Research On Vision Guiding Key Technique Of Industrial Robot

Industrial robot has become an irreplaceable important equipment in advanced manufacturing industry, and its application is expanding into the general industry from the automotive industry. As part of the research focus of robot technology, the industrial robot vision guiding technology lays the foundation for the application of industrial robot in the high flexible and intelligent production line. The calibration technology is the key to the vision guiding technique of industrial robot, which will directly affect the grasping accuracy of the robot. Meanwhile, the robot vision calibration technology is the premise of the off-line programming technology. In this paper, different types of robot vision products applied in practice are studied as the research object, the operational principle of the vision system is analyzed, the calibration process of the vision system is discussed and the algorithm of transformation between different coordinate systems is explored. The primary contents are following:1. The working principle of robot vision guiding system applied in different situations is analyzed and the mathematical model of2D,2.5D and3D vision system is created, the appropriate image features are chosen to obtain the spatial position offset of the object to be grasped. The method to get the deep information of the object is verified by comparing the different images shot at each position.2. A method to calibrated the position and orientation between robot and target is presented, the coordinates of the feature points on the target are got by a special made tool mounted on end of the robot, transformation between target coordinate and robot world coordinate is figured out by the knowledge of spatial geometry and the method is validated by data collected at the scene.3. A hand-eye calibration model of2D vision guiding system is established based on2points, three parameters of the transformation between object coordinate and image coordinate are solved by two features on the object.4. Transformation equations using quaternion between camera coordinate and object coordinate are established, the algorithm to solve the rotation of three-dimensional space and the method to handle the data error are quoted from photogrammetry, camera collinearity equation is created and then linearized, the error equation is given by the existing indirect adjustment method of measurement, finally the iterative solution of the parameters in the transformation equation are solved.