A sliding mode control strategy based on RBF neural network and fuzzy switching gain regulation is proposed to solve the stability problem of vehicle-mounted weapon firing while moving. The mathematical model of the vehicle-mounted weapon servo system is established, and the sliding mode controller based on the new reaching law is designed to make the system converge to the equilibrium state quickly. The fuzzy control is integrated on the basis of the sliding mode control, and the fuzzy rules are used to adjust the switching gain of the controller in real time. The switching gain is used to eliminate the disturbance of the system and weaken the system chattering; RBF neural network is used to adaptively estimate the time-varying items of the system to improve the control precision. The simulation results show that the designed controller is insensitive to the disturbance, and can effectively improve the position accuracy of the system, weaken the chattering, and make the system have strong robustness.