No abstract

No abstract

The main purpose of this article is to propose a linear model-based approach for improving the tracking performance of linear motors. In particular, P-and PI-controllers, augmented with a linear-model-based feedforward controller and disturbance observer (DOB), were utilized in this research. Experimental results show that the feedforward controller can substantially reduce the phase-delay problem resulting from the limitation of closed-loop bandwidth. In addition, the DOB can reject nonlinear behaviors to keep a motor-system close to our nominal linear model. Compared with the results of using P-and PI-controllers only, tracking errors were reduced from 4.44% to 0.67% of the desired trajectory. Therefore, the proposed method provides a solution for improving the tracking performance of linear motors.

The primary purpose of this study is to improve the positioning performance of shape-memory-alloy (SMA) actuators. In order to achieve this goal, the system nonlinearity was reduced with the inversion of a nonlinear model. The system could then be approximated with a linear model. It is easy to construct the corresponding model-reference-adaptive-system (MRAS) based on this linear model. Experimental results show that the MRAS is robust with respect to external disturbances and improves the positioning performance. In addition, with the proposed control scheme, the simulation results will closely match experimental results, which is useful to predict the system performance at the controller-design stage.

No abstract