Autonomous Navigation of Wheelchairs in Indoor Environments using Deep Reinforcement Learning and Computer Vision

• Paulo De Almeida Afonso UFPEL
• Paulo Roberto Ferreira UFPEL

Abstract

We present a novel method for autonomous navigation of motorized wheelchairs in shared indoor environments with humans, utilizing Deep Reinforcement Learning and Computer Vision. The primary goal of this research is to enhance the quality of life for individuals with disabilities who depend on such assistance for mobility. Our approach combines the Deep Deterministic Policy Gradient (DDPG) algorithm with computer vision techniques, enabling motorized wheelchairs to autonomously navigate environments containing both static and dynamic obstacles. We conducted experiments within a simulated environment, and the results obtained showcase the promising potential of our proposed solution, thus contributing to the advancement of research in this field.

Keywords: deep reinforcement learning, collision avoidance, robotic wheelchair, autonomous navigation