Inertial measurement unit slowly-changing fault detection method and system in dynamic environment and storage medium

By combining Gaussian process regression and particle swarm optimization, the dynamic output envelope of the inertial measurement unit (IMU) is predicted and the extreme values ​​of multidimensional statistical characteristics are calculated. This solves the problem of identifying slowly changing faults in IMUs and achieves fault detection with high accuracy and low false alarms.

CN122329362APending Publication Date: 2026-07-03PLA PEOPLES LIBERATION ARMY OF CHINA STRATEGIC SUPPORT FORCE AEROSPACE ENG UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202510998636.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-18
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Existing technologies are difficult to effectively identify slowly changing faults in inertial measurement units in dynamic environments. Traditional methods are susceptible to noise interference, have a high false alarm rate, and are highly dependent on fault samples, making it difficult to achieve high-precision detection in practical applications.

Method used

By combining a Gaussian process regression model with a particle swarm optimization algorithm, the dynamic output envelope is estimated using normal data from training the inertial measurement unit. The extreme values ​​of multidimensional statistical features are calculated and used as fault detection thresholds to determine whether a signal is faulty in real time.

Benefits of technology

It achieves high-precision detection of slowly changing IMU faults in dynamic environments, with an average fault detection rate of 97.80%, reducing false alarm rate and missed alarm rate, and is suitable for practical application scenarios with sparse fault samples.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an inertial measurement unit slowly-changing fault detection method and system in a dynamic environment and a storage medium. In order to solve the technical problems of difficulty in slowly changing fault identification and high missed alarm rate and false alarm rate in an inertial navigation system, a Gaussian process regression model and a particle swarm algorithm are introduced to carry out multi-dimensional feature estimation of an inertial measurement unit; a dynamic envelope model of a normal signal is constructed through a Gaussian process regression model, a multi-dimensional statistical characteristic threshold is optimized in an envelope by using a particle swarm algorithm, high-precision detection of a slowly varying fault can be realized without fault data, and the method is suitable for a small-data-volume and high-noise scene. Experiments prove that the method can meet the diagnosis requirements of various faults of the IMU, can be applied to different types of IMUs, and is better in generalization ability and higher in fault diagnosis accuracy.
Need to check novelty before this filing date? Find Prior Art