摘要: |
针对单目视觉SLAM在复杂场景下跟踪失败导致地图丢失的问题,提出了一种利用RTK信息辅助单目视觉的自动化地图构建与恢复方法。在视觉跟踪失败后,设计了一种自动化恢复地图构建方法,基于RTK信息的辅助,采用坐标变换算法处理丢失前的地图并将其融合到当前地图构建中,在提高建图效率的同时,最大程度地减少了地图信息的缺失。实验仿真结果表明,该方法可以解决系统无法继续跟踪建图的问题,与原始地图相比,其融合后构建的地图在保证全局一致性和完整性的基础上,可以保持米级的估计精度。 |
关键词: 视觉导航 RTK技术 地图融合 尺度估计 坐标变换 |
DOI: |
|
基金项目:江苏省政策引导类计划(产学研合作)基金(BY2016003-16) |
|
Map Construction and Recovery Method Based on RTK-aided Monocular Vision |
XUE Yuan,SUN Yong-rong,ZHAO Ke-dong,ZHAO Wei |
(Navigation Research Center, College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China) |
Abstract: |
To solve the problem of map loss caused by monocular SLAM tracking failure in complex scenes, an automatic map construction and recovery method based on RTK (Real-Time Kinematic) information aided monocular vision is proposed. With the aid of RTK information, the map before visual tracking fails is processed by coordinate transformation algorithm and is integrated into the current map construction, so as to further improve the efficiency of map construction and minimaize the loss of map information. Simulation results show that the proposed method can help with the problem of map loss caused by visual tracking failure. Compared with the original intact map, the integrated map preserves its consistency and integrity with meter level estimation accuracy. |
Key words: Visual navigation RTK technology Map fusion Scale estimation Coordinate transformation |