基于视觉的机器人定位精度提高方法

定位是确定机器人在其工作环境中所处位置的过程，精确、可靠地实现定位是移动机器人应该具备的一项最基本也是最重要的功能之一。针对机器人在完成穿过门的导航任务过程中，现有的视觉定位方法存在精度不高，不能为机器人的导航提供精确位置信息的问题，提出了一种基于离线学习的方法来得到摄像机的内部参数随机器人位置的变化关系，使机器人在粗定位的基础上，自适应地调节摄像头的内部参数，修正定位结果，达到提高定位精度的目的。实验证明了该方法的有效性。

Improvement Method of Localization Precision for Vision based Robot

Localization is the process of finding the position and the orientation of the robot in its working space. Precise and reliable localization is one of the most fundamental and important tasks for autonomous mobile robot. To solve the problem of low precision of the robot localization which cannot meet the requirement for precisely positional information when the robot implements the door-traversal task, a method based on off-line learning to improve the precision of robot localization is presented. The proposed method first implements the rough localization, and then autonomously adjusts the internal parameters of camera to improve the precision of localization. Experiments show the effectiveness.