Singular-Perturbation-based Intelligent Midcourse Guidance for Air-to-Air Missiles Equipped with Dual-Pulse Motor

This paper presents an intelligent midcourse guidance law applicable to a long-range air-to-air missile equipped with a dual-pulse motor. Based on the traditional singular perturbation approach, deep reinforcement learning is introduced to generate secondary ignition commands and acceleration corrections, which enhance the performance of the guidance law at long guidance loop sampling periods and when parameters are not tuned. The approach of singular perturbation midcourse guidance is first reviewed, and then only one neural network is employed to perform the tasks of ignition and correction sequentially to save storage and computational resources. The simulation results show that the proposed method can be applied to attack long-range targets in beyond visual range air-to-air engagement.