Research on the Fault Diagnosis Methods for UUV Utilize Multi-sensor Information Fusion Technology

摘要

For UUV fault detection, we propose a multi-sensor information fusion diagnosis method based on D-S evidence theory. We use the feature information which is extracted by fuzzy membership function to construct the basic probability functions. Then use the weighted distribution method for basic probability functions to avoid the evidence conflict. Finally we use the combining method which was proposed by Murphy to combine the weighted probability functions and give out the fault diagnosis. In the final paper, we will give the performance evaluation by using the measured data.