Integrated Guidance-and-Control Design for Three-Dimensional Interception Based on Deep-Reinforcement Learning

This study applies deep-reinforcement-learning algorithms to integrated guidance and control for three-dimensional, high-maneuverability missile-target interception. Dynamic environment, reward functions concerning multi-factors, agents based on the deep-deterministic-policy-gradient algorithm, action signals with pitch yaw fins as commands were constructed in research, which missile order intercept targets. Firstly, missile-interception system includes dynamics such inertia of missile, aerodynamic parameters, fin delays. Secondly, improve convergence speed accuracy, a factor angular velocity target line sight deep dual-filter methods introduced into design function. The method proposed this paper was then compared traditional proportional navigation. Next, many simulations carried out targets different initial conditions by randomization. numerical-simulation results showed that strategy has higher accuracy stronger robustness generalization capability against parameters.