Robotic Arm System for Assisting Persons with Physical Disabilities to Drink Water

This paper presents a robotic arm system for the purpose of assisting persons with physical disabilities to drink water. We have developed a seven-degree-of-freedom robotic arm with a thermoplastic polyurethane hand. We also implemented functions of an object detection with YOLOv5 algorithm and a user’s mouth detection based on MediaPipe machine learning platform in the proposed system. Firstly, tasks of grasping and moving each detected object (a paper cup or a plastic bottle) to the mouth of a user were conducted to evaluate the system. Secondly, tasks of grasping the plastic bottle laid down on an experimental table and letting it stand on the table were carried out. From the experimental results, it was found that the object detection and the detection of the user’s mouth were successfully performed, and that the success rates of these tasks were $80 \%$ and more.