Aiming at the problems of parameters uncertainty and time variability of electro-hydraulic servo system of mine sweeping device, a control strategy combining fuzzy control and neural network is designed by using the characteristics of strong robustness of fuzzy control and strong adaptive ability of neural network. The mathematical model of mine sweeper electro-hydraulic servo system is established, the structure of fuzzy neural network controller is analyzed, and the genetic algorithm and conjugate gradient method are used to optimize the learning algorithm; the control system model is established in Simulink, and the performance of traditional PID controller, fuzzy neural network controller and fuzzy neural network controller optimized by genetic algorithm and conjugate gradient is compared. The simulation results show that the optimized controller has fast response speed, strong robustness and stability.