Manipulator as a Tail: Promoting Dynamic Stability for Legged Locomotion

TL;DR

This paper demonstrates how a manipulator can enhance the dynamic stability of legged robots, especially at high speeds and under external disturbances, by acting as a tail-like stabilizer through a novel reinforcement learning approach.

Contribution

It introduces an incremental reinforcement learning method inspired by neurophysiological theories to enable legged robots to use manipulators as stabilizing tails, improving locomotion success rates.

Findings

Increased success rate by up to 61 percentage points over baselines.

Reduced failure rate by up to 43.6% during high-speed turning.

Enhanced stability of quadruped robots under external forces.

Abstract

For locomotion, is an arm on a legged robot a liability or an asset for locomotion? Biological systems evolved additional limbs beyond legs that facilitates postural control. This work shows how a manipulator can be an asset for legged locomotion at high speeds or under external perturbations, where the arm serves beyond manipulation. Since the system has 15 degrees of freedom (twelve for the legged robot and three for the arm), off-the-shelf reinforcement learning (RL) algorithms struggle to learn effective locomotion policies. Inspired by Bernstein's neurophysiological theory of animal motor learning, we develop an incremental training procedure that initially freezes some degrees of freedom and gradually releases them, using behaviour cloning (BC) from an early learning procedure to guide optimization in later learning. Simulation experiments show that our policy increases the success rate by up to 61 percentage points over the baselines. Simulation and real robot experiments suggest that our policy learns to use the arm as a tail to initiate robot turning at high speeds and to stabilize the quadruped under external perturbations. Quantitatively, in simulation experiments, we cut the failure rate up to 43.6% during high-speed turning and up to 31.8% for quadruped under external forces compared to using a locked arm.