| Summary: | 碩士 === 國立高雄應用科技大學 === 機械與精密工程研究所 === 98 === High precision motion has become to be the most important part of mechanical systems in present. The traditional linear control had not been to provide the satisfying control result when the requirement of performance became stricter. Although many control methods have been proposed, system parameter uncertainty, external interference, and high-order dynamic problems are still to be overcomed to improve the performance. This dissertation combines neural networks and adaptive control to deal with external disturbances of a nonlinear system with friction. Furthermore, we use backstepping design method to derive an adaptive law of inverse backlash parameters to estimate backlash parameters. Through Lyapunov stability analysis, it is proved that the designed control method can ensure the stability of the controlled system. Finally, some computer simulations are provided to demonstrate and evaluate the control prerformance of the designed controller in this dissertation.
|
|---|
