Group target firepower distribution decision-making method based on deep reinforcement learning

The invention provides a group target firepower distribution decision-making method based on deep reinforcement learning. The method comprises the steps of obtaining to-be-input information; inputting to-be-input information into the trained deep reinforcement learning network model to obtain a firepower distribution result; a construction method of the trained deep reinforcement learning network model comprises the following steps: acquiring a multi-source information packet, and performing data analysis and fusion to obtain fused data; calculating a threat coefficient of each unmanned aerial vehicle based on the multi-source information packet; sorting the threat coefficients of the unmanned aerial vehicles to obtain a threat assessment result; based on the multi-source information, obtaining a defense party equipment action vector; based on an action shielding vector formula, obtaining a defending party equipment action shielding vector according to the defending party equipment action vector; and based on