A Vibration Suppression Method of Series Elastic Actuator Based on Particle Swarm Algorithm

Abstract

Series elastic actuator (SEA) are widely used in exoskeleton robots. For its vibration suppression problem, this paper proposes a method of "planning and controlling" in the SEA system. First, according to the kinematics constraint equation of the connecting rod, the position of motor is planned based on the second-order low-pass filter, which used particle swarm algorithm to optimize the filter parameters in the process. Second, a position tracking controller is used to make the motor accurately track the desired trajectory according to dynamical model of the motor. The optimal filter is finally found to plan the motor trajectory by iterative simulation based on particle swarm algorithm. The experimental results show that this "planning and controlling" method can effectively suppress residual vibration while ensuring that the connecting rod reaches the desired position accurately.