Nonlinear tracking controllers for brushless DC motors

Abstract

This paper summarizes recently developed nonlinear controllers designed to yield high-performance position tracking control of a brushless DC (BLDC) motor driving a mechanical load. The design procedure centers on the use of the integrator backstepping technique to extend nonlinear control design techniques to include the BLDC electrical dynamics. Six controllers are presented that represent a range in modeling and state measurement assumptions. A full state feedback controller, a partial state feedback controller, and an output feedback controller are discussed under the assumption of exact-model knowledge. An adaptive full state feedback, a robust full state feedback, and an adaptive partial state feedback controller are discussed for the case of parametric uncertainty in the electromechanical model. Five of the controllers have been successfully implemented on a state-of-the-art DSP based mechatronics testbed. The experimental results verify that a nonlinear model-based controller can exhibit greatly improved position tracking performance over that of a standard linear controller. >