Learning Bipedal Walking for Humanoid Robots in Challenging Environments with Obstacle Avoidance

TL;DR

This paper introduces a reinforcement learning approach enabling humanoid robots to walk dynamically and avoid obstacles in complex environments, extending prior simple terrain locomotion methods.

Contribution

It presents a novel reward function modification that allows policy-based reinforcement learning to handle obstacle avoidance during bipedal walking.

Findings

Successful navigation without collisions in obstacle-rich environments

Enhanced reward function improves obstacle avoidance capabilities

Demonstrates viability of RL for complex humanoid locomotion

Abstract

Deep reinforcement learning has seen successful implementations on humanoid robots to achieve dynamic walking. However, these implementations have been so far successful in simple environments void of obstacles. In this paper, we aim to achieve bipedal locomotion in an environment where obstacles are present using a policy-based reinforcement learning. By adding simple distance reward terms to a state of art reward function that can achieve basic bipedal locomotion, the trained policy succeeds in navigating the robot towards the desired destination without colliding with the obstacles along the way.