Polar region unmanned aerial vehicle navigation method and system based on image-assisted positioning

By constructing an optical residual strain field and an error state extended Kalman filter, the problem of dynamic cooperative positioning in polar environments without fixed base stations was solved, achieving high-precision navigation and positioning, eliminating aurora interference and refraction distortion, and ensuring the navigation reliability of UAVs in polar regions.

CN122237609AActive Publication Date: 2026-06-19POLAR RES INST OF CHINA
9 Cites 3 Cited by

Patent Information

Application Number
CN202610710980.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-05-22
Publication Date
2026-06-19
Estimated Expiration
2046-05-22

AI Technical Summary

Technical Problem

Existing navigation technologies are ineffective in dynamic cooperative positioning in polar environments without fixed base stations. Traditional stellar image processing algorithms face difficulties in adaptive feature extraction and multi-source kinematic cross-validation under conditions of strong optical noise and distortion in polar environments, resulting in low navigation accuracy and unreliability.

Method used

By constructing an optical residual strain field, aurora interference areas and refractive distortion areas are identified and eliminated. Combined with the three-dimensional translational velocity vector calculated by the mobile local area network, image features are extracted and topological regions are divided. The error state extended Kalman filter is used for precise positioning.

Benefits of technology

It achieved highly reliable navigation accuracy in harsh polar environments, eliminated false starry sky areas, and ensured the absolute safety of the UAV during long-term flight.

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

Abstract

The invention provides a polar region unmanned aerial vehicle navigation method and system based on image-assisted positioning, and relates to the field of computer vision, and the navigation method comprises the following steps: obtaining a rough three-dimensional position and a rough attitude quaternion of a polar region unmanned aerial vehicle; a real polar star sky image is collected, and a continuous optical residual strain field is constructed; the divergence and the rotation of an optical residual strain field are calculated, a refraction distortion area and an aurora interference area are identified and proposed, and an initial starry sky area is obtained; extracting multi-dimensional features in the initial starry sky area, and iteratively correcting an effective starry sky area boundary based on a mahalanobis distance to obtain a candidate sky area; an actual motion optical flow field and a theoretical optical flow field in the candidate sky area are extracted, verification is carried out in combination with a three-dimensional translation velocity vector measured and calculated by a mobile local area network, a pseudo starry sky area is eliminated, and an optimal starry sky area is obtained; and final accurate positioning is completed in the optimal starry sky area.
Need to check novelty before this filing date? Find Prior Art