| Precision assembly is very common in both industry and national defense.However,industrial robots mainly use traditional position control,which lacks the ability of environment perception and compliance.As a result,the overall automation level of this kind of operation is still relatively low,which is mainly completed by human or human-robot cooperation.In order to improve productivity,free our hands,complete the upgrading of manufacturing industry in China as soon as possible,and become a powerful manufacturing country,it is necessary to realize robot independent assembly operation.In this paper,computer vision and compliance control technology are applied to industrial six-axis robot,and the autonomous compliance assembly method of high precision large aperture optical elements is studied.Through the industrial camera,six-axis force sensor and active compliance algorithm,the robot can recognize the position and pose of the workpiece and perceive and comply with the environment,and complete the assembly successfully,which provides theoretical support for the subsequent application to the actual industrial assembly environment.The main research contents include the following aspects:1)Aiming at the target assembly workpiece,the 3D model and reference coordinate system are established,and the possible contact state in the assembly process is analyzed.According to the idea of active compliance control,an assembly experimental platform is built,which is mainly composed of YASKAWA 6-DOF robot,supplemented by vision and torque sensor,which gives the robot the ability of visual perception and force perception;in order to complete the assembly smoothly,the tool coordinate system is also established.2)The visual perception and positioning system is developed.The camera calibration is completed based on the imaging principle of the camera;for the square workpiece with unknown pose,firstly,the image is preprocessed by the filter and enhancement algorithm,and the pixel coordinates of the four feature points of the workpiece are found by the contour recognition algorithm,then the pose of the workpiece is solved by the Pn P algorithm;the hand eye calibration is completed,and the calculated pose coordinates are converted to the base coordinates so as to guide the robot to assemble.3)The compliant assembly strategy based on admittance control algorithm is designed.The calibration of force sensor and gravity compensation algorithm are studied.The contact force information between robot and environment is obtained by completing calibration experiment.The admittance control principle is analyzed.For irregular square assembly,the assembly process is divided into two stages: alignment and leveling,and the admittance control algorithm is used to complete the two stages.4)The development of system application software and human-robot interaction interface is completed.The system software is divided into five modules,which can achieve different functions.Based on the built system hardware,the flexible assembly experiment is carried out,and the contact force and pose adjustment data in the assembly process are analyzed.The experimental results verify the effectiveness of the flexible assembly strategy based on admittance control. |
PDF Full Text Request