Tuning of Control Motion for a three link robot manipulator using Particle Swarm Optimization Technique

In this paper, the control motion of a three link robot manipulator with un-powerd first joint is discussed. The subjected system is a highly nonlinear, underactuated system. It mimics the acrobat which is trying to move from the down position to upside position through pumping energy to the two DC motors which are located on powered joints (active joints). The main challenge of this study is to show how to tune the control input signals applied to the two DC motors in order to move all robot links near neighborhood of the upright balance point. The optimum values of the control actions were obtained by using the Particle Swarm Optimization (PSO) algorithm. The objective function of this optimization method was determining the reasonable time to move the system to the desired upright position. The swing-up of the system was successfully achieved and the simulation outcomes showed the efficiency of the suggested control method.