Development of an artificial vision system for the automatic evaluation of the cutting angles of worn tools

Abstract

This article presents a new method to evaluate the geometry of dull cutting tools in order to verify the necessity of tool re-sharpening and to decrease the tool grinding machine setup time, based on a laser scanning approach. The developed method consists of the definition of a system architecture and the programming of all the algorithms needed to analyze the data and provide, as output, the cutting angles of the worn tool. These angles are usually difficult to be measured and are needed to set up the grinding machine. The main challenges that have been dealt with in this application are related to the treatment of data acquired by the system's cameras, which must be specific for the milling tools, usually characterized by the presence of undercuts and sharp edges. Starting from the architecture of the system, an industrial product has been designed, with the support of a grinding machine manufacturer. The basic idea has been to develop a low-cost system that could be integrated on a tool sharpening machine and interfaced with its numeric control. The article reports the developed algorithms and an example of application.