Robotized unit and method for controlling a metal product in movement in an apparatus, and method to produce metal reels
A robotized unit for controlling a metal product in motion driven in an automated manner, in accordance with a corresponding computerized method, in order to manipulate or cut a portion of the metal product includes an apparatus and method to produce metal reels.
1 . Computerized method to control a metal product in motion, comprising:
acquiring, by a video monitoring system of an acquisition and control unit, a sequence of frames of said metal product which are processed by a processing system, and
transmitting an actuation signal to a robotized articulated arm driving it according to said actuation signal,
wherein an artificial intelligence implemented in said processing system at least:
a) recognizes a tail portion of said metal product; and
b) identifies whether said tail portion has a macroscopic defect,
wherein according to the type of defect detected, moving said robotized articulated arm in order to reposition said tail portion on-the-fly by means of its own gripping tool, or to cut said tail portion on-the-fly by means of its own cutting tool.
2 . Method as in claim 1 , wherein said artificial intelligence comprises a recognition algorithm which identifies on each frame of said sequence of frames a first point and a second point which define between them a control strip, the width and position of which identify whether said frame is a central frame or a tail frame of said metal product, or an empty frame.
3 . Method as in claim 1 , wherein said artificial intelligence comprises a convolutional neural network comprising an input level or layer which receives, on each occasion, a frame of said sequence of frames, one or more convolutional levels or layers having the function of identifying macroscopic geometric and visual characteristics of said tail portion, at least one completely connected level configured to classify said frame on the basis of the information received from a last one of said one or more convolutional levels or layers, and an output level or layer defined by a binary class vector comprising a first class and a second class, both classes characterized by a respective confidence index of absence of defect and presence of defect.
4 . Method as in claim 3 , wherein if the confidence index relating to said second class exceeds a determinate preestablished threshold, the convolutional neural network establishes that said analyzed frame relates to a metal product with a defect, it generates a specific actuation signal which is transmitted to said robotized articulated arm for the correction of the defect.
5 . Method as in claim 3 , wherein said at least one completely connected level is configured to classify said frame on the basis of the information received from a last one of said one or more convolutional levels or layers into respective classes of belonging that comprise tail frames with shape defects and tail frames with cobble defects.
6 . Robotized unit for controlling a metal product in motion, comprising:
at least one robotized articulated arm and an acquisition and control unit comprising a video monitoring system and a processing system configured to acquire and process a sequence of frames of said metal product and transmit an actuation signal to said robotized articulated arm in order to manipulate or cut a portion of said metal product,
wherein said processing system implements an artificial intelligence configured at least to:
a) recognize a terminal portion of said metal product; and
b) identify whether said terminal portion has a macroscopic defect, said processing system being configured to selectively manage movements of said robotized articulated arm on the basis of the defect identified.
7 . Robotized unit as in claim 6 , wherein said robotized articulated arm can be moved in at least six degrees of freedom.
8 . Robotized unit as in claim 6 , wherein said video monitoring system comprises at least one video camera positioned on a manipulation head of said robotized articulated arm.
9 . Robotized unit as in claim 6 , wherein said video monitoring system comprises at least one video camera positioned on a fixed support.
10 . Robotized unit as in claim 6 , further comprising an interception member vertically mobile between a resting position in which it is located below a movement plane of said metal product without interfering with the latter, and an interception position in which it is incident to said movement plane in order to temporarily stop a part of said metal product which is intended to be manipulated/cut with said arm.
11 . Method to produce metal in which comprising:
receiving, by a winding unit, a metal product in linear form and a laying head conforms it into coils,
forming, by an automatic reel forming unit, said metal product in coils and forms a reel, and
moving, by a roller way, said metal product in coils from an initial end, in correspondence with which said winding unit is disposed, to a terminal end, in correspondence with which said automatic reel forming unit is disposed,
wherein the robotized unit, as in claim 1 , is moved laterally into a position between said initial end and said terminal end of said roller way in order to manipulate, reposition or cut said metal product.
12 . Apparatus for the production of metal reels comprising:
a winding unit having a laying head and configured to receive a metal product in linear form and to conform it into coils,
an automatic reel forming unit able to receive said metal product conformed in coils and to form a reel,
a roller way configured to move said metal product conformed in coils from an initial end in correspondence with which said winding unit is disposed, to a terminal end in correspondence with which said automatic reel forming unit is disposed, and
the robotized unit, as in claim 1 , disposed in a position comprised between said initial end and said terminal end of said roller way.