一种基于深度学习的途中景象匹配导航方法和系统
By employing a deep learning-based scene matching navigation method that utilizes tile reference maps and feature point matching, the positioning error problem of inertial navigation systems under GPS rejection is solved, achieving high-precision navigation and positioning applicable to various environments and aircraft models.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- XIAN FLIGHT SELF CONTROL INST OF AVIC
- Filing Date
- 2024-04-28
- Publication Date
- 2026-07-17
AI Technical Summary
Under GPS rejection conditions, the cumulative positioning error of the inertial navigation system cannot be effectively corrected, resulting in a decrease in positioning accuracy. Traditional scene matching navigation technology has poor usability in terms of lighting and viewing angle changes, making it difficult to meet the requirements of high-precision navigation.
A deep learning-based scene matching navigation method is adopted. By pre-preparing a tile reference map, feature points are extracted and matched, the homography matrix is calculated, and inertial navigation data is combined to achieve accurate position calculation.
It improves positioning accuracy and reliability under GPS denial conditions, overcomes the effects of lighting and seasonal changes, adapts to various environments, and is suitable for various aircraft models.
Smart Images
Figure CN118351184B_ABST