A Guidance Law for the Boost Phase of Dual-Pulse Missile Based on Neural Network

To address the challenges of pulse interval optimization and guidance law design for long-range air-to-air missiles equipped with dual-pulse rocket motor, a fast calculation method for the pulse interval and a robust boost-phase guidance law are proposed in this paper. To reduce the complexity of solving for the pulse interval, the original problem is transformed into one of determining the predicted impact point (PIP) and the corresponding flight time. Under given conditions of the PIP, pulse interval, and trajectory constraints, the Radau pseudospectral method is employed to search for an optimal set of trajectories. Subsequently, an offline lookup table is employed to calculate the pulse interval. The influence factors on the optimal trajectory are analyzed, and the main variables are determined, which are then used to fit the nonlinear control in the boost phase based on a back propagation neural network. Based on this BPNN, a boost-phase guidance law is developed by incorporating flight path angle feedback. Simulation results demonstrate that the proposed method exhibits strong generalization capability and robustness.