Feature Extraction Based Margin Optimization Analysis For Robot Scribing

For the processing of aerospace precision forging castings,the scribing process is the pioneering process,and currently most of them are done manually.The processing of the small margin of the complex structural parts of the casting blank determines that it is difficult to align it by manual scribing by the fitter,the positioning datum is difficult to grasp,and the manual scribing is difficult to meet the processing requirements.With the rapid development of intelligent manufacturing and the emergence of three-dimensional digital measurement technology,the integration of informatization and industrialization has become an inevitable trend.The three-dimensional digital measurement technology has been gradually developed at home and abroad,and the optimization analysis of casting blank machining allowance is the standard.Combining digital and integrated precision machining technology with precision CNC machining technology.In order to improve the accuracy and efficiency of the scribing process,and realize the automation and intelligence level of the scribing process,this paper collects the point cloud data of the casting blank through a three-dimensional laser scanner,and processes the point cloud data through the margin optimization analysis system to output the scribing trajectory.Industrial robot technology realizes automatic marking of robots.(1)Research on point cloud data acquisition and preprocessing technology.By comparing contact and non-contact three-dimensional point cloud data acquisition,it is difficult to obtain complete point cloud data in traditional methods.By adopting a three-dimensional laser scanner,a solution based on non-contact blank point cloud data acquisition is proposed.Through point cloud data de-noising and simplification and model discrete preprocessing,the digital display of casting blanks and ideal models is completed.(2)According to the different requirements of the machining allowance of each surface of the casting blank,a process-based point cloud feature extraction technology is proposed.The important feature profiles are mainly plane point cloud data and cylindrical point cloud data.Different types of point cloud data can be segmented according to different geometric properties.It is necessary to calculate the normal vector and curvature information of each point cloud data,establish a process guide model and realize the automatic extraction of typical characteristic objects through the regional growth method.(3)Based on the process requirements,a comprehensive analysis of the process allowance of complex blanks based on features is proposed.The initial registration of the blank and the ideal model is realized by the principal component analysis method.Relevant constraints are imposed on the tolerance surfaces of each machining surface,an accurate registration algorithm based on process characteristics is proposed,and the registration is quickly constrained by a progressive method that simplifies the set of registration points.Through the optimization analysis of the margin,the problem of difficulty in aligning during the scribing process of the casting is solved.(4)Independent design and development of margin optimization analysis system software.Combining example analysis to verify the margin optimization analysis process,output the scribing trajectory,design the robot end execution device and make path planning for the scribing trajectory,and realize the robot's autonomous scribing process through the robot's motion control.The analysis of the marking result verifies the rationality and effectiveness of the technique proposed in this article.