In order to solve the problem of redundant torque in the electric load simulator of the gun control system, the neural network adaptive sliding mode control is studied. Combined with the sliding mode controller features, establish electric load simulator system model, the controller of the complex nonlinear system is established by using sliding mode controller. Based on combined control method of RBF neural network and sliding mode, the RBF neural network can make adaptive approximation of system perturbation parameters and un-modeled dynamics, which can reduce the switching gain and effectively suppress chattering. Carry out simulation analysis and verification. The simulation results show that the control strategy has high control precision and good robustness, which meets the system control requirements.