预测性安全课程用于鲁棒腿式运动
Predictive Safety Curricula for Robust Legged Locomotion
查看机构详情
- ANYbotics AG（ANYbotics公司）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
针对腿式机器人运动策略中罕见但严重的失败，提出预测性安全课程（PSC），利用学得的未来安全成本预测来分配训练经验，在崎岖地形和生产系统中显著提高可靠性并减少碰撞。
中文摘要 AI 辅助
即使平均任务性能很高，学习型腿式机器人运动策略中仍可能持续存在罕见但后果严重的失败，部分原因在于标准课程主要调整任务难度，而非安全关键经验的分布。我们引入了预测性安全课程（PSC），这是一个利用对未来安全成本的学得预测来分配运动训练经验的框架。PSC从策略回放中训练一个分布式安全评论员，并利用其预测来优先处理地形环境和先前遇到的随机事件。由此产生的课程修改了训练分布，同时保持任务奖励和策略优化损失不变。我们在受控的崎岖地形运动和生产级运动系统中评估了PSC。相对于标准地形推进、基于优势的回放和学习进度课程，PSC提高了可靠性，在困难地形和观测退化的情况下增益最大。相同的分配原则可迁移到两个生产级运动栈。在ANYmal-D硬件上，相对于学习进度课程，在三个匹配的训练种子中，PSC将小腿碰撞发生率降低了63%，且每个种子都有所降低。在一个生产级爬楼梯平台上，PSC在评估的硬件试验中消除了观察到的小腿碰撞。这些结果表明，对未来安全成本的学得预测可以为将训练经验分配给罕见失败模式并提高运动可靠性提供有效的信号。
英文摘要
Rare but consequential failures can persist in learned locomotion policies for legged robots even when average task performance is high, in part because standard curricula primarily adapt task difficulty rather than the distribution of safety-critical experience. We introduce Predictive Safety Curricula (PSC), a framework for allocating locomotion training experience using learned predictions of future safety cost. PSC trains a distributional safety critic from policy rollouts and uses its predictions to prioritize both terrain contexts and previously encountered randomized events. The resulting curriculum modifies the training distribution while leaving the task reward and policy-optimization loss unchanged. We evaluate PSC in controlled rough-terrain locomotion and in production locomotion systems. PSC improves reliability relative to standard terrain progression, advantage-based replay, and learning-progress curricula, with the largest gains on difficult terrain and under degraded observations. The same allocation principle transfers to two production locomotion stacks. On ANYmal-D hardware, PSC reduces shank-collision incidence by $63\%$ relative to the learning-progress curriculum across three matched training seeds, with a reduction in every seed. On a production stair-climbing platform, PSC eliminates observed shank collisions in the evaluated hardware trials. These results show that learned predictions of future safety cost can provide an effective signal for allocating training experience toward rare failure modes and improving locomotion reliability.
