Robot Porcine Abdomen Cutting Point Planning Method Based On 3D Point Cloud

With the growth of China population and the change of consumption propensity,the demand for pork products is also increasing year by year.The porcine abdomen cutting process is an important slaughtering process and there are problems such as low manual operation efficiency,great labor intensity,and poor working environment.Moreover,considering the reduction of employees and the increase of labor costs,it has become an urgent demand in the slaughter industry to realize the automation of the slaughtering process and reduce the proportion of manual participation.To solve the practical problems of porcine abdomen cutting process in slaughter production line,a porcine abdomen cutting point planning method based on 3D point cloud is proposed to realize the automation of the porcine abdomen process and improve the automation level of the slaughtering production line.Firstly,the porcine abdomen cutting process is analyzed to determine the action flow of the robot to cut porcine abdomen,and clarify the system architecture and workflow.Then,calibration of the tool and user coordinate frame of the porcine abdomen cutting robot was carried out to realize transformation and unification of frame.3D point cloud of pig carcasses is collected by the industrial laser sensor,and preprocessed using truncation,voxel filtering downsampling and statistical filtering.After preprocessing,the background environment and noise point set is removed and the calculation efficiency and accuracy of the algorithm have been improved.Based on the point cloud Euclidean clustering algorithm,the 3D point cloud of the pig carcass is divided into several point cloud clusters of single pig carcass to avoid interference between multiple pig carcass data.Secondly,Principal Component Analysis algorithm was used to calculate the primary and secondary eigenvectors and the symmetry plane where the two eigenvectors are located.Next,the points within the preset distance threshold are projected onto the symmetry plane as the center line point set of the pig carcass.Through the statistical analysis of the relevant data of multiple pig carcasses,the position ratio of the porcine abdomen cutting starting and ending points on the center line of pig carcass is obtained.According to the ratio,the points within the cutting range are separate from the center line point set of the pig carcass.And these points are discretized to get the final sparse cutting points.Finally,the robot porcine abdomen cutting experimental platform is built and test experiments about the algorithm validation in planning the cutting points and the actual performance in instructing robot to cutting porcine abdomen.The experiment results show that the cutting points extracted using proposed method are evenly spaced and could well reflect the contour curve of pig carcass.According to the extracted cutting points,porcine abdomen cutting could be successfully performed without organ damage and superior to manual cutting in efficiency and precision.