A Multisensor Navigation System Based on an Adaptive Fault-Tolerant GOF Algorithm

This paper describes an adaptive fault-tolerant multi-sensor integrated navigation system. The proposed system uses a decentralized filtering architecture to fuse inertial navigation system (INS), GNSS, and Locata sensor subsystems. In order to improve system accuracy, the global optimal filtering (GOF) algorithm is implemented. The GNSS and Locata subsystems are separately integrated with the INS to obtain the local prediction and local estimation based on the GNSS/INS and Locata/INS combinations. The GOF algorithm is then applied to fuse the local and global information to generate the optimal state estimation of the GNSS/Locata/INS navigation system. The adaptive fault-tolerant algorithm is based on the innovation covariance discrepancy, which mainly adapts to the changes in sensor measurement statistical properties and mitigates the adverse influence caused by these changes. It is found that the GOF algorithm does improve the accuracy of the navigation solution compared with conventional filtering. To evaluate the fault-tolerant ability of the proposed system, a series of GNSS failures is simulated. The results show that the proposed system can mitigate the effect of the failures, which verify the higher reliability and fault-tolerant capability of the proposed system.