Positioning and navigation system based on machine vision intended for laser-electrochemical micro-hole processing

Abstract

In order to improve efficiency and quality of micro-hole processing, a laser-electrochemical micro-hole processing is applied to the micro-hole processing of aero engine blade, i.e., laser drilling and electrochemical machining are performed. However, when electrode wire is inserted into micro-hole, due to the positioning error, it can collide with blade, which could result in short circuit or electrode wire damage. Hitherto, the positioning of electrode wire and micro-hole was achieved by artificial observation, but its efficiency is low, especially for micro-holes with small diameter. Namely, a large amount of asymmetric micro-holes is produced due to the low-accuracy positioning by artificial observation; therefore, machining accuracy and consistency of micro-hole cannot be guaranteed. In this study, a positioning and navigation system based on machine vision is applied for the first time to the laser-electrochemical micro-hole processing. The proposed system is based on parallel installation of camera and electrode wire. A novel hand-eye calibration method, wherein rotation angle and translation vector are calibrated step by step, is proposed for the case when electrode wire is beyond the view area of camera. The tree traversal binary image boundary tracking algorithm is adopted to image processing. The data stem definition is proposed to detect micro-hole edge with burrs. Before electrochemical machining, the micro-hole image is captured by camera that is navigated to the micro-hole top, and then, the captured image is processed in order to get the coordinates of micro-hole center, which enables navigation of electrode wire to the micro-hole center; thus, the accurate positioning and navigation of micro-hole are achieved.