Adaptive Tracking of a Single-Rigid-Body Character in Various Environments

TL;DR

This paper introduces a sample-efficient deep reinforcement learning method that uses a single rigid body model to adaptively track complex motions in various unobserved environments, enabling rapid training and robust performance.

Contribution

It presents a novel SRB-based reinforcement learning approach that achieves environment adaptation and policy transition without extra training, using a reduced state-action space for efficiency.

Findings

Policy adapts to unseen environments like uneven terrain and object pushing.

Training completes within 30 minutes on a portable laptop.

Full-body motion is generated in a physically plausible manner.

Abstract

Since the introduction of DeepMimic [Peng et al. 2018], subsequent research has focused on expanding the repertoire of simulated motions across various scenarios. In this study, we propose an alternative approach for this goal, a deep reinforcement learning method based on the simulation of a single-rigid-body character. Using the centroidal dynamics model (CDM) to express the full-body character as a single rigid body (SRB) and training a policy to track a reference motion, we can obtain a policy that is capable of adapting to various unobserved environmental changes and controller transitions without requiring any additional learning. Due to the reduced dimension of state and action space, the learning process is sample-efficient. The final full-body motion is kinematically generated in a physically plausible way, based on the state of the simulated SRB character. The SRB simulation is formulated as a quadratic programming (QP) problem, and the policy outputs an action that allows the SRB character to follow the reference motion. We demonstrate that our policy, efficiently trained within 30 minutes on an ultraportable laptop, has the ability to cope with environments that have not been experienced during learning, such as running on uneven terrain or pushing a box, and transitions between learned policies, without any additional learning.