Cooperative Target Allocation for UCAV Team Air-to-Ground Attack Based on Decision Graph Bayesian Optimization Algorithm

Abstract

In order to control and optimize cooperative air-to-ground attack decision-making of the unmanned combat aerial vehicle (UCAV) team, the principle of income maximum and loss minimum of UCAV team is built firstly. Accordingly, the mathematical model of cooperative target allocation is built based on the decision variables and constraints. Then Bayesian optimization algorithm (BOA) is introduced which is one kind of the evolution algorithm. For improving the ability of the BOA, decision graph is introduced to enhance the represent and learn of Bayesian network and compress the parameter saving. Finally decision graph Bayesian optimization algorithm (DBOA) is utilized to optimize and analyze the model. The simulation results verify that the mathematical model of cooperative target allocation can reflect the importance of cooperative decision-making, the DBOA can converge quickly to the global optimal solution and can effectively solve the cooperative target allocation problem of UCAV team air-to-ground attack.