Inertial baseline integrated navigation method, and product, medium, device and autonomous underwater vehicle

By introducing a Lie group navigation state error model and a generalized maximum likelihood estimation model into the inertial-based integrated navigation system, and combining them with the robust Kalman filter algorithm for integrated navigation, the filtering failure problem of the inertial-based integrated navigation system under non-Gaussian noise is solved, and higher filtering accuracy and positioning accuracy are achieved.

WO2025228457A1 Publication Date: 2025-11-06HARBIN ENG UNIV +1
5 Cites 6 Cited by

Patent Information

Application Number
PCT/CN2025/104577
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2024-04-29
Filing Date
2025-06-27
Publication Date
2025-11-06

AI Technical Summary

Technical Problem

Existing inertial-based integrated navigation systems fail to filter when faced with non-Gaussian noise, leading to a decrease in the positioning and orientation accuracy of autonomous underwater vehicles and even endangering navigation safety.

Method used

We employ a navigation state error model based on Lie groups and a generalized maximum likelihood estimation model, combined with a robust Kalman filter algorithm for integrated navigation. By using the Huber function, we construct a more robust filtering algorithm, reducing dependence on state variables and improving filtering accuracy.

Benefits of technology

It effectively handles Gaussian and non-Gaussian noise, improves the filtering accuracy and positioning accuracy of inertial-based integrated navigation systems, and enhances the robustness of the system, especially when non-Gaussian noise is severe.

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

Abstract

Disclosed are an inertial baseline integrated navigation method, and a product, a medium, a device and an autonomous underwater vehicle, relating to the field of navigation. The method comprises: in an initial stage of the starting of an inertial baseline integrated navigation system, an SINS using a USBL to assist with initial alignment and obtaining initial attitude information (1); when the inertial baseline integrated navigation system enters a formal navigation operating state, the SINS performing pure inertial navigation solving on the basis of the initial attitude information, so as to obtain navigation parameter calculation values including errors (2), wherein the navigation parameter calculation values comprise the attitude, velocity and position; on the basis of the navigation parameter calculation values including the errors, constructing a navigation state error model based on a Lie group (3); on the basis of a Huber function, constructing a generalized maximum likelihood estimation model (4); according to the navigation state error model based on a Lie group and the generalized maximum likelihood estimation model, establishing an integrated navigation robust Kalman filtering algorithm (5); on the basis of the integrated navigation robust Kalman filtering algorithm, performing filtering estimation on the inertial baseline integrated navigation system, so as to obtain a state estimation value (6); and on the basis of the state estimation value, performing feedback updating on navigation parameters, and on the basis of the updated navigation parameters, performing navigation. The method can improve the integrated filtering precision and positioning and orientation precision.
Need to check novelty before this filing date? Find Prior Art