Research On Motion Planning Strategy Of 6-DOF Industrial Robot Based On Lidar

With the development of laser measurement technology,lidar has been widely used in automobile,industrial production and other fields.However,for the workpiece which is not easy to be clamped and has a large volume in the industrial field,due to its complex configuration and many surface features,the grinding accuracy cannot be guaranteed by manual measurement.Therefore,in order to improve the level of industrial automation in China and meet greater production demand,this paper conducts in-depth research on the motion planning of grinding robot based on point cloud data processing technology.In this paper,3d data measurement system of workpiece is designed based on lidar,and the conversion relationship between robot and lidar is established through hand-eye calibration of robot.The kinematics model is established by D-H method to facilitate the planning of robot movement.In order to achieve the precision of the workpiece surface using robot grinding,this paper USES the point cloud model and standard workpiece model of point cloud registration way implementation artifacts the establishment of accurate position,for industrial grinding operation,this paper put forward the point cloud based on double threshold filter the coarse registration algorithm with fast access to high-quality matching point pairs,and then improve the positioning precision and efficiency of workpiece.Since the point cloud model can not describe the complex surface features of the workpiece surface well,this paper uses the moving least square method to smooth the point cloud model,and then uses greedy projection triangulation algorithm to reconstruct the complex surface of the workpiece.Based on the reconstructed model,grinding path points were extracted based on NURBS surface.Since the points are independent from each other and no trajectory is formed,this paper adopts 3-5-3 piecewise polynomial interpolation method to fit the grinding trajectory.In order to improve the efficiency of industrial grinding,particle swarm optimization algorithm based on dynamic learning factor is used to optimize the trajectory in time.At the end of the paper has set up a platform experiment,has carried on the workpiece based on point cloud registration precision position for experiment and grinding trajectory planning simulation experiment,the experimental results show that the proposed point cloud registration algorithm is superior to other algorithms in the precision and efficiency,from the simulation results can be obtained using time optimization algorithm can shorten the travelling time of 10.3 seconds,The result of time optimization is achieved.