| 引用本文: | 罗世彬,刘海桥,胡茂青,董晶.无人飞行器异源图像匹配辅助惯性导航定位技术综述[J].国防科技大学学报,2020,42(6):1-10.[点击复制] |
| LUO Shibin,LIU Haiqiao,HU Maoqing,DONG Jing.Review of multi-modal image matching assisted inertial navigation positioning technology for unmanned aerial vehicle[J].Journal of National University of Defense Technology,2020,42(6):1-10[点击复制] |
|
| |
|
|
| 本文已被：浏览 441次 下载 395次 |
| 无人飞行器异源图像匹配辅助惯性导航定位技术综述 |
| 罗世彬1,2,刘海桥1,胡茂青2,董晶3 |
|
(1. 中南大学 自动化学院, 湖南 长沙 410083;2. 中南大学 航空航天学院, 湖南 长沙 410083;3. 国防科技大学 空天科学学院, 湖南 长沙 410073)
|
| 摘要: |
| 无全球定位系统下高精度定位与导航是飞行器实现自主侦查、巡航与打击的关键。视觉导航具有被动、低成本、能避免累积误差等优点。视觉导航与惯性导航融合更能够发挥出各自的优势,达到高精度定位的目的。总结了异源图像匹配辅助惯性导航的飞行器定位技术的发展历程; 从相机-惯导标定技术、异源图像匹配、姿态解算、数据融合和后端优化五个方面详细阐述了异源图像匹配辅助惯性导航的飞行器定位的关键技术; 指出了基于深度学习的异源图像匹配与惯性导航两种无源定位组合导航系统融合技术等四个未来可能的发展方向,可为实现异源图像匹配辅助惯性导航飞行器定位技术提供参考。 |
| 关键词: 异源图像 图像匹配 组合导航 飞行器定位 |
| DOI：10.11887/j.cn.202006001 |
| 投稿日期：2019-12-26 |
| 基金项目:国家自然科学基金资助项目(11272349,61802423) |
|
| Review of multi-modal image matching assisted inertial navigation positioning technology for unmanned aerial vehicle |
| LUO Shibin1,2, LIU Haiqiao1, HU Maoqing2, DONG Jing3 |
|
(1. School of Automation, Central South University, Changsha 410083, China;2. School of Aeronautics and Astronautics, Central South University, Changsha 410083, China;3. College of Aerospace Science and Engineering, National University of Defense Technology, Changsha 410073, China)
|
| Abstract: |
| High-precision positioning and navigation in GPS(global positioning system) denied environments is a key technology for aircraft achieving autonomous scout, cruise, and strike. Vision navigation has the advantages of passive-type, low cost, and avoidable accumulated errors avoidable, etc. The fusion of vision and inertial navigation can give full play to their advantages and achieve the purpose of high-precision positioning. Firstly, the development of aircraft positioning technology based on multi-modal image matching assisted inertial navigation was summarized. Then this technology was elaborated in five aspects:the vision-internal calibration, multi-modal image matching, attitude algorithm, data fusion, and back-end optimization. Finally, four possible future directions were proposed as follows, two types of passive positioning combined navigation systems based on deep learning, multi-modal image matching and inertial navigation. The four possible future directions provide a reference for realizing multi-modal image matching assisted inertial navigation aircraft positioning technology. |
| Key words: multi-modal image image matching integrated navigation aircraft positioning |
|
|
|
|
|
