Goods verification control device and method

Abstract

The invention provides a storage system which can more quickly and effectively verify which goods are grabbed by an automatic system, and meanwhile provides an extensible solution. And the goods which are grabbed are verified by measuring the dynamic characteristics of the goods. More specifically, the present invention provides a storage system comprising a control unit, the control unit operating with an automatic system comprising a sensor unit, the automatic system being arranged to grab goods from a first container, and the control unit comprising: a verification unit, the verification unit is configured to receive identification information of goods expected to appear in the first container; receiving information from the sensor unit, the information prompting a dynamic characteristic caused by movement of the grabbed goods relative to the sensor unit; on the basis of the received identification information and the received dynamic characteristic information, whether the grabbed goods are matched with goods expected to appear in the first container or not is verified; and a command unit configured to command the automatic system to execute a preset operation in response to the verification unit failing to verify the grabbed goods, so that the verification unit verifies the grabbed goods based on further dynamic characteristic information corresponding to the preset operation.

Description

[0001] This application is a divisional application of the Chinese invention patent application with application number 201980036228.7, application date May 29, 2019, and name “Control device and method for goods verification” submitted by the same applicant.

[0002] This application claims priority to UK Patent Application No. 1809020.9 filed on 1 June 2018, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present invention generally relates to the field of automatic systems, and more particularly to a control device and method for controlling an automatic system. Technical Background

[0004] In the field of picking systems, automated systems are generally used to selectively grab items from a first container and transport them to a second container. For example, Figure 1 An automated system 200 is shown including an end effector 201, such as a suction cup end effector. The automated system is configured to transfer items from a first container 300 to a second container 400, such as a cube item 301, a first cylindrical item 302 and / or a second cylindrical item 303. Generally, the picking system has information about the contents of the first container 300 from other automated processes. However, in order to ensure that the wrong item is not transferred to the second container 400 due to defective information about the first container 300, the picking system needs to verify which item has been picked before placing it in the second container 400. For example, if a cube item 301 is mistakenly placed in the first container 300, and the second container 400 is required to contain only cylindrical items 302 and 303, but not the cube item 301, then the picking system must verify which item has been picked.

[0005] In existing systems, the photographing unit 500 is generally used to verify which item has been picked. In one example, the photographing unit 500 is a barcode reader configured to read a barcode placed on each item. The identity of the picked item determined by the barcode can be compared with the expected identity of the item in the first container 300. Alternatively, the photographing unit 500 can be an image-based object classifier configured to verify the item 301, 302, or 303 based on computer vision technology that takes a picture of the item and thereby identifies the item.

[0006] However, in existing systems, there is a significant time delay between the automatic system 200 grabbing the goods, verifying the identity of the goods, and allowing the goods to be placed in the second container 400. For example, the automatic system may need to convey the goods to the front of the barcode scanner, which may require the automatic system to pause in front of the barcode scanner to enable the barcode of the goods to be scanned. Sometimes the automatic system 200 covers the barcode, hindering the successful scanning of the barcode of the goods and / or requiring the automatic system 200 to spend extra time maneuvering the goods to a direction that is easier to scan the barcode. Similarly, the automatic system 200 may need to pause when taking a picture of the goods to perform image-based object classification on the goods. In addition, existing image-based object classification can only handle a few thousand categories of goods, so it is unclear whether such image classification can be scaled up to handle a range of industrial products that may be one or more orders of magnitude larger.

[0007] Therefore, there is a need to verify which item has been picked up by the automated system faster and more efficiently, while also providing a scalable solution. Summary of the invention

[0008] In view of the problems in the known picking systems, the present invention aims to provide a device and method for reliably verifying the goods picked up by the automatic system.

[0009] In general, the present invention verifies the items that have been grabbed by measuring the dynamic characteristics of the items. In addition, measuring the dynamic characteristics of the items can also verify the number of items currently held by the automatic system. Similarly, measuring the dynamic characteristics of the items can also determine whether the automatic system has picked up a whole item, rather than a part of an item.

[0010] The present invention provides a control unit for operating with an automated system, the automated system including a sensor unit and configured to grab a product from a container. The control unit includes a verification unit configured to receive identification information of the product expected to be present in the container, receive information indicating the dynamic characteristics of the product grabbed by the automated system from the sensor unit, and verify that the product grabbed by the automated system matches the product expected to be present in the container based on the received identification information and the received dynamic characteristics information.

[0011] The present invention also provides a system, comprising an automatic system, the automatic system comprising a sensor unit and configured to grab goods; and a control unit as described above.

[0012] The present invention also provides a storage system, the storage system includes a first group of parallel tracks or rails extending in the X direction and a second group of parallel tracks or rails extending in the Y direction, the second group of parallel tracks or rails cross-cutting the first group in a substantially horizontal plane to form a grid pattern including a plurality of grid spaces, a plurality of container stacks are located below the tracks, and are arranged so that each stack is located within the coverage area of ​​a single grid space, the storage system also includes at least one transport device, at least one transport device is arranged to selectively move laterally in the X and Y directions above the stack and on the tracks, and is arranged to transport containers, the storage system also includes a picking station arranged to receive containers transported by at least one transport device, and the storage system also includes a system as described above, wherein the system is arranged to grab goods from the container.

[0013] The present invention also provides a method for use with an automated system, the automated system including a sensor unit and configured to grab an item from a container. The method includes the following steps: receiving identification information of an item expected to be present in the container, receiving information from the sensor unit indicating the dynamic characteristics of the item grabbed by the automated system, and verifying that the item grabbed by the automated system matches the item expected to be present in the container based on the received identification information and the received dynamic characteristics information. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Embodiments of the present invention will now be described with reference to the accompanying drawings, which are merely examples, in which like reference numerals denote identical or corresponding parts. In the accompanying drawings: Figure 1 is a schematic diagram of a picking system according to a known system; Figure 2 is a schematic diagram of a control device according to an embodiment of the present invention; Figure 3 for a graph showing a force detected by a sensor unit over time when an automated system picks up a first item and a second item, wherein the first item and the second item are different; Figure 4 A graph showing a torque detected by a sensor unit over time when an automated system picks up a first item and a second item, wherein the first item and the second item are different; Figure 5 for a graph showing a force detected by a sensor unit over time when an automated system picks up a first item and a second item, wherein the first item and the second item are similar; Figure 6 A graph showing a torque detected by a sensor unit over time when an automated system picks up a first item and a second item, wherein the first item and the second item are similar; Figure 7a and 7b is a schematic diagram showing an automatic system performing a preset operation and grabbing goods at the same time; Figure 8 A schematic diagram showing an automated system picking up damaged goods; Fig. 9 A flowchart of a process executed by a control unit according to a first embodiment of the present invention; FIG10 is a schematic diagram of a framework structure according to a known system; FIG11 is a schematic top view showing a stack of boxes arranged in the frame structure of FIG10; 12(a) and 12(b) are schematic perspective views of a load handling device storing boxes, and FIG. 12(c) is a front schematic perspective view of the load handling device lifting a box; FIG. 13 is a system schematic diagram of the system showing the load handling equipment operating on the frame structure. DETAILED DESCRIPTION First embodiment

[0015] Figure 2 A system for grabbing goods from a container 300 and a control unit 100 according to a first embodiment of the present invention are shown. In the system, the container 300 is arranged at the highest level and is arranged to store at least one item. The automatic system 200 is arranged to grab goods from the container 300 and is also arranged to record information about the goods grabbed by the automatic system 200. The control unit 100 according to the first embodiment of the present invention is arranged to receive the information recorded by the automatic system 200 and is also arranged to verify that the dynamic characteristics of the goods picked by the automatic system 200 match the dynamic characteristics required for the goods expected to appear in the container 300.

[0016] In more detail, the illustrated container 300 contains three items, at least one of which needs to be removed from the container 300 by the automated system 200. In one example, the container 300 may be expected to contain only one type of item, wherein the automated system 200 removes at least one of the one type of items from the container 300. In this way, the picking system knows the type of item expected to be present in the container 300. However, the container 300 may include other items that are different types from the other items in the container 300. For example, other types of items may be mistakenly placed in the container 300, or the container 300 may be configured to contain more than one type of item. It should be understood that the container 300 may contain any items or no items at all. The container generally includes a lower surface, i.e., a bottom, for supporting the items and walls, such as to prevent the items from falling. However, it should be understood that the container may be of any size or configuration to accommodate the items, such as including three walls, five walls, etc. Alternatively, the container 300 may have no walls and thus only include a bottom, in which configuration the bottom may be referred to as a tray.

[0017] Ideally, the goods contained in the container 300 are located below the top of the wall, but the first embodiment can also be applied to containers in which the goods extend above the top of the wall, for example, because large goods are placed in the container 300 or because a large number of goods are stacked in the container 300. Figure 2 The container 300 shown includes three items, a cubic item 301, a first cylindrical item 302 and / or a second cylindrical item 303. It should be understood that the items can be of any shape and do not necessarily have to be cubic or cylindrical.

[0018] Furthermore, the container 300 may include identification means, such as a barcode, for uniquely identifying the container 300 among other containers. In this way, the picking system may identify the container 300 among other containers. For example, in a warehouse comprising a plurality of containers, the picking system may identify each container based on the identification means. In this way, when goods are placed in each container, the picking system may record the goods placed in each container. Before the automatic system 200 is allowed to perform a picking action on the container 300, the picking system may be configured to identify the container 300, thereby recalling the goods that the container 300 is intended to contain. The control unit 100 is configured to verify whether the goods 300-302 belong to the container 300 / should be removed from the container 300 based at least on the dynamic characteristics of each item.

[0019] The automated system 200 is configured to grab items from a container 300. The automated system 200 particularly comprises an end effector 201 configured to grab items. Figure 2 , the end effector is shown to be a suction cup end effector 201, which relies on vacuum to grasp the goods. However, other end effectors besides suction cup end effectors can also be used. For example, a flat-nose pliers end effector can be used to grasp the goods from the container 300. The automatic system can thereby grasp the goods 301-303 from the container 300. The automatic system 200 also includes a sensor unit 202, which is configured to measure the dynamic characteristics of the goods 301-303. In this regard, it is envisioned that the sensor unit 202 is configured to measure any characteristics caused by the movement of the goods relative to the sensor unit 202, including but not limited to, pressure, torque and / or force applied by the goods to the sensor unit 202, acceleration caused by the goods relative to the sensor unit 202, the sound of the contents of the goods moving, and / or shear forces applied to the sensor unit 202 when the goods slide.

[0020] The automated system 200 may also be configured to move the grabbed items 301-303 from the first container 300 to a second container (not shown). Thus, the first container 300 may contain items in the warehouse inventory system, while the second container may be configured to contain those items that the customer has ordered from the warehouse inventory system. The automated system 200 may therefore be configured to selectively grab those items that the customer has ordered from the first container 300 and then place them in the second container for transport to the customer. Thus, the automated system moves the items from the first container 300 to the second container. Accordingly, in order to determine the correctness of the customer's order, it is important to grab the correct items.

[0021] The control unit 100 according to the first embodiment of the present invention receives measurement information from the sensor unit 202, the measurement information indicating the dynamic characteristics of the goods 301-303. The control unit 100 is also configured to receive information identifying the goods expected to be present in the container 300. Based on the received information, the control unit 100 is configured to verify that the dynamic characteristics of the goods 301-303 picked by the automatic system 200 are the required dynamic characteristics of the goods expected to be present in the container. It is envisaged that the dynamic characteristics are at least one of the inertial characteristics of the goods, the damping characteristics of the goods and / or the sound of the goods and their contents. Similarly, it is envisaged that the dynamic characteristics include characteristics caused by movement. For example, it is envisaged that the dynamic characteristics include, for example, shear force characteristics caused by the automatic system 200, when the automatic system 200 moves the goods, the goods slide off the end effector, and the sliding is used as a measure of the friction between the end effector and the goods. For example, when the end effector clamps the goods, the end effector can actively rub the surface of the goods with its fingers, or cause the goods to slide. It should be understood that although dynamic characteristics are involved, one or more dynamic characteristics can be measured at any time to provide information about the goods suitable for verification. For example, inertial characteristics and damping characteristics can be measured simultaneously to provide information required for goods verification.

[0022] For example, the picking system may have recorded that the container 300 contains only first cylindrical items 302, wherein each first cylindrical item 302 includes a solid item, such as a canned tomato sauce product. However, due to an error in filling the container 300 (e.g., because someone put the wrong item into the container 300), there may be a second cylindrical item 303 in the middle of the first cylindrical items 302, wherein the second cylindrical item 303 contains a canned pickle product. Accordingly, the items cannot be distinguished based solely on their shape. Conventional technology is slow and generally requires the items to pause in front of a barcode scanner.

[0023] In contrast, the first embodiment of the present invention relies on the dynamic characteristics of the goods to verify that the goods currently grasped are the expected goods. In particular, the goods verification is performed by measuring how the goods react when the goods are moved. For example, when the automatic system 200 moves the goods from the first container 300 to the second container (not shown), the goods verification can be performed while the automatic system 200 moves the goods. In this way, the speed is increased because the automatic system 200 does not need to pause in front of the barcode scanner. The necessary movement of the goods from the first container 300 to the second container is also used to verify the identity of the goods. The first embodiment of the present invention improves the efficiency of the goods transfer by eliminating the need for the automatic system 200 to pause for the goods verification of the camera unit.

[0024] In one example, by moving the goods 301-303, a force / torque is applied to the goods 301-303, which causes the goods to accelerate. The acceleration of the goods corresponds to the dynamic characteristics of the goods 301-303. In this example, the canned tomato sauce product will move differently from the canned pickle product, for example, the (same) canned tomato sauce product will exhibit different dynamic characteristics (such as different shaking characteristics) than the canned pickle product (containing larger solids in the shape of pickles). The shaking will thereby apply a force / torque to the sensor unit 202, and the sensor unit 202 is configured to measure such force / torque. Based on the measured force / torque, the control unit 100 is configured to verify the goods by comparing the dynamic characteristics of the goods with the dynamic characteristics of the goods expected to be present in the container. For example, a statistical model can be used to compare using machine learning to determine the similarity between the dynamics measured by the sensor unit 202 and the expected dynamics of the goods expected to be present in the container. The statistical model is preferably configured to handle noise and drift (such as temperature drift) in the measurements obtained from the sensor unit 202. For example, the output of the sensor unit 202 may be derived from the change in the pose of the goods relative to the end effector. In addition, when different sensor units 202 are used, the statistical model may be set to generalize multiple sensor units 202. Alternatively, the statistical model may be trained for a specific sensor unit 202.

[0025] In the above examples, the disclosed sensor unit 202 is configured to measure force and / or torque. However, measurement of other characteristics is contemplated, namely characteristics caused by the movement of the goods relative to the sensor unit 202, including but not limited to, pressure applied by the goods to the sensor unit 202, acceleration caused by the goods relative to the sensor unit 202, the sound of the contents of the goods moving, and / or shear forces applied to the sensor unit 202 when the goods slide.

[0026] Although the above example mentions a comparison of solid goods to liquid goods, it is contemplated that any type of goods may be compared. For example, if the cube-shaped goods 301 contained different solid goods such as bread, the dynamics of the bread goods and the cake goods of the first cylindrical goods 302 would be more similar than the dynamics of the water goods of the second cylindrical goods 303. However, with a properly trained statistical model, the differences between the goods can be identified and the control unit 100 can verify the picked goods. In addition, counting the number of goods transferred from one container to another can be used to determine that the correct number of items have been picked for a particular order.

[0027] Figure 2 The system shown may further include a photographing unit 500 as required. The photographing unit is configured to receive information indicating the goods. For example, the photographing unit 500 may include a barcode scanner and / or a computer vision imager. Such a photographing unit 500 may be used when training the control unit 100 to recognize the goods 301-303. In particular, as in a conventional system, during the movement of the goods from the first container 300 to the second container, its dynamic characteristics are recorded by the sensor unit 202 and transmitted to the control unit 100. However, at first, the control unit 100 does not know which goods the automatic system 200 has grabbed, and therefore cannot verify that the picked goods are the goods expected to appear in the container. Therefore, the automatic system 200 will be instructed to photograph the goods in a conventional manner by means of a barcode scanner and / or a computer vision imager. As a result, the control unit 100 can receive a prompt of the identity of the goods from the photographing unit 500. Therefore, once sufficient training information has been collected, the control unit 100 will be able to successfully verify the goods when the goods are subsequently grasped by the automated system 200, without the automated system 200 having to pass the goods through the capture unit 500.

[0028] The control unit 100 according to the first embodiment includes a verification unit 101. The control unit 100 may include a training unit 102 and / or a command unit 103 as necessary.

[0029] The verification unit 101 is configured to receive information indicating the dynamic characteristics of the goods picked by the automatic system 200 from the sensor unit 202. In addition, the verification unit 101 is configured to receive identification information of the goods expected to appear in the container 300. The verification unit 101 is also configured to verify that the goods picked by the automatic system 200 are the same as the goods expected to appear in the container 300. This operation can be performed by using a statistical model, which is configured to find the similarity between the dynamic characteristics of the goods expected to appear in the container and the goods currently picked by the automatic system 200.

[0030] The statistical model may be configured to take as input an identifier of the expected item / item category and the measured dynamics. The statistical model may receive the measurements in batches after the automated system 200 has transported the item to a certain point, or it may receive the measurements in a stream, processing the measurements as they are given. In the first case, the statistical model may be configured to give as output a probability that the measured dynamics indicate that the automated system 200 has picked up the expected item. In the second case, the statistical model may be configured to give as output a string of such probabilities, such as one probability for each incoming measurement. In the first case, the output probabilities may be compared to a preset threshold to determine whether the automated system 200 has picked up the expected item. In the second case, the stream of probabilities may be evaluated to determine whether the automated system 200 has picked up the expected item.

[0031] For any given expected item / category identifier, the statistical model may take the form of a standard machine learning model, such as an artificial neural network or a Gaussian process classifier. The statistical model may optionally include prior information about the dynamics of the accelerating item, such as acceleration-invariant dynamic properties.

[0032] The control unit 100 is thus configured to receive dynamic characteristic information from the automated system 200, verifying the item or type of item being grasped by the automated system 200. In particular, verifying that small changes to the position of the automated system 200 and the route taken by the automated system 200 from the first container to the second container are invariant.

[0033] The control unit 100 may further include a training unit 102 as needed. The training unit 102 is configured to train the statistical model based on the measured sensor information and the product identifier when the verification unit 101 cannot verify the product being grasped. More specifically, the training unit 102 is configured to receive information indicating the dynamic characteristics of the product being grasped by the automatic system 200. In addition, the training unit 102 is also configured to receive the product identifier. For example, the product identifier in the form of a barcode or an image-based object classification may be received from the photographing unit 500. Based on the received information, the training unit 102 is configured to retrain the statistical model. In this way, when the same product / product category is grasped at a future time, the verification unit 101 will be able to correctly verify the product based on the information trained by the training unit 102. For example, the training unit 102 may be configured to update the statistical model by retraining the statistical model using the received product identifier and the information indicating the dynamic characteristics of the product. In this way, new items that the automated system 200 will pick up are introduced into the statistical model so that when the automated system 200 picks up the items at a future time, the verification unit 102 will verify the items.

[0034] The trained statistical model can include additional information to increase its accuracy. For example, the nominal weight of the goods, dimensions, etc. In this way, it is possible to find the type of statistical model found, where each statistical model in a category is defined by calibration parameters. This advantageously reduces the amount of storage space required and also reduces the amount of training data required.

[0035] Alternatively, the verification unit 101 may be set to detect only the category of goods, such as all boxed tea leaves are one category and all bottled beverages are another category. Therefore, the training unit 102 does not train the statistical model for a specific inventory measurement unit "SKU" (for example, a boxed tea leaf of a specific size of a specific brand), but trains the statistical model for the SKU category. This eliminates the need to capture training information for each SKU. Instead, the size and specified weight of certain SKUs can be used to automatically classify the SKU into a specific SKU category, which will be applicable to the same associated measurement sensor information as other SKUs in the same category. In this regard, it is envisioned that a relationship will need to be established between the nominal weight / size of the SKU and the associated dynamic characteristics. SKUs can be manually clustered into various types, in which case it is sufficient to examine a small number of SKUs in a category to determine the nature of the relationship. Alternatively, the SKUs can be automatically clustered, which may require capturing training information for each SKU, or a simpler statistical model that requires less training information needs to be used in each category. An automated system 200 including force sensing means / torque sensing means (e.g., a sensor unit 202 configured to measure force and / or torque) can be used to automatically determine the nominal weight of a SKU. In particular, the automated system 200 can statically determine the weight of a SKU by measuring the force applied to the end effector due to the gravitational field when the automated system 200 has grasped and lifted the item but has not moved it. Knowing the strength of the gravitational field at the location of the automated system 200 allows the weight of the item to be calculated.

[0036] While the term "item" is used to refer to both types of items (e.g., a box of tea of ​​a particular brand and size, which will have a unique stock keeping unit (SKU) code) as well as specific items (e.g., a specific box of tea that has been picked), it is contemplated that "item" may also be used more broadly. For example, while two items may have the same SKU code, i.e., referring to the same brand and size of box of tea, they may have slight differences in weight and / or dynamic properties due to manufacturing differences and / or the distribution of the contents of the box. To this end, the above-mentioned verification method may take these factors into account when determining the verification. For example, an average weight or a lower weight bound may be used.

[0037] The control unit 100 may further include a command unit 103 as required. The command unit 103 is configured to command the automatic system 200 to move in a preset manner. For example, the command unit 103 may be configured to command the automatic system 200 to move in a preset manner before the verification unit 101 verifies the goods. In this way, a certain movement pattern of the automatic system 200 will be used to highlight the distinguishing features of the measured dynamic characteristics of a given product. Additionally or alternatively, the command unit 103 is configured to command the automatic system 200 to move in a preset manner after the verification unit 101 fails to verify the goods. For example, if a first preset movement fails to identify the goods, a second preset movement different therefrom enables the verification unit 101 to verify the goods, and the corresponding dynamic characteristic information has been trained in the statistical model for the second preset movement. Therefore, a more accurate verification of the goods can be achieved without using the shooting unit 500.

[0038] Figure 3 2 is a graph illustrating the relationship between time and force measured by the sensor unit 202. The graph shows the measured force for a first product and a second product, where the first product and the second product are different. For example, the first product may be a bread product and the second product may be a beverage product. The force measurements are taken while the automatic system 200 performs a preset movement. The plotted line for each product is substantially different, so that a relatively independent force and time characteristic is indicated for each product / product category.

[0039] Similarly, Figure 4 2 is a graph showing the relationship between time and torque measured by the sensor unit 202. Figure 4 In this study, torque measurements were performed on two different products. Figure 3 Again, the plotted lines are basically different, indicating that each item has different torque characteristics. Figure 3 and Figure 4 In the present invention, both force and torque are measured about one axis. However, it is also possible to measure force and / or torque about additional axes and to generate more independent characteristics for each product / product category based on the force and / or torque measurements about at least one axis.

[0040] Figure 5 The figure shows the relationship between time and force for two similar items. Figure 5 As shown, the plots of measured force for the first and second goods are substantially similar. Therefore, the control unit 100 including information about the force versus time plot for the first goods will be able to determine from the force versus time plot for the second goods that the first and second goods are of the same type / category.

[0041] Similarly, Figure 6 is a graph showing the relationship between time and torque for two similar goods. Figure 5Again, the two torque measurement plots are substantially similar, thus suggesting that each commodity / class of commodities has characteristics that can be associated with it.

[0042] As mentioned above, the optional command unit 103 can be configured to command the automatic system 200 to perform a preset operation. In one example, in order to collect information about the dynamic characteristics of the goods, the automatic system 200 can be commanded to rotate the end effector 201 while the end effector 201 moves from the first container to the second container. In this way, additional dynamic information can be collected beyond the simple trajectory without spending more time to reach the second container.

[0043] Figure 7a An example of a preset operation is shown, which can be performed by the automatic system 200 to move the goods 301 grasped by the end effector 201 to the second container 400 (i.e., the destination container). In this example, the automatic system 200 may have grasped the goods 301 from the first container (not shown) for transportation to the second container 400. With the control unit 100 according to the first embodiment, the automatic system 200 does not have to stop at a location to perform the barcode scanning process, which is a little advantage. Instead, the automatic system 200 can transport the goods 301 to the second container 400 without stopping in the middle. Although a simple arc movement from the first container to the second container 400 is envisioned (including the upward movement of the end effector, the lateral translation of the end effector, and the downward movement), other alternatives may also be used. For example, the measurement sensor information provided by the arc movement may not be sufficient for the verification unit 101 to accurately verify the goods / goods category. Accordingly, other preset movements are envisioned. Figure 7a The movement that can be performed before the verification unit 101 verifies the goods / goods category is shown. In this example, multiple accelerations and decelerations are performed along different axes, providing the possibility to measure more sensor information to assist in the identification of goods. Figure 7a In the preset operation shown in , the route to the second container 400 is slightly longer.

[0044] Figure 7b Another preset operation that can be commanded by the command unit 103 is shown. For example, Figure 7a After the verification unit 101 fails to identify the item / item category, the command Figure 7b Instead of relying on the capture unit 500, another operation is envisioned to collect more sensor information with more acceleration / deceleration. Therefore, Figure 7b The operations in the Figure 7a In the middle, the starting position and the ending position of the product 301 are the same. Figure 7bAfter the verification unit 101 has successfully verified the goods / goods category, the automatic system 200 can put the goods 301 into the second container 400. However, this operation may take too much time, so in some cases, it may be faster to use the shooting unit 500 to detect the identity of the goods.

[0045] Figure 8 Another application of the control unit 100 according to the first embodiment of the present invention is shown. In particular, the verification unit 101 can be configured to detect damaged / incomplete goods. For example, the automatic system 200 can grab the goods 303. Figure 8 As shown, the item 303 may be damaged and separated into a first portion 303a and a second portion 303b. Because the dynamic characteristics measured by the sensor unit 202 are different from the dynamic characteristics of the item expected to be present in the container 300, the control unit 100 can detect the damage. Therefore, the automatic system 200 can also be configured to place the item 303 in a location to discard or otherwise repair the item 303. Similarly, if the item is incomplete, such as the second portion 303b is completely separated from the first portion 303a, the control unit 100 can also be able to detect that the item is incomplete. Similar to damaged items, the verification unit 101 can be configured to detect that the dynamic characteristics of the item have changed based on information from the sensor unit 202, and the change indicates that the item is incomplete.

[0046] In addition, the control unit 100 may be configured to identify whether the goods grasped by the end effector 201 include multiple goods. For example, the goods being picked may have additional goods attached. Accordingly, when the automatic system 200 picks up the first goods, the second goods attached to the first goods will move with the first goods as the automatic system 200 moves the first goods from the first container to the second container. The attachment of the second goods changes the dynamic characteristics of the first goods, and this change can be detected by the control unit 100.

[0047] Furthermore, the control unit 100 may be configured to count the number of items that are transferred from one container to another. In this way, the number of items transferred may be compared to the number of items that are expected to be placed in a particular customer order. Therefore, the control unit 100 may be configured to cause the automated system to continue picking items until the expected number of items that will be placed in the customer order is reached. At this point, the control unit 100 may be configured to cause the automated system to stop picking more items.

[0048] Fig. 9 A process of identifying goods using dynamic characteristics received from the sensor unit 202 of the automatic system 200 according to the first embodiment is shown.

[0049] refer to Fig. 9In step S901, information about an item identifier (e.g., a barcode, product name, etc.) is received. For example, the item identifier may be received from a picking system configured to manage containers and their contents within a warehouse. The picking system may be aware of the container and its expected contents. The identity of the item in the container may thereby be transmitted to step S901.

[0050] At step S902, the method receives information indicative of dynamic properties of the item from the sensor unit 202. In this regard, it is contemplated that the sensor unit 202 is configured to measure any properties caused by the movement of the item relative to the sensor unit 202, including, but not limited to, pressure, torque, and / or force applied by the item to the sensor unit 202, acceleration caused by the item relative to the sensor unit 202, sound of movement of contents of the item, and / or shear forces applied to the sensor unit 202 when the item slides down.

[0051] In step S903, the picked goods / goods categories are verified based on the received dynamic characteristics and the information received in steps S901 and S902. In other words, the information received in step S902 is compared with a statistical model of the dynamic characteristics of the goods / goods categories expected to appear in the container. When the received information matches the expected value of the statistical model, step S903 verifies the picked goods as the goods expected to appear in the container. Although this can be applied to individual goods, it can also be applied to categories of goods, such as boxed tea for one category and bottled beverages for another category.

[0052] The time required to verify goods in this way is shorter than with traditional techniques. Modifications and Changes

[0053] Numerous modifications and variations may be made to the above-described embodiments without departing from the scope of the present invention.

[0054] Online retail businesses that sell multiple product lines, such as online grocery stores and supermarkets, require systems that can store dozens or even hundreds or thousands of different product lines. Using single-product pallets in this situation is impractical because the floor space required to accommodate all the required pallets would be very large. Additionally, for some items, you may only want to store small quantities, such as perishable goods or items that are not ordered frequently, making single-product pallets an inefficient solution.

[0055] International patent application WO 98 / 049075A (Autostore), which is incorporated herein by reference, describes a system in which multi-product stacks of containers are arranged within a frame structure.

[0056] PCT Publication WO2015 / 185628A (Ocado) describes another known storage and order fulfillment system in which a stack of boxes or containers is arranged within a frame structure. A load handling device, which is running on tracks located above the frame structure, retrieves the box or container. The load handling device lifts the box or container from the stack, and multiple load handling devices work in coordination to retrieve the box or container at the lowest position of the stack. Such a system is illustrated in Figures 10 to 13 of the accompanying drawings.

[0057] As shown in Figures 10 and 11, stackable containers, also known as boxes 10, are stacked on top of each other to form a stack 12. The stack 12 is arranged in a grid frame structure 14 in a warehouse or manufacturing environment. Figure 10 is a schematic perspective view of the frame structure 14. Figure 11 is a top view showing the stack 12 of boxes 10 arranged in the frame structure 14. Each box 10 generally holds a plurality of product items (not shown), and the product items in the box 10 may be the same, or may be different product types, depending on the application.

[0058] The frame structure 14 includes a plurality of upright members 16 supporting horizontal members 18, 20. A first set of parallel horizontal members 18 is arranged perpendicular to a second set of parallel horizontal members 20 to form a plurality of horizontal grid structures supported by the upright members 16. The members 16, 18, 20 are generally made of metal. The boxes 10 are stacked between the members 16, 18, 20 of the frame structure 14 such that the frame structure 14 prevents the stack 12 of boxes 10 from horizontal movement and guides vertical movement of the boxes 10.

[0059] The top layer of the frame structure 14 includes rails 22 arranged in a grid pattern across the top of the stack 12. Referring also to FIGS. 12 and 13, the rails 22 support a plurality of automated load handling devices 30. A first set 22a of parallel rails 22 guide movement of the load handling devices 30 over the top of the frame structure 14 in a first direction (X), and a second set 22b of parallel rails 22 arranged perpendicular to the first set 22a guide movement of the load handling devices 30 in a second direction (Y), the second direction (Y) being perpendicular to the first direction (X). In this manner, the rails 22 enable two-dimensional lateral movement of the load handling devices 30 in a horizontal XY plane, enabling the load handling devices 30 to be moved to a position above any stack 12.

[0060] A form of loading handling equipment 30 is further described in Norwegian Patent No. 317366, the contents of which are incorporated herein by reference. Figures 12(a) and 12(b) are schematic cross-sectional views of the loading handling equipment 30 of the storage box 10, and Figure 12(c) is a front schematic stereoscopic view of the loading handling equipment 30 of the lifting box 10. However, there are other forms of loading handling equipment that can be used in combination with the system described in the present invention. For example, PCT patent publication WO2015 / 019055 (Ocado) describes another form of automatic loading handling equipment, in which each automatic loading processor only covers one grid space of the frame structure, which allows the density of the loading processors to be higher, thereby enabling a system of a given size to have a higher workload. The patent publication is incorporated herein by reference.

[0061] Each load handling device 30 includes a vehicle 32 that is configured to travel in the X and Y directions on the rails 22 of the frame structure 14 over the stack 12. A first set of wheels 34 is composed of a pair of wheels 34 in front of the vehicle 32 and a pair of wheels 34 at the rear of the vehicle 32, the first set of wheels 34 being configured to engage two adjacent rails of the first set 22a of rails 22. Similarly, a second set of wheels 36 is composed of two pairs of wheels 36 on either side of the vehicle 32, the second set of wheels 36 being configured to engage two adjacent rails of the second set 22b of rails 22. Each set of wheels 34, 36 can be raised or lowered so that either the first set of wheels 34 or the second set of wheels 36 is engaged with the respective set of rails 22a, 22b at any time.

[0062] When the first set of wheels 34 is engaged with the first set of rails 22a and the second set of wheels 36 is lifted from the rails 22, the wheels 34 can be driven by a drive mechanism (not shown) disposed in the vehicle 32 to move the load handling device 30 in the X direction. In order to move the load handling device 30 in the Y direction, the first set of wheels 34 is lifted from the rails 22 and the second set of wheels 36 is lowered to engage with the second set of rails 22b. The second set of wheels 36 can then be driven using the drive mechanism to achieve movement in the Y direction.

[0063] The load handling device 30 is equipped with a lifting device. The lifting device 40 includes a clamping plate 39 suspended from the body of the load handling device 32 by four cables 38. The cables 38 are connected to a winding mechanism (not shown) arranged in the vehicle 32. The cables 38 can be reeled in or unreeled from the load handling device 32, so that the position of the clamping plate 39 relative to the vehicle 32 can be adjusted in the Z direction.

[0064] The clamping plate 39 is adapted to engage the top of the box 10. For example, the clamping plate 39 may include pins (not shown) that cooperate with corresponding holes (not shown) in a rim forming the upper surface of the box 10, and sliding clamps (not shown) that can engage the rim to clamp the box 10. The clamps are driven to engage the box 10 by a suitable drive mechanism disposed in the clamping plate 39, which is powered and controlled by a signal transmitted through the cable 38 itself or through a separate control cable (not shown).

[0065] In order to remove the box 10 from the top of the stack 12, if necessary, the loading handling device 30 is moved in the X and Y directions so that the clamping plate 39 is located above the stack 12. As shown in Figure 12c, the clamping plate 39 is then lowered vertically in the Z direction to engage with the box 10 at the top of the stack 12. The clamping plate 39 clamps the box 10 and then pulls it upward on the cable 38 with the box 10 attached to the clamping plate 39. At the top of its vertical travel, the box 10 is contained in the vehicle body 32 and placed above the high position of the track 22. In this way, the loading handling device 30 can be moved to different positions in the XY plane, carrying the box 10 during movement to transport the box 10 to another position. The length of the cable 38 is sufficient to allow the loading handling device 30 to retrieve the box from any high position (including the ground position) of the stack 12, and place the box at any high position (including the ground position) of the stack 12. The weight of the vehicle 32 may be comprised in part of the batteries used to power the drive mechanisms of the wheels 34 , 36 .

[0066] As shown in FIG. 13 , a plurality of identical loading handling devices 30 are provided so that each loading handling device 30 can be operated simultaneously, thereby increasing the system's workload. The system shown in FIG. 13 may include specific locations, also known as ports, at which boxes 10 can be moved into or out of the system. An additional conveyor system (not shown) is associated with each port so that the boxes 10 transported to the port by the loading handling device 30 can be moved to another location, such as a picking station (not shown), by the conveyor system. Similarly, the conveyor system can move boxes 10 from an external location, such as a box filling station (not shown), to the port and transport them to the stack 12 through the loading handling device 30 to replenish the system inventory.

[0067] Imagine a picking station as a workstation. In the workstation, items are removed from a container and are set to be placed in other containers or bags for transport to customers. Therefore, it is necessary to ensure that the items removed from the container are correctly identified so that the customer only receives the ordered items. The container can be identified when it enters the picking station, for example, by a barcode. Therefore, the load handling system can identify the expected contents of the container at the entrance to the picking station.

[0068] The picking station may include the automated system 200 and the control unit 100 according to the first embodiment of the present invention. In this way, the picking of goods is faster than in conventional methods, since the use of the camera unit 500 is greatly reduced. Instead, the action of moving the picked goods from the first container to the second container / bag is sufficient to verify that the picked goods match the goods expected to appear in the container, thereby reducing the picking time.

[0069] Each load handling device 30 is capable of lifting and moving one box 10 at a time. If it is necessary to retrieve a box 10b that is not on top of the stack 12 ("target box"), the overlying box 10a ("non-target box") must first be moved to make it possible to obtain the target box 10b. This is achieved in an operation hereinafter referred to as "digging".

[0070] 13, in the excavation operation, one of the load handling devices 30 sequentially lifts the non-target box 10a from the stack 12 containing the target box 10b and places it on an empty space in another stack 12. The target box 10b can then be picked up by the load handling device 30 and moved to the port for further transportation.

[0071] Each load handling device 30 is controlled by a central computer. Each individual box 10 in the system is tracked so that the appropriate box 10 can be retrieved, transported and replaced when necessary. For example, during an excavation operation, the location of each non-target box 10a is recorded so that the non-target box 10a can be tracked.

[0072] The system described with reference to Figures 10 to 13 has numerous advantages and is suitable for a wide range of storage and retrieval operations. In particular, it enables very dense product storage and provides a very economical way to store a large number of different goods in boxes 10 while being able to obtain all boxes 10 at a reasonable economic cost when picking is required.

[0073] However, this type of system still has disadvantages, which are caused by the digging operation described above. When the target box 10b is not on the top of the stack 12, a digging operation must be performed.

[0074] The foregoing description of the embodiments of the present invention has been presented for the purpose of explanation and description. The foregoing description is not intended to be exhaustive or to limit the invention to the precise form disclosed. Modifications and changes may be made without departing from the spirit and scope of the present invention.

Claims

1. A storage system, comprising: a first set of parallel tracks or rails extending in the X direction and a second set of parallel tracks or rails extending in the Y direction, the second set of parallel tracks or rails intersecting the first set of parallel tracks or rails in a substantially horizontal plane to form a grid pattern including a plurality of grid spaces; a plurality of container stacks located beneath the track, arranged so that each stack is within the footprint of a single grid space; at least one transport device, the at least one transport device being configured to selectively move laterally in the X-direction and the Y-direction over the stack on the track and configured to transport containers; a picking station configured to receive the container transported by the at least one transport device; as well as a control unit, the control unit operating together with an automatic system including a sensor unit, the automatic system being arranged to grab the goods from the first container, and the control unit comprising: A verification unit, wherein the verification unit is configured to: receiving identification information of items expected to be present in the first container; receiving information from the sensor unit indicating dynamic characteristics caused by movement of the grasped item relative to the sensor unit; and Verifying whether the grabbed item matches the item expected to be in the first container based on the received identification information and the received dynamic characteristic information; and A command unit is configured to command the automatic system to perform a preset operation in response to the verification unit failing to verify the grasped goods, so that the verification unit verifies the grasped goods based on further dynamic characteristic information corresponding to the preset operation.

2. The storage system according to claim 1, characterized in that: The verification unit is configured to verify the grabbed product before the automated system places the product in a second container.

3. The storage system according to claim 1, characterized in that: Further dynamic characteristic information corresponding to the preset operation is trained in the statistical model.

4. The storage system according to claim 1, characterized in that: The preset operation is a second preset operation of the automatic system; and The command unit is configured to command the automatic system to perform a different first preset operation so that the verification unit verifies whether the grasped goods match the goods expected to appear in the first container based on the dynamic characteristic information corresponding to the first preset operation.

5. The storage system according to claim 4, characterized in that: The first predetermined operation includes an arcuate movement of an end effector of the robotic system from the first container.

6. The storage system according to claim 5, characterized in that: The arcuate motion includes upward motion, lateral translation, and downward motion of the end effector.

7. The storage system according to claim 4, characterized in that: The first predetermined operation includes multiple accelerations and decelerations along different axes.

8. The storage system according to claim 4, characterized in that: The second preset operation includes more movements than the first preset operation.

9. The storage system according to claim 1, characterized in that: The information indicating the dynamic characteristics of the cargo includes at least one of the following items: force information, torque information, shear force information, acceleration information, sound information, pressure information, inertial characteristics of the cargo, and damping characteristics of the cargo.

10. The storage system of claim 1, comprising the automated system configured to grasp the items.

11. The storage system of claim 1, further comprising an imaging unit configured to capture identification information of the item.

12. The storage system according to claim 10, characterized in that: The sensor unit includes at least one of the following: a force sensor, a torque sensor, a shear force sensor, an acceleration sensor, a microphone, and a pressure sensor.

13. A storage system according to any preceding claim, characterised in that The at least one transport device has a footprint that occupies only a single grid space in the storage system such that a transport device occupying one grid space does not obstruct a transport device occupying or traversing an adjacent grid space in the X and Y directions.

14. A method of operating an automated system including a sensor unit in a storage system, the storage system comprising: a first set of parallel tracks or rails extending in the X direction and a second set of parallel tracks or rails extending in the Y direction, the second set of parallel tracks or rails intersecting the first set of parallel tracks or rails in a substantially horizontal plane to form a grid pattern including a plurality of grid spaces; a plurality of container stacks located beneath the track, arranged so that each stack is within the footprint of a single grid space; at least one transport device, the at least one transport device being configured to selectively move laterally in the X-direction and the Y-direction over the stack on the track and configured to transport containers; a picking station configured to receive the container transported by the at least one transport device; wherein the automatic system is configured to grab the goods from the first container, The method comprises: receiving identification information of items expected to be present in the first container; receiving information from the sensor unit indicating dynamic characteristics caused by the movement of the item grasped by the automated system; and Verifying whether the grabbed item matches the item expected to be in the first container based on the received identification information and the received dynamic characteristic information; and In response to failure to verify the grasped item, the automated system is commanded to perform a preset operation to verify the grasped item based on further dynamic characteristic information corresponding to the preset operation.

15. The method of claim 14, comprising verifying the grabbed item before placing the item into a second container.

16. The method according to claim 14 or 15, characterized in that The preset operation is a second preset operation of the automatic system, and the method comprises: The automated system is commanded to perform a different first preset operation to verify whether the grasped item matches the item expected to be present in the first container based on the dynamic characteristic information corresponding to the first preset operation.

Images