POSITION AND LOCATION DETECTION OF OBJECTS
A method for position and location determination of objects, particularly uncooperative objects. A 3D sensor delivers data of the target object, which are compared with target object data stored in a model database and are used both for an initial position and location estimation and for a subsequent more accurate position and location estimation of the target object. The 3D sensor, preferably a LIDAR, delivers the data relating to a sensor coordinate system in the form of a 3D point cloud, which is converted to a depth image in which each pixel receives a gray value, which corresponds to a depth within a cuboid surrounding the target object. The method for initial position and location estimation is based on 2D template matching By using depth data, three of six degrees of freedom can be eliminated, resulting in a robust and fast method for initial position and location estimation.
1 . A method for position and location determination of objects, comprising
delivering data of a target object with a 3D sensor,
comparing the delivered data with data of the target object stored in a model database, and
using the delivered data for an initial position and location estimation and a subsequent more accurate position and location estimation of the target object.
2 . The method according to claim 1 , wherein the 3D sensor delivers data relating to a sensor coordinate system in the form of a 3D point cloud, which 3D point cloud is converted to a depth image in which each pixel receives a gray value, the gray value corresponding to a depth within a cuboid surrounding the target object.
3 . The method according to claim 1 , wherein the sensor is a 3D LIDAR.