Shared Control of Holonomic Wheelchairs through Reinforcement Learning

TL;DR

This paper introduces a reinforcement learning approach for shared control of holonomic wheelchairs, enabling intuitive, safe, and smooth omnidirectional navigation with real-world implementation.

Contribution

It presents the first RL-based shared control method for holonomic wheelchairs, improving user comfort and safety, and demonstrates successful sim-to-real transfer.

Findings

Ensures collision-free navigation with better orientation control.

Achieves smoother movement compared to previous non-learning methods.

First real-world RL-based shared control for omnidirectional wheelchairs.

Abstract

Smart electric wheelchairs can improve user experience by supporting the driver with shared control. State-of-the-art work showed the potential of shared control in improving safety in navigation for non-holonomic robots. However, for holonomic systems, current approaches often lead to unintuitive behavior for the user and fail to utilize the full potential of omnidirectional driving. Therefore, we propose a reinforcement learning-based method, which takes a 2D user input and outputs a 3D motion while ensuring user comfort and reducing cognitive load on the driver. Our approach is trained in Isaac Gym and tested in simulation in Gazebo. We compare different RL agent architectures and reward functions based on metrics considering cognitive load and user comfort. We show that our method ensures collision-free navigation while smartly orienting the wheelchair and showing better or competitive smoothness compared to a previous non-learning-based method. We further perform a sim-to-real transfer and demonstrate, to the best of our knowledge, the first real-world implementation of RL-based shared control for an omnidirectional mobility platform.