Cargo Handling System
The loading and handling system uses cameras and neural networks to continuously monitor and correct load placement, ensuring operations continue uninterrupted even if incorrect loading occurs, thereby addressing the issue of halted workflows due to errors.
Patent Information
- Application Number
- JP2023005854
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-01-18
- Publication Date
- 2025-04-28
- Estimated Expiration
- 2043-01-18
AI Technical Summary
Existing loading and handling systems halt operations if incorrect loading and unloading occur, requiring manual intervention and disrupting the workflow.
A loading and handling system equipped with cameras, a neural network-based load specifying unit, and a management device that continuously monitors and corrects the placement of loads, allowing operations to continue uninterrupted.
Enables continuous loading and handling operations without halting, even if incorrect loading occurs, by quickly identifying and correcting load placement errors, thus minimizing downtime and improving efficiency.
Smart Images

Figure 0007669402000001 
Figure 0007669402000002 
Figure 0007669402000003
Abstract
Description
[Technical field]
[0001] The present invention relates to a loading system including a loading vehicle. [Background technology]
[0002] Conventionally, there is known a delivery system that manages cargo handling in cargo handling work in which cargo is transported by a forklift. For example, the delivery system disclosed in Patent Document 1 aims to reduce the occurrence of cargo handling that differs from the cargo handling instructions, and is configured so that the forklift (work vehicle) is equipped with an imaging means and performs correct cargo handling by determining whether or not the shelf registration mark placed on the shelf matches the shelf registration mark of the instructed cargo handling.
[0003] However, even with the above-mentioned loading and unloading system, incorrect loading and unloading can occur. When an incorrect loading and unloading occurs, even if the forklift heads towards the shelf to be loaded, another load may be placed on the shelf, and the load may not be placed there, or there may be no load on the shelf, and the load may not be picked up. In this case, the commanded loading and unloading must be stopped, and the next loading and unloading must be performed according to the next loading and unloading command. [Prior art documents] [Patent documents]
[0004] [Patent Document 1] JP 2010-222121 A Summary of the Invention [Problem to be solved by the invention]
[0005] Therefore, an object of the present invention is to provide a loading and unloading system that allows loading and unloading to continue without being stopped even if an incorrect loading and unloading operation is performed. [Means for solving the problem]
[0006] In order to solve the above problems, the loading and unloading system of the present invention is a loading and unloading system that manages the loading and unloading of cargo onto a shelf by a loading vehicle, the shelf having a plurality of storage spaces, and the loading and unloading system is characterized in comprising: a first memory unit that stores the position of each cargo corresponding to each storage space, a camera that captures an image of the planned position of the storage space where the cargo is to be loaded and unloaded and the positions of other storage spaces adjacent to the planned position to generate a shelf image, and a cargo identification unit that identifies the position where the cargo has been incorrectly placed by referring to the generated shelf image and the position of each cargo based on the loading and unloading management.
[0007] The cargo handling system preferably further includes a second storage unit and a writing unit that writes cargo handling information related to other storage spaces adjacent to the planned position in the second storage unit before the camera captures the image. The cargo identification unit refers to the generated shelf image and the cargo handling information stored in the second storage unit to identify the position where the cargo was erroneously placed.
[0008] The cargo handling system preferably further includes a replacement unit which, when the cargo identification unit identifies an incorrectly placed position of the cargo, replaces the incorrectly placed position with a correct position in the cargo handling information.
[0009] The cargo handling system preferably further includes an unmanned aerial vehicle, and the camera is mounted on the unmanned aerial vehicle.
[0010] In the above-mentioned cargo handling system, the unmanned aerial vehicle is preferably configured to head to the shelf before the cargo handling vehicle arrives at the shelf.
[0011] In the above-mentioned cargo handling system, the cargo handling vehicle is preferably configured to enable takeoff and landing of the unmanned aerial vehicle.
[0012] In the cargo handling system, for example, a camera is provided in the cargo handling vehicle.
[0013] In the cargo handling system, for example, a camera is disposed in equipment within a facility where shelves are arranged.
[0014] In the above-mentioned cargo handling system, for example, a camera is disposed on the ceiling of a facility in which shelves are arranged.
[0015] In the above-mentioned cargo handling system, for example, the cargo identification unit has a neural network, and the neural network is configured to learn images of the storage space in which cargo is placed as training data, and when an image of the storage space is input, to output whether or not cargo is placed in the storage space.
[0016] In the above-mentioned cargo handling system, the cargo handling vehicle preferably has an alarm device, which notifies an operator of the position where the cargo has been erroneously placed. Effect of the Invention
[0017] The loading and unloading system according to the present invention allows loading and unloading to be continued without being stopped even if an incorrect loading and unloading operation is performed. [Brief description of the drawings]
[0018] [Figure 1] 1 is a side view showing a loading and unloading system according to an embodiment of the present invention. [Diagram 2] FIG. [Diagram 3] FIG. 2 is a functional block diagram of the loading and unloading system. [Figure 4] FIG. 2 is a front view showing the shelf and the unmanned air vehicle. [Diagram 5] FIG. 4 is a flow chart showing a series of operations of the cargo handling system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0019] Hereinafter, an embodiment of a cargo handling system of the present invention will be described with reference to the accompanying drawings.
[0020] Fig. 1 and Fig. 2 show a cargo handling system S according to this embodiment, with Fig. 1 being a side view and Fig. 2 being a plan view. Fig. 3 is a functional block diagram of the cargo handling system S. As shown in Fig. 1 and Fig. 2, the cargo handling system S includes a shelf R, a cargo handling vehicle 1, an unmanned aerial vehicle 2, and a management device 3. First, each component of the cargo handling system S will be described.
[0021] <Shelf> FIG. 4 is a front view showing a shelf R and an unmanned aerial vehicle 2 of the present invention. As shown in FIG. 4, the shelf R has a plurality of storage spaces SS. The plurality of storage spaces SS are configured to store cargo W related to loading and unloading. In this embodiment, the cargo W is stored in the storage space SS together with a pallet. FIG. 4 shows a planned position SP where the cargo W will be loaded and unloaded, and adjacent positions AP. The adjacent positions AP are positions adjacent to the planned position SP, and in the case of the planned position SP in FIG. 4, there are eight adjacent positions AP.
[0022] <Loading vehicle> As shown in Fig. 1 and Fig. 2, the cargo handling vehicle 1 is configured as a manned forklift. As shown in Fig. 1 to Fig. 3, the cargo handling vehicle 1 is equipped with wheels, a vehicle body 10, left and right masts, left and right forks 11, a head guard 12, and an alarm device 13. The cargo handling vehicle 1 is configured to be able to communicate with a management device 3.
[0023] The left and right masts are disposed in front of the vehicle body 10, and the left and right forks 11 are configured to rise and fall along the masts. The loading vehicle 1 has the forks 11 raised and lowered by the operation of the operator O, whereby the forks 11 scoop up a load W, transport the scooped up load W to the shelf R, and unload it into the storage space SS.
[0024] The head guard 12 is positioned above the vehicle body 10 and extends horizontally, allowing the unmanned aerial vehicle 2 to take off and land.
[0025] The notification device 13 is composed of a monitor or a speaker, or both. When the notification device 13 receives a loading command, it notifies the loading command to the operator O by voice, image, or both. The operator O drives the loading vehicle 1 to the loading position in accordance with the loading command.
[0026] <Unmanned Aerial Vehicle> The unmanned aerial vehicle 2 includes a main body, a propeller, and, as shown in Fig. 3, a flight control unit 20 and a camera 21. The unmanned aerial vehicle 2 is configured to be able to communicate with the management device 3.
[0027] The flight control unit 20 controls the rotation of the propellers to hover the unmanned aerial vehicle 2, and causes the unmanned aerial vehicle 2 to fly in accordance with a loading command from the management device 3. When the unmanned aerial vehicle 2 receives the loading command, it takes off from the loading vehicle 1 and arrives at the planned position SP before the loading vehicle 1.
[0028] The camera 21 is disposed on the main body of the unmanned aerial vehicle 2. The camera 21 captures images of the planned position SP and an adjacent position AP adjacent to the planned position SP in advance to generate a shelf image. When the unmanned aerial vehicle 2 arrives at the planned position SP, it flies in front of the planned position SP and the adjacent position AP as shown by the arrow in FIG. 4, so that the camera 21 can capture images of the planned position SP and the adjacent position AP from the front. The generated shelf image is transmitted to the management device 3.
[0029] <Management device> The management device 3 includes a computer and is configured to be able to communicate with the loading vehicle 1 and the unmanned aerial vehicle 2. The computer may be, for example, a server computer, and is configured with a storage device, a calculation device, and a memory.
[0030] As shown in FIG. 3, the management device 3 includes a first memory unit 30, a cargo handling management unit 31, a writing unit 32, a second memory unit 33, a replacement unit 34, a cargo identification unit 35, and a notification unit 36.
[0031] The first storage unit 30 stores a loading / unloading schedule and the position of each load W corresponding to each storage space SS based on loading / unloading management. The loading / unloading schedule is a schedule related to loading / unloading, and the position of each load W based on loading / unloading management is the position of each storage space SS corresponding to each load W to be unloaded, and the position of each storage space SS in which each load W has already been loaded and placed.
[0032] The cargo handling management unit 31 transmits a cargo handling command to the notification device 13 of the cargo handling vehicle 1 and the unmanned aerial vehicle 2 based on the cargo handling schedule stored in the first storage unit 30. The cargo handling command includes the planned position SP.
[0033] When a cargo handling command is transmitted by the cargo handling management unit 31, the writing unit 32 writes in advance in the second storage unit 33 cargo handling information, such as whether or not a cargo W is placed at the adjacent position AP.
[0034] The article identification unit 35 has a neural network. The neural network is trained on an image of the storage space SS in which the article W is placed as training data, and is configured to output whether or not the article W is placed in the storage space SS when the image of the storage space SS is input.
[0035] When the load identification unit 35 receives a shelf image from the unmanned aerial vehicle 2, it inputs the shelf image into a neural network to determine whether load W is placed in the planned position SP and the adjacent position AP, a total of nine storage spaces SS. The load identification unit 35 then identifies the position where the load W is erroneously placed by referring to the result of the determination and the position of each load W based on the cargo handling management stored in the second storage unit 33. A load W is erroneously placed refers to a state where a load W is not placed at the planned position SP when it should be, or a different load W is placed at the planned position SP when it should not be.
[0036] For example, as shown in Fig. 4, when the load W is not placed at the planned position SP where the load W should be placed, and the load W is placed at an adjacent position AP where the load W should not be placed, the load identification unit 35 determines that the load W is mistakenly placed at the adjacent position AP, and identifies the position where the load W is mistakenly placed (in this example, the adjacent position AP directly above the planned position SP). This identification is based on an analysis that when the load W is placed at a wrong position, it is often placed at an adjacent position.
[0037] In addition, since the cargo handling information of the adjacent position AP is written in advance in the second storage unit 33, the cargo identification unit 35 does not need to search for the cargo handling information of the adjacent position AP from the first storage unit 30 in which all cargo handling information is stored in order to refer to the cargo handling information of the adjacent position AP. Therefore, the cargo identification unit 35 can quickly determine that the cargo W is placed in an incorrect position and quickly identify the correct position of the cargo W by quickly referring to the cargo handling information of the adjacent position AP.
[0038] When the cargo identification unit 35 identifies a position where the cargo W has been erroneously placed, the replacement unit 34 replaces the position in the cargo handling schedule with the scheduled position SP. That is, the cargo identification unit 35 corrects the position where the cargo W is actually placed in the cargo handling schedule to the position where the cargo W is actually placed. Then, the replacement unit 34 transmits the corrected cargo handling information to the cargo handling management unit 31 and stores it in the first storage unit 30.
[0039] The cargo handling management unit 31 transmits the corrected cargo handling information to the notification unit 36, and the notification unit 36 transmits the corrected cargo handling information to the cargo handling vehicle 1. The notification device 13 notifies the operator O of the corrected cargo handling information.
[0040] This allows the operator O to recognize the correct position of the load W before it arrives at the scheduled position SP, even if the load W is stored in the wrong storage space SS. As a result, the operator O can continue the loading and unloading operation without stopping the instructed loading and unloading operation and without wasting time correcting the loading and unloading schedule.
[0041] <Loading system operation> Next, a series of operations of the cargo handling system S will be explained again with reference to the flow chart of FIG.
[0042] (1) First, the cargo handling system S transmits a cargo handling command to the unmanned aerial vehicle 2 and the cargo handling vehicle 1 via the cargo handling management unit 31 (S1 in FIG. 5).
[0043] (2) Having received the loading / unloading command, the unmanned aerial vehicle 2 and the loading vehicle 1 move to the scheduled position SP (S2 in FIG. 5). As shown in FIG. 2, the unmanned aerial vehicle 2 can fly above the shelf R, so it travels a shorter distance than the loading vehicle 1, and its movement speed is faster than that of the loading vehicle 1, so it arrives at the scheduled position SP earlier than the loading vehicle 1.
[0044] (3) The cargo handling system S writes the cargo handling information of the adjacent position AP in the second storage unit 33 by the writing unit 32 (S3 in FIG. 5).
[0045] (4) Next, when the unmanned aerial vehicle 2 arrives at the planned position SP, the loading and unloading system S captures images of the planned position SP and the adjacent position AP using the camera 21 to generate a shelf image (S4 in FIG. 5).
[0046] (5) Next, the cargo handling system S determines whether or not a cargo W is placed in each storage space SS based on the shelf image using the cargo identification unit 35 (S5 in FIG. 5).
[0047] (6) Next, if the cargo W is not placed in the correct storage space SS (No in S6 of FIG. 5), the cargo handling system S determines, via the cargo identification unit 35, that the cargo W is placed in an incorrect location and identifies the correct location of the cargo W (S7 of FIG. 5).
[0048] (7) Next, the cargo handling system S corrects the cargo handling information based on the position of the incorrectly placed cargo W by the replacement unit 34 (S8 in FIG. 5).
[0049] (8) Next, the cargo handling system S transmits the corrected cargo handling information to the cargo handling vehicle 1 by the notification unit 36, and notifies the operator O of the corrected position by the notification device 13 (S9 in FIG. 5). This allows the operator O to recognize the correct cargo handling position, so that even if the cargo handling is performed incorrectly, the cargo handling can be continued without stopping the cargo handling.
[0050] Although one embodiment of the cargo handling system of the present invention has been described above, the cargo handling system of the present invention is not limited to the above embodiment. For example, the cargo handling system of the present invention may be embodied in the following modified examples.
[0051] <Modification> For example, the camera 21 may be disposed in equipment within the facility where the loading vehicle 1 or the shelf R is disposed. In this case, the loading system S does not need to include the unmanned aerial vehicle 2. When the camera 21 is disposed in the loading vehicle 1, the camera 21 captures images of the planned position SP and the adjacent position AP automatically or by operation of the operator O when the loading vehicle 1 arrives at the planned position SP. Also, when the camera 21 is disposed in equipment within the facility, for example, on the ceiling, the camera 21 may be configured to automatically capture images of the planned position SP and the adjacent position AP when a loading command is transmitted by the loading management unit 31.
[0052] The loading and unloading system S may be configured, for example, so that a label bearing an identifier for identifying the load W is affixed to the load W and the identifier can be recognized from a shelf image, and the load identification unit 35 may be configured to identify the location where the load W has been mistakenly placed by recognizing whether or not a specific load W corresponding to each storage space SS is placed using the identifier.
[0053] The unmanned aerial vehicle 2 may be configured to be able to take off and land, for example, from a dedicated takeoff and landing site provided within a facility. In this case, the unmanned aerial vehicle 2 may head from the takeoff and landing site to the scheduled position SP based on a loading command from the loading system S.
[0054] The cargo handling system S may not include, for example, the replacement unit 34. In this case, the cargo handling system S is configured to notify the operator O of the position where the cargo W was erroneously placed and the correct position. This allows the operator O to continue the cargo handling work by following this notification and repositioning the cargo W that was placed at the scheduled position SP to the correct position, or by picking up the cargo W that was not at the scheduled position SP from another position.
[0055] The cargo handling system S does not necessarily have to include the second storage unit 33 and the writing unit 32. In this case, the cargo identification unit 35 refers to the position of each cargo W stored in the first storage unit 30 and identifies the position where the cargo W was erroneously placed.
[0056] The cargo handling vehicle 1 may be an unmanned guided vehicle. In this case, the cargo handling vehicle 1 performs cargo handling in accordance with the cargo handling command from the cargo handling management unit 31 and the cargo handling information corrected by the replacement unit 34. [Explanation of symbols]
[0057] S Cargo Handling System R shelf SS Storage space 1. Loading vehicles 10. Body 11. Fork 12 Head Guard 13. Notification device 2 Unmanned Aerial Vehicles 20 Flight Control Unit 21 Camera 3 Management device 30 1st memory section 31 Cargo Handling Management Department 32 Writing section 33 2nd memory section 34 Replacement Section 35 Cargo identification department 36 Notification Department SP planned position AP adjacent position W load O Operator
Claims
1. A loading and unloading system for managing loading and unloading of cargo onto a shelf by a loading vehicle, comprising: The shelf has a plurality of storage spaces, The loading and unloading system includes: A first storage unit that stores the positions of the items corresponding to the storage spaces; a camera that captures an image of a planned position of the storage space where the load is to be loaded and unloaded and the positions of other storage spaces adjacent to the planned position to generate a shelf image; a load identification unit that identifies a position where the load has been erroneously placed by referring to the generated shelf image and the position of each load based on load management; A second storage unit; a writing unit that writes cargo handling information relating to the other storage space adjacent to the scheduled position in advance to the second storage unit before the camera captures an image, A loading and unloading system, wherein the load identification unit identifies the location where the load was mistakenly placed by referring to the generated shelf image and the loading and unloading information stored in the second memory unit.
2. 2. The cargo handling system according to claim 1, further comprising a replacement unit that, when the cargo identification unit identifies an incorrectly placed position of the cargo, replaces the incorrectly placed position with a correct position in the cargo handling information.
3. Further comprising an unmanned aerial vehicle, The camera is provided on the unmanned aerial vehicle, The loading and unloading system according to claim 1 , wherein the loading and unloading vehicle is configured so that the unmanned aerial vehicle can take off and land on the loading and unloading vehicle.
4. The loading and unloading system according to claim 1 , wherein the camera is provided on the loading and unloading vehicle.
5. The cargo handling system according to claim 1 , wherein the camera is disposed in a facility in which the shelves are disposed.
6. The cargo handling system according to claim 5 , wherein the camera is disposed on a ceiling within a facility in which the shelves are disposed.
7. The package identification unit has a neural network, 2. The loading and unloading system according to claim 1, wherein the neural network is configured to learn an image of the storage space in which the load is placed as training data, and when an image of the storage space is input, to output whether or not the load is placed in the storage space.
8. A loading and unloading system for managing loading and unloading of cargo onto a shelf by a loading vehicle, comprising: The shelf has a plurality of storage spaces, The loading and unloading system includes: A first storage unit that stores the positions of the items corresponding to the storage spaces; a camera that captures an image of a planned position of the storage space where the load is to be loaded and unloaded and the positions of other storage spaces adjacent to the planned position to generate a shelf image; a load identification unit that identifies a position where the load has been erroneously placed by referring to the generated shelf image and the position of each load based on load management; An unmanned aerial vehicle, The camera is provided on the unmanned aerial vehicle, A loading and unloading system, wherein the unmanned aerial vehicle is configured to head to the shelf before the loading and unloading vehicle arrives at the shelf.
9. A loading / unloading system for managing loading and unloading of cargo onto a shelf by a loading vehicle, comprising: The shelf has a plurality of storage spaces, The loading and unloading system includes: A first storage unit that stores the positions of the items corresponding to the storage spaces; a camera that captures an image of a planned position of the storage space where the load is to be loaded and unloaded and the positions of other storage spaces adjacent to the planned position to generate a shelf image; a load identification unit that identifies a position where the load has been erroneously placed by referring to the generated shelf image and the position of each load based on load management, The loading vehicle has an alarm device, The warning device informs an operator of the position where the load has been erroneously placed.
Citation Information
Patent Citations
Method for goods input and output management in automatic warehouse
JP2005138957A
Working assistant system for fork lift
JP2007320734A
Delivery system
JP2010222121A
Forklift work assisting system
JP2017036102A
Autonomous robotic navigation in storage site
WO2022010681A1