通过历史感知课程学习，实现机器人高速稳定运动
HACL: History-Aware Curriculum Learning for Fast Locomotion
- 用RNN建模关节速度历史，捕捉运动奖励与速度的时序关系
- 实测达6.7m/s峰值速度，比现有算法快近20%
- 适合需要高速动态行走的四足/双足机器人研究者
我们针对四足和双足机器人敏捷快速运动这一关键问题，提出一种新算法，通过考虑运动的时间特性，保持稳定并生成高速轨迹。该方法基于新颖的历史感知课程学习（HACL）算法，利用循环神经网络（RNN）建模关节速度指令的历史信息，结合观测到的线性与角速度奖励。隐藏状态帮助课程学习在给定时间步内捕捉前向线速度、角速度命令与奖励之间的关系。我们在MIT Mini Cheetah、Unitree Go1和Go2机器人上进行仿真验证，并在真实环境中的Unitree Go1上测试。实验表明，HACL在给定命令速度7m/s下达到6.7m/s峰值速度，相比先前算法性能提升近20%。
原文摘要 · Abstract (English)
We address the problem of agile and rapid locomotion, a key characteristic of quadrupedal and bipedal robots. We present a new algorithm that maintains stability and generates high-speed trajectories by considering the temporal aspect of locomotion. Our formulation takes into account past information based on a novel history-aware curriculum Learning (HACL) algorithm. We model the history of joint velocity commands with respect to the observed linear and angular rewards using a recurrent neural net (RNN). The hidden state helps the curriculum learn the relationship between the forward linear velocity and angular velocity commands and the rewards over a given time-step. We validate our approach on the MIT Mini Cheetah,Unitree Go1, and Go2 robots in a simulated environment and on a Unitree Go1 robot in real-world scenarios. In practice, HACL achieves peak forward velocity of 6.7 m/s for a given command velocity of 7m/s and outperforms prior locomotion algorithms by nearly 20%.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。