Automatically tuned hybrid motion controller in the presence of mechanical non-linearities

This paper presents a study of the control of mechanical nonlinearities in a 2-axis XY-table. A model is derived for the system with nonlinear coupling. Conventional control techniques such as PI and P control are compared with an automatically tuned hybrid fuzzy and disturbance observer controller. A genetic algorithm is used to tune the fuzzy controller. These designs are then tested and compared. The results showed that the fuzzy controller improves the performance of the disturbance observer controller by 5%. The improvement was measured with respect to stiction duration and RMS error. The paper finally presents some discussion of possible expansion of the techniques such as a MRAC-fuzzy controller.