Feedback Linearization of an Electrostatic Actuator by Particle Swarm Optimization

A method of searching for an input to state linearizing controller is presented. The problem of finding the appropriate weights of the control law´s terms is treated as an optimization problem. Given the highly nonlinear surfaces that are likely to be searched, particle swarm optimization is applied. This effort serves to explore the use of particle swarm optimization as a means of tuning a nonlinear controller. The MEMS parallel plate actuator is used to examine the effectiveness of this technique. The resulting tuning of the controller is then compared to that of the analytically designed solution. Using this comparison, conclusion are drawn as to the viability of this technique for partially automated design of a linearizing control