A framework for an interactive robot-based tutoring system and its application to ball-passing training

This paper describes a framework for an interactive robot-based tutoring system (IRTS). The proposed IRTS is based on combining aspects of intelligent tutoring systems (ITSs) which are computer-based expert systems to simulate aspects of a human tutor, and robot-assisted systems which help users with physical interaction using robotic devices. The IRTS involves the robot interacting with a user, generating the user model, providing training tasks for guiding the user, and giving feedback on his/her performance. The proposed system is applied to ball-passing training. In the training, a human trainee passes a ball to a robot tutor and receives feedback from the robot on how to improve the kick, based on the angle and velocity of the ball. In the experiment, the system is implemented in a real robot and demonstrates that the system is able to provide user-adaptive training tasks and feedback to improve his/her ball-passing skill.