Robust Biped Locomotion Using Deep Reinforcement Learning on Top of an Analytical Control Approach

TL;DR

This paper introduces a modular framework combining analytical control and deep reinforcement learning to achieve robust, stable, and efficient bipedal locomotion, validated through simulations in RoboCup environment.

Contribution

It presents a novel hierarchical framework integrating an analytical dynamics model with reinforcement learning for adaptive biped locomotion control.

Findings

Framework produces fast, stable gait in simulations.

It learns to improve upper body efficiency.

Demonstrates robustness and adaptability in RoboCup environment.

Abstract

This paper proposes a modular framework to generate robust biped locomotion using a tight coupling between an analytical walking approach and deep reinforcement learning. This framework is composed of six main modules which are hierarchically connected to reduce the overall complexity and increase its flexibility. The core of this framework is a specific dynamics model which abstracts a humanoid's dynamics model into two masses for modeling upper and lower body. This dynamics model is used to design an adaptive reference trajectories planner and an optimal controller which are fully parametric. Furthermore, a learning framework is developed based on Genetic Algorithm (GA) and Proximal Policy Optimization (PPO) to find the optimum parameters and to learn how to improve the stability of the robot by moving the arms and changing its center of mass (COM) height. A set of simulations are performed to validate the performance of the framework using the official RoboCup 3D League simulation environment. The results validate the performance of the framework, not only in creating a fast and stable gait but also in learning to improve the upper body efficiency.