Material determination method and material determination device

Abstract

To provide a material quality determination method and a material quality determination method, capable of high purity selection by correctly determining a material even under an environment where a material determination of an object to be selected is difficult.SOLUTION: A workpiece to be determined among a workpiece group 2 obtained by mixing metal or resin work-pieces 2a is held by a holding device 6, and is brought into contact with a sensor group 8 constituted of a plurality of kinds of sensors 11 for detecting a specific metal while being held by the holding device. Accordingly, contact detection devices 19 provided individually in sensors constituting the sensor group, for detecting contact with the work-piece, use a detection signal to store a position of the sensor that detects contact with the work-piece, and determine a material quality of the work-piece from the detection result of the sensor that detects contact among the sensor group.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method and device for determining the material of a workpiece to be determined that is extracted from a group of workpieces, such as a plurality of types of parts made of metal or resin. [Background technology]

[0002] In recent years, picking devices that use a detection means to extract specific parts to be selected from a group of multiple types of parts transported on a transport device and pick and select them using a robot or the like are widely used in various industries around the world. For example, in the field of home appliance recycling, specific parts are recognized and picked from dismantled or crushed products that have become unnecessary and have been dismantled or crushed into parts, and are sorted, collected, and recycled.

[0003] A conventional picking device that performs sorting and recovery is disclosed, for example, in Patent Document 1. FIG. 10 is a diagram showing the conventional picking device that performs sorting and recovery described in Patent Document 1.

[0004] In Fig. 10, the conveyor 100 conveys the workpiece 200 placed thereon in one direction. A detection means 300 for detecting the material of the workpiece 200 is provided at the upper part of the conveyor 100 at a position corresponding to the upstream side, and sensors S71 to S73 are installed inside the detection means 300. The sensors S71 to S73 detect the metal 400 contained in the workpiece 200 during conveyance, and the control unit CS issues instructions to the robots R1 to R3 arranged downstream based on the detection results, so that the robots R1 to R3 can pick up the workpiece 200 on the conveyor 100 and move it to the predetermined storage locations F1 to F4. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Special Publication No. 2017-524524 Summary of the Invention [Problem to be solved by the invention]

[0006] However, in waste home appliance recycling plants, the conveyor that transports the parts to be sorted that are generated from waste home appliances generally uses a metal belt because durability is important. Furthermore, since the transported workpieces have various shapes, including thin parts such as metal plates, the configuration of Patent Document 1 makes it difficult to distinguish between the workpieces and the metal belt, and there is a problem that the picking device cannot be applied to a conveyor with a metal belt.

[0007] The present invention is intended to solve the above-mentioned problems in the conventional art, and aims to provide a material determination method and material determination device that can correctly determine the material of workpieces and perform high-purity sorting, even in environments where material determination is difficult from an operational standpoint, such as metal conveyor belts that transport a wide variety of dismantled workpieces. [Means for solving the problem]

[0008] In order to achieve the above object, the present invention is configured as follows.

[0009] According to one aspect of the present invention, there is provided a method for determining the material of a workpiece to be determined from a workpiece group including a plurality of workpieces made of metal or resin, comprising the steps of: The workpiece to be judged is held by a holding device, The present invention provides a material determination method in which the workpiece to be determined, held by the holding device, is brought into contact with a sensor group having multiple types of sensors that each detect different metals, and the material of the workpiece to be determined is determined in a determination unit based on a contact detection signal from a contact detection device that is individually provided on each of the sensors constituting the sensor group and detects contact with the workpiece, and on the material detection results of the sensors.

[0010] According to another aspect of the present invention, a holding device for holding a workpiece to be judged; an image processing unit that recognizes the workpiece to be judged and derives a holding position by the holding device from shape information of the workpiece to be judged; A group of sensors for detecting the material of the workpiece to be judged is provided, The sensor group includes a plurality of types of sensors for detecting different metals, The sensor includes a lifting section that can be raised and lowered independently, a biasing means that biases the lifting section to an upper position in a standby state, and a detection section that detects the lifting section that has been lowered from the upper position due to contact with the workpiece, The material determining device further includes a determining section that performs the material determination based on a metal detection result of the sensor that detects the descent of the lifting section by the detection section, among the sensors. Effect of the Invention

[0011] According to the material determination method and material determination device of the above aspect of the present invention, the material of the workpiece to be determined is determined by a determination unit based on a contact detection signal from the contact detection device individually provided in the sensor that detects the material of the workpiece and the material detection result of the sensor. By configuring in this way, even in an environment where material determination is difficult, such as an environment where a metal conveyor belt is used, it is possible to improve the accuracy of determining the material of each workpiece for a mixed workpiece group consisting of workpieces made of metal or resin, and to increase the purity of the selection. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic configuration diagram of an automatic picking device according to a first embodiment of the present invention. [Diagram 2] FIG. 1 is a diagram showing a material determination device according to a first embodiment of the present invention; [Diagram 3] FIG. 1 is a diagram showing a sensor unit according to a first embodiment of the present invention; [Figure 4] Operation flowchart of the material determination method according to the first embodiment [Diagram 5] FIG. 1 is a side view of a material determination device according to the first embodiment. [Figure 6] FIG. 1 is a diagram illustrating a top view of a material determination device according to the first embodiment. [Figure 7] Operation flowchart of the material determination method according to the second embodiment [Figure 8] FIG. 13 is a diagram showing the top view of the material determination device according to the second embodiment. [Figure 9] FIG. 13 is a diagram showing a top view of a material determination device according to the third embodiment. [Figure 10] Schematic diagram of a conventional picking device described in Patent Document 1 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, a first embodiment of the present invention will be described in detail with reference to the drawings.

[0014] (Embodiment 1) Hereinafter, a first embodiment of the present invention will be described with reference to FIG.

[0015] FIG. 1 is a schematic configuration diagram of an automatic picking device as an example of a material determination device according to a first embodiment of the present invention.

[0016] The automatic picking device includes a holding device 6, an image processing unit 51, and a material detection device 8 as an example of a sensor group.

[0017] The holding device 6 holds the workpiece 2a to be judged, and is, for example, a robot 6 having a robot arm 6a.

[0018] The image processing unit 51 recognizes the workpiece 2a and derives the holding position on the transport unit 1 by the robot 6 from the shape information of the workpiece 2a. More specifically, the image processing unit 51 derives the position coordinates, angle (e.g., inclination angle with respect to the transport direction, etc.) and type (e.g., whether or not the workpieces are only resin based only on the image of color or shape, etc.) of each of the workpieces 2a, 2b, 2c required for sorting by the robot 6.

[0019] The sensor group 8 is configured to detect the material of the workpiece 2a and includes a plurality of types of sensors 11a and 11b that detect specific different metals.

[0020] Each sensor 11a, 11b is equipped with a lifting section 30 which can be raised and lowered independently, a biasing means 17 which biases the lifting section 30 to an upper position in a standby state, and a detection section 19 which detects the lifting section 30 which has been lowered from the upper position due to contact with the workpiece 2a.

[0021] 1, the transport unit 1 is, for example, configured with a conveyor. The conveyor 1 transports a work group 2 including various kinds of work pieces 2a, 2b, 2c such as product parts or disassembled products, placed on a work placement surface of a conveyor belt 1a of the conveyor 1 in one direction. Here, the work pieces 2a, 2b, 2c of the work group 2 are not aligned at equal intervals on the transport unit 1, and the work pieces 2a, 2b, 2c are randomly supplied to the transport unit 1 and transported.

[0022] The object detector 3 is attached to the upstream end of the conveying section 1 and detects passing objects. That is, the object detector 3 detects the timing when the conveyed workpiece group 2 passes a specific point on the conveying section 1. A transmission type sensor is generally used as an example of the object detector 3. A transmission type sensor is a sensor that is composed of a pair of sensor units, a light projector and a light receiver, and detects an object by blocking the light between the light projector and the light receiver. In addition, the conveying unit 1 is equipped with a movement amount detection device 4 that detects the movement amount of the conveying unit 1. The signal from the object detector 3 and the signal from the movement amount detection device 4 are calculated by the signal calculation unit 50, so that the position in the conveying direction of the work group 2 conveyed on the conveying unit 1 can be known, and the calculation result is sent to the control unit 52, so that the robot 6 can know the holding position. One example of the movement amount detection device 4 can be configured with an encoder attached to the driving motor of the conveyor. The encoder is often attached separately from the motor. For example, when the motor has two shafts, the encoder can be connected to one of the shafts by a coupling. Alternatively, a coupling or pulley can be attached to one of the rotating shafts of the conveyor, and the encoder can be connected via a timing belt. The imaging device 5 is attached above the transport section 1 facing the workpiece placement surface of the conveyor belt 1a of the transport section 1, and captures image information from which shape information and the like of each of the workpieces 2a, 2b, 2c can be obtained.

[0023] By grasping the positions of the workpieces 2 to be judged among the workpiece group 2 by the above method, the imaging device 5 can be controlled to start imaging at the timing when the workpieces 2a, 2b, and 2c fall within the imaging device field of view 5a. Note that the object detector 3 is desirably installed upstream of the upstream end of the imaging device field of view 5a in the conveying direction.

[0024] The robot 6 is equipped with a gripping claw 9 at the tip of the robot arm 6a, and is capable of gripping and picking up each of the workpieces 2a, 2b, and 2c with the gripping claw 9, and of moving the workpieces to a collection box 7 assigned according to their type.

[0025] The image processing unit 51 processes the image information of each of the workpieces 2a, 2b, and 2c captured by the imaging device 5 by image recognition. The image processing unit 51 derives the position coordinates, angle (for example, the inclination angle with respect to the conveying direction), and type (for example, whether or not the workpieces are only resin based on the image of the color or shape) of each of the workpieces 2a, 2b, and 2c required for sorting by the robot 6, and sends the processing results to the control unit 52. The control unit 52 sends information on the workpiece group 2 to the robot 6, and also sends a signal from the movement amount detection device 4. With this configuration, the robot 6 picks only the specific workpiece (in other words, the workpiece to be judged) 2a of a specific type to be sorted from the workpieces 2a, 2b, and 2c by the gripping claws 9 provided at the tip of the robot arm 6a, and moves it to the collection box 7 assigned according to the type, thereby making it possible to sort the workpieces.

[0026] The control point 9a of the robot 6 is generally set at the center between the tips of the gripping jaws 9 when the tips of the gripping jaws 9 that grip the workpiece are closed together, and the gripping operation of the gripping jaws 9 is performed by controlling the position of the control point 9a using the gripping jaws 9 with respect to the position coordinates specified by the control unit 52.

[0027] When the image processing unit 51 specifies the center of gravity of the workpiece 2a to be evaluated as the position coordinates, the robot 6 is controlled to align the control point 9a with the center of gravity of the workpiece 2a to be evaluated, and the workpiece 2a to be evaluated is grasped.

[0028] As the imaging device 5, an area camera that obtains two-dimensional information is generally used, but a three-dimensional imaging device that can obtain more information such as height information may also be used.

[0029] The control unit 52 also includes a memory device 52a that stores the metal detection signals and the descent (contact) detection signals from the sensors 11 and 19, and a calculation device 52b that performs calculations based on the information stored in the memory device 52a.

[0030] As an example of a workpiece, there are various kinds of parts generated by disassembling waste home appliances in a recycling plant, and the difficulty of determining the type by image processing also varies. For example, the workpiece group 2 in a thin television includes a board with many electronic components mounted thereon, a fan, and a hard disk, and the characteristics of the parts are relatively easy to obtain from the information such as the external shape and color that can be obtained by the imaging device 5, and there are parts with high accuracy of type determination such as whether they are metal or resin. On the other hand, sheet metal parts made of iron or aluminum, or parts with a high luster or black resin parts have little color information and have a mixture of shapes from simple ones such as rectangular ones to complex ones including notches or holes, so that it is difficult to accurately determine the type such as whether they are metal or resin only with the information from the imaging device 5.

[0031] The higher the purity of the classification of the type, such as whether it is metal or resin, the higher the value when recycling. Therefore, in order to increase the purity of the classification, in addition to the imaging device 5, a material detection device 8 is installed near the conveying section 1. In the present embodiment 1, the accuracy of type determination is improved by using the material detection device 8 described below. 2 shows a material detection device 8 according to the first embodiment of the present invention. As an example, the material detection device 8 has a plurality of sensor units 10 arranged in a matrix at intervals of, for example, about 30 mm. There are two types of sensor units 10: a sensor unit 10a for detecting magnetic materials and a sensor unit 10b for detecting non-magnetic materials. The sensor units 10a for detecting magnetic materials and the sensor units 10b for detecting non-magnetic materials are arranged alternately in a staggered manner, for example, in a matrix of 3 rows by 4 columns, to configure the material detection device 8.

[0032] Fig. 3 shows each sensor unit 10 in the first embodiment of the present invention. The configuration of each sensor unit 10 will be described with reference to Fig. 3. Fig. 3(a) shows a side view of the sensor unit 10, and Fig. 3(b) is a partial cross-sectional view of the sensor 11 and the cover 12 for the purpose of explanation. Fig. 3(c) shows the sensor 11 in a state where it has been lowered from the upper position.

[0033] The sensor 11 is a sensor that detects a specific metal. In the present embodiment 1, a proximity sensor that detects a detection target object by utilizing the principle of electromagnetic induction is used as an example of the sensor 11. The magnetic body detection sensor 11a of the magnetic body detection sensor unit 10a is a type of proximity sensor that detects a magnetic body such as iron by a change in impedance. The non-magnetic body detection sensor 11b of the non-magnetic body detection sensor unit 10b is a type of proximity sensor that detects a non-magnetic body such as aluminum by detecting a change in transmission frequency. Proximity sensors have the advantage of being able to detect objects without contact without having a moving part, but generally have a short stable detection distance of about 2 mm. The cover 12 is made of an insulating resin material, and forms a protective layer with a thickness of, for example, about 1 mm near the surface of the sensor 11 as shown in FIG. 3(b), and serves to protect the surface of the sensor 11. The sensor 11, the cover 12, and the shaft 13 are attached to a bracket 14 to constitute a lifting unit 30. The lifting unit 30 is attached to a guide 16 fixed to a plate 15 that is fastened to a stand (not shown) so that it can be raised and lowered as a unit. A spring 17 is passed through the outside of the shaft 13 to constantly urge the bracket 14 upward, thereby urging the sensor 11 to an upper position. A stopper 18 at the lower end of the shaft 13 abuts against the lower end of the guide 16 and is stopped there, thereby determining the position of the upper surface of the cover 12, i.e., the upper position of the sensor 11.

[0034] A descent detection sensor 19 is disposed near the stopper 18 as an example of a contact detection device or detection unit. The descent detection sensor 19 is a reflective sensor, and a sensor optical axis 19a is set, for example, about 2 mm below the lower end of the fixing bolt 18a of the stopper 18. When the fixing bolt 18a descends, for example, by 2 mm or more, the descent detection sensor 19 detects this. That is, when the specific work 2a that is held by the gripping jaws 9 and is the type judgment target comes into contact with the cover 12 and the specific work 2a pushes the cover 12 further downward, the fixing bolt 18a also descends together with the cover 12 and the shaft 13. When the fixing bolt 18a descends, for example, by 2 mm or more, the fixing bolt 18a enters the detection area of ​​the sensor optical axis 19a as shown in FIG. 3(c), and the descent detection sensor 19 detects the descent of the fixing bolt 18a.

[0035] That is, each sensor unit 10 includes a lifting section 30 that can be raised and lowered independently, a coil spring 17 as an example of a biasing means for biasing the lifting section 30 to an upper position in a standby state, and a descent detection sensor 19 that detects the lifting section 30 that has descended from the upper position. Therefore, by receiving the detection of the descent detection sensor 19 by the control section 52, it is possible to detect that a specific workpiece 2a to be subjected to type determination has come into contact with the sensor unit 10 of the material detection device 8.

[0036] Fig. 4 shows an operation flow of the material determination method by the material determination device in the present embodiment 1. Fig. 5 shows the material detection device 8 as viewed from the side. Fig. 6 shows the material detection device 8 as viewed from above. The material determination method will be described with reference to Figs. 4, 5, and 6.

[0037] FIG. 4 shows the flow of the part related to this embodiment 1 after the image processing unit 51 performs position coordinates, angle, and type determination based on the image information acquired by the imaging device 5, and the robot 6 picks up a specific work 2a based on the information output by the control unit 52.

[0038] First, in step S1, whether or not the gripping jaws 9 of the robot 6 have successfully gripped the specific workpiece 2a (FIG. 4(a)) is notified to the control unit 52. Whether or not the gripping has been successful is detected by a known method based on information from a sensor or the like provided on the gripping jaws 9. If the gripping has been successful, the process proceeds to step S2, and if the gripping has failed, the process proceeds to step S7.

[0039] When the control unit 52 is notified by the robot 6 that it has successfully grasped a specific workpiece 2a (step S1 in FIG. 4(a)) and proceeds to step S2, if the control unit 52 determines that the grasped workpiece is the workpiece 2a whose material is to be determined (step S2 in FIG. 4(b)), a material detection operation is performed (step S3 in FIG. 4(c)). That is, when the control unit 52 determines that the grasped workpiece is the workpiece 2a whose material is to be determined, for example, if it can be determined from the image information from the imaging device 5 that the grasped workpiece is clearly resin, the process proceeds to step S6, and the workpiece is placed in the collection box 7 for resin as the designated position. If not, it is determined that the workpiece may be a metal such as iron or aluminum, and the process proceeds to step S3, and a material detection operation is performed.

[0040] In the material detection operation of step S3, the robot 6, while holding the specific workpiece 2a with the gripping jaws 9, moves the workpiece control point 9a of the gripping jaws 9 to a preset upper position 20 above the material detection device 8 shown in Fig. 5 and Fig. 6. As shown in Fig. 6, the upper position 20 is located near the center line 31 of the three rows of the arrangement of the sensor units 10 of the material detection device 8 when viewed from above the material detection device 8. After the specific workpiece 2a held by the gripping jaws 9 has moved to the upper position 20, the gripping jaws 9 are lowered to a level height 21 where the upper surface of the cover 12 and the lower surface of the gripping jaws 9 are at the same height as shown in Fig. 5.

[0041] After the control point 9a of the gripping jaws 9 gripping the specific workpiece 2a moves to the level height 21, the control point 9a of the gripping jaws 9 is further lowered to the push-in height 22, so that the specific workpiece 2a and the sensor 11 are in a state of contact with each other reliably through the cover 12. The specific workpiece 2a often has an uneven portion 2e formed on the underside thereof, and the specific workpiece 2a often does not reach the detectable distance of the sensor 11 simply by moving the control point 9a to the level height 21. For this reason, in order to bring the specific workpiece 2a and the sensor 11 into contact with each other without any gaps, it is necessary to push the specific workpiece 2a downward against each sensor unit 10 of the material detection device 8. The position of the sensor unit 10 that has been lowered when pushed downward is stored in the storage device 52a in the control unit 52 (step S4 in FIG. 4(d)) by the control unit 52 receiving a contact detection signal from the descent detection sensor 19 installed in each sensor unit 10. Moreover, the detection results of the magnetic material detection sensor 11a and / or the non-magnetic material detection sensor 11b are notified to the control unit 52 and stored in the storage device 52a.

[0042] In this way, the control unit 52 determines the material of the specific workpiece 2a being gripped by the gripping jaws 9 from the material detection result of the sensor 11 that has a contact detection signal among the sensors 11 in the group of sensors installed in the material detection device 8 (step S5 in Figure 4(e)).

[0043] That is, when the control unit 52 determines that the magnetic body detection sensor 11a with the contact detection signal has reacted, the control unit 52 overwrites the type determination result derived by the image processing unit 51 and determines that the material is iron. Similarly, when the control unit 52 determines that the non-magnetic body detection sensor 11b with the contact detection signal has reacted, the control unit 52 overwrites the type determination result derived by the image processing unit 51 and determines that the material is aluminum. When the control unit 52 determines that both the magnetic body detection sensor 11a with the contact detection signal and the non-magnetic body detection sensor 11b with the contact detection signal have not reacted, the control unit 52 overwrites the type determination result derived by the image processing unit 51 and determines that the material is resin. Here, it is the control unit 52 that determines the material. The proximity sensor only notifies the control unit 52 of a value of 1 or 0 with respect to the set threshold value, and the sensor does not determine the material. Therefore, the magnetic body detection sensor 11a for detecting iron as an example of a magnetic body reacts to the presence of iron nearby using impedance changes and notifies the control unit 52 of a signal of 1. The non-magnetic material detection sensor 11b for aluminum, an example of a non-magnetic material, reacts to the change in transmission frequency when aluminum is nearby and notifies the control unit 52 of a signal of 1. In the material determination operation, the control unit 52 determines the material based on the values ​​of the signals from the sensors 11a and 11b.

[0044] Thereafter, the robot 6 stores the specific workpiece 2a held by the gripping jaws 9 in a collection box 7 according to the material determination result (step S6 in FIG. 4(f)). After that, this series of material determination operations is completed, and the robot 6 moves on to material determination operations for the next workpiece. Thus, according to the first embodiment, the control unit 52, which is an example of a determination unit, is configured to determine the material of the workpiece 2a to be determined from the contact detection signal from the descent detection sensor 19, which is provided separately for the sensor 11 (11a, 11b) that detects the material of the workpiece 2a and detects contact with the workpiece 2a, and the material detection result of the sensor 11. In other words, by using only the material detection information of the sensor 11 that has a contact detection signal and not using the material detection information of the sensor 11 that does not have a contact detection signal, it is possible to eliminate erroneous information from the sensor 11 that reacts to the metallic conveyor belt 1a.

[0045] By configuring in this manner, even in an environment where a metal conveyor belt 1a is used, the accuracy of determining the material of each individual workpiece 2a can be improved for a work group 2 that is a mixture of workpieces 2a made of metal or resin, thereby making it possible to increase the purity of material selection.

[0046] (Embodiment 2) In the above-mentioned material detection operation (step S5 in FIG. 4(e)) in the first embodiment, if there is no reaction from the sensor 11 even though the descent detection sensor 19 detects a push, the specific workpiece 2a is determined to be resin, but there is also the possibility that the specific workpiece 2a is a metal different from the detection target of the sensor 11, or that the contact state with the sensor 11 is insufficient. In this case, a case where the material detection operation is performed multiple times rather than being performed only once to make a determination will be described as the second embodiment with reference to FIGS. 7 and 8.

[0047] 7 is a flowchart illustrating an operation flow when no reaction is obtained from the descent detection sensor 19 in the material detection operation (step S5 in FIG. 4(e)), although the operations from step S1 in FIG. 7(a) to step S6 in FIG. 7(f) are the same as the operations from step S in FIG. 4(a) to step S6 in FIG. 4(f), respectively, and also illustrating the operations from step S10 in FIG. 7(g) to step S12j in FIG. 7(i).

[0048] A method of deriving the upper position 20a when the material detection operation is performed multiple times will be described with reference to Fig. 8(a) and Fig. 8(b). Fig. 8(a) illustrates the positions of the three sensor units 10 that detected the descent in the first material detection operation surrounded by a circle 35. The type and position of the three sensor units 10 that detected the descent are stored in the storage device 52a in the control unit 52. In the second embodiment, it is assumed that two magnetic body detection sensors 11a for detecting magnetic bodies and one non-magnetic body detection sensor 11b for detecting non-magnetic bodies are lowered, and that the two magnetic body detection sensors 11a do not react to magnetic bodies. In Fig. 8(a), it is also meant that the workpiece 2a to be judged is present in the inverted L-shaped area formed by these sensors 11a, 11b, and 11a. The arithmetic device 52b in the control unit 52 incorporates a logic circuit that determines that the more contact points there are, the more opportunities for detection, and since there was no reaction of the magnetic material to the two descending magnetic material detection sensors 11a, the arithmetic device 52b of the control unit 52, which is an example of a determination unit, determines that the material of the specific workpiece 2a is more likely to be aluminum than iron. In this case, during the second material determination operation, the specific workpiece 2a is moved so that there is a greater opportunity for the specific workpiece 2a to come into contact with the non-magnetic material detection sensor 11b for aluminum than with the magnetic material detection sensor 11a. Therefore, since the arrangement of the sensor unit 10 in the material detection device 8 is known, when continuing the material detection operation, as shown in Figure 8(b), the workpiece 2a to be judged, which was located in the inverted L-shaped area described above, is updated to a position where it is in more contact with the non-magnetic material detection sensor 11b for detecting aluminum than in the case of Figure 8(a), that is, a new upper position 20a which is a position shifted to the left by about 30 mm, which is equivalent to the distance between the sensors (step S12 in Figure 7(h)), and the robot 6 moves the workpiece 2a with the gripping jaws 9, and then the material detection operation (the operation of step S3 in Figure 7(c)) is continued.

[0049] As shown in FIG. 7, the material detection operation is performed the number of times set by the control unit 52 in step S10 of FIG. 7(g). When either the sensor 11a or the sensor 11b reacts, the process proceeds to step S6 of FIG. 7(f), and this repeated material detection operation is terminated.

[0050] 7(g), if no signal is generated from the sensors 11a and 11b even after a set number of times, it is determined to be resin and moved to a predetermined discharge position (step S11 in FIG. 7(i)). That is, in the case where the material detection operation is performed multiple times in the control unit 52, a new upper position 20a is derived from contact position information based on the contact detection signal of the sensor unit 10, and the material detection operation is performed while moving the workpiece.

[0051] Thus, according to the second embodiment, in addition to the effects of the first embodiment, the following operational effects can be achieved. That is, even in an environment where a metallic conveyor belt 1a is used, when a material detection operation of the workpieces 2a made of metal or resin is performed multiple times on the sensor group (i.e., the material detection device 8) before judgment is performed on a workpiece group 2 including mixed workpieces 2a, the sensor 11a that detects the specific metal (e.g., iron) that detects contact with the workpiece 2a from the sensor group 8 for the Nth time (where N is an integer of 1 or more) and the position of the sensor 11a that detects the specific metal (e.g., iron) are stored in the storage device 52a. Then, in the (N+1)th material detection operation, the contact position of the workpiece 2a with the sensor group 8 is derived by the control unit 52 so that the number of contacts between the workpiece 2a and the sensor 11b that detects the specific metal (e.g., aluminum) different from the metal detected in the Nth material detection operation is equal to or greater than the number of contacts between the workpiece 2a and the sensor with the most contacts in the Nth operation, i.e., the sensor 11a that detects the specific metal (e.g., iron), and the (N+1)th material detection operation is performed at that position, and then the judgment is performed based on the results of multiple material detection operations, for example, the Nth and N+1th material detection operations. By configuring in this way, the accuracy of judging the material of each workpiece 2a is further improved, and the purity of material selection can be further increased. (Embodiment 3) 9 shows the top surface of the material detection device 8 in the embodiment 3. In the embodiment 3, the image processing unit 51 processes the image information of the captured specific workpiece 2a by image recognition, and when deriving the information of the specific workpiece 2a required for sorting by the robot 6, the upper position 20 in the material detection operation is derived based on information on the shape of the captured specific workpiece 2a in addition to information on the gripping position coordinates, angle, and type.

[0052] The position data of each sensor 11 (11a, 11b) in the material detection device 8 is stored in the storage device 52a of the control unit 52. In the control unit 52, the position data of the sensor 11 and the shape information of the specific workpiece 2a are superimposed using, for example, pattern matching, and the calculation device 52b of the control unit 52 performs calculations so that the overlap between the sensor 11 and the specific workpiece 2a is maximized, and the upper position 20 of the specific workpiece 2a relative to the material detection device 8 is derived. For example, FIG. 9(a) shows the upper position 20 of the specific workpiece 2a-1 relative to the material detection device 8 when a rectangular specific workpiece 2a-1 is held. FIG. 9(b) shows the upper position 20 of the specific workpiece 2a-2 relative to the material detection device 8 when an L-shaped specific workpiece 2a-2 is held. FIG. 9(c) shows the upper position 20 of the specific workpiece 2a-3 relative to the material detection device 8 when a square-shaped workpiece 2a-3 with an opening in the center is held. By extracting the shape of a specific workpiece 2a in this manner using the image processing unit 51, it becomes possible to bring the workpiece 2a to be judged into contact with more sensor units 10 according to the extracted shape, thereby minimizing the time required for material judgment.

[0053] Thus, according to embodiment 3, in addition to the effects of embodiment 1, shape information of the workpiece 2a to be judged is extracted by the image processing unit 51 for the workpiece 2a to be judged from the workpiece group 2 and stored in the memory device 52a, and position data of each sensor 11a, 11b in a sensor group 8 having multiple types of sensors 11a, 11b for detecting specific metals is stored, and the position data of the sensors 11a, 11b and the shape information of the workpiece 2a are superimposed by the arithmetic device 52b of the control unit 52, and a calculation is performed by the arithmetic device 52b so that the overlap between the position data of the sensors 11a, 11b and the workpiece 2a is maximized, and the contact position of the workpiece 2a with the sensors 11a, 11b is derived by the arithmetic device 52b and moved to that position to perform a material detection operation, thereby reducing the number of material detection operations.

[0054] (Modification) In the present embodiments 1 to 3, a material detection method has been described in which only the underside of a specific workpiece 2a is brought into contact with the material detection device 8, but it is also possible to perform the material detection operation by bringing the side surface of the specific workpiece 2a, etc. into contact with the material detection device 8.

[0055] In the present embodiments 1 to 3, the upper position 20 when continuing the material detection operation is updated based on the logic of comparing the number of reactions of the sensor 11 during the initial material detection operation, but various methods can be considered for deriving a new upper position 20 when continuing the material detection operation, such as predicting the work shape from the position information of the sensor 11 that detected the descent.

[0056] In the present embodiments 1 to 3, material determination was performed using the material determination method only for specific workpieces 2a for which accurate type determination is difficult based on information obtained from an image alone, but if there is sufficient time, the range of target types can also be expanded to include those subject to material determination.

[0057] Furthermore, in the first to third embodiments, the gripping jaws 9 are used as an example of a holding device for the specific workpiece 2a, but the material can also be determined in the same manner using a suction pad.

[0058] Furthermore, in the first to third embodiments, a metallic conveyor belt 1a is used, but the material can be determined in the same manner even when picking from a resin conveyor belt or rack.

[0059] In addition, by appropriately combining any of the various embodiments or modifications described above, it is possible to achieve the effects of each of them. In addition, combinations of embodiments, combinations of examples, or combinations of embodiments and examples are possible, and combinations of features of different embodiments or examples are also possible. [Industrial Applicability]

[0060] The material determination method and material determination device according to the above aspect of the present invention can eliminate the influence of the surrounding environment and achieve high-purity sorting of specific workpieces to be sorted from a group of workpieces such as multiple types of parts or dismantled products. Although the material determination method and material determination device according to the above aspect of the present invention have been described using picking from a conveyor as an example in the above embodiment, the present invention is not limited to this and can be widely applied to requests for high-purity sorting according to material, such as sorting from a container. [Explanation of symbols]

[0061] 1: Transport section 1a: Conveyor belt 2: Work groups 2a: Specific work 2b, 2c: Work 2e: Uneven part 3: Object detector 4: Movement detection device 5: Imaging device 5a: Imaging device field of view 50: Signal calculation section 51: Image processing unit 52: Control unit 52a: Storage device 52b: Arithmetic device 6: Robot 6a: Robot arm 9: Gripping claw 7: Collection box 8: Material detection device 10: Sensor unit 10a: Sensor unit for detecting magnetic material 10b: Sensor unit for detecting non-magnetic material 11: Sensor 11a: Magnetic material detection sensor 11b: Sensor for detecting non-magnetic material 12:Cover 13: Shaft 14: Bracket 15: Plate 16: Guide 17: Spring 18: Stopper 19:Down detection sensor 20:Upper position 20a: Upper position when performing material detection operations multiple times 21: Level Height 22: Push-in height 30: Lifting section 31: Center line 35: Circle

Claims

1. A method for determining the material of a workpiece to be evaluated from a workpiece group including a plurality of workpieces made of metal or resin, comprising: The workpiece to be judged is held by a holding device, A material determination method in which the workpiece to be determined, held by the holding device, is brought into contact with a sensor group having multiple types of sensors that each detect different metals, and the material of the workpiece to be determined is determined in a determination unit based on a contact detection signal from a contact detection device that is individually provided on the sensors that make up the sensor group and detects contact with the workpiece, and the material detection result of the sensor.

2. When a material detection operation of the workpiece to be judged is performed a plurality of times for the sensor group before the judgment, the material detection operation is regarded as a first material detection operation, and the sensor from the sensor group that has come into contact with the workpiece to be judged and detected the contact detection signal by the contact detection device during the first material detection operation and the position of the sensor are stored in the judgment unit, The material determination method of claim 1, wherein, in a second material detection operation following the first material detection operation, the determination unit derives a contact position of the workpiece to be determined with the sensor group so that the number of contacts between the sensor that detects a metal different from the metal detected in the first material detection operation and the workpiece to be determined is equal to or greater than the number of contacts between the sensor that detected the metal detected in the first material detection operation and the workpiece to be determined, and after performing the second material detection operation at the derived contact position, the determination is made based on the results of the first and second material detection operations.

3. An image processing unit extracts shape information of the workpiece to be judged from the workpiece group and stores the shape information in the judgment unit; The material determination method according to claim 1 or claim 2, further comprising: storing position data of each sensor in a sensor group consisting of multiple types of sensors that detect specific metals in a storage device; superimposing the position data of the sensor group with shape information of the workpiece to be determined; and performing a calculation in the determination unit so as to maximize the overlap between the position data of the sensor group and the workpiece to be determined; and deriving the contact position of the workpiece to be determined with respect to the sensor group in the determination unit.

4. A holding device that holds a workpiece to be judged; an image processing unit that recognizes the workpiece to be judged and derives a holding position by the holding device from shape information of the workpiece to be judged; A group of sensors for detecting the material of the workpiece to be judged is provided, The sensor group includes a plurality of types of sensors for detecting different metals, The sensor includes a lifting section that can be raised and lowered independently, a biasing means that biases the lifting section to an upper position in a standby state, and a detection section that detects the lifting section that has been lowered from the upper position due to contact with the workpiece, The material determination device further includes a determination unit that performs the material determination based on a metal detection result of the sensor that detects the descent of the lifting unit by the detection unit, among the sensors.

5. The determination unit is a storage device that stores the descent detection signals of the sensors that constitute the sensor group; a calculation device that derives the contact position of the workpiece with respect to the sensor group based on the lowering detection signal of the storage device and the metal detection information of the sensor that stores the signal, and sends the contact position to the holding device; The material determination device according to claim 4 , wherein the holding device moves the held workpiece to be determined to a contact position of the workpiece with the group of sensors.

6. Further comprising an image processing unit that extracts the shape of the workpiece to be judged by image processing, The determination unit is 5. The material determining device according to claim 4, further comprising a calculation device that superimposes the data on the shape of the workpiece extracted by the image processing unit and the positions of the sensors of the sensor group with reference to the holding position of the workpiece.

Images