Robot Contact In-situ Measurement Technology For Blades With Complex Surfaces

Aero engine blades are the most important components of aircraft power system.The processing quality of blades have a great influence on the performance and the life of aircraft.The complex surface of blades makes it very difficult to process and measure.At present,many of blades are still measured by coordinate measuring machine(CMM).The CMM can't realize in-situ measurement,which leads to low efficiency and high cost of measurement.In order to improve the processing quality and efficiency of blade,an automatic "machining-measurement" scheme for blades is proposed based on industrial robot technology.In this paper,we focus on the in-situ measurement technology for blades,and the main research works are as follows:1.Construction of in-situ measurement system for blade.A novel automatic "machining-measurement" scheme for blade based on robot technology is constructed.A light pen in-situ measurement method based on force and vision is proposed.The light pen is grasped by robot and touch to blade with force control,and the blade is scanned according to the planned measurement trajectory.At the same time,the non-contact measurement is realized by binocular vision system.This method combines the high accuracy of contact measurement and the high efficiency of visual measurement,which avoids the influence of motion error in traditional contact measurement device to measurement accuracy,and this method can realize in-situ measurement for blade.2.Three-dimensional reconstruction of light pen based on the binocular vision technology.The visual imaging model is established and the calibration of binocular vision system is completed by Zhang's calibration method.The pixel coordinates of feature circle centers in left and right images are obtained by image processing.Three-dimensional reconstruction of light pen is realized by stereo geometry method,and the spherical coordinates of probe are calibrated by least square method.The repetitive precision measurement experiment was carried out on the light pen in-situ measurement platform,and the results show that the repetitive measurement error is less than 0.282 mm,and the depth direction measurement error is larger than other directions.3.Measuring trajectory planning for blade surface.Based on the three-dimensional model of blade,the measuring points are planned according to the inverse process of blade design.The measuring trajectory of blade surface is extracted by the method of equal cross-section.By incorporating the curvature information of contours into the equal step sampling method,an adaptive sampling method is proposed to plan the measuring points on the blade surface.According to the measurement path,the trajectory of robot is planned,and the stability of robot motion is verified by simulation.4.Compliance control for contact measurement.The force control model of robot contact measurement is established,and the impedance controller is used to control robot.The force control algorithm of robot is simulated and experimented on the sinusoidal surface,and the results show that the controller is stable and the force tracking error is less than ±1N.The blade measurement experiment was carried out on the light pen in-situ measurement experimental platform,and the feasibility of the measurement method is verified.The measurement results were compared with CMM.The results show that the measurement error is less than 0.4 mm.The measurement error may be caused by environmental illumination,camera resolution,camera calibration accuracy,image contour extraction accuracy and excessive contact force.