Development Of Industrial Robot's End-Effector Pose Control System With Integrated Vision

With the rising of labor costs and the need for industrial upgrading,the application of industrial robots is becoming more and more common,especially in the field of industrial production which is harmful to people,such as automobile parts surface painting,parts grinding and so on.In the past,the robot trajectory was generated by manual teaching or offline programming.It was prerequisite to ascertain the relative position of the machining coordinate system and the world coordinate system,and it usually chose the plane features with high machining accuracy on the workpiece as the positioning basis,which was time-consuming,laborious and not high precision.However,as we all know,each workpiece blank has different degrees of deviation from the standard CAD model in the process of parts grinding,so it is not possible to achieve the ideal processing effect only by part positioning and off-line programming.In addition,the robot control system with good open source and strong expansibility is needed to realize the information interaction between the robot arm and the working environment.In addition to the motion control functions,it should also have logic control and process management functions.In view of the above requirements,this paper carried out the following research work:The corresponding D-H parameters are obtained according to the existing mechanical structure of the robot.Combined with the kinematics of the robot,the corresponding soft PLC control system of the robot is developed under the development environment of Codesys.It has good open source features and strong expansibility,which can realize the customized monitoring of the robot state.In addition to the basic motion control function,it also has the function of NC code file processing,which can load and identify the international standard G code motion instructions,so as to control the movement trajectory of the robot.A vision system is designed based on the principle of laser triangulation.The line laser is projected onto the surface of the object to be measured,and the position of the line laser is obtained by using an industrial camera to calculate the 3D contour data of the object.With the movement of the object,the contour data are synthesized into 3D data of the object.The distortion coefficient and internal parameter matrix of the camera are obtained by calibrating the industrial camera,so as to get the coordinate transformation matrix between the camera coordinate system and the pixel coordinate system.The calibration accuracy is analyzed by designing experiments.Then,according to the measurement principle of laser triangulation and perspective transformation principle,the transformation matrix is calculated,and the trapezoidal distortion caused by the plane of structured light which is not perpendicular to the camera axis is solved.Finally,a software system is developed to turn the collected image data into the data of the corresponding surface,and to display the calculation results with the help of Matlab.According to the surface tool path planning algorithm and the calculation method of surface normal vector in point cloud data processing,the contour tracking control of scanned surface is completed.The tool path topology with parallel structure is used to slice the machining surface according to the set tool path spacing,so the initial machining path is obtained.On this basis,the tool points are calculated and the cutting step length is calculated with the method of equal chords height error.The normal vector is calculated by the local fitting method based on K neighborhood.The standard NC code is finally output to complete the robot trajectory planning,and the cumbersome process of teaching and offline programming has been solved.