Multiple sets of INS fault detection based on information fusion

As measurement noise exists in the output of gyroscopes and accelerometers in practical systems, traditional fault detection methods based on hardware redundancy easily cause misjudgment after determining the threshold. This paper presents a fault detection algorithm based on information fusion of multiple sets of inertial navigation system (INS). Combined with wavelet analysis, firstly fuse the information with weighted average algorithm, then reduce the noise interference of the error by the given threshold denoising method, finally, use the maximum coefficient of wavelet transform to determine the location of failure. The experimental results show that the proposed algorithm can effectively reduce the misjudgment rate and greatly improve the reliability of fault detection and the stability of fusion performance of multiple sets of INS.