Collaborative path planning for a robotic wheelchair

Abstract

Generating a path to guide a wheelchair's motion faces two challenges. First, the path is located in the human environment and that is usually unstructured and dynamic. Thus, it is difficult to generate a reliable map and plan paths on it by artificial intelligence. Second, the wheelchair, whose task is to carry a human user, should move on a smooth and comfortable path adapted to the user's intentions. To meet these challenges, we propose that the human operator and the robot interact to create and gradually improve a guide path. This paper introduces design tools to enable an intuitive interaction, and reports experiments performed with healthy subjects in order to investigate this collaborative path learning strategy. We analyzed features of the optimal paths and user evaluation in representative conditions. This was complemented by a questionnaire filled out by the subjects after the experiments. The results demonstrate the effectiveness of this approach, and show the utility and complementarity of the tools to design ergonomic guide paths.