A motion phase-based hybrid assistive controller for lower limb exoskeletons

Human motions are realized not only from voluntary contractions of muscles, but also by the sophisticated and variable impedance characteristics of the muscles. Such variable impedance characteristics are also required in assistive robotics, in particular wearable robots, in order to help humans in a natural and comfortable way. As the human muscles are controlled to exhibit a certain impedance characteristic in each motion phase, an assistive controller, i.e., the controller of an assistive robot, should also be able to realize different impedance characteristics. Moreover, the different impedance characteristics should be seamlessly switched for the sake of comfort, safety, and stability of a human interacting with an assistive robot. In this paper, a hybrid assistive controller that incorporates various assistive algorithms enabling the continuous and smooth change of impedance characteristics is proposed. The proposed algorithm has many degrees of freedom in the viewpoint of controller functionality. For the verification of the effectiveness of the proposed method in practice, it is applied to a lower limb exoskeleton actuated by compact rotary series elastic actuators.