Sensor Fault Diagnosis for Autonomous Underwater Vehicles
Sensor Fault Diagnosis for Autonomous Underwater Vehicles
复制标题
DOI：
10.1166/sl.2011.1535
复制
发表时间：
2011-10
期刊：
影响因子：
--
通讯作者：
Xiao Liang;Jun-dong Zhang;Wei Li-;Jianguo Lin
中科院分区：
文献类型：
--
作者：
Xiao Liang;Jun-dong Zhang;Wei Li-;Jianguo Lin
Sensor reliability is one of the preconditions for autonomous underwater vehicles (AUVs) achieving their tasks successfully, and sensor fault diagnosis system has been a line of defense for AUVs miss operation. Aiming at sensor faults which may appear on AUVs, novel fault diagnosis methods corresponding to three kinds of sensor faults are proposed in this paper based on analyzing fault forms for AUV sensors. By monitoring signal gather, we diagnose sensor faults of the gathered signals keeping unchanged if there have be not sensor output data for a long time. We adopt linear smoothness to solve output vibration of the fixed position sonar. We measure velocity information by using Doppler Velocity Log to estimate position information. We judge there is a fault if their difference is greater than the given threshold value by comparing the measured value and the estimated value, and we substitute the estimated value for the measured value of the fixed position sonar. We also adopt wavelet transform to detect accident signals. Finally, sea experiments are carried out to verify the feasibility. The results prove the means are valid enough, and we can achieve fault diagnosis for AUV sensors successfully.
