Servo System PID Control of Neural Network Algorithm Based on LuGre Model

Considering the parameters of the traditional PID control algorithm are not capable of online adjustment and do not have adaptive capability once they are determined, which can not satisfied the precision control requirement of servo system with complex nonlinear friction, a PID controller based on radial basis function neural network (RBFNN) is designed, the advantages of RBFNN such as infinity approaching nonlinear system, little operation quantity and speedy constringency are combined with PID control technology organically to obtain high position tracking accuracy. The simulate results show that compared with common PID control, the RBFNN PID controller can adaptively adjust the controller parameters, greatly improving the control precision of the system.