中文摘要：

在组合导航系统故障检测中，针对残差χ检验法对于检测缓变故障的不足，提出了用基于卡尔曼滤波模型的速度检测模块的输出值辅助残差χ2检验，并采用模糊逻辑和加权平均来处理检验阈值，从而更有效地检测缓变故障，并在GPS／JTIDS／IFDL／INS组合导航系统中各个子滤波器中进行仿真验证，在加入不同大小的缓变故障时与残差χ2检验法进行了比较。仿真结果表明，该方法有效缩短了检测延迟，提高了故障检测的灵敏度，并且此优势在故障越小时，体现越明显。因此，改进残差χ2检验法提高了组合导航系统对于缓变故障的检测效率，使系统的可靠性得到了改善。

英文摘要：

In the integrated navigation system failure detection,in view of the residual chi square test for detection of graded fault insufficiency, proposed based on Kalman filtering model with the speed of detection module output value of the auxiliary residual chi square test,and the fuzzy logic and the weighted average to deal with inspection threshold, in order to achieve a more effective detection graded fault,and in the GPS/JTIDS/IFDL/INS integrated navigation system in each child filter in the simulation test and verify,in join different size of the graded fault with residual chi square test were compared. The simulation results show that this method is effective shorten the detecting delay and improve the fault detection sensitivity, and the advantage in the fault more hours, reflect the more obvious. Therefore, improving the residual chi square test to improve the integrated navigation system for graded fault detection efficiency, make the reliability of the system is improved.