Multi-missile Path Planning algorithm based on Reinforcement Learning.

Aiming at the problem that the on-line path planning of multiple targets hit by multiple missiles requires high real-time performance and optimal results of the algorithm, the Deep Q-network (DQN) algorithm is designed and used based on shared experience. Firstly, the Markov decision-making model of multi-missile path planning is established. Secondly, the DQN theory based on shared experience is established to design the multi-missile path planning algorithm. Finally, the DQN algorithm is simulated and verified in multiple scenarios. The simulation results show that the DQN algorithm can make multi-missile bypass the threat and get a better path to achieve online path planning.