Method for assisting visual SLAM (Simultaneous Localization and Mapping) by using mark

Abstract

The invention discloses a method for assisting visual SLAM by using a mark, and the method comprises the steps: obtaining a preset mark, and determining initial map data according to the mark; tracking and positioning according to the mark to obtain pose data of the shooting camera, and updating the initial map data according to the pose data of the shooting camera to obtain map data; and adjusting the map data according to the mark to obtain target map data. The same set of preset marks are used for assisting in positioning and mapping in the steps of initialization, tracking and positioning, map optimization and the like of the visual SLAM, so that the precision and robustness of the visual SLAM system are improved, and the time efficiency and the space efficiency are improved. The problem that in the prior art, due to the fact that robustness of a visual SLAM system is limited by a complex environment and a dynamic environment, tracking fails is solved. 本发明公开了一种使用标志辅助视觉SLAM的方法，所述方法包括：获取预设的标志，根据所述标志确定初始地图数据；根据所述标志进行跟踪定位，得到拍摄相机的位姿数据，并根据所述拍摄相机的位姿数据对所述初始地图数据进行更新，得到地图数据；根据所述标志对所述地图数据进行调整，得到目标地图数据。本发明通过在视觉SLAM的初始化、跟踪定位以及地图优化等步骤中使用预先设置的同一套标志辅助定位与建图，提高了视觉SLAM系统的精度和鲁棒性，提高了时间效率与空间效率。解决了现有技术中由于视觉SLAM系统的鲁棒性受限于复杂环境与动态环境，从而导致跟踪失败的问题。