Multi-failure isolation method for redundant strapdown inertial navigation system

Abstract

The invention discloses a multi-failure isolation method for a redundant strapdown inertial navigation system. The method comprises the following steps: acquiring output data of inertial devices of the redundant inertial navigation system at first, performing failure detection by virtue of a generalized likelihood ratio method, estimating predicted values output by the inertial devices at failure moments by virtue of a linear estimation method when failures are detected to occur to the redundant strapdown inertial navigation system, finally comparing the predicted values of the inertial devices with output values to obtain difference values, locating the failing inertial devices, and isolating the failing inertial devices. According to the method, the generalized likelihood ratio method is combined with the linear estimation method, the characteristics of high sensitivity and convenience for implementation of the generalized likelihood ratio method and small calculated amount, high accuracy and the like of the linear estimation method are fully utilized, and the failing inertial devices are timely and accurately isolated when a plurality of inertial devices of the redundant strapdown inertial navigation system simultaneously fail, so that the reliability of the inertial navigation system is ensured.

Description

technical field [0001] The invention relates to the field of strapdown inertial navigation systems, in particular to a redundant type strapdown inertial navigation system multi-fault isolation method. Background technique [0002] The aerospace and aviation fields attach great importance to the reliability of autonomous navigation systems, and improve the reliability of inertial navigation systems through system-level redundancy or device redundancy. Therefore, redundancy technology has become the most commonly used system measure to improve the reliability of inertial navigation systems. Failure of any accelerometer or gyroscope inertial sensor in a redundant SINS could have serious consequences. In order to prevent the navigation task from being paralyzed due to a fault in the navigation process, it is necessary to detect and isolate the fault in the operating state of the redundant SINS to ensure the safety and reliability of the navigation process. [0003] Fault detect...

AI Technical Summary

Problems solved by technology

Although the singular value decomposition method proposed by Duk-Sun Shim et al. in "Geometric FDI based on SVD for Redundant Sensor Systems" (published in "Asian Control Conference, 2004, Issue 02") can successfully detect and isolate redundant SINS Double faults in the system, but the amount of calculation is large, the delay in fault detection is large, and it is impossible to accurately isolate the situation where two or more inertial devices fail at the same time

Images