Development of a Standing Support Arm with PPO Method for a Robotic Wheelchair

Abstract

To address the global concern of an aging population, robotic wheelchairs emerge as a promising solution for enhancing elderly mobility. Sit-to-stand movement is a pivotal task for individuals using wheelchairs. However, the natural decline in the physical strength of the elderly, especially in the waist and lower limb strength, brings challenges to independent sit-to-stand movement from the wheelchair. This study focuses on utilizing proximal policy optimization (PPO), a reinforcement learning (RL) method, to develop a standing support arm motion for aiding sit-to-stand movements for elderly wheelchair users. This study explores the design and potential of a standing support arm to assist with the sit-to-stand motion. Through RL training in a simulated environment, we have demonstrated the capability of a standing support arm to support users in sit-to-stand movement. This proves the substantial potential of the support arm to effectively aid elderly individuals. The implementation of such a device holds promise in reducing the workload of caregivers and significantly enhancing the independence of the elderly.