论文标题
了解对双向运动的深控制政策的稳定性
Understanding the Stability of Deep Control Policies for Biped Locomotion
论文作者
论文摘要
实现稳定性和鲁棒性是双重运动控制的主要目标。最近，Deep Fearforce学习（DRL）引起了人们的极大关注，作为构建双头控制策略的一般方法，并且对先前的最新技术表现出了显着改善。尽管深度控制政策比以前的控制器设计方法具有优势，但许多问题仍未得到解决。深层控制政策是否像人类步行一样强大？模拟步行是否使用类似的策略与人类步行以保持平衡？特定的步态模式是否会类似地影响人类和模拟的步行？深层政策学会了如何实现步态稳定性？这项研究的目的是通过评估与人类受试者和先前的反馈控制者相比的深度政策的推进恢复稳定性来回答这些问题。我们还进行了实验，以评估DRL算法变体的有效性。
Achieving stability and robustness is the primary goal of biped locomotion control. Recently, deep reinforce learning (DRL) has attracted great attention as a general methodology for constructing biped control policies and demonstrated significant improvements over the previous state-of-the-art. Although deep control policies have advantages over previous controller design approaches, many questions remain unanswered. Are deep control policies as robust as human walking? Does simulated walking use similar strategies as human walking to maintain balance? Does a particular gait pattern similarly affect human and simulated walking? What do deep policies learn to achieve improved gait stability? The goal of this study is to answer these questions by evaluating the push-recovery stability of deep policies compared to human subjects and a previous feedback controller. We also conducted experiments to evaluate the effectiveness of variants of DRL algorithms.
