Device and method for detecting position of edge of cutting tool

A method for detecting a position of an edge of a cutting tool comprising the steps of:

• positioning a reference plate in front of the edge of the cutting tool to form a slit extending in a direction transverse to the longitudinal direction of the cutting tool between the edge of the cutting tool and said reference plate;
• irradiating a laser light to said slit;
• detecting, along at least two nonparallel directions, light intensity values of a diffraction pattern created through diffraction of the laser light by said slit;
• and
• determining the position of the edge of the cutting tool based on a comparison of the detected light intensity values of the diffraction pattern.

A device for detecting the position of an edge of a cutting tool comprising:

• a reference plate positioned to form a slit between the edge of the cutting tool and said reference plate;
• a laser light source for irradiating a laser light toward said slit;
• sensing means receiving said laser light passing through said slit and including at least one group of light sensitive elements which are arranged and spaced in a direction perpendicular to said slit for enabling said light sensitive elements to output a combination of signals corresponding to a diffraction pattern of said laser light received by said sensing means;
• and
• processing means including a neural network responsive to said combination of signals output from said light sensitive elements for detecting the position of said edge of said cutting tool relative to said reference plate in accordance with an input-to-input relation which has been in advance taught into said neural network.

A method for detecting the position of an edge of a cutting tool comprising the steps of:

• positioning a reference plate to form a slit between the edge of the cutting tool and said reference plate;
• irradiating a laser light from a laser light source toward said slit;
• receiving on sensing means said laser light passing through said slit;
• said sensing means including at least one group of light sensitive elements which are arranged and spaced in a direction perpendicular to said slit so that said sensing means outputs a combination of signals corresponding to a diffraction pattern of said laser light;
• and
• detecting the position of said edge of said cutting tool relative to said reference plate based upon data signals which a neural network outputs upon receipt of said combination of signals from said sensing means in accordance with an input-to-output relation which has been taught in advance into said neural network.