Publication
In this work a solution for the design of an assistive system for both muscular effort compensations and muscular effort generations for physical and rehabilitation tasks is presented. The proposed human-in-the-loop (HITL) control directly exploits the subject muscle sEMG signals measures to produce a specified and repeatable muscular response, without the need for human joint torque estimations. A set of experimental tests addressing different assistive tasks are proposed to validate the control design. Moreover different robotic devices, both grounded and wearable, are considered to assess the control under different working scenarios. The experimental results, involving four healthy subjects, show the efficacy of the proposed approach and the successful compensation/generation of the subject effort in the different assistive tasks considered.