Research On Path Planning And Correction Of Robotic Arc Wire Additive Manufacturing Based On Line Laser Vision Sensor

Metal wire additive manufacturing has been a hotspot of extensive research due to its high forming efficiency,good mechanical properties,short manufacturing cycle and high material utilization.However,due to its complex forming process,numerous influencing factors and lack of effective feedback control methods,its application is greatly restricted.In order to improve the forming accuracy of metal wire additive manufacturing,this paper takes robotic arc wire additive manufacturing as the research object,and builds a set of arc additive manufacturing forming control system based on welding robot and laser vision sensor system.The focus is on the path planning and correction of robotic arc wire additive manufacturing.First of all,to solve the problem of difficulty in teaching the forming process of metal wire arc additive manufacturing,this paper builds a robotic arc wire additive manufacturing forming control system,including optimizing and storing 3D models,realizing equal layer thickness slicing algorithms,path planning algorithms,etc.It is suitable for solid and complex structural parts and rotating structural parts,and control robot surfacing forming.It realizes the automation of the whole process of additive manufacturing,avoids the teaching phase of complex paths,and reduces labor costs.Aiming at the lack of feedback links in the forming process of metal arc wire additive manufacturing,the forming process is unstable,and the robustness is extremely poor.In this paper,the vision sensor is used to collect the point cloud of the workpiece in real time,perform point cloud data preprocessing,and then select welding to obtain the position and geometric characteristics of the surfacing workpiece,and perform slicing and path planning corrections to realize the accurate accumulation of the workpiece and reduce the error between the actual workpiece and the model.Finally,for the filling defect problem of irregular contour path planning in the layer,this paper first conducts a single-pass process experiment,and uses machine learning algorithms to establish a model of the relationship between welding process parameters and weld bead geometric features to assist the setting of process parameters.Then carry out multiple single-layer experiments,and establish a model to obtain appropriate weld bead spacing parameters.Finally multi-layer and multi-pass experiments were performed,by correcting the path,the defect of insufficient filling in the center of the workpiece was avoided,and the complicated workpiece surfacing was finally realized.