JP2019501033A - Method and equipment for composing batches of parts from parts placed in different storage areas - Google Patents

Abstract

  Using an articulated system (7) subordinate to at least one vision system (2), from parts placed in at least two different storage areas (5, 13), at least parts (6) in the receiving area (10) A method for automatically configuring a batch, wherein the articulated system comprises at least two separate gripping members each capable of gripping a part (6) by at least one specific gripping region. According to the invention, the method comprises the step of automatically moving the articulated system (7) to the first storage area (5, 13), and (i) using the visual system (2). A stage consisting of capturing an image of the storage area (5, 13), and (ii) a stage consisting of processing all the information resulting from the image and identifying all gripping areas of the part (6) that can be fitted with the gripping member (Iii) comprising locating the identified gripping region in an appropriate position and orientation; (iv) selecting a gripping region that has the greatest likelihood of being gripped by the gripping member and is free of collision constraints; For the corresponding gripping member, comprising automatically calculating a path for gripping the part by the selected gripping area; and (v) following the calculated path for the corresponding part (6). The The articulated system (7) is automatically moved to the receiving area (10) to place the part (6) in the receiving area (10), At least including the step of automatically moving to the second storage area (5, 13) and repeating steps (i) to (v).

Description

  The present invention relates to the field of handling parts by articulated systems. The expression “articulated system” is understood to mean, for example, a six-axis robot system.

  More particularly, the present invention relates to the separation of discrete parts from separate parts located in at least two different storage areas (collection areas) automatically and using an articulated system controlled by a vision system. The present invention relates to a method using an articulated system in which a batch is configured in a receiving area (arrangement area).

  The present invention has an advantageous application in so-called kitting methods, in which separate but related individual parts are grouped together to form a kit meaning a batch or collection of parts. The parts can then be assembled together and / or packaged and / or delivered.

  More generally, the present invention is advantageously applicable to so-called picking methods, which consist for example of collecting parts one by one in preparation for ordering.

  In the application involved, the storage area can be of various types without departing from the scope of the present invention. The storage area consists of, for example, a fixed area for receiving stacked bulk parts or a movable area such as a conveyor for transporting the bulk parts. The expression “bulk” is understood to mean a collection of parts that are randomly oriented in space or plane. Each storage area includes a plurality of identical parts.

  In general, at the current state of the art, so-called picking or kitting tasks are either performed manually by an operator or by an articulated system controlled by a visual system for gripping and placing the parts. It is.

  However, the type of articulated system used is a type that requires learning of the gripped component and the step of recognizing the component.

This type of multi-joint system is known in particular from US Pat. No. 6,057,056, which describes a control method for a robot for the movement of at least one component arranged on a support in a three-dimensional space. ing. In particular, this method
-Acquiring at least two 2D images of the 3D space;
-Segmenting each previously acquired 2D image for the purpose of extracting at least one 2D geometric primitive from each 2D image;
-Looking for correspondence between 2D geometric primitives extracted from separate 2D images;
-For each group of corresponding 2D geometric primitives, calculating the coordinates of the 3D geometric primitives contained in the 3D space;
-Associating each 3D geometric primitive with a known part;
Looking for the position and orientation in 3D space of each known part associated with at least one 3D geometric primitive;
-Using a robot to move at least one of the known parts;
including.

  In this way, this method provides the ability to detect, position and orient parts in three-dimensional space.

  However, as suggested, this method has the disadvantage that it is not suitable for all kinds of moving parts. In fact, this method only works with parts that were previously known and identified by the learning software. In this method, references to unknown parts cannot be added during the method, nor can the configuration of the batch be changed. Effectively, another program that can drive the articulated system must be loaded to change the configuration of the batch.

French patent application No. 2998785

  The present invention ameliorates the aforementioned drawbacks, thereby automatically configuring a batch of parts without prior learning of the parts and without prior learning of the gripping path of the parts and The purpose is to provide equipment.

  For this purpose, an improved method uses an articulated system controlled by at least one vision system to at least one batch of parts in a receiving area from parts located in at least two different storage areas. Developed from automatic configuration, the articulated system comprises at least two separate gripping members each capable of gripping a part by at least one specific gripping region.

According to the invention, the method comprises:
-The step comprising automatically moving the articulated system towards the first storage area;
(I) comprising capturing an image of the storage area using a vision system;
(Ii) processing the information generated from the image and identifying all gripping areas of the part that are compatible with the gripping member;
(Iii) comprising locating the identified gripping region in an appropriate position and orientation;
(Iv) selecting a gripping region that has the greatest likelihood of being gripped by a gripping member and without any collision constraints between the gripping member and the environment, and for the corresponding gripping member, depending on the selected gripping region A step comprising automatically calculating a path for gripping the part;
(V) gripping the corresponding part following the calculated path and automatically moving the articulated system toward the receiving area to place the part in the receiving area;
-Automatically moving the articulated system towards the second storage area and repeating steps (i) to (v);
At least.

  Thus, the articulated system itself determines the best method for collecting parts in real time. The articulated system is moved or moved from one area to another in order to continuously collect parts with very different shapes to form a batch of parts.

  The articulated system operates without learning so that the introduction of new part references can be done without difficulty and without constraints. The articulated system does not attempt to recognize the part being gripped, but attempts to recognize a specific gripping region in the part being gripped. The part can be of any type and it is only sufficient that the gripping area is detected so that the part can be gripped by a gripping area that can be fitted with the gripping member.

  If the gripping region is positioned with the proper position and orientation, the method does not need to learn any about the path for gripping the corresponding part. The grip path is calculated for each part being gripped in real time. And very quick gripping and positioning operations can be performed with unknown parts.

  The advantage of the present invention lies in the fact that the articulated system is autonomous. After capturing and processing the images, the articulated system finds and reviews all possible gripping areas of the gripped part that can be fitted with the gripping members, and reviews all available gripping members and considers them. And, for the grip provided, consider all collision constraints in the gripping member that can interfere with said gripping, and finally, the best gripping region to grip and the best gripping member to use And decide.

  After gripping the part, the articulated system is moved or moved toward the receiving area to place the part in the receiving area, and the articulated system moves its position to grip another part in another storage area. Move to another storage area or proceed to a batch that has been moved and prepared.

  Advantageously, if the storage area comprises a storage container for the part, the method comprises a step consisting of detecting an empty storage container, a step consisting of gripping the empty storage container, and an empty storage container Comprising the steps of:

  If the storage container includes a package component such as a spacer, a cross brace, etc., the method consists of recognizing the package component, comprising grasping the package component, and removing the package component Comprising the steps of:

  Preferably, when the empty storage container is removed, the storage container containing the parts to be gripped is automatically replenished in the storage area.

  Advantageously, if the receiving area comprises a receiving container for a part, the method is satisfied with the steps consisting of detecting the filled receiving container and the step consisting of gripping the filled receiving container. Consisting of removing the receiving container.

  Preferably, an empty receiving container is automatically replaced in the receiving area when the filled receiving container is removed.

  Thus, it refers to intelligent independent robotics that can be adapted.

  According to a particular embodiment, the step consisting of identifying a gripping area that can be adapted consists of identifying a continuous flat area when the gripping member is of suction cup type.

  The expression “suction cup” is understood to mean both air suction and a suction cup operating in a magnetic manner, for example a magnet. Thus, to be compatible with the suction cup, it is clear that the identified continuous region must have a continuous region with a minimum surface area corresponding to the active surface area of the suction cup.

  According to another particular embodiment, the step consisting of identifying a specific area that can be adapted comprises identifying edges, busbars, or parallel surfaces when the gripping member is clamped. The maximum spacing between the edges, the busbars, or the surfaces should be compatible with the kinematic characteristics of the clamp. For example, the maximum spacing should be less than the maximum possible opening between the clamp jaws.

  The invention also relates to an installation for automatically composing at least one batch of parts in a receiving area from parts located in at least two different storage areas, said equipment taking in an image of the storage area An articulated system controlled by at least one possible visual system, and at least two separate gripping members each capable of gripping a part by at least one specific gripping region.

  According to the invention, the articulated system comprises means of movement between a storage area and a receiving area, the visual system and the articulated system identifying all gripping areas that can be fitted with a gripping member; Positioning the identified gripping region in the appropriate position and orientation and selecting the gripping region that has the greatest likelihood of being gripped by the gripping member and without any collision constraints between the gripping member and the environment And automatically calculating the path for gripping the part by the selected gripping area for the corresponding gripping member, and the storage area and the receiving area to place the gripped part in the receiving area. Are controlled using processes and calculations that can process information generated from the captured images.

  This installation can thus implement the method according to the invention in a way that has all the aforementioned advantages of the method.

  According to certain embodiments, the vision system is mounted movably between storage areas, for example in an articulated system or on a rail. The vision system may be movable between a storage area and a receiving area. According to another embodiment, the installation comprises several visual systems, each fixed above the storage area, preferably also fixed above the receiving area.

  Other features and advantages of the present invention will become apparent from the description provided below, without any limitation for information, with reference to the accompanying drawings.

1 is a schematic perspective view showing a first example embodiment of an installation for automatically configuring a batch of parts according to the present invention; FIG. FIG. 6 is a schematic perspective view showing a second example embodiment of the equipment according to the present invention.

  The present invention uses an articulated system controlled by a vision system to move at least one batch of parts in a receiving area from parts located in at least two different storage areas, especially for so-called picking and / or kitting applications. It relates to a method and equipment for automatic configuration.

  The facility comprises an articulated system, at least one receiving area for a batch of parts, and at least two storage areas for various parts to constitute the batch.

  The articulated system is, for example, a six-axis robot that is well known from the prior art. The multi-joint system can grip a part by a gripping region unique to the part. For this purpose, the articulated system comprises at least two and preferably three gripping members and is controlled by at least one vision system.

  The gripping member can be of any suitable type. The essence is that the gripping member can grip the part by a specific gripping area. This is, for example, in the form of a clamp with two jaws that can be separated and approached to each other to grip the part. Particular gripping areas of the parts that can be held by the clamps are edges, busbars, or surfaces that are preferably parallel. Obviously, the maximum spacing between the edges, the busbars, or the surfaces needs to be compatible with the kinematic characteristics of the clamp. For example, the maximum spacing should be less than the maximum possible opening between the clamp jaws.

  The gripping member is of the type of suction or magnetic suction cup with a sleeve-shaped elastic member that can be applied in a specific area of the part at the end, for example to grip the part by suction or magnetism And can. The specific area of the part that can be gripped by the suction cup is a continuous area, clearly a substantially flat surface, sized to receive the elastic sleeve of the suction cup for that specific area. Surface.

  The parts to be gripped are randomly stored in bulk in the storage area. Thus, these storage areas can comprise a storage container for the parts, which can include package components such as spacers, cross braces, etc., for arranging and separating the parts within the container, if possible. Accompanying the inside of the container. The storage area can be fixed or movable using a conveyor. The parts can be, for example, different sizes, shapes, and types in one storage area and another storage area.

  According to the present invention, the articulated system is movable between a storage area and a receiving area for a batch of parts. The means of movement is of any suitable type. For example, the movement of the articulated system can be autonomous, the articulated system can move by itself, and / or the articulated system can be a cart mounted on a rail or automatically driven, for example. It can be controlled by other means of separate movement, such as mounted on a vehicle. The articulated system may be attached to an arm that is movable from one region to another.

  The vision system is of any suitable type, for example a 3D camera or a camera in stereo mode. The vision system can be placed directly on the articulated system so that it can move with the articulated system to capture an image of the area from one region to another. According to another illustrative embodiment, the vision system is mounted on a rail that is movable between storage areas and advantageously movable between a storage area and a receiving area. According to another example, the installation comprises several visual systems that are each fixed above the storage area and the receiving area.

  In addition, the equipment comprises means for processing and calculation controlled by the articulated and visual systems.

  These processing and calculation means are of any suitable type and may take the form of, for example, a microprocessor incorporating processing and calculation software.

  These processing and calculation means include, for example, edges, busbars, or preferably parallel surfaces, or elastic sleeves of the suction cups whose spacing can be matched to the kinematic characteristics of the jaws of the clamp. It is suitable for processing information generated from captured images to identify a gripping area in the gripped part that can be fitted with a gripping member, such as a flat surface that is dimensioned to accept.

  The identified specific gripping area that can be adapted may be positioned in an appropriate position and orientation by the means of processing and calculation.

  Processing and calculation means can automatically select the gripping region with the greatest likelihood of being accurately gripped by one of the gripping members when the gripping region is positioned. Therefore, various gripping members are compared.

  The selection of the gripping region with the greatest likelihood of being gripped correctly can be done, for example, by weighting each identified gripping region with a factor that depends on the likelihood of successful gripping by the corresponding gripping member, Selecting the region with the highest probability of success. Thus, the gripping members are compared and the gripping member with the best chance of success is actuated to grip the part. In practice, this involves assigning a weighting factor to each identified conformable gripping region, depending on one or more parameters. The parameters used can be of any type, for example, the height of the gripped area, the area of the viewable surface, and the slope of the gripping area. Thus, each gripping region includes weighting for each selected parameter. Next, an average is performed from each weighting for the grip region to obtain a score. Depending on the nature of the weighting, the gripping region with the maximum score or the minimum score is selected for gripping.

  According to another characteristic of the invention, the processing and calculation means, for the selected gripping area, impose a constraint from the collision between the gripping member that can grip the part by the selected area and the environment. It can also be specified. Effectively, depending on the positioning of the part for which the gripping area has been selected, the conformable member may have constraints in holding. The part to be gripped may be positioned, for example, too close to the edge of the storage container or another part, so that the gripping member will be interfered and hit the edge or adjacent part of the container The parts cannot be gripped. There may be other collision constraints.

  In fact, the space occupied by the gripping member is known by processing and calculation means, so from the captured image of the storage area, the space occupied by the gripping member and the environment around the selected gripping area The problem is to verify the common part.

  If there is a collision constraint, specifically if there is a risk of interference between the gripping member and the environment of the selected gripping area, the gripping area will be abandoned and the second best chance of being gripped Is selected. When it is determined that there is no collision constraint, the processing and calculation means automatically calculates a path for gripping the part by the selected gripping region for the corresponding gripping member.

  Thus, a learning-free method for automatically configuring at least one batch of parts in the receiving area from parts located in at least two different storage areas can be used in the installation according to the invention.

According to the present invention, the above equipment is
-Automatically moving the articulated system towards the first storage area;
(I) comprising capturing an image of the storage area using a vision system;
(Ii) processing the information generated from the image and identifying all gripping areas of the part that are compatible with the gripping member;
(Iii) positioning the identified gripping region in an appropriate position and orientation;
(Iv) selecting a gripping region that has the greatest likelihood of being gripped by a gripping member and without any collision constraints between the gripping member and the environment of the gripping region, and selected for the corresponding gripping member Automatically calculating a path for gripping the part by the gripping area;
(V) gripping the corresponding part following the calculated path and automatically moving the articulated system toward the receiving area to place the part in the receiving area;
-Automatically moving the articulated system towards the second storage area and repeating steps (i) to (v);
Can be executed continuously.

  This method moves an articulated system from one region to another and allows any part to join the articulated system to form a batch without learning the part or the path to grip the part. This is advantageous in that it can be gripped and placed.

  With reference to FIG. 1, the application of an example of the method according to the invention is described. The facility (1) is a vision system mounted on a straight rail (3) along two conveyor lines (4) that receive several different storage containers (5) of bulk parts (6) and A vision system is provided which can be moved next to the conveyor lines (4). In the example shown, the conveyor (4) comprises two lines of three storage containers (5) each having a different part (6), for example to constitute a batch of three parts. The facility is controlled by a vision system (2) and has an articulated system (7) mounted on a vehicle (8) which is movable in a rail (9) parallel to the conveyor (4) and adjacent to the conveyor (4). I have. Also mounted on the vehicle (8) is a receiving container (10) for a batch of components to be constructed. Preferably, the receiving container (10) can be positioned towards the outlet (12) for the filled receiving container by movement of the vehicle (8) or, for example, the vehicle (8). ) Placed on a conveyor intended to supply an empty receiving container. It is also possible for the empty or filled receiving container (10) to be removed or supplied by the operator.

  Thus, the vision system (2) moves along the rail (3) to capture an image of the storage container (5). When the image is captured and processed, the articulated system (7) grips the part (6) and is positioned in the same reference coordinate system as the articulated system (7), so that the receiving container has known coordinates. In order to place the part (6) in bulk in (10), it can be moved. A second vision system for determining the location and orientation of the receiving container can be considered.

  The articulated system (7) repeats the gripping and placement operations for two other parts (6) present in two other storage containers (5).

  Thus, the batch of parts is gripped and placed in the receiving container (10), which can be removed and a new empty receiving container (10) can be provided.

  The vision system (2) recognizes the presence of package components such as spacers or cross braces between parts in the storage container (5). These package components have been stored beforehand by means of processing and calculation and are recognized, grasped and removed by the articulated system (7). Similarly, the vision system (2) detects the presence of an empty storage container (5), locates it, then detects its position via computational and processing means, and the articulated system (7) is empty. Grasp and remove the storage container (5).

  Referring to FIG. 2, which shows a second facility (1) for carrying out the method according to the invention, the articulated system (7) controlled by the vision system (2) is in a separate storage area (13). The various parts to be positioned (6) are gripped one by one for placement in the receiving container (10) to form a batch. The vision system (2) is mounted so as to be rotatable about an axis in order to move through various storage areas (13) arranged around the articulated system (7). The filled receiving container (10) is then emptied, for example, into a bagging machine for packaging together parts from a single batch.

  In this embodiment, the means of movement of the articulated system (7) is formed from an arm of the articulated system (7) that can move between the storage area (13) and the receiving container (10). .

  It is clear that the method and equipment (1) according to the invention can use multiple articulated systems (7), each of which can be controlled by the same vision system (2), or each of which can be controlled by its own vision system. It is. Thus, one of the articulated systems moves one gripped part while another articulated system can grip another of the parts simultaneously with the movement of the gripped part, etc. , Repeat in order. The operation of gripping and placing the unknown parts can thus be carried out optimally and quickly.

DESCRIPTION OF SYMBOLS 1 Equipment 2 Vision system 3 Rail 4 Conveyor line 5 Storage container 6 Parts 7 Articulated system 8 Vehicle 9 Rail 10 Receiving container 11 Conveyor 12 Exit 13 Storage area

Claims (11)

Using an articulated system (7) controlled by at least one vision system (2), from parts located in at least two different storage areas (5, 13), the parts (6) in the receiving area (10) A method for automatically configuring at least one batch, wherein the articulated system comprises at least two separate gripping members each capable of gripping a part (6) by at least one specific gripping area;
Automatically moving the articulated system (7) towards the first storage area (5, 13);
(I) capturing an image of the first storage area (5, 13) using the vision system (2);
(Ii) processing the information generated from the image and identifying all the gripping areas of the part (6) that are compatible with the gripping member;
(Iii) comprising positioning the identified gripping region in an appropriate position and orientation;
(Iv) selecting the gripping region that has the greatest likelihood of being gripped by a gripping member and without any collision constraints between the gripping member and the environment, and for the corresponding gripping member, the selected gripping Comprising automatically calculating a path for gripping the part by area;
(V) gripping the corresponding part (6) following the calculated path and facing the receiving area (10) to place the part (6) in the receiving area (10) Comprising automatically moving the articulated system (7);
-The step comprising automatically moving said articulated system (7) towards the second storage area (5, 13) and repeating steps (i) to (v);
A method characterized by comprising at least.   The storage area comprises a storage container (5) for parts, the method comprising a step consisting of detecting an empty storage container (5) and gripping the empty storage container (5) 2. Method according to claim 1, characterized in that it comprises a step and a step consisting of removing the empty storage container (5).   The storage container (5) includes a package component, and the method comprises the steps comprising recognizing the package component, comprising grasping the package component, and removing the package component. The method of claim 2, comprising:   3. The storage container (5) comprising a part (6) to be gripped is automatically replenished in the storage area when the empty storage container (5) is removed. Method.   The receiving area comprises a receiving container (10) for a part, the method comprising detecting a filled receiving container (10) and gripping the filled receiving container (10) 5. A method according to any one of the preceding claims, characterized in that it comprises a step consisting of: and a step consisting of removing the filled receiving container (10).   Method according to claim 5, characterized in that when the filled receiving container (10) is removed, an empty receiving container (10) is automatically replaced in the receiving area.   7. The method according to claim 1, wherein the step comprising identifying a gripping area that can be adapted comprises identifying a continuous flat area when the gripping member is a suction cup type. The method according to one item.   The step consisting of identifying a specific area that can be adapted comprises identifying edges, busbars, or parallel surfaces when the gripping member is clamped, the edges, busbars 8. A method according to any one of the preceding claims, characterized in that the maximum spacing between the surfaces needs to be compatible with the kinematic properties of the clamp. Equipment (1) for automatically configuring at least one batch of parts (6) in a receiving area (10) from parts (6) located in at least two different storage areas (5, 13) The equipment (1) comprises an articulated system (7) controlled by at least one visual system (2) capable of capturing images of the storage area (5, 13), and at least one specific grip Comprising at least two separate gripping members each capable of gripping the part (6) by region,
The articulated system (7) comprises means (8) for movement between the storage area (5, 13) and the receiving area (10), the visual system (2) and the articulated system (7). ) Identifies all the gripping areas that can be fitted with the gripping member, locates the identified gripping area in an appropriate position and orientation, and has the greatest likelihood of being gripped by the gripping member; Selecting the gripping region without any collision constraints between the gripping member and the environment, and for the corresponding gripping member, a path for gripping the part (6) by the selected gripping region. The automatic calculation and the multiple between the storage area (5, 13) and the receiving area (10) to place the gripped part (6) in the receiving area (10). For the instruct a moving section system (7), characterized in that it is controlled by using the process and the calculation can process information generated from the captured image; facility (1).   10. Equipment according to claim 9, characterized in that the vision system (2) is mounted movably between the storage areas (5, 13).   10. Equipment according to claim 9, characterized in that it comprises several vision systems (2) each fixed above the storage area (5, 13).

Images