Objetive: In order to improve the actuator precision，a method for eliminating nonlinear no-smooth hysteresis characteristic of piezoceramic actuator is proposed. Method :An inner product-based dynamic neural network nonlinear and no-smooth hysteresis inverse model for piezoceramic is obtained

in which the feedback error learning method is used to avoid obtaining Jacobian information by model for piezoceramic. On dSPACE system platform，the piezoceramic is controlled by combined with a PID control and an neural networks adaptive inverse control. In order to satisfy the requirement of real time control

the programs is designed by a high efficiency and fast C-MEX S function. Result: Experimental results indicate that the precision of the proposed adaptive inverse control based on neural networks is 0.13μm

PID control precision is 0.32μm. Conclusion: The proposed control method removes efficiently the hysteresis characteristic of piezoceramic and of higher control precision.