【摘要】 针对在室内工作的自动引导运输车(automated guided vehicle,简称AGV)面临的障碍物实时检测问题,提出一种基于Kinect相机的检测方法。在深度相机标定的基础上,将相机获得的深度图像准确地变换成相机坐标系内的点云数据,利用点云数据,通过制定障碍物检测规则,实现AGV小车行驶过程中障碍物的快速实时检测,实试验证了该方法的准确性。更多还原

【Abstract】 In this paper, a Kinect camera based method is proposed for the real-time detection of obstacles faced by automated guided vehicle(AGV) working indoors.On the basis of the calibration of the depth camera, the depth image obtained by the camera is accurately transformed into the point cloud data in the camera coordinate system. By using the point cloud data, the obstacle detection rules are formulated to realize the fast real-time detection of obstacles in the process of AGV vehicle running. The accuracy of the method is verified by experiments.更多还原