An Improved Integrated Navigation Fault Tolerance Algorithm Based on Dynamic Evaluation

Integrated navigation systems often need to work in harsh and complex environments. In this case, the navigation instruments may output wrong information, such as abnormal values and data loss. This will cause wrong information to participate in the filtering operation, which will affect the navigation accuracy. Aiming at the deficiency that traditional residual chi-square test is not easy to detect slowly-variable faults, this paper proposes a fault-tolerant algorithm for combined navigation faults based on dynamic evaluation. The algorithm uses Mahalanobis distance to establish a dynamic threshold evaluation method to achieve fault detection, and constructs a fuzzy fault-tolerant algorithm based on the T-S model, and uses an improved adaptive gradient descent algorithm to solve this issue to elevate the accuracy of fault diagnosis. The simulation results show that the algorithm can detect faults earlier than traditional methods, and the accuracy can be improved after fault tolerance.