Intelligent decision-making method of UAV cooperative operation based on situation awareness

In this paper, an intelligent decision-making method for UAV cooperative operations based on situational awareness is proposed. The overall framework of this method includes situational awareness module, information fusion module, intelligent decision-making module and task execution module. Through the cooperation of all modules, real-time awareness of battlefield situation, efficient information fusion, intelligent decision-making and accurate task execution can be realized. In the aspect of situation awareness and information fusion, UAV obtains real-time battlefield situation information through various sensors, and uses D-S evidence theory to fuse multi-source data, effectively dealing with redundancy and conflict between data, and improving the accuracy and reliability of target detection. In the design of intelligent decision-making algorithm, a decision-making algorithm based on deep reinforcement learning (DRL) is proposed, and the operation strategy of UAV group is optimized by Q-learning method, so that it can learn independently and make the best decision in complex environment. In the stage of task allocation and target cooperation, the task allocation method and consistency algorithm based on auction mechanism are adopted to realize efficient resource allocation and cooperative operation of UAV group and ensure the efficient completion of tasks. The experiment compares the rule-based decision-making algorithm, the traditional Q-learning algorithm and the method proposed in this paper by constructing typical combat scenarios such as static target attack, dynamic target tracking and complex electromagnetic environment. The results show that this method is superior to the traditional methods in task completion rate, response time, resource consumption rate and collaborative error, especially in dynamic and complex scenes. This research provides new ideas and methods for intelligent decision-making of UAV cooperative operations, and has important theoretical and application value.