摘要
为了提高水下航行器组合导航系统长时间远航程水下导航的可靠性和精度,采用了捷联惯性导航系统、多普勒速度声纳、地形匹配模块和TCM2电子磁罗经构成组合导航系统。采用容错联邦卡尔曼滤波对水下航行器组合导航系统进行信息融合、故障诊断与系统重构,在设计联邦滤波算法基础上,结合组合导航信息融合方案,详细分析了χ2故障检测方法,给出了水下航行器容错联邦滤波结构和故障检测算法,进行了仿真实验,并对仿真结果进行了分析。仿真结果表明:联邦滤波方法对系统故障能够及时检测并且有效隔离了故障传感器,对系统进行了重构,提高了系统的导航精度和可靠性。
To improve the long-term precision and reliability of integrated navigation system of an underwater vehicle during a long voyage, an system which integrated the SINS, Doppler velocity sonar（DVS）, terrain-aided navigation block and TCM2 magnetic compass, and a fault-tolerant federated Kalman filter was used to fuse the various navigation sensors, detect the system fault and reconstruct the navigation system. χ^2 fault-detecting method was analyzed in detail, and simulation experiments were carried out based on the federated filtering method and information fusion architecture. The simulation results were analyzed in detail. Simulation experiments results show that the fault-detection and system reconstruction in time can effectively isolate the fault sensors and improve the stability and precision of the navigation system.
出处
《中国惯性技术学报》
EI
CSCD
2008年第2期167-170,共4页
Journal of Chinese Inertial Technology
基金
"十一五"总装备部预研项目(51309020503)
国防973项目(973-61334)
国家自然科学基金项目(50575042)
教育部博士点专项科研基金项目(20050286026)
关键词
水下航行器
联邦卡尔曼滤波
信息融合
故障诊断
underwater vehicle
federated Kalman filter
information fusion
fault diagnosis
