Foundational Policy Acquisition via Multitask Learning for Motor Skill Generation

TL;DR

This paper introduces a multitask reinforcement learning approach inspired by human sensorimotor adaptation to acquire foundational policies for motor skill generation, demonstrating superior performance in locomotion and robotic kicking tasks.

Contribution

The study presents a novel multitask reinforcement learning pipeline with encoder-decoder networks for foundational policy acquisition adaptable to various tasks and environments.

Findings

Outperforms previous multitask RL methods in locomotion tasks

Successfully adapts to new target positions and physical parameters

Generates new motor skills like overhead kicking from learned policies

Abstract

In this study, we propose a multitask reinforcement learning algorithm for foundational policy acquisition to generate novel motor skills. \textcolor{\hcolor}{Learning the rich representation of the multitask policy is a challenge in dynamic movement generation tasks because the policy needs to cope with changes in goals or environments with different reward functions or physical parameters. Inspired by human sensorimotor adaptation mechanisms, we developed the learning pipeline to construct the encoder-decoder networks and network selection to facilitate foundational policy acquisition under multiple situations. First, we compared the proposed method with previous multitask reinforcement learning methods in the standard multi-locomotion tasks. The results showed that the proposed approach outperformed the baseline methods. Then, we applied the proposed method to the ball heading task using a monopod robot model to evaluate skill generation performance. The results showed that the proposed method was able to adapt to novel target positions or inexperienced ball restitution coefficients but to acquire a foundational policy network, originally learned for heading motion, which can generate an entirely new overhead kicking skill.