In order to solve the problem that the trajectory of mobile robot is easily disturbed by uncertain external factors, the control system of mobile robot was designed by using the inverse neural network model. The inverse neural network controller and the traditional PI controller model were introduced to track the velocity and angular velocity of two wheeled differential mobile robot. In the traditional PI controller model, approximately linear equivalent load driver was used, while in the inverse neural network control,the feedforward multilayer perceptual neural network model was used. The inverse neural network control model combined the mathematical model of kinematics and dynamics, and it was trained discretely in a specific working area. In the plane, the speed tracking control of the mobile robot was simulated. The results show that with the PI controller, the wheel speed and angular speed of the mobile robot have a large error with the theoretical value, while the error is small when the inverse neural network model is adopted. By using inverse neural network model to design the speed control loop of mobile robot, the performance of the mobile robot can be improved and it can better adapt to the changes of the external environment.