Learning Multiple Gaits within Latent Space for Quadruped Robots

Jinze Wu, Yufei Xue,Chenkun Qi

CoRR(2023)

引用 0|浏览2
暂无评分
摘要
Learning multiple gaits is non-trivial for legged robots, especially when encountering different terrains and velocity commands. In this work, we present an end-to-end training framework for learning multiple gaits for quadruped robots, tailored to the needs of robust locomotion, agile locomotion, and user's commands. A latent space is constructed concurrently by a gait encoder and a gait generator, which helps the agent to reuse multiple gait skills to achieve adaptive gait behaviors. To learn natural behaviors for multiple gaits, we design gait-dependent rewards that are constructed explicitly from gait parameters and implicitly from conditional adversarial motion priors (CAMP). We demonstrate such multiple gaits control on a quadruped robot Go1 with only proprioceptive sensors.
更多
查看译文
关键词
quadruped robots,multiple gaits,learning
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要