JP7011063B2 - Parts serving device for kitting tray - Google Patents

Description

The present invention relates to a parts serving device for serving a plurality of types of parts having different sizes to a kitting tray provided with a plurality of accommodating portions.

The kitting tray is a tray that houses a set of parts including fastening parts such as screws, bolts and washers used for assembling one machine product, and sealing parts such as O-rings and gaskets. Patent Document 1 discloses picking of parts by a robot hand from such a kitting tray, and application of machine learning for accurately performing the picking.

The kitting tray includes a plurality of storage compartments divided by a frame. Parts are served in these storage compartments from storage trays that store various parts. Large parts that do not fit in the storage compartment may be placed on or leaned against the upper edge of the frame. It is desirable to automate the serving of parts to this kitting tray using a robot hand. Patent Document 1 only discloses a technique for picking parts, and does not mention serving parts to a kitting tray.

Japanese Unexamined Patent Publication No. 2017-30135

An object of the present invention is to provide a parts serving device capable of accurately serving various parts to a kitting tray using a robot hand.

The parts serving device for a kitting tray according to one aspect of the present invention is a parts serving device for serving a plurality of types of parts having different sizes to a kitting tray provided with a plurality of accommodating portions, and picking and releasing of the parts can be performed. Serving that has a possible head portion, picks and transports the target component from the plurality of types of components by the head portion at the storage position of the component, and releases the target component from the head portion to the kitting tray. A robot hand that performs an operation, a control unit that controls the operation of the robot hand , an image pickup device that acquires a three-dimensional image of the kitting tray after the serving operation by the robot hand is executed, and the three-dimensional image. The control unit includes an evaluation unit for evaluating the serving state of the parts in the kitting tray based on the above, and the control unit is the kitting tray of the parts according to the mode of the plurality of types of parts and the plurality of accommodating parts. The rule setting unit is provided with a rule setting unit for setting a serving rule to the robot hand and a drive control unit for causing the robot hand to execute the serving operation based on the serving rule . While acquiring the ideal serving information in the state where the parts of the above are ideally served, the serving rule is initially set based on the ideal serving information, and the serving rule is modified according to the evaluation of the evaluation unit.
Alternatively, the rule setting unit acquires information on the kitting tray and its accommodating unit and information on the size of parts, and initially sets the serving rule based on the information, and responds to the evaluation of the evaluation unit. Modify the serving rule.

FIG. 1 is a block diagram showing a configuration of a component serving device for a kitting tray according to an embodiment of the present invention. FIG. 2A is a plan view of a top view showing an example of a kitting tray in which parts are arranged, and FIG. 2B is a sectional view taken along line IIB-IIB of FIG. 2A. FIG. 3 is a flowchart showing the operation of the parts serving device. 4 (A) to 4 (D) are diagrams for explaining an example of the serving rule. 5 (A) and 5 (B) are diagrams for explaining an example of a serving rule. 6 (A) to 6 (D) are diagrams for explaining an example of a serving rule. 7 (A) to 7 (D) are diagrams for explaining an example of a serving rule. 8 (A) to 8 (C) are diagrams for explaining an example of the serving rule. FIG. 9 is a flowchart showing an example of the learning operation of the serving rule. FIG. 10 is a flowchart showing an example of the learning operation of the serving rule. 11 (A) and 11 (B) are diagrams for explaining an example of a serving rule.

[Overall configuration of parts serving equipment]
Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. First, the configuration of the component serving device 1 for the kitting tray according to the embodiment of the present invention will be described with reference to the block diagram of FIG. The parts serving device 1 is a device that serves a plurality of types of parts having different sizes to the kitting tray 40, and is a robot hand 10 that serves the parts and a three-dimensional measuring device 20 that captures a three-dimensional image of the parts. An image pickup device) and a control unit 30 that collectively controls the operations of the robot hand 10 and the three-dimensional measurement device 20.

The robot hand 10 is a robot device that picks target parts from the parts storage trays Ta, Tb ... That store a plurality of types of parts Wa, Wb ... Of different sizes, and transfers them to the kitting tray 40. The parts storage trays Ta, Tb ... Are trays in which the parts Wa, Wb ... Are individually stored, and various parts Wa, Wb ... Are stored in bulk or in an aligned state. .. The kitting tray 40 is a tray capable of individually accommodating a plurality of types of parts Wa, Wb, Wc, Wd ... In a plurality of pre-divided accommodating portions. The robot hand 10 picks and transports the target parts, and performs a serving operation to release the target parts picked in the accommodating portion.

The robot hand 10 is a multi-axis articulated robot including a base portion 11, a body portion 12, a first arm 13, a second arm 14, a head portion 15, and a hand portion 16. The base portion 11 is a housing that is fixedly installed on a floor surface, a pedestal, or the like. The body portion 12 is arranged on the upper surface of the base portion 11 so as to be rotatable in both forward and reverse directions around the axis of the first axis 1A extending vertically. The first arm 13 is an arm member having a predetermined length, and its base end portion is rotatably attached to the body portion 12 around the axis of the second axis 1B extending in the horizontal direction. The second arm 14 is an arm member connected to the first arm 13 so as to be rotatable around the axis of the third axis 1C whose base end portion extends in the horizontal direction, and is attached to the tip end portion of the first arm 13. It is attached.

The head portion 15 is rotatably attached to the tip end side of the second arm 14 around the axis of the fourth axis 1D extending in the horizontal direction. The hand portion 16 is attached to the head portion 15 via a fifth axis 1E perpendicular to the fourth axis 1D. The hand unit 16 can individually pick the parts Wa, Wb ... From the parts storage trays Ta, Tb ... And release them to the kitting tray 40. The hand portion 16 is composed of a pair of gripping pieces and is rotatable around the axis of the fifth axis 1E, and the spacing between the pair of gripping pieces is adjustable for picking and releasing. The head portion 15 may be provided with a mechanism capable of picking and releasing parts. For example, instead of the hand portion 16, an electromagnet or a negative pressure generator that generates an attractive force for the parts is used as the head. It may be provided in the portion 15.

The three-dimensional measuring device 20 includes a first camera 21, a second camera 22, and a camera control unit 23. The first camera 21 is arranged above the component storage trays Ta, Tb ..., And captures an image including the components Wa, Wb ... Stored in these trays. The second camera 22 is arranged above the kitting tray 40 and captures an image including the parts Wa, Wb, Wc, and Wd housed in the kitting tray 40. The three-dimensional measuring device 20 or the first and second cameras 21 and 22 may be mounted on the robot hand 10.

The camera control unit 23 causes the first and second cameras 21 and 22 to execute an image pickup operation, and executes three-dimensional measurement of parts based on the obtained image. The image pickup control unit 24 and the image processing unit 25 including. The image pickup control unit 24 causes the first camera 21 to perform an operation of photographing the component storage trays Ta, Tb ... When the components Wa, Wb ... Are picked. Further, the image pickup control unit 24 operates to image the kitting tray 40 on the second camera 22 when confirming the position of the storage unit of the kitting tray 40 or when acquiring the information of the kitting tray 40 after serving the parts. Let it run.

The image processing unit 25 generates image data including three-dimensional position information of each component by performing image processing on the images acquired by the first and second cameras 21 and 22. The three-dimensional position information of each component is represented by, for example, coordinate values (X, Y, Z) using an XYZ Cartesian coordinate system. From the image acquired by the first camera 21, the position information of the parts Wa, Wb ... Stored in the component storage trays Ta, Tb ... Is acquired. This position information is used when picking parts Wa, Wb ... By the robot hand 10. Further, the position information of the parts Wa, Wb, Wc, and Wd served on the kitting tray 40 is acquired from the image acquired by the second camera 22. Based on this position information, it is possible to evaluate the parts serving state.

The control unit 30 includes a drive control unit 31, a rule setting unit 32, an information acquisition unit 33, and a learning unit 34. The drive control unit 31 causes the robot hand 10 to execute the serving operation of the parts based on the serving rule set by the rule setting unit 32. The drive control unit 31 operates a drive motor (not shown) provided in the robot hand 10 so as to sequentially execute picking of parts, holding and transportation of the parts, and release of the parts according to the serving rule. To control. When machine learning related to the serving operation is executed in the learning unit 34, information on how the drive control unit 31 operates the robot hand 10 in the picking and release is output to the learning unit 34.

The rule setting unit 32 is set to the parts Wa, Wb ... Depending on the size and shape of the plurality of parts Wa, Wb ... Set the serving rule for the kitting tray 40. This serving rule is an arrangement regarding the concept of serving such as matters to be prioritized and matters to be avoided when serving parts to the kitting tray 40. For example, the serving rule is a rule that small-sized parts are served preferentially to the small accommodating portion, and which parts are served first when the parts are served so as to overlap in the vertical direction (Figs. 4 to 4). It will be described in detail later based on 8.). This serving rule may be a serving rule taught by the operator via the input unit 26 or the like, or a serving rule in which the taught serving rule is evaluated and revised based on the serving result, or the learning unit 34. It may be a serving rule created or revised as a result of machine learning by.

The information acquisition unit 33 acquires information input by the operator from the input unit 26 and three-dimensional measurement information from the camera control unit 23. From the input unit 26, for example, attribute information such as the size and shape of the component, information regarding the shape of the kitting tray 40, and the like are given. From the camera control unit 23, three-dimensional position information of parts Wa, Wb ... In the parts storage tray Ta, Tb ..., three-dimensional position information of the storage part of the kitting tray 40, and parts in the kitting tray 40. Three-dimensional position information of Wa, Wb, Wc, Wd and the like are given. The rule setting unit 32 sets or revise the serving rule based on various information given to the information acquisition unit 33.

The learning unit 34 is a functional unit that executes a learning process for learning the movement of the robot hand 10. When the serving rule is set by machine learning, the learning unit 34 has control information of the robot hand 10 by the drive control unit 31 and three-dimensional measurement information indicating the serving result to the kitting tray 40 input from the camera control unit 23. And are acquired for each learning cycle. Then, the learning unit 34 learns the optimum behavior pattern of the robot hand 10 when serving each component from these information, and reflects this in the serving rule. The behavior pattern is, for example, at what position the component is gripped and picked, at which three-dimensional position of the kitting tray 40 the component is released, and in what order the plurality of components are released. It is the action of the robot hand 10 regarding whether to serve at. The learning unit 34 includes a displacement amount observation unit 35, a reward setting unit 36, and a value function update unit 37. These will be described in detail when the embodiments to which machine learning is applied will be described later.

[Kitting tray]
2 (A) is a plan view of a top view showing an example of a kitting tray 40 in which parts are arranged, and FIG. 2 (B) is a sectional view taken along line IIB-IIB of FIG. 2 (A). The kitting tray 40 is a tray that houses a set of parts including fastening parts such as screws, bolts and washers used for assembling one mechanical product, and sealing parts such as O-rings and gaskets. The kitting tray 40 is a rectangular parallelepiped container having a relatively shallow bottom with an open upper surface, and has an outer frame portion 41 forming an outer frame portion, an inner frame portion 42 arranged inside the outer frame portion 41, and a rectangular parallelepiped portion 42. It includes a bottom plate 43 that forms the bottom surface of the kitting tray 40.

The outer frame portion 41 has a rectangular shape as shown in the top view of FIG. 2 (A), and as shown in the cross-sectional view of FIG. 2 (B), the upper end edge 41T thereof is in the kitting tray 40. The highest side plate. The inner frame portion 42 extends vertically and horizontally in the outer frame portion 41 in a plan view, and is a frame plate for partitioning a plurality of accommodating portions A1, A2, and A3. Here, one large accommodating portion A1 having the largest accommodating space, four medium accommodating portions A2 having the next largest accommodating space, and eight small accommodating portions A3 having the smallest accommodating space form an inner frame portion. An example of being partitioned by 42 is shown. The height of the upper end edge 42T of the inner frame portion 42 is lower than that of the upper end edge 41T of the outer frame portion 41. As a result, the upper accommodating portion A0 having the upper end edge 42T as the bottom surface and the outer frame portion 41 as the partition wall is formed.

The same parts are served in each of the accommodating portions A1, A2, and A3. Here, the large accommodating portion A1 has a large component W1 composed of a large-diameter seal ring, the medium accommodating portion A2 has medium-sized parts W21 and W22 such as a medium-diameter C ring, and the small accommodating portion A3 has small parts such as bolts and nuts. An example in which the parts W31 and W32 are housed is shown. These parts W1, W21, W22, W31, and W32 are housed in the accommodating portions A1, A2, and A3 in a manner of contacting the bottom plate 43. Further, an example is shown in which a super-large component W0 such as a gasket that does not fit into any of the accommodating portions A1 to A3 is accommodated in the upper accommodating portion A0. The super-large component W0 is in contact with the upper end edge 42T of the inner frame portion 42.

The kitting tray 40 to which the parts W0 to W32 are served is transported to a predetermined work place by an operator or a transport robot or the like. Alternatively, the parts W0 to W32 may be taken out from the kitting tray 40 by another robot hand. Although mechanical parts are exemplified here as parts, the parts accommodated in the kitting tray 40 are not particularly limited. For example, electric power / electronic parts, material chips, material rods, tools, and the like can be stored in the kitting tray 40 as parts.

[Basic flow of serving operation]
FIG. 3 is a flowchart showing the basic operation of the parts serving device 1. First, the information acquisition unit 33 of the control unit 30 acquires information regarding the kitting tray 40 and the parts W0 to W32 (step S1). This information includes information on the shape of the kitting tray 40, aspects of the housing portions A1 to A3 (opening size, depth, placement position, etc.), placement position with respect to the robot hand 10, type, shape, size of parts W0 to W32. It is an arrangement position with respect to the robot hand 10. The information acquisition unit 33 acquires the above-mentioned information based on an input operation via the input unit 26 or a measurement result by the three-dimensional measuring device 20.

Next, the rule setting unit 32 sets the serving rules for the target parts W0 to W32 accommodated in the kitting tray 40 based on the above information acquired by the information acquisition unit 33 (step S2). Later, some specific examples of the serving rule will be described. As described above, the serving rules were formulated from the information acquired by the information acquisition unit 33, or by machine learning in which the result of the actual serving was evaluated by the measurement of the three-dimensional measuring device 20. Things can be used.

Subsequently, the drive control unit 31 drives the robot hand 10 according to the set serving rule, and executes a serving operation of serving the target parts W0 to W32 to the accommodating portions A1 to A3 of the kitting tray 40 (step S3). That is, the drive control unit 31 rotates the robot hand 10 around the axis of the first axis 1A, and also appropriately operates the second axis 1B to the fifth axis 1E to move the hand unit 16 to the parts storage tray Ta. The stored parts Wa, Wb ... Are individually picked toward Tb ..., and are carried to the designated accommodating portions A1 to A3 of the kitting tray 40 and released. The drive control unit 31 confirms whether or not all the scheduled parts W0 to W32 have been served for one kitting tray 40 (step S4), and if the serving has not been completed (step S4). NO) in S4 continues the serving operation in step S3.

When the serving is completed (YES in step S4), the control unit 30 determines whether or not to execute the serving result evaluation (step S5). When executing the result evaluation (YES in step S5), the information acquisition unit 33 causes the second camera 22 of the three-dimensional measuring device 20 to capture an image of the kitting tray 40 after serving, and the served parts W0 to W32. Acquire the 3D measurement result. Then, the rule setting unit 32 evaluates the serving result from the state information of the parts W0 to W32 in the kitting tray 40 based on the three-dimensional measurement result (step S6).

Based on the evaluation result, the rule setting unit 32 determines whether or not the current serving rule needs to be modified (step S7). When the evaluation result is at a level exceeding a predetermined threshold value, the rule setting unit 32 determines that it is necessary to correct the serving rule (YES in step S7), and corrects the serving rule (step S8). As the correction here, the picking position, the grip force, the grip direction, the height position of the release, the correction of the serving order, and the like by the hand portion 16 of the parts W0 to W32 can be exemplified.

On the other hand, when the serving defect is equal to or less than the threshold value, the rule setting unit 32 determines that the serving rule does not need to be modified (NO in step S7), and ends the process. Further, even when it is determined in step S5 that the evaluation of the result of serving is not executed (NO in step S5), the process is also terminated. The steps S5 to S8 may be manually executed by the operator or may be executed according to a predetermined correction rule. Further, the process substantially the same as in steps S5 to S8 can be replaced with machine learning by the learning unit 34. An example of this machine learning will be described later based on FIGS. 9 and 10.

[Specific example of serving rules]
Hereinafter, a specific example of the serving rule set by the rule setting unit 32 will be described with reference to FIGS. 4 to 8. 4 (A) to 4 (D) are diagrams for explaining an example of the serving rule, and are diagrams showing the serving rule when the parts W11 and W12 are served so as to overlap each other in the vertical direction. be. 4 (A) is a top view showing a desirable serving state of the parts W11 and W12 to the kitting tray 40, and FIG. 4 (B) is a side view of FIG. 4 (A) in the direction of arrow a.

The component W11 is a ring-shaped component such as a packing, and the component W12 is a long rod-shaped component such as a bolt. The parts W11 are arranged on the bottom plate 43 of the accommodating portion A11 partitioned by the inner frame portion 42 of the kitting tray 40 in a manner of arranging 2 × 3 in a matrix. On the other hand, the component W12 is served in such a manner that it is supported by the upper end edge 42T of the inner frame portion 42 that partitions the accommodating portion A11.

When forming the serving state shown in FIGS. 4A and 4B, the rule setting unit 32 sets a serving rule for serving the component W11 whose serving position is downward in advance. That is, as shown in FIG. 4C, first, the robot hand 10 is served with each component W11 at a predetermined position in the accommodating portion A11. In this case, in the case of the ring-shaped component W11, it is desirable to arrange them in a matrix as illustrated in FIG. 4A so that the section of the accommodating portion A11 can be fully utilized.

After the serving of all the parts W11 scheduled to be served to the accommodating portion A11 is completed, the robot hand 10 is subsequently served with the rod-shaped parts W12 as shown in FIG. 4 (D). At this time, the component W12 is released from the hand portion 16 so that one end side and the other end side of the component W12 are supported by the upper end edges 42T of the pair of inner frame portions 42 facing each other. As a result, both parts are served in the kitting tray 40 in such a manner that the parts W12 are overlapped on the parts W11. By adopting the serving rules of FIGS. 4C and 4D, the part W11 whose serving position should be lower is served on the part W12 whose serving position is upper, or the part W11 is served. Can be prevented from being hindered by the component W12.

5 (A), 5 (B), and 6 (A) to 6 (D) are diagrams for explaining an example of the serving rule, and the overlap of the parts accommodated in the same accommodating portion is reduced as much as possible. It is a figure which shows the serving rule. In the kitting tray 40, when both the serving in which the parts are arranged in the horizontal direction and the serving in which the parts are overlapped in the vertical direction are feasible, the rule setting unit 32 serves the serving in which the parts are arranged in the horizontal direction rather than the serving in which the parts are overlapped in the vertical direction. Set a serving rule that gives priority to execution.

FIGS. 5A and 5B show an example in which six parts W21 having a hexagonal cross section are housed in the housing part A12 partitioned by the inner frame part 42. The six parts W21 with respect to the accommodating portion A12 can be arranged in a horizontal row as shown in FIG. 5 (A), or can be served vertically as shown in FIG. 5 (B). It is also possible to stack them in multiple stages (two stages) and serve them in a bale-like manner. That is, both a flat serving and a three-dimensional serving can be adopted because of the size of the parts W21 and the shape of the hexagonal cross section that can be easily stacked.

In such a case, the rule setting unit 32 gives priority to the flat serving of FIG. 5A. That is, instead of having the parts W21 pile up in the vertical direction, the horizontal space of the accommodating portion A12 is effectively utilized and the serving is performed in a well-balanced manner. Further, even in the case where the number of parts W21 is large and it is necessary to serve them in the vertical direction, as many parts W21 as possible are served directly above the bottom plate 43 in the accommodating portion A12, and then two stages are provided. Let them serve their eyes. Thereby, the accommodation stability in the accommodation portion A12 of the plurality of parts W21 can be improved. Further, when the parts W21 are stacked in a bale, the parts W21 served at a relatively high position from the bottom surface 43 are generated, and the parts W21 easily pop out from the accommodating portion A12 when the kitting tray 40 is transported after the serving. Will be. In view of these, the two-dimensional serving of FIG. 5 (A) is more advantageous than the three-dimensional serving of FIG. 5 (B).

6 (A) to 6 (D) also show the same example as in FIG. 6 (A) is a top view showing a desirable serving state of the bolt-shaped parts W31 to W34 to the accommodating portion A13 of the kitting tray 40, and FIG. 6 (B) is a top view showing the desired serving state of the bolt-shaped parts W31 to W34. It is a side view. Here, the parts W31 to W34 are arranged in a horizontal row so that the bolt heads are alternately oriented in opposite directions. By serving the parts W31 to W34 in this way, the horizontal space of the accommodating portion A13 can be effectively utilized, and the serving height (height of the center of gravity) of the parts W31 to W34 in the accommodating portion A13 is suppressed. be able to.

On the other hand, FIG. 6C is a top view showing the serving state of the parts W31 to W34 according to the comparative example, and FIG. 6D is a side view in the direction of arrow c of FIG. 6C. Here, of the four parts W31 to W34, the parts W33 and W34 are served on the bottom plate 43, and the parts W31 and W32 are served so as to be placed on the upper side of the parts W31 and W34, whereby the parts W31 and W32 are arranged in a grid pattern. An example is shown. In this way, when the parts W31 to W34 are served to the accommodating portion A13, it is possible to serve them three-dimensionally in a grid pattern instead of serving them in a plane. However, the rule setting unit 32 sets a serving rule that gives priority to a flat serving as shown in FIGS. 6A and 6B. As a result, it is possible to improve the accommodation stability of the parts W31 to W34 in the accommodating portion A13 and prevent the parts W31 to W34 from popping out.

7 (A) to 7 (D) are diagrams for explaining an example of the serving rule, and are diagrams showing the serving rule for setting the serving order according to the external area. It is similar to the serving rule shown in FIG. 4, but the rule setting unit 32 precedes the component having a small occupied area in a plan view when serving the components in the kitting tray 40 in which the components overlap in the vertical direction. Set the serving rules to serve. The occupied area here is an area determined by the outer contour in a plan view, and if a space exists in the outer contour, it is an area including the space.

FIG. 7A is a plan view showing a state in which parts W41 and W42 having different occupied areas are served on the kitting tray 40. The component W41 is a bolt-shaped component and is served to the storage portion A14 partitioned by the inner frame portion 42. That is, the component W41 has a small occupied area that can be accommodated in one storage unit A14. On the other hand, the component W42 is a large ring-shaped component such as a gasket, and has a large occupied area that cannot be accommodated in any of the accommodating portions partitioned by the inner frame portion 42. Therefore, the component W42 is served so as to be supported by the upper end edge 42T of the inner frame portion 42.

In such a case, the rule setting unit 32 serves the component W41 having a small occupied area in the storage unit A14 in advance, and then serves the component W42 having a large occupied area. That is, as shown in FIG. 7B, first, the robot hand 10 is served with the component W41 in the storage portion A14, and is brought into contact with the bottom plate 43. If there are other parts to be accommodated in the compartment to the other accommodation, the other parts will be served as well. After that, as shown in FIG. 7C, the component W42 having a large occupied area is served at a predetermined position in contact with the upper end edge 42T of the inner frame portion 42.

When the component W42 having a large occupied area is served in advance, the component W42 closes the upper opening of the storage portion A14. This causes a problem that the part W41 having a small occupied area cannot be served, and a problem that the part W41 is served on the part W42 as shown in FIG. 7 (D). By setting the serving rules as shown in FIGS. 7 (B) and 7 (C), these problems can be prevented.

8 (A) to 8 (C) are diagrams for explaining an example of the serving rule, and are views showing the serving rule to be accommodated in the kitting tray 40 so that the center of gravity of the parts is as low as possible. The rule setting unit 32 has a first serving state in which the height position of the center of gravity of the component is the first position and a height position of the center of gravity of the component in the first position as the serving state of the component to the kitting tray 40. When the second serving state, which is a higher second position, can be obtained, the serving rule is set so that the component takes the first serving state.

FIG. 8A is a top view showing the serving state of the parts W51 and W52 to the accommodating portion A15 partitioned by the inner frame portion 42. The parts W51 and W52 are the same bolt parts having the bolt head B1 and the bolt body B2, and have a length that the entire parts cannot fit in the compartment of the accommodating portion A15. The component W51 is served with the side of the bolt head B1 accommodated in the accommodating portion A15 (contacting the bottom plate 43) and the side of the bolt main body B2 riding on the upper end edge 42T of the inner frame portion 42. On the contrary, the component W52 is served in a state where the side of the bolt main body B2 is housed in the accommodating portion A15 and the side of the bolt head B1 protrudes upward from the upper end edge 42T of the inner frame portion 42. Reference numerals G1 and G2 in the figure indicate the positions of the centers of gravity of the parts W51 and W52, respectively. Since the bolt head B1 is a heavy portion in the parts W51 and W52, the centers of gravity G1 and G2 are located close to the bolt head B1 of the bolt body B2.

FIG. 8B is a side view showing a state in which the component W51 is accommodated in the accommodating portion A15. The component W51 is served to the accommodating portion A15 in a state where the bolt head B1 side is downward and the vicinity of the tip end side of the bolt body B2 is inclined so as to be lifted by the upper end edge 42T (first serving state). The center of gravity G1 of the component W51 is located at a height h1 (first position) determined by the size of the bolt head B1 with respect to the bottom plate 43.

On the other hand, FIG. 8C is a side view showing a state in which the component W52 is accommodated in the accommodating portion A15. The component W52 is served to the accommodating portion A15 in a state where the tip of the bolt body B2 is downward and the vicinity of the root portion of the bolt body B2 near the bolt head B1 is inclined so as to be lifted by the upper end edge 42T (second). Serving state). In this case, since the bolt head B1 side is upward, the center of gravity G2 of the component W52 is located at a height h2 (second position) higher than the center of gravity G1 of the component W51 by Δh.

In such a case, the rule setting unit 32 sets the serving rule so as to take the serving state adopted for the component W51 of FIG. 8 (B). That is, the serving rule is set so that the serving with the lowest height position of the center of gravity G1 can be served in the state of being accommodated in the accommodating portion A15. As shown in FIG. 8C, in a storage mode such as the component W52 in which the center of gravity G2 is present at a relatively high position, the component W52 is likely to fall from the accommodating portion A15. On the other hand, as shown in FIG. 8B, by adopting a storage mode such as the component W51 in which the center of gravity G1 is lower, the component W51 can be prevented from falling from the accommodating portion A15.

[Specific examples of machine learning]
Subsequently, an example in which the rule setting unit 32 sets the serving rule by machine learning by the learning unit 34 (evaluation unit) will be described. Here, an example in which the three-dimensional measuring device 20 (imaging device) acquires a three-dimensional image of the kitting tray 40 after the serving operation by the robot hand 10 is executed, and the learning unit 34 evaluates based on the three-dimensional image. Is shown.

<Structure of learning department>
When machine learning is executed, the learning unit 34 is based on the control information of the robot hand 10 when a certain serving operation is executed and the position information of the parts in the kitting tray 40 on which the serving operation is executed. Learn the optimal behavior pattern of the robot hand 10 when serving. The learning result acquired by the learning unit 34 is reflected in the serving rule set by the rule setting unit 32. As described above, the learning unit 34 includes the displacement amount observation unit 35, the reward setting unit 36, and the value function update unit 37 (FIG. 1).

The displacement amount observing unit 35 includes three-dimensional image data (hereinafter referred to as basic image data) of the comparison source kitting tray 40 in which the parts are served and a three-dimensional image of the comparison target kitting tray 40 in which the parts are newly served. Compare with data (hereinafter referred to as comparative image data). Then, the displacement amount observing unit 35 derives the displacement amount of the three-dimensional position of the component to be compared with respect to the three-dimensional position of the comparison source component. For the basic image data, for example, a state in which the target component is ideally served on the kitting tray 40, which serves as a serving sample, is imaged by the three-dimensional measuring device 20, and the image processing unit 25 derives the image by the imaging. Image data including three-dimensional position information (X, Y, Z coordinate values) for the target component. The comparative image data is also image data including similar three-dimensional position information acquired by imaging the kitting tray 40 on which the target component is served in the learning process with the three-dimensional measuring device 20.

The reward setting unit 36 associates the serving operation (behavior pattern) executed by the robot hand 10 with the serving state of the target component served by the action pattern, and performs a process of giving a reward R to the action pattern. .. Specifically, the reward setting unit 36 acquires control data of an action pattern executed by the robot hand 10 at the time of picking and releasing a certain target component from the drive control unit 31. Further, the reward setting unit 36 acquires the displacement amount data derived by the displacement amount observation unit 35 for the target component served according to the behavior pattern. A reward R is given to the behavior pattern based on the control data of the behavior pattern and the data of the displacement amount.

The reward R can be set so that the smaller the displacement amount, the larger the value is given. For example, when the displacement amount is larger than the preset threshold value Th, the reward R = 0, but when the displacement amount is smaller than the threshold value Th, the reward R> 0 is set. Then, when the displacement amount is smaller than the threshold value Th, the smaller the displacement amount, the larger the reward R can be set. Further, the reward R is given a larger value when the center of gravity of the parts is not served lower, the overlap between the parts is less, or the tact time of one serving cycle is shorter. Can be set to be.

The value function updating unit 37 updates the value function that defines the value Q (s, a) of the action pattern of the robot hand 10 according to the reward R set by the reward setting unit 36. The value function update unit 37 updates the value function using the update formula of the value Q (s, a) represented by the following formula (1).

In the above equation (1), "s" represents the state of the robot hand 10, and "a" represents the action of the robot hand 10 according to the action pattern. The state of the robot 2 shifts from the state "s" to the state "s'" by the action "a". R (s, a) represents the reward R obtained by the transition of the state. The term with "max" is the value Q (s', a') when the most valuable action "a'" is selected in the state "s'" multiplied by "γ". “Γ” is a parameter called the attenuation factor, and is in the range of 0 <γ ≦ 1 (for example, 0.9). Further, “α” is a parameter called a learning rate, and is in the range of 0 <α ≦ 1 (for example, 0.1).

In the above equation (1), the value Q (s, a) of the action “a” in the state “s” is based on the reward R (s, a) set by the reward setting unit 36 for the action “a”. Represents an update expression that updates. That is, the above equation (1) has a value Q (s', a') of the action "a'" in the state "s'" rather than a value Q (s, a) of the action "a" in the state "s". If the total value of and the reward R (s, a) is larger, the value Q (s, a) is increased, and if it is smaller, the value Q (s, a) is decreased. That is, the value function updating unit 37 updates the value function using the updating formula represented by the above formula (1), so that the value Q (s, a) of a certain action “a” in a certain state “s” is used. ) To the reward R set for the action "a" and the value Q (s', a') of the best action "a'" in the next state "s'" by the action "a". I try to get closer.

<Machine learning process by the learning department>
9 and 10 are flowcharts showing an example of the learning operation of the serving rule. First, the information acquisition unit 33 of the control unit 30 acquires the basic image data of the kitting tray 40 to which the parts are served and the shape data of the kitting tray 40 itself (step S11). For the basic image data, for example, as shown in FIG. 2, the kitting tray 40 in a state where a plurality of types of parts W0 to W32 are ideally served is imaged by the second camera 22 of the three-dimensional measuring device 20. It is acquired based on the three-dimensional image data (ideal serving information) obtained. Further, the shape data is data relating to the size of the outer frame portion 41 of the kitting tray 40, the size and depth of each inner frame portion 42, and the like. The information acquisition unit 33 acquires the three-dimensional image data obtained by imaging the empty kitting tray 40 with the second camera 22 or the data given from the input unit 26 as the shape data.

The rule setting unit 32 initially sets the serving rule based on the above ideal serving information and the shape data of the kitting tray 40. That is, which parts W0 to W32 are served to which accommodating portions A1 to A3 of the kitting tray 40 are determined. When setting the serving order, the control unit 30 stores in advance the basic rules regarding the serving order and the serving mode, as described above based on FIGS. 4 to 8, and is used for the initial setting of the serving rule. Ru. This initially set serving rule is modified according to the learning result (evaluation of the evaluation unit) by the learning unit 34.

Next, the information acquisition unit 33 acquires the shape and storage position information of the target component served on the kitting tray 40 (step S12). That is, the information regarding the shapes of the parts W0 to W32 and the position information of the parts storage tray for storing the parts W0 to W32 with respect to the robot hand 10 are acquired. These information can be acquired from the three-dimensional image data based on the image pickup result of the first camera 21 of the three-dimensional measuring device 20 or the input data given from the input unit 26. The above steps S11 and S12 are advance preparations for the learning process.

When the learning process is started, the information acquisition unit 33 acquires the position information of the kitting tray 40 from which the parts are to be served from the image pickup result of the second camera 22 (step S13). That is, the position information of the kitting tray 40 with respect to the robot hand 10 is acquired. Subsequently, the information acquisition unit 33 causes the first camera 21 to take an image of the parts storage tray accommodating the parts to be served, and obtains the three-dimensional position information of the parts to be served based on the result of the object recognition processing of the image processing unit 25. Acquire (step S14). As a result, the coordinate values of the target parts to be served in the parts storage tray are acquired. The position information acquired by the information acquisition unit 33 is given to the drive control unit 31 through the rule setting unit 32.

The drive control unit 31 operates the robot hand 10 based on the serving rule set by the rule setting unit 32 and the position information acquired by the information acquisition unit 33, and sequentially picks the target parts (step S15). Then, based on the image pickup result of the component recognition camera (not shown) that captures the component gripped by the hand portion 16 of the robot hand 10 from the lower surface side, the control unit 30 determines whether or not the target component is gripped by the hand portion 16. Determination (step S16). When the target component is not gripped, that is, when the hand portion 16 fails to grip the component (NO in step S16), the process returns to step S14 and the picking retry of the target component is executed.

When the target part is gripped (YES in step S16), the drive control unit 31 drives the robot hand 10 to carry the picked target part to the kitting tray 40, and also according to the serving rule and the position information. Based on this, the target component is released at a predetermined XYZ position (step S17). This completes the serving of one target part.

After that, it is confirmed whether or not all the scheduled serving of the parts has been completed (step S18). When all the servings have not been completed (NO in step S18), it is subsequently confirmed whether or not the success rate of gripping the parts by the hand portion 16 is good (step S19). When the gripping success rate is good (YES in step S19), it can be said that good picking is executed based on the object recognition processing result acquired in step S14. In this case, the process proceeds to step S15, and the drive control unit 31 executes picking of the next target component. On the other hand, when the gripping success rate is not good (NO in step S19), it can be said that there is a discrepancy between the object recognition processing result and the actual picking. In this case, the process returns to step S14, and the first camera 21 is made to take an image of the component storage tray again to perform the object recognition process.

When all the servings are completed (YES in step S18), the process proceeds to the flow of FIG. 10, and the information of the kitting tray 40 after the serving is completed is acquired (step S21). Specifically, the image pickup control unit 24 of the three-dimensional measuring device 20 causes the second camera 22 to take an image of the kitting tray 40 after serving, and the image processing unit 25 indicates a three-dimensional position indicating the serving position of the parts in the kitting tray 40. Derive information. The displacement amount observation unit 35 of the learning unit 34 acquires image data including the three-dimensional position information of such a component from the camera control unit 23 as the above-mentioned comparative image data. After releasing the part gripped by the hand portion 16, it is unknown how the part behaves in the kitting tray 40. In this step S21, the execution result of one action pattern of the robot hand 10 in which one part is gripped at a specific position, picked, and released at a specific height position is grasped.

The displacement amount observing unit 35 compares the comparative image data acquired in step S21 with the basic image data acquired in step S11, and the three-dimensional position of the component in the comparative image data and the three-dimensional position of the component in the basic image data. The amount of displacement with respect to is derived (step S22). The image data of the kitting tray 40 after serving obtained in the learning process up to that point can be used as the basic image data of the comparison source. In this case, the position stability of the parts in the kitting tray 40 after serving can be evaluated. Further, it is desirable that the displacement amount is mainly obtained from the displacement amount at the position of the center of gravity of the component.

Subsequently, the reward setting unit 36 determines whether or not the displacement amount is larger than the preset threshold value Th (step S23). The fact that the amount of displacement is large means that the parts are not served at the desired positions or the parts cannot be served stably in the behavior pattern of the robot hand 10 this time. When the displacement amount is equal to or greater than the threshold value Th (YES in step S23), the reward setting unit 36 gives a reward R of “0; zero” to such an action pattern of the robot hand 10 (step S24). On the other hand, when the displacement amount is less than the threshold value Th (NO in step S23), the reward setting unit 36 gives a reward R larger than "0; zero" to such an action pattern of the robot hand 10. (Step S25).

After that, the value function updating unit 37 updates the value function that defines the value Q (s, a) of the action pattern of the robot hand 10 by using the updating formula (1) above (step S26). Each of the processes shown in steps S13 to S26 above is a process executed in one cycle of the learning process by the learning unit 34. The learning unit 34 determines whether or not the number of learnings has reached the predetermined number N (step S27). If the predetermined number of times N has not been reached (NO in step S27), the learning unit 34 returns to step S13, causes the next kitting tray 40 to serve parts, and repeats the learning process. On the other hand, when the predetermined number of times N is reached (YES in step S27), the learning unit 34 ends the learning process.

<Variation example of machine learning processing>
In the above machine learning processing example, in step S11 of FIG. 9, an example is shown in which the information acquisition unit 33 acquires the basic image data (ideal serving information) of the kitting tray 40 in which the parts are served. Alternatively, the ideal serving information may not be given to the information acquisition unit 33, and only the information regarding the kitting tray 40 on which the parts are not served and the accommodating portion thereof and the information regarding the size of the parts may be acquired.

That is, in this modification, in step S11 of FIG. 9, the information acquisition unit 33 acquires only the shape data of the kitting tray 40 itself. Then, the rule setting unit 32 initially sets a provisional serving rule based on the shape data of the target part acquired in step S12. This serving rule is modified according to the learning result (evaluation of the evaluation unit) of the learning unit 34.

According to this modification, it is possible to detect a serving rule close to the ideal serving by machine learning even if the ideal serving information is not given. In ideal serving information, serving rules tend to be set according to fixed assumptions. For example, there is a tendency for a serving rule to be set based on the premise that small-sized parts are served in a small-sized accommodating portion prepared in the kitting tray 40. However, based on such a premise, it may not be possible to serve good parts. This point will be described with reference to FIG.

11 (A) and 11 (B) are views showing an example of serving a relatively large large-sized component W61 and a relatively small-sized small-sized component W62 to the kitting tray 40. The kitting tray 40 is provided with a wide accommodating portion A16 having a relatively wide accommodating space and a narrow accommodating portion A17 having a relatively narrow accommodating space.

FIG. 11A shows a state in which a large-sized component W61 is served in the wide accommodating portion A16 and a small-sized component W62 is served in the narrow accommodating portion A17. The serving example shown in FIG. 11A is in line with the general concept of serving, in which small-sized parts are served to a small-sized accommodating portion. However, in the actual serving result, since the number of small size parts W62 served is large, the small size parts W62 are piled up in the narrow accommodating portion A17. That is, the position of the center of gravity of many small-sized parts W62 exists at a position higher than the bottom plate 43 in the narrow accommodating portion A17. In this case, the small-sized component W62 tends to overflow from the kitting tray 40 (narrow accommodating portion A17), which is not preferable.

On the other hand, in the serving example shown in FIG. 11B, the narrow accommodating portion A17 is provided with the large-sized component W61, and the wide accommodating portion A16 is provided with the small-sized component W62 having a large number of parts. In this serving example, the positions of the centers of gravity of the small-sized parts W62 and the large-sized parts W61 are located lower than the bottom plate 43 in the accommodating portions A16 and A17, respectively. As a result, it is possible to realize a serving in which the parts W61 and W62 are unlikely to fall from the kitting tray 40. It is possible to detect the serving mode as shown in FIG. 11B by learning the serving mode in which the center of gravity is lowered as much as possible by the learning process by the learning unit 34 without giving the ideal serving information.

[Invention included in the above embodiment]
The specific embodiments described above mainly include inventions having the following configurations.

The parts serving device for a kitting tray according to one aspect of the present invention is a parts serving device for serving a plurality of types of parts having different sizes to a kitting tray provided with a plurality of accommodating portions, and picking and releasing of the parts can be performed. Serving that has a possible head portion, picks and transports the target component from the plurality of types of components by the head portion at the storage position of the component, and releases the target component from the head portion to the kitting tray. The robot hand is provided with a robot hand that performs an operation and a control unit that controls the operation of the robot hand. It is provided with a rule setting unit for setting a serving rule to the robot hand, and a drive control unit for causing the robot hand to execute the serving operation based on the serving rule.

According to this parts serving device, a serving rule is set according to the mode of the plurality of accommodating portions of the kitting tray and the mode of the parts served in the kitting tray. That is, it is possible for the rule setting unit to flexibly determine the serving rule according to the mode of the parts and the accommodating portion, instead of strictly determining the serving procedure by the robot hand by programming or the like. Therefore, it is possible to properly serve the parts to the kitting tray without performing programming that requires labor and skill.

In the above-mentioned parts serving device, the rule setting unit sets a serving rule for serving parts whose serving position is lower in advance when serving parts that overlap each other in the vertical direction in the kitting tray. Is desirable.

According to this parts serving device, it is possible to prevent a problem that a part whose serving position should be lower is served on a part whose serving position is upper.

In the above-mentioned parts serving device, the rule setting unit is more than a serving in which the parts are overlapped in the vertical direction when both the serving in which the parts are arranged in the horizontal direction and the serving in which the parts are overlapped in the vertical direction can be performed in the kitting tray. It is desirable to set a serving rule that gives priority to serving the parts arranged in the horizontal direction.

According to this parts serving device, it is possible to serve by effectively utilizing the horizontal space of the kitting tray without having to serve the parts in the vertical direction.

In the above-mentioned parts serving device, the rule setting unit sets a serving rule for serving parts having a small occupied area in a plan view in advance when the parts are served in the kitting tray in which the parts overlap in the vertical direction. It is desirable to do.

According to this parts serving device, large parts with a large occupied area block the accommodation part of the kitting tray, making it impossible to serve small parts with a small occupied area, and small parts are served on top of the large parts. It is possible to prevent problems that occur.

In the above-mentioned parts serving device, the rule setting unit determines the first serving state in which the height position of the center of gravity of the parts is the first position and the height of the center of gravity of the parts as the serving state of the parts to the kitting tray. When it is possible to take a second serving state in which the position is higher than the first position, it is desirable to set the serving rule so that the component takes the first serving state.

According to this component serving device, each component can be served in a kitting tray with a lower center of gravity. This makes it difficult for the parts to fall from the kitting tray when the kitting tray is moved after serving.

In the above-mentioned parts serving device, an image pickup device that acquires a three-dimensional image of the kitting tray after the serving operation by the robot hand is executed, and a serving state of parts in the kitting tray based on the three-dimensional image. It is desirable that the rule setting unit further includes an evaluation unit for evaluating the above, and the rule setting unit sets the serving rule according to the evaluation of the evaluation unit.

According to this component serving device, it is possible to make the rule setting unit perform machine learning about a method of performing better serving through the evaluation of the recording and evaluation unit, for example, the mode of picking by the head unit and the release position. Then, based on the machine learning, the rule setting unit can set the serving rule, so that it is possible to set the serving rule excellent in serving performance without programming.

In the above-mentioned parts serving device, the rule setting unit acquires the ideal serving information in a state where the plurality of types of parts are ideally served to the kitting tray, and the serving based on the ideal serving information. It is desirable to initialize the rules and modify the serving rules according to the evaluation of the evaluation unit.

According to this parts serving device, the serving rule initially set based on the serving information can be modified according to the machine learning of the serving method by the rule setting unit, and the ultimate optimum serving rule can be set. ..

In the above-mentioned parts serving device, the rule setting unit acquires information on the kitting tray and its accommodating portion and information on the size of the parts, and initially sets the serving rule based on the information, and the evaluation unit. It is desirable to modify the serving rule according to the evaluation of.

According to this component serving device, the serving rule is initially set based on the information regarding the size of the kitting tray and the component, and the serving rule is modified according to the machine learning of the serving method by the rule setting unit. Therefore, even if the ideal serving information as described above is not given, it is possible to detect a serving rule close to the ideal serving by machine learning.

According to the present invention as described above, it is possible to provide a parts serving device for a kitting tray that can accurately serve various parts to the kitting tray by using a robot hand.

[Explanation of code]
Wa, Wb, Wc, Wd, W1 to W62 parts (target parts)
Ta, Tb parts storage tray (storage position)
A1 to A3, A11 to A17 Accommodating parts G1, G2 Center of gravity h1, h2 First position, second position 1 Parts serving device 10 Robot hand 15 Head part 16 Hand part 20 Three-dimensional measuring device (imaging device)
21 1st camera 22 2nd camera 23 Camera control unit 24 Image control unit 25 Image processing unit 30 Control unit 31 Drive control unit 32 Rule setting unit 33 Information acquisition unit 34 Learning unit (evaluation unit)
35 Displacement amount observation unit 36 Reward setting unit 37 Value function update unit 40 Kitting tray 41 Outer frame unit 41
42 Inner frame 42
42T Top edge 42T
43 bottom plate 43

Claims (6)

A parts serving device that serves multiple types of parts of different sizes to a kitting tray equipped with multiple accommodating parts.
It has a head part that can pick and release parts, and at the storage position of the parts, the target part is picked and transported from the plurality of types of parts by the head part, and the target part is picked and transported from the head part. A robot hand that performs serving operations to be released to the kitting tray,
A control unit that controls the operation of the robot hand,
An image pickup device that acquires a three-dimensional image of the kitting tray after the serving operation by the robot hand is executed, and
An evaluation unit for evaluating the serving state of parts in the kitting tray based on the three-dimensional image is provided.
The control unit
A rule setting unit for setting a serving rule for the parts to the kitting tray according to the mode of the plurality of types of parts and the plurality of accommodating portions.
A drive control unit that causes the robot hand to execute the serving operation based on the serving rule is provided .
The rule setting unit acquires ideal serving information in a state where the plurality of types of parts are ideally served to the kitting tray, and initially sets the serving rule based on the ideal serving information. A parts serving device for a kitting tray that modifies the serving rule according to the evaluation of the evaluation unit. A parts serving device that serves multiple types of parts of different sizes to a kitting tray equipped with multiple accommodating parts.
It has a head part that can pick and release parts, and at the storage position of the parts, the target part is picked and transported from the plurality of types of parts by the head part, and the target part is picked and transported from the head part. A robot hand that performs serving operations to be released to the kitting tray,
A control unit that controls the operation of the robot hand,
An image pickup device that acquires a three-dimensional image of the kitting tray after the serving operation by the robot hand is executed, and
An evaluation unit for evaluating the serving state of parts in the kitting tray based on the three-dimensional image is provided.
The control unit
A rule setting unit for setting a serving rule for the parts to the kitting tray according to the mode of the plurality of types of parts and the plurality of accommodating portions.
A drive control unit that causes the robot hand to execute the serving operation based on the serving rule is provided.
The rule setting unit acquires information on the kitting tray and its accommodating unit and information on the size of parts, initially sets the serving rule based on the information, and sets the serving rule according to the evaluation of the evaluation unit. A parts serving device for the kitting tray that modifies the rules. In the parts serving device for the kitting tray according to claim 1 or 2 .
The rule setting unit sets a serving rule for serving parts whose serving position is lower when serving parts that overlap each other in the vertical direction in the kitting tray. Device. In the parts serving device for the kitting tray according to claim 1 or 2 .
In the rule setting unit, when both the serving in which the parts are arranged in the horizontal direction and the serving in which the parts are overlapped in the vertical direction are feasible in the kitting tray, the serving in which the parts are arranged in the horizontal direction is more than the serving in which the parts are overlapped in the vertical direction. A parts serving device for a kitting tray that sets a serving rule that gives priority to execution. In the parts serving device for the kitting tray according to claim 3 ,
The rule setting unit sets a serving rule for serving parts having a small occupied area in a plan view in advance when serving parts that overlap each other in the vertical direction in the kitting tray. Serving device. In the parts serving device for the kitting tray according to any one of claims 1 to 5 .
In the rule setting unit, as the serving state of the component to the kitting tray, the first serving state in which the height position of the center of gravity of the component is the first position and the height position of the center of gravity of the component are the first position. A component serving device for a kitting tray that sets the serving rule so that the component takes the first serving state when the second serving state, which is a higher second position, can be obtained.

Images