Quantum Neural Network Algorithm Based on Multi-agent in Target Fusion Recognition System

The paper presents the quantum neural network algorithm based on multi-agent in target fusion recognition system. Firstly, it discusses the distributed cooperation and solution synthesis in MAS and describes the design of target recognition system. In that, the key techniques of task cooperation, task distribution and solution synthesis based on information fusion are demonstrated. Then quantum neural network is introduced to the MAS, by synthesizing the infrared and radar features of target from the 2 heterogeneous recognition agents, the membership function value is calculated, and the fusion membership function value is gained by using multi-laver excitation function. According to the fusion data, the true target is found out. The experiment shows its satisfactory performance and effectiveness. So the solution presented by the authors is valuable for designing distributed fusion target recognition system under heterogeneous sensors and target characteristics being imitated each other.