Object positioning method based on laser radar and monocular vision

The invention relates to an object positioning method based on laser radar and monocular vision. According to the method, a camera and a laser radar are installed according to a certain relative position relation, a target object is recognized in a motion process of a robot by utilizing a depth learning algorithm, the principle of the camera is referred to, the angle of the target object under thecamera coordinate system is calculated by utilizing the position of the target object in a picture, and the corresponding laser beam of the angle is found in the laser radar data so as to obtain therelative position relation between the target object and the robot, in the process of real-time positioning and map constructing, the position of the target object in the map is obtained. 本发明涉及种基于激光雷达与单目视觉的物体定位方法，将摄像头与激光雷达按照定的相对位置关系安装，利用深度学习算法在机器人运动过程中对目标物进行识别，参照摄像头原理，利用目标物在图片中的位置计算得到目标物在摄像头坐标系下的角度，并在激光雷达数据中寻找该角度的对应的激光束以得到目标物与机器人的相对位置关系，在实时定位与地图构建过程中，获取得到目标物在地图中的位置。