一种飞行器定位方法
By combining inertial navigation and image matching, using the Superpoint neural network and Superglue algorithm for feature point extraction and matching, and combining Kalman filtering for positioning result fusion, the problem of autonomous navigation of aircraft in environments without GPS signals was solved, achieving high-precision and real-time navigation results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 武汉华中旷腾光学科技有限公司
- Filing Date
- 2023-08-18
- Publication Date
- 2026-07-17
AI Technical Summary
The need for autonomous navigation of aircraft in areas without GPS signals or with interfered GPS signals has not been met. Inertial navigation positioning errors accumulate rapidly, visual navigation has a large computational load and its accuracy is affected by the richness of the natural scene. Common visual navigation methods lack the use of map information.
Combining inertial navigation and image matching, feature point extraction and matching are performed using a superpoint neural network and a superglue matching algorithm. Kalman filtering is used to fuse the positioning results, and satellite maps are used as a reference for navigation.
It achieves high-precision autonomous navigation in environments without GPS signals, reduces the accumulation of inertial navigation errors, improves the real-time performance and anti-interference capabilities of navigation, and is suitable for long-distance, long-duration missions.
Smart Images
Figure CN117073669B_ABST