen, Zijun Xu et al. · 2026
Siyuan Ma, Boshi Zhang et al. · 2026
…and 16 more papers.
