Assist-as-needed Policy For Movement Therapy Using Telerobotics-mediated Therapist Supervision

In this paper, a new impedance-based teleoperation strategy is proposed for assist-as-needed tele-rehabilitation via a multi-DOF telerobotic system where a patient and a therapist interact with the master and slave robots, respectively. Unlike a regular teleoperation system, the therapist’s hand only follows the patient’s deviation from the target trajectory. The admissible deviation of the patient’s limb from a reference target trajectory is governed by an impedance model and related to the patient’s and therapist’s applied forces. In this framework, two sources of assistance to the patient delivered through the master robot are the adjustable impedance model and the force applied by the therapist to the slave robot. Bilateral nonlinear control laws with two types of adaptation laws are designed for the nonlinear teleoperation system. The Lyapunov stability proof of the teleoperation system and the stability of the impedance model enhance the patient’s and therapist’s safety even in the presence of structured (parametric) and unstructured (non-parametric) modeling uncertainties in the multi-DOF telerobotic system. The performance of the proposed bilateral impedance-based strategy is experimentally investigated using different impedance parameters adjusted based on the patient’s characteristics (e.g., involuntary tremor) and disabilities (e.g., insufficient actuation forces). The experiments are performed by a healthy person (as the therapist), a mechanical test-bed and a volunteer (simulating the patients’ characteristics).