Abstract-The traditional modeling method of multi-Agent system (MAS) for weapon systems is to build the underlying Agent unit in micro perspective, and then integrate the units to make a larger system. The MAS built by this way cannot show the emerged characteristics of the macro system in the simulation, the effectiveness has been questioned. To solve this problem, the traditional weapon system Agent modeling method has been improved, to put forward an emergence-imitation-based MAS modeling method. Firstly, the framework of Agent combat capability model is designed to meet the needs of system simulation. Then, a method of constructing Agent behavior model based on emergence is proposed. Lastly, the method is used to build an MAS model for an armored brigade. The simulation results demonstrate that the MAS model constructed by this method can produce emerged characteristics of real weapon systems in the simulation, which proved the effectiveness of the method.

Abstract. In order to improve the accuracy and reliability of the multi-sensor data fusion, a new modified reciprocal fuzzy neartude based approach to calculate the weights of the fusion model is proposed. Through the research of the fusion model, it is found that the fuzzy neartude is more practical than fuzzy membership for the calculation of the weights. The fusion performance of the five types of fuzzy neartude is analyzed, and the reciprocal fuzzy neartude is proved to be one of the best for the resolution and mount of calculation. However, the neartude cannot suppress the singular data well. To address the problem, the reciprocal fuzzy neartude is modified. The simulation analysis shows that the modified reciprocal fuzzy neartude based approach can fuse the multi-sensor data with high accuracy and reliability.