Over the last years, object recognition has become a more and more active field of research in robotics. 3D perception is a promising technology for automatic control, manufacturing and robotics. Compared to 2D vision systems, 3D range sensors can provide direct geometric information of the environment. This study presents a high quality, low cost 3D laser range finder designed for object recognition. The 3D laser is built on the base of a 2D laser range finder by the extension with a servo motor and a rotation module. The servo is controlled by an embedded computer running Linux. The survey yield a digital data set, which is essentially a dense “point cloud”, where each point is represented by a coordinate in 3D space. A complete 3D scan data of an area of 360° with 48,598 points are grabbed in 0.5 seconds. A systematic error is 5 cm. While scanning, different online algorithms for line and surface detection are applied to the data. Object segmentation and detection are done in real-time after the scan. The implemented software modules detect objects.
展开▼