Assistive Gym: A Physics Simulation Framework for Assistive Robotics

TL;DR

Assistive Gym is an open-source physics simulation framework that models multiple assistive tasks and human preferences, enabling reinforcement learning-based training of robots for diverse daily living assistance activities.

Contribution

It introduces a multi-task simulation environment with human modeling, facilitating research and development of assistive robots across various activities.

Findings

Modeling human motion improves assistance quality.

Baseline policies trained on Assistive Gym demonstrate effective robot assistance.

Comparison of different robots shows varying performance levels.

Abstract

Autonomous robots have the potential to serve as versatile caregivers that improve quality of life for millions of people worldwide. Yet, conducting research in this area presents numerous challenges, including the risks of physical interaction between people and robots. Physics simulations have been used to optimize and train robots for physical assistance, but have typically focused on a single task. In this paper, we present Assistive Gym, an open source physics simulation framework for assistive robots that models multiple tasks. It includes six simulated environments in which a robotic manipulator can attempt to assist a person with activities of daily living (ADLs): itch scratching, drinking, feeding, body manipulation, dressing, and bathing. Assistive Gym models a person's physical capabilities and preferences for assistance, which are used to provide a reward function. We present baseline policies trained using reinforcement learning for four different commercial robots in the six environments. We demonstrate that modeling human motion results in better assistance and we compare the performance of different robots. Overall, we show that Assistive Gym is a promising tool for assistive robotics research.