从MPC安全学习运动技能
Safe Learning of Locomotion Skills from MPC
AI总结：
本研究利用MPC作为专家并结合SafeDAGGER框架，安全地学习运动技能，相比行为克隆和标准DAGGER，训练失败更少且策略更鲁棒。
摘要：
运动技能的安全学习仍然是一个开放性问题。实际上，运动系统开环动力学固有的不稳定性使得从零开始的朴素学习在现实世界中容易导致灾难性失败。在本工作中，我们研究了使用迭代算法从模型预测控制（MPC）中安全学习运动技能。在我们的框架中，我们将MPC作为专家，并借鉴安全数据聚合（SafeDAGGER）框架的思想，以最小化策略训练期间的失败次数。通过与行为克隆和标准DAGGER等其他标准方法的比较，我们表明，我们的方法不仅在训练期间的失败次数显著更少，而且得到的策略对外部扰动也更具鲁棒性。
英文摘要：
Safe learning of locomotion skills is still an open problem. Indeed, the intrinsically unstable nature of the open-loop dynamics of locomotion systems renders naive learning from scratch prone to catastrophic failures in the real world. In this work, we investigate the use of iterative algorithms to safely learn locomotion skills from model predictive control (MPC). In our framework, we use MPC as an expert and take inspiration from the safe data aggregation (SafeDAGGER) framework to minimize the number of failures during training of the policy. Through a comparison with other standard approaches such as behavior cloning and vanilla DAGGER, we show that not only our approach has a substantially fewer number of failures during training, but the resulting policy is also more robust to external disturbances.
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。
