A Novel Antagonistic Weapon-Target Assignment Model Considering Uncertainty and its Solution Using Decomposition Co-Evolution Algorithm

Abstract

The weapon-target assignment (WTA) problem is a crucial decision issue in the process of cooperative aerial warfare (CAW). The decision strategy of fighter teams involved in the CAW is susceptible to the influence of the enemy fire attack and electronic interference, which will lead to both the antagonism and uncertainty of the decision making. In this paper, a novel antagonistic game WTA (AGWTA) model with uncertainty is introduced. The antagonism is described by a non-cooperative zero-sum game model conducted by two fighter teams. Then, a modified sensor data fusion method using belief entropy and similarity of sensor data is presented to manage the uncertainty of AGWTA. According to the characteristics of the AGWTA model, a decomposition co-evolution algorithm (DCEA-AGWTA) is proposed to obtain the non-cooperative Nash equilibrium (NCNE) strategy. The experimental results show that the modified sensor data fusion method contributes to higher reliability of target type identification and the AGWTA model is meaningful in the antagonistic and uncertain situation of CAW. In addition, the DCEA-AGWTA is effective and has a promising ability in finding the closest strategy to the NCNE strategy compared with the other three intelligent evolution-based algorithms.