Fault detection in SINS/CNS/GPS integrated system based on federated filter

In order to improve the reliability and accuracy of the SINS/CNS/GPS integrated system, a fault-tolerant integrated navigation algorithm based on federated Kalman filter is proposed. The architecture of the federated Kalman filter includes a master filter, two local filters, and two fault detection functions which are based on the residuals of the local filter. Because the choosing of threshold for fault detection is very important to fault-tolerant integrated navigation system, a new method of Dual-threshold detection is built and used in fault detection. The simulation results show that the proposed algorithm using Dual-threshold detection based on federated Kalman filter is effective, and the reliability and accuracy of the integrated navigation system are improved.