A residual detection method based on the maximum average residual x2 vector eigenvalue is proposed to address the shortcomings of the traditional residual x2 detection method in detecting faults in aircraft integrated navigation,such as missed detection of sudden faults and low sensitivity in detecting gradual faults.For the problem of overlapping failure probability density curves,fuzzy logic is used to generate dual inspection thresholds,and the detec-tion amount is normalized and weighted average processing to improve the sensitivity of slowly changing fault detec-tion.Based on this algorithm,the Aircraft SINS/BD/RP/TAN fault-tolerant integrated navigation system was simula-ted and verified.The simulation results show that the improved residual detection method can detect abrupt faults ac-curately and without delay compared with the traditional residual detection method,and it can also significantly im-prove the sensitivity of slowly changing fault detection.
万方数据