Shield-Loco：基于预测性安全过滤的防护运动策略
Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering
AI总结：
提出一种预测性安全过滤器，通过全物理模型优化接触序列，减少四足机器人在密集杂乱环境中的安全违规，同时保持任务性能。
摘要：
强化学习（RL）策略能够实现动态腿部运动，但缺乏避免训练中未出现的约束违反的机制。大规模离线安全学习对于覆盖所有边缘情况是不切实际的。现有的安全框架要么依赖无法推理全身行为的降阶模型，要么需要保守的恢复控制器，这会降低任务性能。我们提出一种预测性安全过滤器，它对输入到RL策略的名义接触位置进行事后过滤。当预测到碰撞时，基于采样的优化器使用全物理模型异步搜索更安全的接触序列，而学习的价值函数则引导长期回报。我们的三个算法组件（采样接触的几何投影、动量增强更新和副本交换）使得在不连续的接触景观中优化变得可行。我们在密集杂乱环境中的四足机器人上验证了该过滤器，无论是在仿真还是真实世界中，都显示出在最小偏离名义输入的情况下大幅减少安全违规。
英文摘要：
Reinforcement learning (RL) policies enable dynamic legged locomotion but lack mechanisms to avoid violations of safety constraints that are absent during training. Large-scale offline safe learning is impractical for covering all edge cases. Existing safety frameworks either rely on reduced-order models that cannot reason about whole-body behaviors or require conservative recovery controllers that degrade task performance. We propose a predictive safety filter that post-hoc filters the nominal contact locations fed to the RL policy. When a collision is predicted, a sampling-based optimizer asynchronously searches for safer contact sequences using a full-physics model, while a learned value function bootstraps long-horizon returns. Our three algorithmic components (geometric projection of sampled contacts, momentum-augmented updates, and replica-exchange) make the optimization tractable in a discontinuous contact landscape. We validate the filter on a quadruped robot in dense, cluttered environments, both in simulation and in the real world, showing substantial reductions in safety violations with minimal deviation from the nominal input.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
