Research On Trajectory Planning Of Drawing Robot Based On Image Edge Detection

Compared with the research on processing and production,the research on robot drawing is less.The development of drawing function is a good carrier for robot research,and robot drawing has a broad application space in the fields of industrial labeling and popular science education.Robot drawing first needs to extract the contour of the image,and the edge detection technology used to extract the contour is an important part of image processing and pattern recognition,as well as the basis of other image operations.It has been widely used in industry,agriculture,medicine and other fields.Similar to industrial scenes such as robot carving,welding and gluing,trajectory planning can also be studied for the drawing robot,which can make the robot move continuously,smoothly,reduce impact and improve the stability and accuracy of the robot operation.Planning the drawing sequence can shorten the path length of robot movement and thus shorten the running time.In this paper,the robot drawing as the entry point,through the image edge detection and other image processing technology combined with the robot drawing sequence planning,trajectory planning,has important research value.The main research contents of this paper are as follows:The classical Canny operator is improved,and the Gaussian filter in the Canny operator is improved adaptively based on the principle of minimum description length to ensure the accuracy of edge recognition.Sobel operator is used to calculate the image gradient.Otsu method is used to determine the image threshold.In order to highlight the main contours during the drawing of the image,smaller areas were deleted.In this paper,an edge thinning algorithm is proposed to avoid drawing the same contour twice,and the edges with single pixel width are obtained.The broken contours are connected to reduce the times of lifting the pen when the robot is drawing.In order to avoid contour tracking error resulting in incomplete image rendering,the burr points in the image were removed.Robotstudio simulation software and ABB IRB1200 robot are used to simulate and experiment the contour image after processing.The results show that the proposed method can effectively filter out noise and irrelevant details.The goal of edge refinement is realized and the integrity of each section contour is improved.Reduce the number of robot lift pen,improve the efficiency of drawing.In order to improve the efficiency of the robot,the ant colony algorithm is used to optimize the drawing sequence of the contours.MATLAB is used to simulate the drawing sequence,and Robotstudio is used to simulate the drawing sequence before and after optimization,and the drawing verification is carried out in the actual robot.The results show that the distance of the robot’s empty travel is greatly shortened and the drawing efficiency is improved.Based on quintic nonuniform B-spline interpolation of joint motion curves,the shortest running time satisfying the kinematics constraints of each joint is obtained by using genetic algorithm.The curves of joint position are smooth,and the curves of velocity,acceleration and additional acceleration are smooth,which shortens the running time of the joint and reduces the joint impact.