Gesture Guided Industrial Robot Teaching Technology And Its Applications

Industrial robots as a multi-disciplinary integration of advanced intelligent machines are widely used in various industrial production areas.Teaching as a key step in controlling industrial robots to complete the tasks,has become an important field of robot technology research.The traditional teaching methods include teaching programming and offline programming,which usually require the instructor to be familiar with the operation flow and the procedure instruction,the teaching profession's professional knowledge is higher,and the teaching process is cumbersome and the interaction is bad.In view of the above problems,this paper proposes gesture guide based teaching method,so that the instructor can focus on the teaching task itself rather than teaching process and instruction learning,make the interaction friendly,improve teaching efficient.The first chapter summarizes the current research status in teaching technology for industrial robots and gesture interaction at home and abroad.Then,we give the research content and the framework of this thesis in the end.In the second chapter,a method of generating trajectory of robot based on gesture recognition is proposed.Firstly,industrial robot kinematics method is discussed and the kinematic model is constructed with RB08 industrial robots.For the problem of multiple solutions in inverse kinematics,an optimal solution extraction method based on link weight and angle two norm is proposed.Then,based on KinectV2,complete the construction of hand coordinate system,and realize the positions and orientations mapping between hand and robot end implement.Finally,Kalman filter is introduced to improve the tracking stability,and the feasibility of this tracking method is verified experimentally,which lays the foundation for follow-up gesture guidance teaching.In the third chapter,we propose a method to optimize the teaching path of hand gestures in free space.Firstly,use the gesture to mark the point control points,to achieve the task of guiding to work area.Secondly,for the problem of dither and clustered of the trajectory.The space-bound sphere method and the Gaussian filter are used to complete the pre-processing of the trajectory.Then the trajectory feature points are?extracted based on the curvature and the torsion threshold and the trajectory is initially fitted.Finally,the Hausdorff distance is used as the criterion of fitting accuracy,and the feature points are added until the accuracy requirements are satisfied,and complete the optimization of teaching trajectory.In the fourth chapter,we propose the gesture guide teaching intention reproduction method for task constraints.Firstly,the characteristics of task object and gesture behavior are analyzed,and a gesture guided trajectory segmentation method based on velocity segmentation domain is proposed.Then the improved equidistant sampling method is used to complete the matching and recognition of the segmentation segment.Aiming at the trajectory of the processing section,the method of quantifying the design feature and gesture characteristics is put forward to realize the quantization intention expression of the gesture on the trajectory of the processing section,and the trajectory of the processing section is generated according to the manufacturing feature.Aiming at the trajectory of the transition section,the automatic transition process of the trajectory of the machining section is realized by the circular arc segment method.In the fifth chapter,we develop the virtual robot simulation system for industrial robots.The system includes the main functions of the original teaching box,including manual teaching,programming teaching and reproduction teaching.And according to the gesture guidance teaching method proposed in this paper for the two operating environment add gesture guide teaching function respectively,which making the teaching process more friendly and efficient.The sixth chapter summarizes the main contents and innovations of this thesis,and puts forward the prospect of the current shortcomings and the need for further research.