一个策略搞定七种人形机器人跌倒恢复，无需重新训练
Learning to Get Up Across Morphologies: Zero-Shot Recovery with a Unified Humanoid Policy
- 用统一的强化学习策略跨形态实现跌倒起身
- 零样本迁移成功率86%±7%，覆盖0.48~0.81米身高差异
- 适合需快速部署、多型号通用的人形机器人应用
在动态环境如RoboCup中，跌倒恢复能力对人形机器人的表现至关重要，长时间停机常导致比赛失利。现有基于深度强化学习（DRL）的方法虽能生成稳健的起身行为，但需为每种机器人构型单独训练策略。本文提出一种单一DRL策略，可在七种不同构型的人形机器人上实现跌倒恢复，其高度范围为0.48–0.81米，重量范围为2.8–7.9千克，动力学特性各异。该策略通过CrossQ训练，在未见过的构型上实现86%±7%（95%置信区间[81, 89]）的零样本迁移成功率，无需针对特定机器人训练。通过留一法实验、构型缩放分析及多样性消融实验表明，针对性地覆盖构型差异可显著提升零样本泛化能力。在某些情况下，共享策略甚至优于专用基线。这些结果验证了形态无关控制在跌倒恢复中的实用性，为人形机器人通用控制奠定了基础。代码已开源：https://github.com/utra-robosoccer/unified-humanoid-getup
原文摘要 · Abstract (English)
Fall recovery is a critical skill for humanoid robots in dynamic environments such as RoboCup, where prolonged downtime often decides the match. Recent techniques using deep reinforcement learning (DRL) have produced robust get-up behaviors, yet existing methods require training of separate policies for each robot morphology. This paper presents a single DRL policy capable of recovering from falls across seven humanoid robots with diverse heights (0.48 - 0.81 m), weights (2.8 - 7.9 kg), and dynamics. Trained with CrossQ, the unified policy transfers zero-shot up to 86 +/- 7% (95% CI [81, 89]) on unseen morphologies, eliminating the need for robot-specific training. Comprehensive leave-one-out experiments, morph scaling analysis, and diversity ablations show that targeted morphological coverage improves zero-shot generalization. In some cases, the shared policy even surpasses the specialist baselines. These findings illustrate the practicality of morphology-agnostic control for fall recovery, laying the foundation for generalist humanoid control. The software is open-source and available at: https://github.com/utra-robosoccer/unified-humanoid-getup
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。