FRASA: An End-to-End Reinforcement Learning Agent for Fall Recovery and Stand Up of Humanoid Robots

TL;DR

This paper presents FRASA, a deep reinforcement learning agent that unifies fall recovery and standing up in humanoid robots, demonstrating superior adaptability and performance over traditional methods in dynamic environments.

Contribution

FRASA is the first DRL-based framework integrating fall recovery and stand-up, reducing training time and enhancing adaptability for humanoid robots.

Findings

FRASA outperforms KFB in fall recovery and stand-up tasks.

FRASA demonstrates robustness against unpredictable disturbances.

Training time is significantly reduced using the Cross-Q algorithm.

Abstract

Humanoid robotics faces significant challenges in achieving stable locomotion and recovering from falls in dynamic environments. Traditional methods, such as Model Predictive Control (MPC) and Key Frame Based (KFB) routines, either require extensive fine-tuning or lack real-time adaptability. This paper introduces FRASA, a Deep Reinforcement Learning (DRL) agent that integrates fall recovery and stand up strategies into a unified framework. Leveraging the Cross-Q algorithm, FRASA significantly reduces training time and offers a versatile recovery strategy that adapts to unpredictable disturbances. Comparative tests on Sigmaban humanoid robots demonstrate FRASA superior performance against the KFB method deployed in the RoboCup 2023 by the Rhoban Team, world champion of the KidSize League.