Combining Learning-based Locomotion Policy with Model-based Manipulation for Legged Mobile Manipulators

TL;DR

This paper presents a method that combines model-based external disturbance predictions with learning-based locomotion policies to improve the stability and adaptability of legged robots with manipulators, enabling zero-shot transfer from simulation to real hardware.

Contribution

It introduces a novel integration of external dynamics plans into reinforcement learning policies for legged mobile manipulators, enhancing robustness and transferability.

Findings

Successful zero-shot transfer to real robots

Stable locomotion with external wrench prediction

Enhanced robustness to external disturbances

Abstract

Deep reinforcement learning produces robust locomotion policies for legged robots over challenging terrains. To date, few studies have leveraged model-based methods to combine these locomotion skills with the precise control of manipulators. Here, we incorporate external dynamics plans into learning-based locomotion policies for mobile manipulation. We train the base policy by applying a random wrench sequence on the robot base in simulation and adding the noisified wrench sequence prediction to the policy observations. The policy then learns to counteract the partially-known future disturbance. The random wrench sequences are replaced with the wrench prediction generated with the dynamics plans from model predictive control to enable deployment. We show zero-shot adaptation for manipulators unseen during training. On the hardware, we demonstrate stable locomotion of legged robots with the prediction of the external wrench.