Research On Automatic Spray Path Acquisition Of Furniture Based On Laser Sensor

With the development of production intelligence,the application of robots in the field of spraying has become more and more intense.However,even if the robot is applied to the spraying scenario,the planning of realizing spraying tasks is still mainly based on the manual demonstration method.This method has obvious disadvantages.The quality of the work depends heavily on the proficiency of the worker,and the repeatability like the machine cannot be guaranteed.The research object of this article is furniture.Furniture production has the characteristics of non-large quantity,complex varieties and easy replacement.It is impossible to set up a special production line for each furniture variety.When the production is replaced,it must be re-planned.Therefore,it is necessary to study the manual and fully automated adaptive spraying path information acquisition for furniture spraying.The automatic acquisition of spraying path information firstly needs to automatically obtain the geometric characteristics of the furniture.The host computer automatically calculates and generates a path based on the surface information of the workpiece and transmits the path point information to the spraying robot.First,we design the overall layout of the furniture spraying production line.First,plan the system structure and operation process.In order to collect point cloud data,a data collection system was built.The system software platform is designed to calculate the acquired point cloud data and pave the way for subsequent test verification.Secondly,we model and parameterize the spraying process and combine actual conditions to optimize spraying parameters.Based on the study of the influencing factors of the spraying process,an appropriate model was selected to model the coating thickness at a point on the plane when spraying,then the coating thickness at a point on the surface was also modeled,and then the spray gun straight line was calculated Dynamic spraying model,and use simulation software to analyze the distribution of paint.After abstracting and simplifying all possible situations in furniture spraying operations,the gun trajectory optimization problem is established for parameter optimization.Solve the optimization problem to obtain important spraying parameters.Then,we study the data collection and processing of the high-precision laser sensor selected in this paper.First,the specific physical basis for the sensor to obtain data is discussed,and the coordinate system conversion relationship in the system is introduced.After that,the data optimization of point cloud was discussed.First,the removal of outliers based on statistical filtering is studied to remove small outliers at the edges of the workpiece.For the slight interference caused by the roughness of the furniture surface,an improved bilateral filtering algorithm is used.Due to the high accuracy of the laser sensor and the high density of the data points obtained,it is difficult to perform real-time calculations,so the original point cloud is down-sampled while retaining the edge characteristics of the point cloud.Finally,the automatic acquisition algorithm of robot spraying path is studied.A point cloud slicing algorithm based on pose recognition is proposed to adaptively select the slice plane direction that is most conducive to subsequent calculations,and sort the obtained slice points to generate a path that can be executed by the robot.Through the calculation and offset of the normal vector on the surface of the workpiece,the expected position and posture of the end spray gun at work are obtained.Aiming at the problem of repairing the edges of furniture,a corresponding algorithm is proposed to make the thickness of the edges meet the requirements of the subject.Finally,the path planning algorithm is verified on the established experimental platform,and the analysis of time and spraying effect proves the superiority of the algorithm in this paper.