Research On Parameters Calibration And Path Generation Methods Of Robot Milling Trimming

Skin parts are a type of parts that form the shape of an aircraft and require good aerodynamic characteristics.Its processing and assembly accuracy have an important impact on many key performance parameters of the aircraft.Due to the accumulation of errors in the manufacture and assembly of the skin skeleton,the no-margin skin processed according to the theoretical design model cannot meet the assembly requirements,so a certain process margin is left in the skin processing stage for the adjacent The skins are trimmed against each other to eliminate accumulated errors.Traditional methods mostly use manual alignment and scoring,and the contrast-trimming method is repeated to remove this margin.The randomness of the manual alignment scribing is large,and the scribing width is about 1mm.In the subsequent manual trimming and trimming process,it is easy to go beyond and poor consistency.To this end,this paper focuses on the calibration of robot-stereoscopic target-external measurement system,high-precision ordered extraction of boundary and reconstruction of boundary feature,generation and smoothing of cutting edge path of skin ring milling,accuracy test of measurement system and processing verification experiment,and strives to It fundamentally solves the problems of easy deviation and poor consistency of artificial edge trimming of aircraft skin,and improves the accuracy of skin manufacturing and assembly.The main results are as follows:(1)Study the closed form in the quaternion form of the hand-eye calibration equation,derive the transformation matrix rotation and translation stepwise solution method of the objective function constructed by the Lagrange multiplier method,and study the left multiplication perturbation BCH(Baker-Campbell-Hausdorff)approximation model,constructing an optimized objective function of projecting all data to the centroid of the measuring point in the first scan pose,and using the GN(Gauss-Newton)iterative method to nonlinearly optimize the pose parameters of the robot-external measurement system to achieve High-precision calibration of robot-three-dimensional target-external measurement system.(2)Study the rough extraction method of point cloud boundary based on point cloud triangle meshing and area growth method,propose an orderly and precise extraction method of non-convex polygonal point cloud boundary,and establish a boundary parameterized feature template that refers to the design model of skin parts Design the segmentation and parameterized reconstruction algorithm of the ordered point cloud boundary of the reference part model,realize the high-precision and orderly extraction of the mouth-like skin boundary,and ensure the high-precision reconstruction of the boundary corner feature.(3)Research the two methods of margin estimation based on line drawing area extraction / point cloud matching,analyze the characteristics of the neighborhood point set,and realize the differential information estimation of the space curve boundary point cloud,and study the attitude of the tool axis on each processing path during the processing process A fine-tuning algorithm is proposed,which generates a layer-by-layer offset machining path generation method that combines the processing boundary parameters of the skin boundary point to the margin area,and realizes the generation of the complete milling edge cutting path of the flaplike skin parts.(4)Carry out the measurement accuracy test experiment of the measurement system.The results show that the measurement system’s single measurement accuracy is better than ±0.02 mm,and the splicing measurement accuracy is better than ±0.05 mm;carry out the in-situ measurement and simulation processing experiments of the skin sample to verify the data processing algorithm and software module The correctness;the experiment of milling and trimming of the skin sample robot was carried out.The results showed that the precision of 1000 mm aviation aluminum alloy straight edge trimming in the processing system was ﹢0.3 mm,and the precision of φ300mm arc edge trimming was ±0.16 mm,and 6 kinds of skin samples were selected for actual processing.Test to verify that the overall function of the system meets the design requirements.