Hidden failure prognosis for navigation inertial platform using belief rule base expert system

The gyro drift is the main factor in affecting the performance of the inertial platform which is an important unit in the navigation system. The belief rule base (BRB) is used to construct the failure prognosis model of the inertial platform which holds the following characteristics: (1) the storage lives are long and working lives are short, (2) a lot of testing data can not be obtained, (3) some expert knowledge may be available, and (4) the failure is hidden. In order to solve the problems, two sub-BRBs are used with one employed to capture the relationship between the external stress and hidden failure and the other to construct the forecasting model of hidden failure. An experimental case study is examined to demonstrate the implementation and potential applications of the proposed failure prognosis model.