一种新颖的基于两个特征点的室内移动机器人定位方法

提出了一种新颖的基于两个特征点的室内移动机器人定位方法。与已有的几何位姿估计方法或航标匹配方法不同，该方法不需要人工航标，也不需要准确的环境地图，只需一幅由传统的CCD相机拍摄的图像。从机器人接近的目标上选取相对于地面等高的两个点作为两个特征点。基于这两点建立一个目标坐标系。在相机平视且这两个特征点与相机投影中心相对于地面不是恰好等高的条件下，就可以根据这两个特征点在图像中的坐标确定机器人相对于目标坐标系的位置和运动方向。该方法非常灵活，适用范围广，可以大大简化机器人定位问题。试验结果表明这一新的方法不仅简单灵活而且具有很高的定位精度。

Novel Two Feature Points Based on Localization Approachfor Indoor Mobile Robot

A novel two feature points based on localization approach for indoor mobile approach is presented. Unlike those reported geometry pose estimation or landmarks matching methods, this approach requires an image taken by a conventional CCD camera but needs neither artificial landmarks nor an accurate map of an indoor environment. Two points are selected as two feature points from the target to which a robot moves. An target coordinate system is created based on the two feature points. Under the conditions that the tilt angle of the camera is zero and the projection center of the camera is not just as high as the two feature points relative to a ground plane, the proposed approach can determine the position and motion direction of the robot according to the coordinates in the image of the two feature points. This approach is flexible and can be extensively applied. Therefore, it can greatly simplify the robot localization problem. Experimental results indicate that this approach is not only flexible but also has high localization precision.