Integrated guidance control with sliding mode differentiator and neural network

To improve the performance of the missile interceptors against modern air-defense threats such as ballistic missiles, a new integrated missile guidance and control with sliding mode differentiator and neural network algorithm is proposed in this paper. The three dimensional target and missile relative kinematics are firstly established based on which the zeroing line-of-sight (LOS) angular velocity interception model is developed. In order to negate the effects caused by the target maneuvers, a sliding mode differentiator is used to estimate the target acceleration along the Y and Z axis of the LOS coordinate. To achieve adaptive performance, neural network is adopted to compensate for the modeled and unmodeled uncertainties of the missile by updating the weight matrices and gains. The stability of the proposed algorithm is proven based on the Lyapunov theory. The six degree of freedom (6DOF) nonlinear numerical simulation results show that the algorithm can ensure hit-to-kill performance and that the sliding mode differentiator can perfectly observe the target maneuvers and the robust stability of neural network compensation.