基于模糊神经网络的多机多目标空战决策

摘要

A decision-making approach based on fuzzy neural network was put forward for air combat of multi-aircraft against multi-target. For simplicity and considering of the air combat environment, regional environment and so on, the large-scale formation air combat was divided into combats within small groups. The missiles were assigned according to our bomb loads and the threat degree of enemy with consideration of our secondary attack. After implementing one attack, the loss of both sides was calculated out, and air combat situation of both sides were analyzed to make decision for the second attacking. To verify the correctness of the algorithm, simulation was carried out by using Matlab. The result shows that the fuzzy neural network can take the advantages of a neural network and fuzzy theory, which is effective for dealing with the multi-objective decision-making problem in air combat.%提出了一种基于模糊神经网络的多机协同多目标空战决策的方法.为简单起见且在不影响战局的情况下,从空战中的环境、地域等方面考虑,将大规模的机群空战划分为小集团内部的小机群作战.根据我机的载弹量和敌机的威胁度,进行分配导弹,分配原则考虑了我方的二次攻击.一次攻击后,计算敌机和我机的损失,再次对敌机和我机进行空战态势分析,进行第二次攻击决策.为验证算法的正确性,采用Maflab软件进行仿真,仿真结果表明,模糊神经网络能发挥神经网络和模糊理论的优点,有效地解决多目标空战决策问题.