Aiming at the problem of autonomous evasion of carrier aircraft facing enemy incoming missiles, a deep reinforcement learning method based on the improved DDPG algorithm is adopted to train and learn the problem, and in addition to considering the evasion performance in the reward function, it also focuses on the original The aircraft's altitude maintenance and speed maintenance, as well as the relative altitude change and approach speed change of the incoming missile, are used to establish a reward model. Finally, a training simulation test analysis was carried out based on the aircraft model. Through simulation, it can be seen that the training results can effectively realize the evasion decision of the incoming missile, and the designed reward function and input parameters can also play a correct role, and the results are available Certain generalization ability.
|