Decentralized control of telerehabilitation systems subject to communication delay using multi-agent systems theory

In this paper, a decentralized controller for telerehabilitation systems is proposed in the framework of multi-agent systems in the presence of time-varying delays between agents’ communication channels. In fact, the telerehabilitation systems can include various scenarios, and in this scenario, a therapist, patient, and one or more trainee are present. Different scenarios can be implemented easily by matching the telerehabilitation system in the framework of multi-agent systems. Also, the time-varying delays of the communication channels between the therapist, the patient, and the trainees can be easily corresponded to multi-agent systems. In fact, we consider therapists, patients, and trainees as different agents in multi-agent systems, and each agent includes a manipulator with n-DOF nonlinear dynamics. The proposed decentralized controller in this paper is also designed based on the estimation of the dynamics of each agent as well as the information between the communication graph in the presence of time-varying delays between the agents. The advantage of using the presented controller is that the dynamics of each agent have been estimated without calculating the regressor matrix and using functional approximation theory based on basic functions, and this estimate has been used in the structure of the controller and using this the control scheme can also easily implement different scenarios in the telerehabilitation systems within the framework of multi-agent systems.