Volume 31, Issue 5 (September 2003)
The Reliability and Self-Diagnosis of Sensors in a Multisensor Data Fusion Diagnostic System
Any error in a data fusion system that is used to diagnose and detect machine faults, especially an error in sensor measurement information, will result in an incorrect diagnostic conclusion. The accuracy and reliability of sensor measurement is a prerequisite for reducing the uncertainty of the fusion system and accurately diagnosing system faults. The measurement error of the multisensor and the degree of reliability of the detecting system can be determined by a multisensor consistency check. In this paper, we use the “probability distance measure” to determine the compatibility of multisensor information and to ensure that the measurements represent the same target (or state). We show our experimental results, which demonstrate a good result.