ADAPTIVE LEARNING OF HUMAN MOTION BY A TELEROBOT USING A NEURAL-NETWORK MODEL AS A TEACHER

This paper describes a model and experiment in which human motion data is used to teach a laboratory telerobot simple human skills. The prot otype learns an adapted motor skill involving pick and place tasks in a microblock world environment. We choose place and pick tasks because of their repetitive and dynamic motor motions involved during both fo rward and backward task execution. By assuming the skill transfer take s place implicitly, we use human skill data to train the robot via a n eural network simulation model. The method has an advantage over the c lassical state-space control models in that the model is adaptive to c hanges in human input data.