Robust Adaptive Guidance Law with Incomplete Information

For the guidance system in the presence of the acceleration saturation constraint and target maneuver, a robust guidance law is designed by employing adaptive dynamic surface control method in combination with neural network (NN). Treating the missile autopilot as second-order dynamics, adaptive dynamic surface control is utilized to reject the effect of external disturbance (i.e., target maneuver and neural network approximation error) as well as regulate the system states, and the neural network is leveraged to approximate the system nonlinearities. A reduced-order observer is also suggested to estimate the missile jerk to support guidance law implementation. The proposed algorithm needs less information and can be used for passive guidance. Simulation results reveal its satisfactory robust performance.