System and method for acquiring and determining axes of valves in tyre vulcanisation moulds

Abstract

The invention relates to a system (100) that implements a method for identifying vents (150) in a tyre vulcanisation mould (10) comprising one or more segments (10A) and an inner surface (10a) over which the vents are distributed in order to enable the corresponding valves (200) to be inserted therein. The invention also relates to a method implemented by the disclosed system (100).

Description

[0001] Description

[0002] Title: SYSTEM AND METHOD FOR ACQUIRING AND DETERMINING VALVE AXES IN TIRE VULCANIZATION MOLDS

[0003] Technical Field

[0004] The invention relates to a system and method for inserting valves into segments of a curing mold for tires. More particularly, the invention relates to a system and method for identifying vents of a vulcanization mold for tires whose vents are dispersed to allow the insertion of the corresponding valves therein.

[0005] Context

[0006] In the field of tires, molds for vulcanization of the segment type are known. Referring to Figure 1, this type of mold is represented by a mold 10 mainly comprising two shells (not shown) which each mold one of the lateral sidewalls of a tire P, a plurality of segments 12 which mold the tread Pio of the tire P along the internal surfaces 12a of the segments. The segments 12 are radially movable between an open position (shown in Figure 1) and a closed position of the mold 10. This type of mold may further comprise at least one clamping ring (not shown) to allow radial movement of the segments. An example of this type of mold is disclosed by the Applicant's patent US10,239,270.

[0007] The manufacture of tires using this type of mold requires that pressure be applied to the green tire in order to press it against the internal surfaces of the mold at the same time as heat is supplied to the mold (e.g., by electrical induction and / or magnetic induction, or by means of a heat transfer fluid such as pressurized steam). For this reason, this type of mold must be ventilated so that the green tire inflates against the internal surfaces of the mold segments.

[0008] It is therefore also known that this type of mold comprises a plurality of ventilation holes (or "vents") to achieve this ventilation during the vulcanization cycles. For example, a typical segment mold may comprise between 4000 and 12000 substantially cylindrical vents distributed along each segment of the mold. In each of the vents there is a valve 20 of the type shown by way of example in Figure 2 (see, for example, patent EP774333B1). The valve 20 comprises a movable insert 22 which moves up and down in a substantially cylindrical housing 24. The movable insert 22 comprises a valve stem 26 with a conical section 26a frustrated towards an internal cavity 28 (see Figure 2) and a flat surface 26b towards the surface of the tire.The tapered section 26a mates with a seating surface 24a of the housing 24 such that, during a vulcanization cycle, the valve is closed by the approaching tire blank surface, and, during tire extraction, the valve reopens after vulcanization. A gasket (not shown) may be disposed between the tapered section 26a and the seating surface 24a in a manner understood by those skilled in the art.

[0009] Valves are small, rigid, tubular mechanical parts (e.g., with a diameter of around 2.5 mm and a length of 5 to 12 mm). Their installation in the mold involves force-fitting them into vents drilled to a diameter that guarantees the valves' fit and hold throughout the mold's life. The installation operation requires:

[0010] Location of the vent to insert the valve;

[0011] The valve socket in the correct direction;

[0012] The positioning of the valve in the vent;

[0013] The force generation required to seat it in the adjustment; and The pressure applied until the valve is flush.

[0014] The valves are placed individually in the mold segments (either by a human operator or a mechanical operator such as a robot). This operation is usually carried out using a tweezer-type tool that grips the valve and inserts it precisely into the corresponding vent in the mold. The valve is then hammered into the vent using a hammer and a mandrel. This type of insertion requires a lot of effort and is time-consuming. Each insertion represents several seconds of work, leading to a tedious, repetitive task that, for a human operator, is not of great interest. This leads to risks of fatigue and forgetting valves, calling into question the proper functioning of the mold. To overcome this problem, there are devices in the prior art for inserting valves into molds.For example, German publication DE102010060901 discloses a tool comprising a tubular guide system in which a valve is arranged. The tubular system is positioned directly above the vent where, by a force in the valve axis, a piston pushes the valve to push it in a guided and regulated manner. By means of a spring, the piston rises, and a new valve engages in the tubular system. Automation therefore lies in the positioning directly above the vents, but the valves must be positioned precisely so that they find their bearings.

[0015] Korean patent KR100845093B discloses a valve assembly system incorporating a machine for manufacturing valves in two parts: a body in which the spring is installed and the valve itself. The machine can be diverted to serve as the basis for a valve press-fitting system for inserting the valves into the vents. However, it lacks the ability to adapt to any mold shape and also to move to position the valves in segments.

[0016] Indeed, the holes that create the vents are not always made as indicated on the plans, and there are variations due to the manufacturing process that cause discrepancies (for example, vents are added, or molds are modified by hand). Since precise knowledge of the position of the vents and / or their axes is not absolutely guaranteed, it is desirable to develop a system that can do without this information, as a human operator would do when detecting and analyzing them themselves.

[0017] Thus, the disclosed invention uses knowledge of the mold segment to perform the insertion of the valves in a repetitive manner. The insertion of the valves is done with a force of up to around 70 kg, which requires good control of the trajectory of a robot to avoid damaging the mold. For this, the disclosed invention uses the coordinates of the vents and the detection of their centers and their normals to give a robot the correct approach and thrust trajectory to facilitate the installation of the valves.

[0018] Summary of the invention

[0019] The invention relates to a system implementing a method for identifying vents of a vulcanization mold for tires comprising one or more segments and an internal surface whose vents are dispersed to allow the insertion of the corresponding valves therein, characterized in that the system comprises: a robot incorporating a detection system with one or more sensors which detect the presence of one or more vents dispersed along the internal surface of the segment of the mold; a communication network which manages the data incoming to the system from the detection system;and one or more communication servers, each comprising one or more processors operatively connected to a memory configured to store an application for analyzing data representative of the imaged molds, and the one or more processors comprising a module for executing the analysis application which performs the processing of the images, the one or more processors of which are capable of executing programmed instructions stored in the memory to perform the following steps: a step of detecting a presence of a vent arrangement in the field of view of the detection system, which triggers to capture at least one image of the internal surface of the mold segment;and a step of searching, in the image captured by the detection system, for the presence of the detected vents, such that the detection system continues to capture the images if no vent is detected, until the search for the mold is exhausted. In some embodiments of the system of the invention, the detection system comprises at least one three-dimensional (3D) camera of the RGB-D type attached to the robot which provides 3D images represented in a set of 3D points with coordinates (X, Y, Z) In some embodiments of the system of the invention, the processor(s) are capable of executing programmed instructions stored in the memory to carry out the following steps: a step of annotating the positions of samples of the vents, this step comprising a step of creating a coordinate reference of the vents searched for in images captured by the detection system of the system;a segment reconstruction step comprising a step of constructing an annotated database storing captured images and pixel coordinates of the captured images; a step of analyzing the contours of the vents performed by the system analysis application execution module, this step comprising a step of determining the surface plane by finding the shape closest to a circle that represents the desired vent, this step further comprising a step of determining the normal vector to the determined surface plane to find the insertion axis of the valve; and a step of determining the diameter of the vent allowing the drilling of a valve of the corresponding diameter, during which each vent is identified by a contour analyzed during the step of analyzing the contours of the vents, and during which a corresponding center is identified by a point;so that the system receives the coordinates of an identified vent to choose a valve of the appropriate diameter.;

[0020] In some embodiments of the system of the invention, the execution module of the analysis application stored in the memory of the system employs annotation software to construct bounding boxes around the vents appearing on the captured image of the mold.

[0021] In some embodiments of the system of the invention, the processor of the system continuously trains at least one neural network whose output is the classification of the coordinates of the vents, such that the captured images reveal the positions of the vents. In some embodiments of the system of the invention, the at least one neural network is selected from convolutional neural networks.

[0022] In certain embodiments of the system of the invention, one or more steps employ the use of a neural network of the deformable transformer type.

[0023] In some embodiments of the system of the invention, the robot includes a gripping device supported by a pivotable elongated arm, the gripping device extending from the elongated arm to a free end where a gripper is disposed along a common longitudinal axis.

[0024] In some embodiments of the system of the invention, the gripper comprises a pivoting clamp incorporating gripping fingers that extend from a platform where attachment of the clamp to the free end of the gripping device is achieved, each finger comprising a member with a predetermined length that extends between an actuation end, where movement of the finger is achieved, and an opposing gripping end, where the finger grips the valve.

[0025] In some embodiments of the system of the invention, the one or more processors are capable of executing programmed instructions stored in the memory to perform a step of moving the robot so that it can place the valve for insertion into an identified vent in a segment of the mold.

[0026] The invention also relates to a method implemented by the disclosed system for identifying vents of a vulcanization mold for tires comprising one or more segments and an internal surface whose vents are dispersed to allow the insertion of the corresponding valves therein, characterized in that the method comprises the following steps: a step of positioning the mold in a field of view of a detection system of the system, so that the vents defined along the internal surface of at least one segment are visible, during which the detection system flies over the mold; a step of detecting a presence of an arrangement of vents in the field of view of the detection system, which triggers to capture at least one image of the internal surface of the segment of the mold;and a step of searching, in the image captured by the detection system, for the presence of the detected vents, so that the detection system continues to capture the images if no vent is detected, until the search for the mold is exhausted. In certain embodiments of the method of the invention, the method further comprises a control step carried out after the insertion of the valves into the vents of the mold. In certain embodiments of the method of the invention, the method further comprises a final step of positioning the robot directly above an identified vent, in the axis of insertion thereof, during which the robot blows the valve.;

[0027] Other aspects of the invention will become apparent from the following detailed description.

[0028] Brief description of the drawings

[0029] The nature and various advantages of the invention will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings, in which like reference numerals designate like parts throughout, and in which:

[0030] Figure 1 shows a perspective view of one embodiment of a segment-type vulcanization mold.

[0031] Figure 2 shows one embodiment of a valve inserted into a vent of the mold of Figure 1.

[0032] Figure 3 represents a schematic view of a system of the invention allowing the insertion of valves into a tire vulcanization mold.

[0033] Figure 4 represents an internal surface of a segment of a tire vulcanization mold whose vents are identified by the system of Figure 3.

[0034] Figure 5 shows an example of annotated bounding boxes that are constructed around vents appearing in an image of a mold captured in a method performed by the system of the invention.

[0035] Figure 6 shows an example of the points that define vents appearing on an image of a mold captured in a process carried out by the system of the invention.

[0036] Figure 7 represents an example of the points extracted in a method carried out by the system of the invention and forming one or more ellipses.

[0037] Figure 8 represents a normal vector of a vent identified in an image of a mold captured in a process carried out by the system of the invention.

[0038] Figure 9 represents vents identified by contours analyzed during a process carried out by the system of the invention.

[0039] Detailed description

[0040] Referring now to the figures, in which like numbers identify like elements, Figure 3 shows a valve insertion system (or "system") 100 of the invention. The system 100 implements a method of the invention for inserting valves (e.g., valves of the type shown in Figure 2) into segments of a vulcanization mold for tires (e.g., a mold 10 of the type shown in Figure 1 and having segments 12). The disclosed method incorporates a machine learning method that is based on data corresponding to images obtained from the mold, the algorithm employed of which analyzes the internal surface of the mold to position and insert the valve into an identified vent.

[0041] Referring to Figure 3, a mold 10 is positioned on a worktable or equivalent support 50 for processing by the system 100. The support 50 may be configured to move in a rotational, vertically alternating manner, and / or horizontally alternating manner, thus allowing processing of a variety of molds. Referring again to Figure 3, in one embodiment of the system 100, the system comprises a robot 102 having a gripping device 104 supported by a pivotable elongated arm 106. The gripping device 104 extends from the elongated arm 106 to a free end 104a where a gripper 108 is disposed along a common longitudinal axis. Attachment of the gripper 108 to the gripping device 104 may be accomplished by screwing an adapter to the free end 104a of the gripping device.It is understood that the attachment of the gripper 108 to the gripping device 104 may be achieved by one or more known attachment means (including, without limitation, welding, gluing and equivalent means).

[0042] In one embodiment of the gripper 108, the gripper includes a pivotable clamp 108a incorporating gripping fingers (or "fingers") 108b that extend from a platform 108c (where the adapter provides attachment of the clamp 108a to the free end 104a of the gripping device 104). Each finger 108b includes a member with a predetermined length that extends between an actuation end (where movement of the finger is provided) and an opposing gripping end (where the finger grips a valve 200 retained by the clamp during the process performed by the system 100). Each finger 108b has an inner gripping surface that engages the valve 200 during insertion into an identified vent and an opposing outer surface. The fingers 108b are arranged so that a predetermined gap is defined between the gripping surfaces, allowing movement of the fingers along a common axis during the method implemented by the system 100.Thus, the robot 102 facilitates the gripping of a variety of valves without interruption of the linear movement of the fingers.

[0043] The reciprocating movement of one or more fingers 108b may be achieved by one or more known cylinders that are actuated by a pressurized fluid (e.g., compressed air) from a conduit (not shown). Accordingly, the movement of each finger 108b achieves the corresponding linear movement of the fingers between a standby position (where the engagement surfaces remain substantially parallel with the gap between them) (not shown) and an engagement position (where the engagement surfaces approach to engage the valve 200 and to place it in an insertion position relative to an internal surface 10a of the mold 10) (see Figure 3). The cylinder(s) are selected from commercial cylinders.

[0044] During the method implemented by the system 100, the robot 102 can be set in motion so that the gripper 108 can perform the gripping of the valve 200 (as described below). By means of the fingers 108b, the gripper 108 performs a gripping to hold the valve 200 during a movement of the gripper between a gripping position (in which the gripper 108 performs the gripping of a selected valve for insertion into a corresponding identified vent) (see FIG. 3) and an insertion position (in which the gripper 108 places the picked valve for insertion into the identified vent) (not shown). In the embodiments of the gripper 108 comprising the gripping fingers 108b, the gripping position of the gripper 108 means that the fingers are in their position for gripping the selected valve. In all embodiments of the robot 102, the robot may be configured to have six degrees of freedom allowing it to move along all six axes.In all embodiments, the robot 102 may be disposed on a support 55 which is configured to move in a rotational, vertically reciprocating manner and / or horizontally reciprocating manner, thus enabling the processing of a variety of molds.

[0045] The robot 102 is set in motion to place the valve 200 for insertion into a vent identified in a segment 10A of the mold 10. In one embodiment of the system 100, the robot 102 may be part of a roving robot that may be set in motion either by integrated motion means (e.g., one or more integrated motors) or by non-integrated motion means (e.g., one or more autonomous mobile carts or other equivalent mobile means). In another embodiment of the system 100, the robot 102 may be attached to a ceiling, a floor, a wall, or any support that allows the method implemented by the system 100 to be carried out (see, for example, the support 55 of FIG. 3). It is understood that such a robot may be a conventional industrial robot or a collaborative robot or even a delta or cable robot.

[0046] The robot 102 includes a sensing system that uses one or more sensors (not shown) to sense information about the physical environment around the robot. In the following description, the terms "sensor," "camera," "camera," and "optical sensor" may be used interchangeably and may refer to one or more devices configured to perform two-dimensional (2D) and / or three-dimensional (3D) image sensing, 3D depth sensing, and / or other types of sensing of the physical environment around the robot 102. In embodiments of the system 100, the sensors of the sensing system incorporated with the robot 102 may be attached to the elongated arm 106 (e.g., at the end 104a) and / or the gripper 108 of the robot.

[0047] The sensor(s) of the robot detection system 102 detect the presence of one or more vents of a mold. For example, an inner surface 10a of a segment 10A of a vulcanization mold 10 is shown in FIG. 4 (FIG. 4 shows a photo of the inner surface 10a taken with an RGB type camera). A plurality of vents 150 are dispersed along the inner surface 10a of the segment 10A, each vent receiving a corresponding valve 200. It is expected that each vent is substantially cylindrical and that all the vents 150 have substantially the same diameters.

[0048] In some embodiments of the robot 102, the sensor is triggered when a segment of a mold enters the field of view of the camera. In cases where a mold portion is not visible in the image obtained by the detection system of the robot 102 (e.g., the camera of the detection system), a hook point may be placed at a known position relative to the sensor (e.g., at a known horizontal distance and a known vertical distance from the position of the sensor).

[0049] The sensing system may determine information about the physical environment around the mold 10 that may be used by a control system of the system 100 (the control system including, for example, software for planning the motions of the robot 102). The control system could be located on the robot 102 or it could be in remote communication with the robot. In embodiments of the system 100, one or more 2D or 3D sensors mounted on the robot 102 (including, without limitation, navigation sensors) may be integrated to provide a digital model of the physical environment (including, where applicable, the side(s), floor, and ceiling). Using the obtained data, the control system may cause the robot 102 to move between the tapping positions of the valves during their insertion into the mold 10.

[0050] In one embodiment of the system 100, the sensing system includes at least one camera that provides 3D images represented as a set of 3D points with coordinates (X, Y, Z), and sometimes red, green, blue color values ​​(the "RGB" or "RGB-D" format) (referred to as an "RGB-D camera"). In this embodiment, an RGB-D camera is attached to the robot 102 (e.g., to the end 104a and / or the gripper 108). Two or more RGB-D cameras may be oriented to provide a predetermined overlap between the cameras' fields of view. As used herein, the term "camera" includes one or more cameras.

[0051] RGB-D cameras generally provide depth information using depth maps, being images where each pixel contains the distance between the camera and the corresponding point in space. Compared to traditional measurement methods such as manual measurement and other measurements based on electronic devices, 3D point cloud data from RGB-D type cameras have a much higher measurement rate. By using a sparser structure, a point cloud can be constructed from the RGB-D images by calculating the real world (e.g., (X, Y, Z) coordinates) with the intrinsic data of a scanning camera. Thus, information about the physical environment around the system 100 is obtained from the 3D point cloud data obtained from sensing technologies that are capable of capturing the 3D surface geometries of the molds accurately and efficiently.These detection technologies could be selected from commercially available devices (selected, for example, from cameras sold under the ZIVID® brand of Zivid AS, machine vision systems sold by Cognex Corp., and their equivalents).

[0052] The term "point cloud" (singular or plural) is used herein to refer to a collection or collections of data points in space. A camera(s) (or equivalent device(s)) can collect three-dimensional (3D) data and detect the surfaces of objects (e.g., a segment 10A of a mold 10) using a series of coordinates. Storing the information as a collection of spatial coordinates can save space because many objects do not fill a large portion of the environment. Even if the information is not visual, interpreting the data as a point cloud helps in understanding the relationship between multiple variables through classification and segmentation.

[0053] It is understood that one or more cameras may include one or more programming modes, including by learning, to feed, modify and train at least one neural network. The detection system of the robot 102 detects the presence of a vent arrangement 150 in the field of view of the detection system (e.g., the field of view of a camera of the system 100), which triggers it to capture the image of an internal surface 10a of the segment 10A of a mold 10 (see FIG. 4). In all embodiments of the system 100, the system “searches”, in the image obtained by the detection system, for the presence of the vents “seen” by the robot 102. If no vents are detected, the detection system continues to obtain the images until the search of the mold 10 is exhausted. The perimeter points of each detected vent are extracted to determine its center in preparation for the insertion of a corresponding valve.The detection system of the system 100 may comprise a rangefinder means which is used in the working space of the mold 10 to deduce its dimensions. In this embodiment, the rangefinder means comprises a scanner (not shown) for scanning the entire internal surface 10a of the mold 10 in real time in the physical environment around the mold. Such a scanner allows for accurate generation of the mold. The scanner may be provided together with a vision system (not shown) configured to precisely locate the vents in a real-time scenario based on the 3D profile generated by the scanner.

[0054] The vision system may receive a CAD file of the mold 10 to match the location of a vent from the CAD file with the identified vent in real time to accurately locate and determine its coordinates. The vision system may receive the CAD file by data transmission methods known to those skilled in the art. The vision system may further include at least one camera and at least one sensor (not shown) to determine the location (i.e., coordinates) of the vents based on the data collected in real time and / or the contour profile generated by the scanner.

[0055] To implement the method of the invention by computer means, the system 100 comprises a communication network (or "network") which manages the data incoming to the system from various sources (for example, from at least one robot 102 and the associated detection system). The communication network incorporates one or more communication servers (or "servers") each comprising one or more processors operatively connected to a memory. The memory is configured to store an application for analyzing data representative of the molds (and segments of the molds) imaged. The one or more processors comprise an analysis application execution module which performs the processing of the images, the one or more processors of which are capable of executing programmed instructions stored in the memory to carry out the steps of the method (as described below).

[0056] The term "processor" (or, alternatively, the term "programmable logic circuit") refers to one or more devices capable of processing and analyzing data and including one or more software programs for processing them (e.g., one or more integrated circuits known to those skilled in the art as being included in a computer, one or more controllers, one or more microcontrollers, one or more microcomputers, one or more programmable logic controllers (or "PLCs"), one or more application-specific integrated circuits, one or more neural networks, and / or one or more other known equivalent programmable circuits). The processor includes one or more software programs for processing the data captured by the detection system of the system 100 (and the corresponding data obtained) as well as one or more software programs for identifying and locating variances and identifying their sources to correct them.

[0057] In the system 100, the memory may include both volatile and non-volatile memory devices. The non-volatile memory may include solid-state memories, such as NAND flash memory, keep-alive memory (KAM) for saving various operating variables while the processor is powered off, magnetic and optical storage media, or any other suitable data storage device that retains data when the system 100 is powered off or loses power. The volatile memory may include static and dynamic RAM that stores program instructions and data, including a learning application.

[0058] Referring again to Figures 1-4, and further to Figures 5-9, a detailed description is given by way of example of embodiments of a method of the invention (or "method") implemented by the system 100. It is understood that the system 100 can implement the method of the invention in any physical environment without prior knowledge of the mold configuration.

[0059] As used herein, the term “method” or “process” may include one or more steps performed by at least one computer system having one or more processors to execute instructions that perform the steps. Unless otherwise indicated, any sequence of steps is exemplary and does not limit the described methods to any particular sequence.

[0060] In the following description, embodiments of the method of the invention are described which differ in the accuracy of the information obtained by the detection system (e.g., the camera).

[0061] In carrying out the method of the invention, the system 100 incorporates a combination of vision and machine learning techniques to correctly and quickly reconstruct the observed scene from three-dimensional (or "3D") scattered point clouds, originating from a view of the segment 10A of the mold 10. The system 100 therefore achieves a continuous improvement in the recognition of the vents and their relative distribution along the internal surface 10a of the mold 10.

[0062] In launching an embodiment of the method of the invention, the method comprises a step of positioning the mold 10 in the field of view of the detection system of the robot 102 (for example, positioning the mold on the support 50 as shown in FIG. 3). The mold 10 is positioned so that the vents 150 defined along the inner surface 10a of at least one segment 10A are visible in the detection field of the sensor (see FIG. 4). During this step, the robot 102 (and particularly the integrated detection system) flies over the mold 10.

[0063] The method further comprises a step of annotating the positions of samples of the vents dedicated to learning. During this step, a reference is created of the coordinates of the vents sought in images captured by the detection system of the robot 102 (for example, a camera of the RGB type). The reference of coordinates of the vents which is created during this step comprises expected images corresponding to the vents 150 distributed along the internal surface 10a of the mold 10. This step can be carried out in advance of other steps of the method of the invention to feed a neural network the true coordinates of the vents and their relative positions with respect to each other. In this embodiment of the method, at least part of the reference of the vents can be created by one or more persons skilled in the art.

[0064] In embodiments of the method, during this step, a neural network may be trained to recognize the true coordinates of the vents and create bounding boxes (or "boxed regions") around the recognized vents. During this training, the coordinates of the bounding box of the recognized vent are correlated with the coordinates of the searched vents to calculate displacements between them. The boxed regions and the displacement calculations are transmitted to a neural network (e.g., one or more CNNs) to jointly learn the representation of a vent in different perspectives of the images taken by the detection system of the robot 102.

[0065] In this embodiment of the method of the invention, the method further comprises a step of capturing images of the mold 10 (and more particularly, capturing images of the inner surface 10a of the segment 10A of the mold). This step, which is performed by the robot 102 (and particularly by the associated detection system), comprises a step of scanning the detection system of the robot 102 over the segment 10A of the mold 10. Each image captured during this step is composed of a matrix of pixels where each pixel has a different color and a brightness that indicates the position of a vent 150 of the mold 10. The images obtained, revealing one or more positions of the vents 150, train at least one neural network to identify all the expected positions of the vents in the imaged mold 10. Thus, these image variations serve as input to the neural network whose outputs are the classification of the coordinates of the vents.

[0066] During this step, the algorithm of the execution module aims to automatically locate and indicate the external profile of the vent as well as the interfaces and perimeters of the vent (for example, its diameter and the angle of its cylindrical axis relative to the curvature of the internal surface 10a of the mold 10). During this step, the execution module therefore uses annotation software to construct bounding boxes around the vents 150 appearing on the image of the mold 10 (see Figure 7).

[0067] The processor of the system 100 continuously trains the neural network from newly captured data of the images of the molds obtained by the detection system of the robot 102. In order to automatically detect the boundaries between the vents and the surrounding metal material of the mold, the robot 102 takes images (which may include videos) and collects an image data set from multiple molds of the same type (e.g., of the type represented by the mold 10 of FIG. 4). Before being recorded, the image data set may be annotated based on the data entered by the operator to create the ground truth data. For example, in some embodiments, to assist the neural network in detecting and identifying the boundaries of the vent 150 and / or the surrounding metal material of the mold, the image data set is annotated.Known variations are manually identified based on the knowledge of mold professionals.

[0068] During this step, the server processor may use the ground truth data to train and / or develop one or more neural networks to automatically detect the space where the object (e.g., the vents 150 of the mold 10) is located. As such, ground truth data as described herein generally refers to information provided by direct observation of professionals in the field as opposed to information provided by inference. It may have data from multiple sources, including multiple professionals located in remote locations, to develop the neural network. A feedback loop of the annotated images may be updated with additional ground truth data over time to improve the accuracy of the system 100.

[0069] In this embodiment of the method of the invention, the method further comprises a step of reconstructing the segment 10A by three-dimensional scanning with a high degree of resolution. During this step, the bias of a 3D camera of the detection system is used in the workspace of the mold 10 to immediately reconstruct the volume of the segment 10A and to determine its dimensions and to delimit its work area. This step comprises the construction of an annotated database storing the RGB images, the coordinates (X,Y,Z) of the pixels of the images obtained as well as the coordinates of the bounding boxes. During this step, the algorithm of the execution module aims to extract the points that define each vent 150 (see Figure 8).Using the extracted points together with their coordinates (X, Y, Z) and the analysis of the surface whose vent is defined by these points, it is possible to find the center C150 of each reconstructed vent (see again Figure 8) and therefore the normal vector of the surface plane. The orientation of the reconstructed vent 150 relative to the internal surface 10a of the mold 10 indicates the corresponding orientation of a valve to achieve its correct insertion into the mold. Thus, the location of the mold 10 and its limits can be deduced as well as its geometry (including the sculpture elements), the internal surface 10a, the normals at any point of the mold, the vents 150 used to insert the valves.

[0070] In this embodiment of the method of the invention, the method further comprises a step of analyzing the contours of each vent 150. In this embodiment of the method of the invention, the 3D camera of the robot detection system 102, by the width of its measurement spectrum, gives better precision by calculating the normal on the homogenized plane on the internal surface 10a of the mold 10. During this step, each of the extracted points (including the center C150) has corresponding coordinates.

[0071] This step includes a step of determining the surface plane by finding the shape closest to the circle that the desired vent represents. Due to the positioning of the vent along the internal surface 10a of the mold 10 (having substantially curved parts), it is understood that the extracted points can form one or more ellipses (see the ellipses given as an example in Figure 9).

[0072] This step also includes a step of determining the normal vector to the surface plane determined during the previous step. Referring to Figure 8, a determined normal vector passes the center C150 of the vent 150 to find the insertion axis X200 of the valve 200 (oriented, for example, at an angle a relative to the internal surface 10a of the mold 10). This step will allow the robot 102 to choose and orient a valve of appropriate diameter (for example, a valve of the type shown in Figure 2) to ensure its correct insertion into the corresponding vent.

[0073] In this embodiment of the method of the invention, the method further comprises a step of determining the diameter of the vent allowing the drilling of a valve of the corresponding diameter. Referring to Figure 9, each vent 150 is identified by a contour 150A analyzed in the previous step. The corresponding center C150 is identified by a point. During this step, the algorithm of the execution module aims to recognize the diameter which corresponds closest to the diameters of the known vents (known, for example, in the reference of the vents created during the annotation step of this embodiment of the method).

[0074] In this embodiment of the method of the invention, the method further comprises a final step of positioning the valve 200 in the identified vent having the corresponding diameter. During this step, the robot 102 receives the coordinates of an identified vent so that the robot can choose the valve of the appropriate diameter (for example, a valve of the type shown in Figure 2). During this step, the robot 102 can choose the valve by means of a tool changer and a valve feed system (both being commercially known). During this step, the robot 102 can position itself directly above the identified vent 150, in the insertion axis thereof and blow the valve.Once the valve is pre-positioned thanks to its shape (either conical or stepped), the robot 102 can proceed with the insertion, either by means of the valve feed head, or by pushing it with a dedicated zone of an effector deposited at the end 104a (not shown). The use of neural networks provides robustness in the determination of the vents and especially speed which is freed from long calculations. Although the incarnations are described here with regard to the use of neural networks (and more particularly convolutional neural networks, or "CNN" in English) as a machine learning model, other types of machine learning models can be used.These include, but are not limited to, models using linear regression, logistic regression, decision trees, support vector machines, naive Bayes, nearest neighbor (knn), K stands for clustering, random forest, dimensionality reduction algorithms, gradient-based algorithms, neural networks (e.g., autoencoders, RNNs, perceivers, logarithmic short-term memory (LSTM), Hopfield, Boltzmann, deep belief, deconvolution, generative adversarial (GAN), etc.) and their complements and equivalents. The CNN(s) may be trained with ground truth data that is generated using sensor data representative of the motion of the robot 102, including the positioning of the gripper 108.

[0075] One or more steps of this embodiment of the method of the invention employs the use of a neural network of the deformable transformer type (or "deformable DETR" or "DETR"). The DETR is used for end-to-end object detection, by combining neural networks of the CNN type and coders-decoders of the "Transformer" type. The DETR first reduces the computations by only being interested in a small set of key sampling points around a reference (for example, the points form the contour 150A around a vent 150 of the mold 10) (see Figure 9). The DETR then uses a deformable attention module to aggregate multi-scale features to facilitate the detection of small objects.Therefore, DETR can model dependencies between distant objects in the observed scene to achieve the ability to automatically and accurately detect, localize, and classify vents that are planned to be inserted with corresponding valves.

[0076] In all embodiments of the method of the invention, the method may further comprise an optional inspection step after the valves 200 are inserted into the vents 150 of the mold 10. During this step, an operator may perform a unitary manual inspection of everything that the robot 102 provides. During this step, a fully automatic inspection may be performed, involving presence detection and / or a feeler to validate the presence as well as the proper functioning of the valves.

[0077] It is understood that all embodiments of the method of the invention can be carried out in the same factory (e.g., by a single installation incorporating the system 100). By using the system 100 of the invention to carry out the disclosed method, any mold presented to the system 100 is analyzed in the same way. There is no need to know the CAD file in advance or to have arrangements in the mold to position it flawlessly. The system 100 is natively designed to accommodate variations, which gives, for example, the possibility of working with third-party molds and / or molds retouched by hand.

[0078] The system 100 of the invention may include pre-programming of information regarding expected events. For example, a setting of the process of the invention may be associated with the parameters of typical physical environments in which the system 100 operates (e.g., tire production facilities). In embodiments of the invention, the system 100 (or another system incorporating the system 100) may receive audio commands (including voice commands) or other audio data representing (e.g., a start or stop of one or more steps of the process of the invention). The request may include a request for the current state of an ongoing process (e.g., the number of valves inserted versus the number of vents 150 in the mold 10 intended to receive a corresponding vent).A generated response can be represented audibly, visually, tactilely (e.g., using a haptic interface), and / or virtually and / or augmented. This response, together with the corresponding data, can be recorded in a neural network.

[0079] It is understood that the system 100 may include multiple computing devices that perform various aspects of learning. In these embodiments, the processor may configure the system 100 to one or more parameters of a vent and its known location. In these embodiments, it is understood that one or more means of reinforcement learning could be employed.

[0080] For all embodiments of the system 100, a monitoring system could be implemented. At least part of the monitoring or alerting system may be provided in a portable device such as a mobile network device (e.g., a mobile phone, a laptop, one or more network-connected portable devices (including “augmented reality” and / or “virtual reality” devices, network-connected wearables, and / or any combinations and / or equivalents). It is conceivable that detection and comparison steps may be performed iteratively.

[0081] The terms "at least one" and "one or more" are used interchangeably. Ranges that are presented as "between a and b" encompass the values ​​"a" and "b".

[0082] Although particular embodiments of the disclosed apparatus have been illustrated and described, it will be understood that various changes, additions, and modifications may be practiced without departing from the spirit and scope of the present disclosure. Accordingly, no limitations should be imposed on the scope of the disclosed invention except those set forth in the appended claims.

Claims

Claims 1. System (100) implementing a method for identifying vents (150) of a vulcanization mold (10) for tires comprising one or more segments (10A) and an internal surface (10a) whose vents are dispersed to allow the insertion of the corresponding valves (200) therein, characterized in that the system comprises: a robot (102) incorporating a detection system with one or more sensors that detect the presence of one or more vents (150) dispersed along the internal surface (10a) of the segment (10A) of the mold (10); a communication network that manages the data incoming to the system (100) from the detection system;and one or more communication servers, each comprising one or more processors operatively connected to a memory configured to store an application for analyzing data representative of the imaged molds, and the one or more processors comprising a module for executing the analysis application which performs the processing of the images, the one or more processors of which are capable of executing programmed instructions stored in the memory to perform the following steps: a step of detecting a presence of an arrangement of vents (150) in the field of view of the detection system, which triggers to capture at least one image of the internal surface (10a) of the segment (10A) of the mold (10); and a step of searching, in the image captured by the detection system, for the presence of the detected vents (150), so that the detection system continues to capture the images if no vent is detected, until the search for the mold (10) is exhausted.; 2. System (100) of claim 1, wherein the detection system comprises at least one three-dimensional (3D) camera of the RGB-D type attached to the robot (102) which provides 3D images represented in a set of 3D points with coordinates (X, Y, Z).

3. The system (100) of claim 2, wherein the processor(s) are capable of executing programmed instructions stored in the memory to perform the following steps: a step of annotating the positions of samples of the vents (150), this step comprising a step of creating a coordinate reference of the vents sought in images captured by the detection system of the system (100); a step of reconstructing the segment (10A) comprising a step of constructing an annotated database storing captured images and coordinates (X,Y,Z) of the pixels of the captured images; a step of analyzing the contours (150A) of the vents (150) carried out by the analysis application execution module of the system (100), this step comprising a step of determining the surface plane by finding the shape closest to a circle which represents the vent (150) sought, this step further comprising a step of determining the normal vector to the surface plane determined to find the insertion axis (X200) of the valve (200);and a step of determining the diameter of the vent allowing the drilling of a valve (200) of the corresponding diameter, during which each vent (150) is identified by a contour (150A) analyzed during the step of analyzing the contours of the vents, and during which a corresponding center (C150) is identified by a point; so that the system (100) receives the coordinates of an identified vent to choose a valve (200) of the appropriate diameter.; 4. The system (100) of claim 3, wherein the execution module of the analysis application stored in the memory of the system employs annotation software for constructing bounding boxes around the vents (150) appearing on the captured image of the mold (10).

5. The system (100) of claim 4, wherein the processor of the system continuously trains at least one neural network whose output is the classification of the coordinates of the vents (150), such that the captured images reveal the positions of the vents (150).

6. The system (100) of claim 5, wherein the at least one neural network is selected from convolutional neural networks (CNNs).

7. The system (100) of claim 6, wherein one or more steps employ the use of a deformable transformer type neural network (DETR).

8. The system (100) of any one of claims 1 to 7, wherein the robot (102) comprises a gripping device (104) supported by a pivotable elongate arm (106), the gripping device (104) extending from the elongate arm (106) to a free end (104a) where a gripper (108) is disposed along a common longitudinal axis.

9. The system (100) of claim 8, wherein the gripper (108) comprises a pivotable clamp (108a) incorporating gripping fingers (108b) extending from a platform (108c) where attachment of the clamp to the free end (104a) of the gripping device (10) is performed, each finger (108b) comprising a member with a predetermined length extending between an actuation end (108b'), where movement of the finger is performed, and an opposite gripping end (108b”), where the finger grips the valve (200).

10. The system (100) of any one of claims 1 to 9, wherein the processor(s) are capable of executing programmed instructions stored in the memory to perform a step of moving the robot (102) so that it can place the valve (200) for insertion into a vent identified in a segment (10A) of the mold (10).

11. Method implemented by a system (100) of any one of claims 1 to 10 for identifying vents (150) of a vulcanization mold (10) for tires comprising one or more segments (10A) and an internal surface (10a) whose vents are dispersed to allow the insertion of the corresponding valves (200) therein, characterized in that the method comprises the following steps: a step of positioning the mold (10) in a field of view of a detection system of the system (100), so that the vents (150) defined along the internal surface (10a) of at least one segment (10A) are visible, during which the detection system flies over the mold (10); a step of detecting a presence of a vent arrangement (150) in the field of view of the detection system, which triggers to capture at least one image of the internal surface (10a) of the segment (10A) of the mold (10);and a step of searching, in the image captured by the detection system, for the; presence of the vents (150) detected, such that the detection system continues to capture images if no vent is detected, until the search for the mold (10) is exhausted.

12. The method of claim 11, further comprising a control step performed after the insertion of the valves (200) into the vents (150) of the mold (10).

13. Method of claim 11 or claim 12, further comprising a final step of positioning the robot (102) directly above an identified vent (150), in the insertion axis (X200) thereof, during which the robot insufflates the valve (200).

Images