论文arXiv 2611.239163 天前
ReLoco框架实现人形机器人统一运动与跌倒恢复单策略部署
本文由 AI 根据公开内容整理,可能有误,请以原文为准:ReLoco: Unified Humanoid Locomotion and Fall Recovery via Behavior-Conditioned Motion Priors(arXiv 2611.23916)
核心内容
针对人形机器人跌倒恢复与运动指令衔接难题,研究提出ReLoco框架,通过行为条件化对抗运动先验将直立运动、任意姿态起身及指令恢复整合至单一本体感知策略。该方法在训练阶段根据机器人姿态动态选择专家池并调节共享判别器,结合任务奖励与临时辅助机制优化探索效率。消融实验证实了条件监督与输出分离的有效性,为复杂地形下的鲁棒运动提供了统一解决方案。
全量策略在匹配训练迭代下实现96.80%独立起身成功率与96.96%推挤恢复成功率,躯干平面速度均方根误差降至0.1266 m/s,较条件单输出版本降低14.7%。该算法已在64.5 kg级Tiangong 3.0人形机器人上完成零样本部署,仅凭单一本体感知Actor即可执行行走、奔跑、跌倒起身及运动恢复全流程,无需控制器切换或机载策略微调,验证了Sim-to-Real迁移的可行性。
关键看点
- 统一策略架构:采用行为条件化对抗运动先验,在单一本体感知策略中同时编码直立运动与任意姿态跌倒恢复能力,消除多控制器切换延迟。
- 量化性能指标:全量策略独立起身成功率达96.80%,推挤恢复成功率96.96%,躯干平面速度RMSE为0.1266 m/s,较基线降低14.7%。
- 真机零样本部署:在64.5 kg Tiangong 3.0上验证行走、奔跑及跌倒恢复,全程无需控制器切换或On-robot微调,证实Sim-to-Real泛化性。
- 训练优化机制:引入基于姿态的专家池选择与双输出共享判别器,配合临时辅助机制解决跌倒恢复等高难度动作的探索效率瓶颈。
工程与产业观察
ReLoco证明了通过行为条件化先验可在单一端到端策略中解决人形机器人“运动-恢复”耦合难题。其在64.5kg全尺寸本体上的零样本部署成功,表明高质量仿真参考数据与条件对抗学习能有效弥合Sim-to-Real鸿沟,为具身智能体摆脱复杂状态机依赖、实现高鲁棒性自主运动提供了可复现的工程范式。
英文摘要 (Abstract)
Humanoid robots must recover from falls and resume commanded motion to sustain locomotion. Learning these behaviors in one policy requires supervision that accommodates both upright movement and ground-contact recovery. We present ReLoco, a framework that uses behavior-conditioned adversarial motion priors to learn locomotion, get-up from arbitrary fallen postures, and command resumption within one proprioceptive policy. During training, robot posture selects the expert pool and conditions a shared discriminator with two behavior-specific outputs. Task rewards and shared regularization support command tracking and recovery, while temporary assistance facilitates exploration. A simulation-based refinement stage supplies robot-executed motion references for this supervision. Ablations examine conditional supervision, output separation, and reference quality under matched settings. At a common training iteration, the full policy achieves $96.80%$ independent get-up success and $96.96%$ push-recovery success, including command resumption. Its torso planar velocity root-mean-square error is $0.1266$~m/s, $14.7%$ lower than the conditional single-output variant, with similar joint-angle discrepancy measured by dynamic time warping. Deployment on the 64.5-kg Tiangong~3.0 humanoid demonstrates walking, running, get-up, and locomotion resumption after falls using one proprioceptive actor, without controller switching or on-robot policy fine-tuning.
作者:Lequn Fu, Yufeng Hu, Yijun Zhong, Jie Zhang, Xiao Li, Lingxiang Ji, Zhiyuan Xu, Jian Tang, Shiqi Li
外部来源
本页事实对应的公开页面,以原文为准。(点击将离开本站)