Picking method

Abstract

To provide a picking method, a picking vehicle, and a picking system for effectively achieving the automation of picking processes.SOLUTION: There are provided a picking method, a picking vehicle, and a picking system. The picking method includes steps of: acquiring current volume data and current weight data of a box; and generating current placement position data of the box based on the current volume data, the current weight data, and a placement position model of an artificial intelligence model. The placement position model includes multiple placement data respectively containing predetermined volume data, predetermined weight data, and predetermined placement position data corresponding to one another and moves the box to a target placement position via the picking vehicle based on the current placement position data.SELECTED DRAWING: Figure 9

Description

Technical Field

[0001] The present disclosure relates to a picking method, a picking vehicle, and a picking system.

Background Art

[0002] In recent years, due to the rapid development of online shopping, a large amount of labor is required in the logistics industry. However, manually picking items not only causes a shortage of manpower but also makes it difficult to control the stacking of items. For example, when manually moving and stacking items, the position of the items in the rack may shift or protrude from the rack, and it is difficult to maintain the operating rate of the rack above a desired level.

Summary of the Invention

[0003] The present disclosure aims to provide a picking method, a picking vehicle, and a picking system that can effectively automate the picking process.

[0004] An embodiment of the present disclosure includes obtaining current volume data and current weight data of an item, generating current placement position data of the item based on the current volume data, the current weight data, and a placement position model of an artificial intelligence model, where the placement position model includes a plurality of placement data, and each of the plurality of placement data includes a plurality of groups of corresponding predetermined volume data, predetermined weight data, and predetermined placement position data, and moving the item to a target placement position via a picking vehicle based on the current placement position data.

[0005] Another embodiment of the present disclosure provides a picking vehicle configured to move on a support surface, comprising a movable base, a mounting platform, at least one fork, and a pressing component. The movable base is configured to move on a support surface. The mounting platform is movably positioned on the movable base and has a mounting surface facing away from the support surface. At least one fork is movably positioned on the mounting surface of the mounting platform. The pressing component is movably positioned on the mounting surface of the mounting platform. At least one fork is positioned between the pressing component and the mounting surface.

[0006] A further embodiment of the present disclosure provides a picking system configured to pick one or more articles, comprising a control device, a supply device, at least one rack, and at least one picking vehicle. The supply device is electrically connected to the control device and configured to transport one or more articles. At least one rack is positioned to the side of the supply device. At least one picking vehicle comprises a movable base, a mounting platform, at least one fork, and a pressing component. The movable base is configured to move between the supply device and at least one rack on a support surface and is electrically connected to the control device. The mounting platform is movably positioned on the movable base and has a mounting surface oriented away from the support surface. At least one fork is movably positioned on the mounting surface of the mounting platform and configured to support one or more articles. The pressing component is movably positioned on the mounting surface of the mounting platform. At least one fork is positioned between the pressing component and the mounting surface. The pressing component is configured to press one or more articles supported by at least one fork into at least one rack.

[0007] According to the picking method, picking vehicle, and picking system disclosed in the above embodiments, current placement data of an item is generated based on current volume data, current weight data, and a placement position model of an artificial intelligence model. Therefore, the trained artificial intelligence model can determine the optimized placement position of an item based on its volume and weight. In this way, not only is it prevented that items shift within the rack or protrude from the rack, but it is also permissible for the rack occupancy rate to be higher than a desired level, and the automation of the picking process is effectively realized.

[0008] Furthermore, the picking vehicle can move and push items into the racks using forks and pressing components movably mounted on the mounting surface of the mounting platform. Thus, automation of the picking operation is effectively achieved through the control of the picking vehicle. In addition, compared to picking vehicles that suck items up using a vacuum chunk, this disclosure allows for the movement of items of various sizes using forks and pressing components without being limited by the size of the vacuum chunk. [Brief explanation of the drawing]

[0009] This disclosure will be better understood from the detailed description and accompanying drawings set forth below in this specification. The accompanying drawings are given for illustrative purposes only and are therefore not intended to limit this disclosure.

[0010] [Figure 1] This is a perspective view of a picking vehicle according to one embodiment of the present disclosure; [Figure 2] Figure 1 is a side view of the picking vehicle. [Figure 3] Figure 1 is a partially enlarged perspective view of the picking vehicle. [Figure 4] Figure 3 is a partially enlarged perspective view of a picking vehicle. [Figure 5] This is a block diagram of a picking system according to one embodiment of the present disclosure. [Figure 6]Figure 5 is a schematic perspective view of the picking system. [Figure 7] Figure 5 is a partially enlarged perspective view of the picking system. [Figure 8] Figure 5 is a partially enlarged perspective view of the picking system. [Figure 9] This is a flowchart of a picking method according to one embodiment of the present disclosure. [Figure 10] This is a flowchart of a picking method according to one embodiment of the present disclosure. [Figure 11] This is a flowchart of a picking method according to one embodiment of the present disclosure. [Figure 12] This is a perspective view of a picking vehicle according to another embodiment of the present disclosure. [Modes for carrying out the invention]

[0011] The following detailed description includes numerous specific details to provide a complete understanding of the disclosed embodiments for illustrative purposes. However, it will be apparent that one or more embodiments may be carried out without these specific details. In other examples, well-known structures and apparatus are shown schematicly for the sake of simplicity in the drawings.

[0012] Please refer to Figures 1 to 4. Here, Figure 1 is a perspective view of a picking vehicle according to one embodiment of the present disclosure, Figure 2 is a side view of the picking vehicle of Figure 1, Figure 3 is a partially enlarged perspective view of the picking vehicle of Figure 1, and Figure 4 is a partially enlarged perspective view of the picking vehicle of Figure 3.

[0013] The picking vehicle 400 is, for example, an automated guided vehicle (AGV) or an autonomous mobile robot (AMR). In this embodiment, the picking vehicle 400 includes a movable base 410, a mounting plate 420, a rotating assembly 430, a lifting assembly 440, a mounting platform 450, a drive assembly 460 for forks, a fixed plate 465, a plurality of forks 470, a drive assembly 480 for pressing parts, and pressing parts 490.

[0014] As shown in Figure 1, the movable base 410 is configured to move on the support surface 30. The movable base 410 has a top surface 411, a first side surface 412, and a second side surface 413. The first side surface 412 and the second side surface 413 face away from each other and are connected to the top surface 411. The top surface 411 is configured to face away from the support surface 30.

[0015] As shown in Figure 1, the mounting plate 420 is rotatably positioned on the upper surface 411 via a rotating assembly 430. The rotating assembly 430 may include a drive unit 431 and a gear assembly 432. The drive unit 431 is, for example, a motor and is positioned on the movable base 410. The gear assembly 432 connects the drive unit 431 and the mounting plate 420, and is configured so that the drive unit 431 rotates the mounting plate 420 via the gear assembly 432, for example, along the Z-axis.

[0016] As shown in Figures 1 and 2, the lifting assembly 440 includes a slide rail 441, a drive unit 442, and a slider 443. The slide rail 441 is, for example, a linear slide rail and is erected on the upper surface 411 of the movable base 410. The drive unit 442 is, for example, a motor and is located on the slide rail 441. The slider 443 is slidably located on the slide rail 441. The drive unit 442 is connected to the slider 443 via, for example, a belt (not shown) inside the slide rail 441 to drive the slider 443 to slide on the slide rail 441, for example, along the Z-axis. The drive unit 442 includes, for example, a reduction drive for converting the rotational speed of the drive unit 442 into torque. The mounting platform 450 is fixed to the slider 443 and has a mounting surface 451. The mounting surface 451 is oriented away from the support surface 30.

[0017] As shown in FIGS. 1 and 2, the picking vehicle 400 can further include a chain 455. Two opposing sides of the chain 455 are respectively fixed to the slide rail 441 and the mounting platform 450.

[0018] As shown in FIGS. 3 and 4, the fork drive assembly 460 includes, for example, a drive unit 461, a connecting rod 462, a slide rail 463, and a slider 464. The drive unit 461 is, for example, a motor and is disposed on the mounting surface 451. The connecting rod 462 connects the drive unit 461 and the slide rail 463. The slide rail 463 is disposed on the mounting surface 451 of the mounting platform 450. The slider 464 is slidably disposed on the slide rail 463. The drive unit 461 is configured to move the slider 464 on the slide rail 463 toward or away from the slide rail 441 via the connecting rod 462. The fixing plate 465 is fixed to the slider 464. The forks 470 are spaced apart from each other and protrude from the side surface of the fixing plate 465. Further, in the present embodiment, the mounting platform 450 is disposed between the fork 470 and the movable base 410. That is, the mounting platform 450 is disposed above the movable base 410. Therefore, the picking vehicle 400 can move an article (an article at a high position) at a position away from the support surface 30.

[0019] The drive assembly 480 for pressing the component (for the pressing component) includes, for example, a drive unit 481, a connecting rod 482, two gears (spur gears) 483, two belts 484, two sliders 485, and two guide rods 486. The drive unit 481 is, for example, a motor and is disposed on the mounting surface 451. The connecting rod 482 is connected to the drive unit 481. The two gears 483 are sleeve-shaped and are fixed to the connecting rod 482. The two belts 484 are respectively engaged with the two gears 483. The two sliders 485 are respectively fixed to the two belts 484 and are slidably disposed on the mounting surface 451 of the mounting platform 450 via the guide rods 486. The drive unit 481 is configured to drive the slider 485 to move toward or away from the slide rail 441 on the guide rod 486 via the connecting rod 482, the gear 483, and the belt 484. The pressing component 490 is fixed on the two sliders 485. Further, the fork 470 is disposed between the pressing component 490 and the mounting surface 451.

[0020] The picking vehicle 400 can further include a control unit 495. The control unit 495 is disposed on the mounting surface 451 and is electrically connected to the drive units 461 and 481 on the mounting (installation) surface 451 via one or more cables (not shown). The one or more cables may be disposed in the chain 455 so as not to interfere with the operation of other components.

[0021] Next, the picking system 10 according to an embodiment of the present disclosure will be described. Referring to FIGS. 5 and 6, FIG. 5 is a block diagram of a picking system according to an embodiment of the present disclosure, and FIG. 6 is a schematic perspective view of the picking system of FIG. 5. In the present embodiment, the picking system 10 includes a control device 100, a supply device 200, a plurality of racks 300, and a plurality of picking vehicles 400.

[0022] The supply device 200 is, for example, a conveyor and is electrically connected to the control device 100. The supply device 200 is configured to transport one or more articles 20, 21. The rack 300 is located to the side of the supply device 200. The structure of the picking vehicle 400 has been described in detail with reference to Figures 1 to 4, so a repeated description will be omitted. The movable base 410 is configured to move between the supply device 200 and the rack 300 on the support surface 30 and is electrically connected to the control device 100. For example, a circuit assembly 496 electrically connected to the control device 100 may be located on the movable base 410. The circuit assembly 496 may include a computing computer and a control module. The control unit 495 is electrically connected to the circuit assembly 496, for example, via the internet. The fork 470 is configured to support one or more articles 21, 22. The pressing component 490 is configured to push the one or more articles 21, 22 supported by the fork 470 into the rack 300.

[0023] In this embodiment, the picking system 10 may further include a code reader 500, a sensor 600, and an imaging device 700. The code reader 500 is electrically connected to the control device 100 and is configured to read the barcode of the article 20. The sensor 600 is, for example, an infrared sensor or an image sensor. The sensor 600 is electrically connected to the control device 100 and is configured to detect the position of the article 21 on the supply device 200. The imaging device 700 is electrically connected to the control device 100 and is configured to capture an image of the rack 300. In other embodiments, at least one of the code reader 500, sensor 600, and imaging device 700 may be omitted depending on the actual requirements.

[0024] A picking method according to one embodiment of the present disclosure will be described below. Please refer to Figures 6 to 11. Figures 7 and 8 are partially enlarged perspective views of the picking system of Figure 5. Figures 9 to 11 are flowcharts of the picking method according to one embodiment of the present disclosure. For illustrative purposes, articles 20 to 23 labeled with different numbers are shown in Figure 6. In practice, articles 20 to 23 may be the same article at different picking stages. The following steps may be performed, for example, by the control device 100 of Figure 5.

[0025] Please refer to Figures 6 to 9. First, step S01 is performed, for example, by supplying articles 20 to the sensing area S of the supply device 200 using a sorter. In other embodiments, articles may be supplied to the sensing area manually.

[0026] Please refer to Figures 6 and 9. Next, step S02 is performed in which the barcode of item 20 is read by the code reader 500 and the current volume data and current weight data of item 20 are obtained. For example, the code may correspond to the code of the item, and the corresponding volume data and weight data may be obtained from a database based on the code of the item corresponding to the barcode.

[0027] Refer to Figure 9. Next, step S03 is performed, comparing the current volume data and current weight data with the reference volume data and reference weight data, respectively. If the current volume data and current weight data do not match the reference volume data and reference weight data, respectively, a warning signal may be generated to notify the technician to address the error.

[0028] Refer to Figures 6 and 9. If the current volume data and current weight data match the reference volume data and reference weight data, respectively, step S04 is performed to move the article 21 from the sensing area S to the pickup area 201 of the supply device 200, and the sensor 600 determines whether the article 21 is located in the pickup area 201. The pickup area 201 is located, for example, at the end of the supply device 200. If the sensor 600 determines that the article 21 is not located in the pickup area 201, it may generate a warning signal to notify the technician to take action regarding the error. If the sensor 600 determines that the article 21 is located in the pickup area 201, step S05 is performed to call the artificial intelligence model learned by the artificial intelligence algorithm.

[0029] Refer to Figure 9. Next, step S06 is performed to generate current placement data for article 21 based on the current volume data, current weight data, and placement position model of the artificial intelligence model. In this embodiment, the current placement data includes target rack data and target stacking position data. The placement model includes a plurality of placement data. Each placement data includes a plurality of groups of predetermined volume data, predetermined weight data, and predetermined placement data that correspond to each other. In this embodiment, the predetermined placement data includes predetermined rack data and predetermined stacking position data. Specifically, the target rack data and predetermined rack data indicate which rack 300 the article 21 should be placed on. The target stacking position data and predetermined stacking position data indicate where on the indicated rack 300 the article 21 should be placed.

[0030] In other embodiments, the picking system may include one rack. In such embodiments, the current placement position data may not include target rack data, and the predetermined placement position data may not include predetermined rack data.

[0031] In this embodiment or other embodiments, the artificial intelligence model may be trained in a supervised manner with a plurality of training datasets and a generative artificial intelligence (GAI) algorithm. Each training dataset includes volumetric data and weight data, and each training dataset may have a label indicating one of a plurality of predetermined placement locations. The artificial intelligence model may be, for example, ChatGPT, Databricks-Dolly 2.0, or Meta-Llama 2, and may be trained with at least one of PEFT (Parameter Efficient Fine Tuning), LoRA (Low-Rank Adaptation), and Cutting-Stock Algorithm. For example, more than 1,000 training datasets corresponding to an utilization rate of 85% or higher can be randomly generated using the Cutting-Stock Algorithm, and such training datasets can be converted into a plurality of predetermined placement location datasets corresponding to an average utilization rate of 94%.

[0032] In this embodiment, step S04, which determines the position of article 21, is performed before step S05, which invokes the artificial intelligence model, but the disclosure is not limited thereto. In other embodiments, step S05 may be performed before step S04, or steps S04 and S05 may be performed simultaneously.

[0033] Please refer to Figure 9. Next, step S07 is performed to transfer the current mounting position data to the database.

[0034] Please refer to Figures 6 and 9. Next, step S08 is performed to select one of the picking vehicles 400.

[0035] Please refer to Figures 7 and 10. Next, step S09 is performed to control the selected picking vehicle 400 to the vicinity of the pickup area 201.

[0036] Refer to Figures 7 and 10. Next, step S10 is performed, in which the height of the mounting platform 450 relative to the support surface 30 is adjusted via the lifting assembly 440 and the angle of the mounting platform 450 along the Z-axis direction is adjusted via the rotating assembly 430.

[0037] Refer to Figures 7 and 10. Next, step S11 is performed to move the fork 470 away from the slide rail 441 via the fork drive assembly 460 so that the fork 470 is pushed into the space between the article 21 and the feeding device 200.

[0038] Please refer to Figures 7 and 10. Next, step S12 is performed, in which the article 21 is supported by the fork 470.

[0039] Refer to Figures 8 and 10. Then, steps S13 to S18 are performed to move the item 22 to the target placement position via the picking vehicle 400 based on the current placement position data.

[0040] For details, please refer to Figures 8 and 10. Step S13 is first performed to control the picking vehicle 400 to move the item 22 to the vicinity of the rack 300 indicated by the target rack data. Furthermore, referring back to Figure 6, the three picking vehicles 400 in Figure 6 may be assigned to the vicinity of the supply device 200, the vicinity of the rack 300, and a waiting area A located to the side of the rack 300 along the assignment direction D, respectively. The picking vehicle 400 located in waiting area A may be assigned to the vicinity of the supply device 200 in the next cycle of the picking method.

[0041] Refer to Figures 7, 8, and 10. Next, step S14 is performed to adjust the height of the mounting platform 450 relative to the support surface 30 via the lifting assembly 440 and adjust the angle of the mounting platform 450 along the Z-axis via the rotating assembly 430, based on the target stacking position data.

[0042] Refer to Figures 8 and 10. Then, based on the target stacking position data, step S15 is performed in which the drive assembly 460 moves the fork 470 toward or away from the slide rail 441. Furthermore, when moving the fork 470, interference between the fork 470 and the rack 300, or between the fork 470 and the articles 23 placed on the rack 300, may be prevented by a dynamic tracking algorithm.

[0043] Refer to Figures 8 and 11. Next, step S16 is performed to move the pressing component 490 away from the slide rail 441 via the pressing component drive assembly 480.

[0044] Refer to Figures 8 and 11. Next, step S17 is performed in which the pressing component 490 is driven via the pressing component drive assembly 480 to push the article 22 out of the picking vehicle 400.

[0045] Please refer to Figures 8 and 11. Next, step S18 is performed, and the articles 23 are stacked on rack 300 at the target placement position indicated by the target stacking position data. The target placement position indicated by the target stacking position data is located on rack 300 indicated by the target rack data.

[0046] Please refer to Figure 11. Next, step S19 is performed to determine whether rack 300 is abnormal or not via the imaging device 700. If rack 300 is normal, step S20 is performed to determine whether rack 300 is full or not via the imaging device 700. If rack 300 is full, the picking method is terminated. If rack 300 is not full, steps S02 to S19 are repeated.

[0047] For example, the imaging device 700 may capture one or more of the following: a depth image, an infrared image, or an RGB image. Alternatively, it may analyze the structure of the rack to construct a rack coordinate system, analyze the differences between multiple images to identify the contours of the items, and track the items to determine their condition. Therefore, it may analyze the vortices and height of the rack, analyze the differences in depth features, and track the position and condition of the items. In this way, the imaging device 700 can determine whether the item 23 is abnormal or whether the rack 300 is full. For example, if the item 23 is misaligned with other items in the rack 300 or is protruding from the rack 300, the rack 300 is determined to be abnormal, and if the space in the rack 300 containing the items occupies a certain percentage or more of the total space in the rack 300, the rack 300 is determined to be full.

[0048] If rack 300 is abnormal, step S21 is performed and a warning signal is generated to notify a technician to address the error.

[0049] In the picking vehicle 400 shown in Figures 1 to 4, the mounting platform 450 is positioned above the movable base 410, but the disclosure is not limited thereto.

[0050] Other embodiments will be described below. In the following embodiments, some of the reference numerals (symbols) and content from the above embodiments will be used, the same reference numerals will be used for identical or similar elements, and the same technical content will be omitted from the explanation. For explanations of the omitted parts, please refer to the above embodiments, and detailed explanations will be omitted in the following embodiments.

[0051] Please refer to Figure 12, which is a perspective view of a picking vehicle according to another embodiment of the present disclosure. In this embodiment, the picking vehicle 400a includes a movable base 410, a mounting plate 420a, a rotating assembly 430, a lifting assembly 440, a mounting platform 450a, a fork drive assembly 460, a fixed plate 465, a fork 470, a press component drive assembly 480, and a press component 490. The mounting plate 420a includes, for example, a first plate portion 421a and a second plate portion 422a. The second plate portion 422a is erected on the side of the first plate portion 421a. The first plate portion 421a is positioned on the upper surface 411 and connected to the rotating assembly 430. The second plate portion 422a is located on the side of the first side 412 furthest from the second side 413. The mounting platform 450a protrudes from the second plate portion 422a, and the first side surface 412 is located between the mounting platform 450a and the second side surface 413. Therefore, the picking vehicle 400a can have a large workspace and can be used to move items (items with a small height) that are located close to the support surface 30.

[0052] According to the picking method, picking vehicle, and picking system disclosed in the above embodiments, current placement data of an item is generated based on current volume data, current weight data, and a placement position model of an artificial intelligence model. Therefore, the trained artificial intelligence model can determine the optimized placement position of an item based on its volume and weight. In this way, not only is it prevented that items shift within the rack or protrude from the rack, but it is also permissible for the rack occupancy rate to be higher than a desired level, and the automation of the picking process is effectively realized.

[0053] Furthermore, the picking vehicle can move items and push them into the racks using forks and pressing components that are movably mounted on the mounting surface of the mounting platform. Thus, automation of the picking operation is effectively achieved through the control of the picking vehicle. In addition, compared to picking vehicles that suck items up using a vacuum chunk, this disclosure allows for the movement of items of various sizes using forks and pressing components without being limited by the size of the vacuum chunk.

[0054] It will be apparent to those skilled in the art that various modifications and variations are possible with respect to the disclosed embodiments. This specification and the examples are intended to be considered illustrative only, and the true scope of this disclosure is indicated by the following claims and equivalents.

Claims

1. Obtain the current volume data and current weight data of the item, Based on the current volume data, the current weight data, and the placement position model of the artificial intelligence model, current placement position data of the article is generated, the aforementioned placement position model includes a plurality of placement data, each of which includes a plurality of groups of predetermined volume data, predetermined weight data, and predetermined placement position data that correspond to each other. This includes moving the article to the target placement location via a picking vehicle based on the current placement location data, The aforementioned picking vehicle is Movable base and A mounting platform having a mounting surface that is movably positioned on the movable base and facing away from the support surface, At least one fork movably positioned on the mounting surface of the mounting platform, A pressing component that is movably positioned on the mounting surface of the mounting platform, wherein the pressing component comprises at least one fork positioned between the pressing component and the mounting surface, The picking vehicle further comprises a lifting assembly, at least one fork drive assembly, and a press part drive assembly, The lifting assembly includes a slide rail, a drive unit, and a slider, wherein the slide rail is erected on the movable base. The drive unit is positioned on the slide rail, The slider is slidably positioned on the slide rail, The drive unit is connected to the slider, The mounting platform is fixed to the slider, The at least one fork drive assembly is connected to the at least one fork, The drive assembly for the pressing component is connected to the pressing component, Based on the current placement position data, moving the item to the target placement position via the picking vehicle is: The at least one fork is moved toward or away from the slide rail via the at least one fork drive assembly, The pressing component is moved away from the slide rail via the drive assembly for the pressing component. The pressing component is driven via the aforementioned drive assembly for the pressing component to push the article out of the picking vehicle. This includes enabling the stacking of the articles at the target placement position. Picking method.

2. The current placement position data includes target rack data and target stacking position data, the predetermined placement position data includes predetermined rack data and predetermined stacking position data, and the picking vehicle moves the item to the target placement position located in one of a plurality of racks. The picking method according to claim 1.

3. The further includes supplying an article to the sensing area of ​​a supply device before acquiring the current volume data and current weight data of the article, The picking method according to claim 1.

4. After obtaining the current volume data and current weight data of the article, the method further includes comparing the current volume data and current weight data with reference volume data and reference weight data, respectively, and if the current volume data and current weight data match the reference volume data and reference weight data, respectively, the current placement position data of the article is generated based on the current volume data, the current weight data and the previously described placement position model of the artificial intelligence model. The picking method according to claim 1.

5. The following further includes, before moving the article to the target placement position via the picking vehicle based on the current placement position data, moving the article from the sensing area to the pickup area on the supply device, determining by a sensor whether the article is located in the pickup area, and if the article is located in the pickup area, moving the article to the target placement position via the picking vehicle based on the current placement position data. The picking method according to claim 3.

6. The process further includes moving the item to the target placement position via the picking vehicle based on the current placement position data, and then determining whether the rack is abnormal via an imaging device. The picking method according to claim 1.

7. The process further includes moving the item to the target placement location via the picking vehicle based on the current placement location data, and then determining whether the rack is full via an imaging device. The picking method according to claim 1.

Images