Design and implementation of a fuzzy output feedback assistive controller for a series-elastic-actuator-driven knee exoskeleton

The series elastic actuators make more comfort in the use of assistive exoskeletons. In this paper, an assistive controller is designed for a series-elastic-actuator-driven knee exoskeleton to restore normative mobility of individuals with weak muscles. The main target of the proposed controller is to modify the dynamics performance of the coupled human-exoskeleton system. In other words, the proposed controller modifies the relationship between the net muscle torque exerted by the human and the resulting angular motion. There are fewer sensors in the proposed intent-independent method relative to other methods. Moreover, there are less controller coefficients to regulate where these coefficients are extracted from a type zero Takagi-Sugeno-Kang fuzzy system. The performance of the controller is evaluated by simulations and experiments. The amplitude of the EMG signals decreased in a healthy person worn the SUT-KneeExo. Moreover, the proposed algorithm has a better performance in comparison with integral admittance shaping mothed and output feedback assistive controller. In other words, the amplitude of the integral admittance is more and the phase lag is less than other methods.