Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion

no code implementations • 31 Jan 2024 • Tairan He, Chong Zhang, Wenli Xiao, Guanqi He, Changliu Liu, Guanya Shi

Legged robots navigating cluttered environments must be jointly agile for efficient task execution and safe to avoid collisions with obstacles or humans.

CoVO-MPC: Theoretical Analysis of Sampling-based MPC and Optimal Covariance Design

1 code implementation • 14 Jan 2024 • Zeji Yi, Chaoyi Pan, Guanqi He, Guannan Qu, Guanya Shi

Sampling-based Model Predictive Control (MPC) has been a practical and effective approach in many domains, notably model-based reinforcement learning, thanks to its flexibility and parallelizability.

Model-based Reinforcement Learning Model Predictive Control