Real-Time Whole-Body Control of Legged Robots with Model-Predictive Path Integral Control

TL;DR

This paper introduces a real-time whole-body control system for legged robots using model-predictive path integral control, leveraging MuJoCo simulation for efficient sampling, enabling robust locomotion and manipulation in real-world scenarios.

Contribution

It presents the first successful real-world deployment of whole-body sampling-based MPC on legged robots, demonstrating effective locomotion and manipulation with a simple control strategy.

Findings

Effective real-world locomotion over varied terrains

Successful climbing and pushing tasks

First deployment of sampling-based MPC on legged robots

Abstract

This paper presents a system for enabling real-time synthesis of whole-body locomotion and manipulation policies for real-world legged robots. Motivated by recent advancements in robot simulation, we leverage the efficient parallelization capabilities of the MuJoCo simulator to achieve fast sampling over the robot state and action trajectories. Our results show surprisingly effective real-world locomotion and manipulation capabilities with a very simple control strategy. We demonstrate our approach on several hardware and simulation experiments: robust locomotion over flat and uneven terrains, climbing over a box whose height is comparable to the robot, and pushing a box to a goal position. To our knowledge, this is the first successful deployment of whole-body sampling-based MPC on real-world legged robot hardware. Experiment videos and code can be found at: https://whole-body-mppi.github.io/