Achieving Stable High-Speed Locomotion for Humanoid Robots with Deep Reinforcement Learning
School of Computer Science and Technology, University of Science and Technology of China(中国科学技术大学计算机科学与技术学院);Ningbo Institute of Digital Twin, Eastern Institute of Technology(浙大宁波理工学院宁波数字孪生研究院)
本文提出结合深度强化学习与运动动力学先验的 KSLC 方法,通过协调手臂运动提升人形机器人高速运动稳定性,在仿真中实现 3.5 m/s 速度跟踪并验证了泛化能力。
摘要:
人形机器人在执行各种任务时具有显著的通用性,但其基本的行走和奔跑能力,尤其是在高速运动时,仍然是一个挑战。本文提出了一种新方法,将深度强化学习与运动动力学先验相结合,以实现稳定的运动控制(KSLC)。KSLC 促进协调的手臂运动来抵消不稳定的力,从而增强整体稳定性。与基线方法相比,KSLC 能够更准确地跟踪指令速度,并在速度控制方面具有更好的泛化能力。在仿真测试中,启用 KSLC 的人形机器人成功跟踪了 3.5 m/s 的目标速度,且波动减小。在高保真环境中的仿真到仿真验证进一步证实了其稳健的性能,凸显了其在实际应用中的潜力。
英文摘要:
Humanoid robots offer significant versatility for performing a wide range of tasks, yet their basic ability to walk and run, especially at high velocities, remains a challenge. This letter presents a novel method that combines deep reinforcement learning with kinodynamic priors to achieve stable locomotion control (KSLC). KSLC promotes coordinated arm movements to counteract destabilizing forces, enhancing overall stability. Compared to the baseline method, KSLC provides more accurate tracking of commanded velocities and better generalization in velocity control. In simulation tests, the KSLC-enabled humanoid robot successfully tracked a target velocity of 3.5 m/s with reduced fluctuations. Sim-to-sim validation in a high-fidelity environment further confirmed its robust performance, highlighting its potential for real-world applications.
State Estimation for Human Motion and Humanoid Locomotion
Istituto Italiano di Tecnologia(意大利理工学院);IRI CSIC-UPC(西班牙国家研究委员会-加泰罗尼亚理工大学机器人研究所);AIST(日本产业技术综合研究所);LAAS-CNRS(法国国家科学研究中心系统分析与架构实验室);DIBRIS-UNIGE(热那亚大学计算机、生物工程、机器人与系统工程系)
Biomechanical Comparisons Reveal Divergence of Human and Humanoid Gaits
Zhejiang University(浙江大学);School of Engineering, Westlake University(西湖大学工程学院);Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences(中国科学院宁波材料技术与工程研究所)
TAGA: Terrain-aware Active Gaze Learning for Generalizable Agile Humanoid Locomotion
MarmotLab, National University of Singapore(马尔莫实验室,新加坡国立大学);Center of X-Mechanics, Zhejiang University(浙大X力学中心);South China University of Technology(华南理工大学)
UniReLo: Learning a Unified Humanoid Policy from Fall Recovery to Locomotion across Diverse Terrains
School of Control Science and Engineering, Shandong University(山东大学控制科学与工程学院);Key Laboratory of Machine Intelligence and System Control, Ministry of Education(教育部机器智能与系统控制重点实验室)
X2-N: A Transformable Wheel-legged Humanoid Robot with Dual-mode Locomotion and Manipulation
Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology(香港理工大学电子与计算机工程系);X-lab Department, Agibot Innovation Company(阿吉博创新公司X实验室)
Trajectory Optimization for Robust Humanoid Locomotion with Sample-Efficient Learning
Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);K. N. Toosi University of Technology(K.N.托西理工大学);Tandon School of Engineering, New York University(纽约大学坦登工程学院)
HWC-Loco: A Hierarchical Whole-Body Control Approach to Robust Humanoid Locomotion
The Chinese University of Hong Kong (Shenzhen)(香港中文大学(深圳));Harbin Institute of Technology(哈尔滨工业大学);DexForce Technology(DexForce科技);Southeast University(东南大学)
GM-Loco: Terrain-Adaptive Humanoid Locomotion on Granular Media
Institute for Robotics and Intelligent Machines, Georgia Institute of Technology(佐治亚理工学院机器人与智能机器研究所);School of Physics, Georgia Institute of Technology(佐治亚理工学院物理学院);Northeastern University(东北大学)
PolygMap: A Perceptive Locomotion Framework for Humanoid Robot Stair Climbing
The Shenzhen Institute of Artificial Intelligence and Robotics for Society(深圳人工智能与机器人社会研究院);Electronic science and technology, Guangdong University of Technology(广东技术大学电子科学与技术学院);Leju robotics(乐居机器人)
Explicit Stair Geometry Conditioning for Robust Humanoid Locomotion
Shenzhen Institute of Artificial Intelligence and Robotics for Society(深圳人工智能与机器人社会研究院);School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)科学与工程学院);Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)(穆罕默德·本·扎耶德人工智能大学)