Decision Support for Beyond Visual Range Air Combat Based on HTN Planning and Machine Learning

Beyond visual range air combat (BVR) has become the main form of modern air combat. In order to solve the problem of pilot information overload in BVR air combat, this paper designs an assistance decision-making model based on Hierarchical Task Network (HTN) planning and machine learning algorithms to provide pilots with tactical strategies. This paper first designs a hierarchical task network in a BVR air combat scenario, as well as an efficient scoring system that can dynamically adjust the mission execution plan based on the situation in real time. Then a machine learning algorithm combined with data generalization was proposed to predict the launch effectiveness of missiles in different situations and optimize part of the scoring process. Finally, a human-in-the-loop confrontation simulation experiment was conducted, and the feasibility of the model and the effectiveness of the algorithm were verified by analyzing the winning rate.