外观
论文arXiv 2610.0585523 天前
用于欠驱动双足机器人无碰撞行走的分层强化学习
原文链接：Hierarchical Reinforcement Learning for Collision-Free Locomotion of an Underactuated Biped（arXiv 2610.05855 · 发布日期：2026-10-05）
Hierarchical Reinforcement Learning for Collision-Free Locomotion of an Underactuated Biped
面向欠驱动双足避障与平衡耦合，提出SAC联合训练的HRL，并在随机PyBullet静态与动态环境中验证。
核心要点
- 欠驱动双足无髋或踝横滚，绕障时路径偏离与平衡扰动严重耦合。
- 高层策略观测位姿、36条射线、移动障碍状态和局部目标，输出机体速度指令。
- 低层策略以速度为条件跟踪PD关节目标，两级策略通过SAC联合训练。
- 随机PyBullet试验中成功率达静态98.0%、动态88.0%，消融表明各观测和奖励项有贡献。
验证范围与局限：验证范围为随机PyBullet仿真环境中每种方法100次静态和动态试验。
英文摘要 (Abstract)
A bipedal robot cannot deviate from its path to avoid an obstacle without disturbing its balance, and this coupling is most severe on underactuated platforms such as the biped considered here, which has four actuated joints per leg and no hip or ankle roll. This paper presents a Hierarchical Reinforcement Learning (HRL) framework in which a High-Level (HL) policy observes the robot pose, 36 raycast proximity measurements, moving-obstacle states, and a receding-horizon local goal, and outputs a body-velocity command $(v_x, v_y, ω_{yaw})$ every ten control steps, while a velocity-conditioned Low-Level (LL) policy tracks each command through PD-controlled joint targets. Both policies are trained jointly with Soft Actor-Critic (SAC). Because the converged gait is task-agnostic, it is frozen and driven by classical planners over the same command interface, yielding three controlled baselines: SAC+A*, SAC+RRT*, and SAC+APF. Across 100 evaluation trials per method in randomized PyBullet environments, the proposed method reaches the goal in 98.0% of static and 88.0% of dynamic trials, against at most 78.0% and 68.0% for the planner hybrids, with path lengths within 4% of the A* reference, and ablations confirm that each observation channel and reward term contributes materially to this performance.
作者：Jagannath Prasad Sahoo, Saurabh Kumar, Surya Prakash S. K., Samiran Datta, Abhay Dwivedi, Amit Shukla
来源
本页事实对应的公开页面，以原文为准。