An optimal control approach for hybrid actuator system

In this study, we consider the control problem of a hybrid actuation system that is composed of more than two different types of actuators. Even though actuators are one crucial component for developing a robot with many degrees of freedom, such as humanoid and exoskeleton robots, technical difficulties and costs complicate the development of new actuators. On the other hand, since different types of actuators provide various strengths and weaknesses, an efficient approach will probably combine different types of actuators to design a novel actuation system. The crucial issue for such a hybrid actuator is a method that distributes the necessary torque/force to different actuators for generating target movements. We propose using an optimal control method to find the torque distribution strategy for hybrid actuators. We consider an optimal torque distribution method for a pneumatic-electric (P-E) hybrid actuator model that is used in our newly developed exoskeleton robot XoR. We present a control approach with a trajectory-based optimal control method with minimum energy criterion. Focusing on the robot´s ankle joint, we apply our control method in numerical simulations and demonstrate that it reasonably finds the optimal torque distribution strategy based on the nature of the given task, and the results of high control performance.