为了提高三视场定位定向设备可靠性，建立三视场系统功能的基本约束条件测试模型，以模型为依据对可见光波段的视场进行优化分析。首先，分析了设备正常完成工作时采用组合星图识别和简单星图识别所需的必要条件，并建立这两种识别方式关于视场大小的概率分布模型。从原理和实验数据两个方面说明了三视场结构识别三角形概率分布不能通过单视场的概率分布简单推导获取的原因。接着，给出用蒙特卡洛法求取此概率分布和视场最优值的方法。然后，通过仿真分析得出了两种识别方法关于视场大小的概率分布，在分析两种识别方法仿真数据的基础上，取识别概率为1时视场大小为其优化值。实验结果表明：以组合星图识别为工作基础的三视场定位定向设备有更好的可靠性，优化的视场大小为4.225°×3.168 75°。它满足三视场定位定向设备识别条件可靠性要求。

In order to improve the three FOVs position and orientation determination equipment reliability, the basic constraints test probability distribution model of the three FOVs system was established. Based on the model, the optimized FOV size of the equipment was analyzed on the visible band. First, the necessary conditions of the composite FOVs star identification and the sample FOV star identification were analyzed, and the probability distribution models of the methods were established. The reason why probability distribution of recognized triangles on three FOVs equipment cannot be simply deduced from the one FOV′s was illuminated in theory and simulated experiment. Then, the optimal value of the probability distribution was obtained by the monte Carlo method. Finally, the optimal design of the FOV size was presented by computer simulation. The experimental results show that: the three FOVs celestial navigation equipment using the composite FOVs star identification is more reliable. The optimal FOV size is 4.225°×3.168 75°. It can satisfy the reliability requirements of the three FOVs position and orientation determination equipment.