A neuro-based adaptive training method for robotic rehabilitation aids

In this paper, a new training method for robotic rehabilitation aids is proposed, in which only one neural network (NN) is simultaneously used in order to identify the dynamic properties of a human-robot system and give an assist to the trainee. The model used for identification of the dynamics of the human-robot system consists of the NN and a reference model which represents the main characteristic of a skilful operator. This paper explains the working principle of the training method and shows the validity of the proposed method through experiments by non-skilled operators.