鲁棒通用四足运动策略的采样策略
Sampling Strategies for Robust Universal Quadrupedal Locomotion Policies
AI总结：
研究通过采样物理参数和关节增益，训练单一强化学习策略以泛化到多种四足机器人配置，比较三种采样策略，并利用先验和参考模型提升鲁棒性，在ANYmal上零样本部署。
摘要：
本文聚焦于生成鲁棒通用四足机器人运动策略的配置变化采样策略。我们研究了采样物理机器人参数和关节比例-微分增益的效果，以训练单一强化学习策略，使其泛化到多种参数配置。比较了三种基本关节增益采样策略：参数采样与（1）质量-增益的线性和多项式函数映射，（2）基于性能的自适应滤波，以及（3）均匀随机采样。通过使用名义先验和参考模型对配置进行偏置，我们提高了策略的鲁棒性。所有训练均在RaiSim仿真环境中进行，在多种不同的四足机器人上进行了仿真测试，并在ANYmal四足机器人上进行了零样本硬件部署。与多个基线实现相比，我们的结果证明了显著随机化关节控制器增益对于鲁棒地缩小仿真到现实差距的必要性。
英文摘要：
This work focuses on sampling strategies of configuration variations for generating robust universal locomotion policies for quadrupedal robots. We investigate the effects of sampling physical robot parameters and joint proportional-derivative gains to enable training a single reinforcement learning policy that generalizes to multiple parameter configurations. Three fundamental joint gain sampling strategies are compared: parameter sampling with (1) linear and polynomial function mappings of mass-to-gains, (2) performance-based adaptive filtering, and (3) uniform random sampling. We improve the robustness of the policy by biasing the configurations using nominal priors and reference models. All training was conducted using the RaiSim simulation environment, tested in simulation on a range of diverse quadrupeds, and zero-shot deployed onto hardware using the ANYmal quadruped robot. Compared to multiple baseline implementations, our results demonstrate the need for significant joint controller gains randomization for robust closing of the sim-to-real gap.
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。
