Hbrid Force/Position Intelligent Control Based On X-Y Table

In the recent decades, many efforts have been devoted to thecontrol of X-Y table. Especially, problems of force/position controlfor X-Y table have attached many researchers and experts. Manyresearchers try many new theories and ways to it form differentpoints of view. It is known that there exists complex nonlinear,coupling and lots of uncertainties in X-Y table system. When it doesmachining the different environment stiffness have great affection onthe system's performance.In this dissertation, some intelligent control strategies on X-Ytable are mainly studied based on force/position hybrid control.A new control strategy based on iterative learning control ispresented based on X-Y table force/position hybrid control. Thedifference with the traditional iterative learning control is that afuzzy controller is inducted to adjust an important parameter Γonline to swell the maturity of iterative learning control strategy.Mostly, iterative learning control is used in position control inmodern researches about iterative learning control. However, theiterative learning control is inducted to force control loop accordingto the character of X-Y table. And it enhances the performance of thesystem by updating the errors of force and position every cycle toupdate foreback compensation. In order to conquer the fall short oflarge calculation, a strategy is presented by combining iterativecontrol compensation and self-adaptive fuzzy control. The controlarithmetic in position part is achieved by combining the traditionalPID controller and iterative control as error compensation. And akind of self-adaptive fuzzy control is used in force control loop. Theformat of control arithmetic is determined by inducting a switchfunction. Fuzzy the variable according to the fuzzy rule we designand then defuzzy it, we can achieve the control arithmetic. On thebase of all these, a new strategy of force/position hybrid control bycombing iterative control compensation and CMAC nerve net controlis introduced. The results of simulation show that all the proposedcontrol methods can obviously improve the impression of positionand force.