Unmanned network communication resource optimization method and system

The invention belongs to the technical field of network communication resource optimization, and discloses an unmanned network communication resource optimization method and system based on multi-target deep reinforcement learning, and the method employs a multi-target deep reinforcement learning Deep Description Policy Gradient algorithm (MORL + DDPG), employs the data feature extraction capability of deep learning and the decision-making capability of reinforcement learning, achieves the optimization of the communication resource of an unmanned network, and achieves the optimization of the communication resource of the unmanned network based on the multi-target deep reinforcement learning. The optimal balance between the throughput of the AOI of the CCH and the throughput of the TCH is realized in the unmanned vehicle network. According to the method, the GPI thought is used, the GPI considers a group of strategies interacting with the same environment, and the goal is to select and improve strategies super