Systems and methods for assembling pallets

Abstract

A system for assembling pallets is disclosed. The system includes a conveyor having an unloading area and a loading area. The conveyor further has at least one staging area that is configured to receive a supply pallet and move the supply pallet to the main conveyor. The system has a vision system configured to determine a quantity of units on the supply pallet and at least one robot. The system further includes at least one controller configured to cause the at least one robot to position a donor pallet on the main conveyor in the loading area and move a number of units of the quantity of units corresponding to an order quantity on the supply pallet to the donor pallet.

Description

PCT Application Attorney Docket No. 19133.0406P1 SYSTEMS AND METHODS FOR ASSEMBLING PALLETSCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to, and the benefit of the filing date of U. S. Provisional Patent Application No. 63 / 771,270, filed March 13, 2025, the entirety of which is hereby incorporated by reference herein.FIELD

[0002] This disclosure relates to systems and methods for preparing pallets for shipment. In exemplary aspects, the pallets can contain units of surface covering materials (for example, floor covering materials).BACKGROUND

[0003] Suppliers of various products can ship pallets having a number of units corresponding to an order. The supplier can store supply pallets, often having a predetermined number of units for a particular type of product, when the supply pallets are considered full. The number of units of a particular order does not always match the number of units on the supply pallet. Accordingly, in conventional pallet preparation processes, one or more individuals manually move the number of units corresponding to an order to a pallet for shipment. This requires expensive labor and introduces ergonomic issues. Further, manual pallet assembly can be subject to error, such as including the incorrect quantity of units and / or including the incorrect unit type (e.g., stock keeping unit (SKU)).SUMMARY

[0004] Described herein are various systems and methods for assembling pallets for shipment. In one aspect, a system includes a conveyor having an unloading area and a loading area. The conveyor further has at least one staging area that is configured to receive a supply pallet and move the supply pallet to the main conveyor. The system has a vision system configured to determine a quantity of units on the supply pallet and at least one robot. The system further includes at least one controller configured to cause the at least one robot to position a donor pallet on the main conveyor in the loading area and move a number of units of the quantity of units corresponding to an order quantity on the supply pallet to the donor pallet.PCT Application Attorney Docket No. 19133.0406P1

[0005] Also disclosed herein is a method comprising the step of moving, to a first supply pallet having at least one unit thereon, by at least one robot, at least one additional unit from a second supply pallet. Optionally, the at least one unit and the at least one additional unit are of a same SKU and a different dye lot.

[0006] Additional advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The advantages of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims. It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.DETAILED DESCRIPTION OF THE FIGURES

[0007] These and other features of the preferred embodiments of the invention will become more apparent in the detailed description in which reference is made to the appended drawings wherein:

[0008] FIG. 1 is a schematic top plan view of an exemplary system for assembling pallets as disclosed herein.

[0009] FIG. 2 is a schematic perspective view of an exemplary system for assembling pallets as disclosed herein.

[0010] FIG. 3 is a schematic top plan view of an exemplary system for assembling pallets as disclosed herein.

[0011] FIG. 4A is a perspective view of a supply pallet with a plurality of stacks of units thereon. FIG. 4B is a perspective view of a supply pallet with a plurality of stacks of units thereon.

[0012] FIG. 5 is a block diagram of an exemplary computing system for use with the exemplary system for assembling pallets as disclosed herein.

[0013] FIG. 6 is a block diagram of a machine learning system.

[0014] FIG. 7 is a flowchart illustrating an example training method.

[0015] FIG. 8 is a schematic top plan view of an exemplary system for assembling pallets as disclosed herein.PCT Application Attorney Docket No. 19133.0406P1 DETAILED DESCRIPTION

[0016] The present disclosure can be understood more readily by reference to the following detailed description, examples, drawings, and claims, and their previous and following description. However, before the present devices, systems, and / or methods are disclosed and described, it is to be understood that this disclosure is not limited to the specific devices, systems, and / or methods disclosed unless otherwise specified, as such can, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular aspects only and is not intended to be limiting.

[0017] The following description is provided as an enabling teaching of the disclosed articles, systems, and methods in their best, currently known embodiments. To this end, those skilled in the relevant art will recognize and appreciate that many changes can be made to the various aspects of the articles, systems, and methods described herein, while still obtaining the beneficial results of the disclosure. It will also be apparent that some of the desired benefits of the present disclosure can be obtained by selecting some of the features of the present disclosure without utilizing other features. Accordingly, those who work in the art will recognize that many modifications and adaptations to the present disclosure are possible and can even be desirable in certain circumstances and are a part of the present disclosure. Thus, the following description is provided as illustrative of the principles of the present disclosure and not in limitation thereof.

[0018] As used throughout, the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, unless the context dictates otherwise, reference to “a pallet” provides disclosure of embodiments in which only a single such pallet is provided, as well as embodiments in which a plurality of such pallets are provided.

[0019] Ranges can be expressed herein as from “about” one particular value, and / or to “about” another particular value. When such a range is expressed, another aspect includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent "about,” it will be understood that the particular value forms another aspect. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint. Optionally, in some aspects, when values are approximated by use of the antecedents “about,” “substantially,” or “generally,” it is contemplated that values within upPCT Application Attorney Docket No. 19133.0406P1 to 15%, up to 10%, up to 5%, or up to 1 % (above or below) of the particularly stated value can be included within the scope of those aspects. In other aspects, when angular values are approximated by use of the antecedents “about,"’ “substantially,” or “generally,” it is contemplated that angular values within up to 15 degrees, up to 10 degrees, up to 5 degrees, or up to one degree (above or below) of the particularly stated angular value can be included within the scope of those aspects.

[0020] As used herein, the terms “optional” or “optionally” mean that the subsequently described event or circumstance may or may not occur, and that the description includes instances where said event or circumstance occurs and instances where it does not.

[0021] Disclosed herein, and with reference to FIGS. 1-3, is a system 10 for assembling pallets. In use. it is contemplated that the disclosed system 10 can limit the amount of manual labor necessary to assemble pallets of units (e.g., surface covering materials), thereby reducing or avoiding ergonomic issues and / or worker turnover challenges. It is further contemplated that the disclosed system 10 can limit or avoid errors in the assembly of pallets (e.g., errors resulting from including incorrect product types within a pallet). It is still further contemplated that the disclosed system 10 can incorporate automation within various portions of the system.

[0022] The system 10 can comprise a conveyor 20 having a loading area 22 and an unloading area 22. In some aspects, the conveyor 20 can comprise a main conveyor 26 (e.g., a powered and / or automation conveyor section) that includes the loading area 22 and the unloading area 24. The conveyor 20 can further comprise at least one staging area 28 (optionally, a plurality of staging areas) configured to receive a supply pallet 12. Each staging area of the at least one staging area 28 can comprise one or more conveyor sections that are configured to move the supply pallet 12 to the main conveyor 26. Alternatively, it is contemplated that the supply pallet 12 can be directly provided to the loading area 22 (without being first positioned in the stating area 28).

[0023] The system 10 can comprise a vision system 30 that is configured to determine a quantity of units 14 on the supply pallet 12. The vision system 30 can comprise at least one optical sensor 32 that is configured to capture optical data associated with the supply pallet 12. For example, in some aspects, the at least one optical sensor 32 of the vision system 30 can comprise a camera for determining the quantity of units on the supply pallet 12. In these aspects, the vision system 30 can perform image processing on images taken by the camera.PCT Application Attorney Docket No. 19133.0406P1 In additional aspects, the at least one optical sensor 32 of the vision system 30 can comprise an optical distance sensor, such as, for example, a laser sensor, (optionally, a LIDAR sensor) for determining the quantity of units on the supply pallet 12. For example, the vision system 30 can use time-of-flight to determine positions of the units on the pallet. In exemplary aspects, the vision system 30 can comprise an overhead optical sensor (e.g., camera) that is configured to determine a topography (e.g., physical features, such as a surface profile) of units stacked on the supply pallet 12. In additional aspects, the vision system 30 can comprise at least one sensor (optionally, a plurality of sensors (e.g., a camera and an optical distance sensor)) for determining quantities and / or positions of units on the supply pallet 12. As further disclosed herein, the vision system 30 can determine, based on the topography (e.g., surface profile) of the units stacked on the supply pallet, a quantity of the units stacked on the supply pallet. For example, the vision system 30 can comprise one or more processors that receive data from the sensor(s) and determine the quantity and locations of units based on the data from the sensor(s). In other aspects, the vision system 30 can communicate with a centralized computing device, such as computing device 1001, described further herein.

[0024] The system 10 can comprise at least one robot 40 and at least one controller 42 in operable communication with the at least one robot 40. The at least one controller 42 can be configured to cause the at least one robot 40 to position a donor pallet 16 on the main conveyor 30 in the loading area 22 and move a number of units 14 of the quantity of units corresponding to an order quantity on the supply pallet 12 to the donor pallet 16. Optionally, the system 10 can have only a single robot 40. In other aspects, system 10 can comprise a plurality of robots 40. For example, the plurality of robots 40 can comprise a first robot 40a and a second robot 40b. The first and second robots 40a, b can be configured to cooperate to move the quantity of units corresponding to the order quantity from the supply pallet to the donor pallet. In some aspects, the first and second robots 40a, b can be controlled by a single controller 42. In other aspects, each robot can have a respective controller. Optionally, in these aspects, the robots 40 can be restricted to specific areas to inhibit crashing. In additional aspects, the system 10 can comprise three or more robots 40. Each of the plurality of robots 40 can be configured to perform one or more of: moving units from the supply pallet to the donor pallet, moving donor pallets to the loading area, or moving slip sheets (e.g., a cardboard, fiber, or polymer sheets) to the donor pallet in the loading area. As should be understood, slip sheets can be positioned on the donor pallet and / or between layers of units stacked thereon, as conventionally performed when assembling pallets. More generally,PCT Application Attorney Docket No. 19133.0406P1 referring also to FIG. 5, the system 10 can be controlled by a single computing device 1001 / controller 42 or a plurality of computing devices 1001 / controllers 42 operating in coordination. The computing device(s) 1001 / controller(s) 42 can each be any suitable device such as, for example, an on-board controller or remote computing device as described further herein.

[0025] In some aspects, the robot(s) 40 can be configured to manipulate units having a length from about 24 inches to about 96 inches. For example, in some aspects, the robot(s) 40 can be configured to manipulate units having a length from about 24 inches to about 60 inches, or from about 60 inches to about 96 inches, or from about 16 inches to about 100 inches, or from about 32 inches to about 64 inches. In further aspects, the robot(s) 40 can be configured to manipulate units from about 46 inches to about 120 inches. In additional aspects, the robot(s) can be configured to manipulate units of any desirable dimensions.

[0026] In some aspects, the units can be rectangular prisms (e.g., units having rectangular top and bottom faces and a thickness extending between the rectangular faces). In other aspects, the units can be any shape, including having non-polygonal sides. For example, one or more sides of the units can include rounded edges. In additional aspects, the units can have triangular sides, oval sides, circular sides, or irregular sides.

[0027] In some aspects, the robot(s) 40 can be configured to manipulate units having a weight from about 1 pounds to about 400 pounds, or from about 5 pounds to about 200 pounds, or from about 5 pounds to 100 pounds, or from about 100 pounds to about 200 pounds, or from about 25 pounds to about 150 pounds.

[0028] In some aspects the robot can comprise a multi-axis robotic arm, such as, for example, a 4-axis robotic arm, a 5 -axis robotic arm, a 6-axis robotic arm, or a 7-axis robotic arm. The robot can further comprise an end effector, which can be or comprise, for example and without limitation, a vacuum gripper and / or two or more grippers that are movable toward and away from each other to compressively engage the units.

[0029] The single robot 40, or the plurality of robots 40 in a coordinated fashion, can assemble a shippable pallet comprising a donor pallet and the number of units of the quantity of units corresponding to an order quantity. Optionally, the shippable pallet can further comprise one or more slip sheets (e.g., a cardboard, fiber, or polymer sheet). For example, the robot(s) 40 can place a slip sheet between the donor pallet and the units. In additional aspects, the robot(s) 40 can place a slip sheet between layers of units.PCT Application Attorney Docket No. 19133.0406P1

[0030] The conveyor 20 can comprise a positioning assembly that is configured to locate the supply pallet in a particular position on the main conveyor 26. For example, the positioning assembly 50 can comprise a pair of pushers 52 that position (e.g., center) the supply pallet relative to an axis that is transverse to a longitudinal axis 27 of the main conveyor 26. Further, the pushers 52 can obtain a desired rotation alignment of the supply pallet (e.g., with side edges of the supply pallet parallel to the longitudinal axis). In this way, the supply pallet can be positioned in an optimal location for the vision system 30 as well as for the at least one robot 40 to interface with the units on the supply pallet. The pushers 52 can comprise, for example, plates having surfaces parallel to the longitudinal axis of the main conveyor. In other aspects, the pushers 52 can comprise any suitable surface that is movable to engage the supply pallet for positioning and alignment. The positioning assembly 50 can comprise, for example, hydraulic or pneumatic cylinders 54 or electric linear actuators that are coupled to the pushers 52 for positioning the supply pallet.

[0031] The conveyor 20 can further comprise a portion 56 that is configured to rotate the supply pallet (e.g., about a vertical axis) to rotationally orient the supply pallet in a particular orientation. For example, said portion 56 of the conveyor 20 can comprise a turntable.

[0032] The system 10 can further comprise a donor staging area 60 having one or more donor pallets 16 thereon. Optionally, in these aspects, the donor staging area can have pallets with different dimensions. For example, the donor staging area can have a first stack of pallets with dimensions of 40 inches by 61 inches, a second stack of pallets with dimensions of 40 inches by 48 inches, and a third stack of pallets with dimension of 40 inches by 72 inches. The system 10 can further comprise a slip sheet staging area 62 having one or more slip sheets 64 thereon. Optionally, the slip sheets, like the donor pallets can be provided in stacks of different dimensions. In one or more examples, the pallets that are handled by the system of the present disclosure may have dimensions ranging in along a first axis from 25 inches to 55 inches and along a second axis perpendicular to the first axis from 55 inches to 85 inches, wherein the axes may represent a length and / or a width. In some examples, the pallets may be square shaped and range in dimension from 25 inches to 85 inches. In some other examples, the pallets may have non-square rectangular shapes, polygonal shapes, or non-polygonal shapes (e.g., a generally rectangular pallet with rounded comers or a customshaped pallet).PCT Application Attorney Docket No. 19133.0406P1

[0033] Referring to FIG. 8, in some aspects, the donor staging area 60 can further comprise one or more pallet dispensers 66. Each pallet dispenser 66 can be configured to receive a stack of donor pallets thereon. The pallet dispenser can be configured to present a pallet from the stack of pallets to the robot(s) 40. For example, the stack of pallets can have at least a first pallet at a bottom of the stack of pallets and second pallet that is second from the bottom of the stack (immediately above the bottom pallet). The pallet dispenser can be configured to grip the second pallet to thereby support any additional pallets thereon while moving the first pallet (e.g., via a conveyor) toward the robot 40. In alternate embodiments, the pallet dispenser 66 can be configured to dispense pallets in any other order, e.g., release a top pallet while maintaining the rest of the stack or dispense from the middle and combining the stacks above and below the middle pallet that is dispensed. Furthermore, the pallets may be of different sizes and shapes and the pallet dispenser 66 may be configured to select (receive instructions and responsively select, or self-determine through own controller and responsively select) a desired pallet of a specific size and shape while maintaining / preserving the rest of the pallets in a stack formation.

[0034] In some aspects, the at least one robot 40 can be configured to move a donor pallet from the donor staging area to the loading area 22 of the conveyor 20. In further aspects, the at least one robot 40 can be configured to move a slip sheet from the slip sheet staging area to the loading area 22 of the conveyor 20. For example, the end effector can comprise a vacuum that is configured to move units can similarly be configured to grip the donor pallet and / or the slip sheet. Optionally, the first robot 40a can be configured to move the donor pallet from the donor staging area to the loading area, and the second robot 40b can be configured to move the slip sheet from the slip sheet staging area to the loading area 22 of the conveyor 20. In other aspects, the first robot 40a can be configured to move both the donor pallet from the donor staging area and the slip sheet from the slip sheet staging area to the loading area.

[0035] Referring to FIGS. 4 A and 4B, the supply pallet 12 can have one or more stacks 18 of units 14. For example, in some aspects, the stacks 18 of units 14 can be arranged in one or more rows and one or more columns. Referring also to FIG. 1, in some aspects, the vision system 30 can be configured to determine the quantity of units on the supply pallet by detecting a height of each stack 18 of at least one stack of units on the supply pallet 12. For example, the vision system 30 can comprise a LiDAR sensor positioned over the conveyor 20PCT Application Attorney Docket No. 19133.0406P1 for detecting distance to the units on the supply pallet, thereby permitting determination of the height of each stack. In additional aspects, the vision system 30 can comprise at least one sensor (e.g., laser sensor) that is configured to determine height(s) of the stack(s) from one more sides of the supply pallet. For example, the vision system 30 can comprise at least one robotic scanner configured to scan opposed sides of the supply pallet (e.g., lateral sides of the pallet parallel to the longitudinal axis of the main conveyor 26). The robotic scanner can comprise at least one sensor that is movable vertically along one or opposite sides of the supply pallet and is configured to determine positions of the units on the supply pallet.

[0036] Referring to FIGS. 1-2, the vision system 30 can be configured to inspect the supply pallet 12. For example, the vision system 30 can be configured to capture an identifier associated with each unit. In this way, the system 10 can determine if any unit on the supply pallet is improperly associated with the supply pallet 12. For example, the system 10 can detect if a unit of an incorrect SKU has been placed on the supply pallet 12. Accordingly, in some aspects, the vision system 30 can be configured to determine whether all units on the supply pallet are of a single stock keeping unit (SKU). For example, the vision system 30 can be configured to generate a condition based on whether all units on the supply pallet are of a single stock keeping unit (SKU). If the units are not of a single SKU, the vision system 30 can generate an error.

[0037] In additional aspects, the vision system 30 can be configured to detect a damaged unit (e.g., a unit comprising a box that is open or broken or a unit that is misshapen). For example, the vision system 30 can compare a shape of a unit to a shape of an undamaged unit. In some aspects, a database can store a shape of an undamaged unit for each type (e.g., SKU) of unit, and the vision system 30 can compare the shape of the unit captured by the optical sensor(s) 32 to the shape of the undamaged unit from the database. A deviation beyond a predetermined threshold between the shape of the unit captured by the optical sensor(s) 32 and the shape of the undamaged unit from the database can correspond to a damaged unit. Optionally, in these aspects, the vision system 30 can be configured to generate a condition based on whether any of the units is damaged. If any unit is damaged, the vision system 30 can generate an error. In one example, upon determining a damaged unit, the vision system 30 may communicate or transmit the error message so that the damaged unit can be moved either manually or by the at least one robot 40.PCT Application Attorney Docket No. 19133.0406P1

[0038] Optionally, the vision system 30 can comprise an inspection region 46 that is spaced from the loading and unloading areas 22, 24. In this way, the vision system 30 can perform an inspection on one supply pallet in parallel with another supply pallet being positioned at the unloading area 24.

[0039] The inspection region 46 of the vision system 30 can comprise at least one optical sensor on each side of the conveyor for capturing lateral sides (side extending between the front and rear sides) of the supply pallet. In some aspects, the vision system 30 can comprise a first arm that is configured to extend forwardly in front of the supply pallet along the longitudinal axis 27 of the main conveyor to position at least one optical sensor for capturing a front-facing side of the supply pallet. In further aspects, the vision system 30 can comprise a second arm that is configured to extend forwardly in front of the supply pallet along the longitudinal axis 27 of the main conveyor to position at least one optical sensor for capturing a rear-facing side of the supply pallet. The first and second arms can be actuatable to move the respective at least one optical sensor in position for capturing optical data of the units on the supply pallet and to move the respective at least one optical sensor away from the conveyor to permit advancement of the supply pallet and receipt of another supply pallet. In this way, the vision system can be configured to inspect the front and / or rear sides of units on the supply pallet.

[0040] In additional aspects, the inspection region 46 can be provided at the portion 56 of the conveyor 20 that is configured to rotate the supply pallet. In this way, the at least one optical sensor on each side of the conveyor can capture the lateral sides, and the portion 56 can rotate the supply pallet 90 degrees to capture the remaining two sides of the supply pallet.

[0041] In some aspects, and with reference to FIG. 4A, no unit is surrounded by other units on the supply pallet so that at least a portion of each unit can be outwardly facing to present a respective identifier to the vision system 30 for confirming that the unit is properly- associated with the supply pallet 12. In other aspects, as illustrated in FIG. 4B. at least one unit can be surrounded by other units on the pallet.

[0042] The vision system 30 can be configured to determine the quantity of units 14 on the supply pallet 12 by using a known unit vertical dimension of the units associated with the supply pallet, and calculating the quantity of units based on the known unit vertical dimension and the detected height of each stack of the at least one stack of units on thePCT Application Attorney Docket No. 19133.0406P1 supply pallet. For example, the vision system 30 can receive the known unit vertical dimension from a database having a known unit vertical dimension for each type of unit (e.g., stock keeping unit (SKU)). In other aspects, the vision system 30 can receive the known vertical dimension from a user via a user interface of the computing device 1001.Accordingly, the vision system 30 can calculate the number of units on the supply pallet by dividing the height of each stack by the know n vertical dimension of the particular unit received from the database to determine the number of units in each stack, and sum the number of units from each stack. By using a database having a known unit vertical dimension for each type of unit, the system can adaptably handle units of various dimensions. That is, the system 10 is not limited to handling units all having a universal vertical dimension.

[0043] The vision system 30 can further determine the locations of the units 14 on the supply pallet 12. For example, the vision system 30 can further be configured to determine a position of each unit on top of each stack of units on the supply pallet 12. Optionally, in these aspects, the vision system can be configured to capture a new image after movement of an entire layer (e.g., after each unit on top of each stack of units is moved to the donor pallet). In additional aspects, the vision system 30 can comprise a plurality of sensors. For example, the vision system 30 can comprise a camera, an optical distance sensor, or a combination of sensors 32 configured to determine the position of each unit on the donor pallet in the loading area 22. In this way. the controller 42 can use the information of the locations of the units on the supply pallet and on the donor for controlling movement of the robot(s) 40. Accordingly, in some aspects, the robot(s) 40 need not have vision thereon. That is, the robot(s) 40 themselves can be free of any vision system (e.g., camera or optical sensors). Thus, the controller 42 can be configured to cause the at least one robot 40 to position the donor pallet 16 on the main conveyor 26 in the loading area 22 and move the number of units of an order from the supply pallet to the donor pallet without feedback from a vision device coupled to the at least one robot (e.g., positioned on the at least one robot).

[0044] The system 10 can further comprise a scanner 44 that is configured to capture an identifier associated with the supply pallet 12. For example, the identifier can be an optical identifier, such as a barcode or a QR-code. Accordingly, in some aspects, the scanner 44 can be, for example, an optical scanner, such as a camera or a laser scanner. In additional aspects, the identifier can be an RFID tag or other suitable identifier, and the scanner 44 canPCT Application Attorney Docket No. 19133.0406P1 be configured to capture the RFID tag or other suitable identifier. In some aspects, the identifier can be located on one or more of the units 14 on the supply pallet 12. In other aspects, the identifier can be located on the underlying pallet itself. The scanner 44 can permit the system 10 to confirm that the units are associated with a particular order. For example, the identifier can be associated with additional information in a database, such as, for example, an SKU or other identity of the units on the supply pallet, a dye lot, at least one dimension of the units thereon (e.g., known unit vertical dimension associated with the units of the supply pallet), or a quantity of units on the supply pallet. The scanner 44 can be in communication with the computing device 1001. Accordingly, the computing device 1001 of the system can access, based on the identifier captured by the scanner, the additional information from the database. Further, the scanner 44 can identify, by the identifier, the units of the supply pallet and, accordingly, the known unit vertical dimension associated with the units of the supply pallet. Optionally, the scanner 44 can be integral to the vision system 30. In other aspects, the scanner 44 can be separate from the vision system 30. For example, in some aspects, the scanner 44 can be a handheld scanner provided at each staging area 28. Once an operator delivers a supply pallet 12 to a staging area, the operator can scan the identifier with the scanner 44. By receiving the identifier from the scanner 44, the computing device 1001 can determine that a particular supply pallet 12 is at a particular staging area 28.

[0045] In some aspects, each staging area 28 can further be provided with a display. The display can indicate to the operator at which staging area 28 a particular supply pallet 12 should be delivered. In additional aspects, each staging area 28 can comprise a humanmachine interface in communication with the computing device 1001 for receiving additional information from the operator.

[0046] The conveyor 20 can define a vision station 34. The vision system 30 can be configured to determine the quantity' of units on the supply pallet at the vision station 34. In some optional aspects, the vision station 34 can be spaced from the unloading area.Optionally, in these aspects, the conveyor 20 can be configured to stop the supply pallet 12 at the vision station 34 prior to delivering the supply pallet to the unloading area 24. In other aspects, the vision station 60 can be at the unloading area 24.

[0047] In some aspects, the computing device 1001 and / or controller 42 can receive an order, the order having a shipment quantity corresponding to the number of units of the shippable pallet. The order can further include the type of unit, such as an SKU or other suchPCT Application Attorney Docket No. 19133.0406P1 category of unit. The computing device 1001 can indicate, by a human-machine interface (HMI) a particular staging area 28 to receive a supply pallet having the type of unit thereon. The computing device 1001 can further receive an indication from the scanner 44 that the supply pallet having the type of unit thereon has been provided to the staging area. The computing device 1001 and / or controller 42 can cause the conveyor 20 to move the supply pallet associated with the unit type (e.g., SKU) from the staging area 28 to the vision station 60. The vision system 30 can determine the quantity of units on the shippable pallet, as well as the locations of the units thereon. Once the conveyor 20 has moved the supply pallet to the unloading station, the computing device 1001 and / or controller 42 can cause the robot(s) 40 to begin assembling the shippable pallet based on the quantity and location of units determined by the vision system.

[0048] In some situations, the most efficient way to form a shippable pallet can be to remove a number of units from the supply pallet to leave the shipment quantity remaining on the supply pallet. For example, an order can require 40 units, and the first supply pallet can have 50 units thereon. It can be more efficient, therefore, to remove 10 units from the supply pallet than to move 40 pallets to a donor pallet. Accordingly, in some aspects, the computing device 1001 / controller 42 can be configured to determine, based on the quantity of units determined by the vision system and a shipment quantity, a reverse shipment condition. In response to determining the reverse shipment condition, the controller can cause the robot(s) 40 to move, to the donor pallet, a number of units that is equal to a difference between the quantity of units determined by the vision system 30 and the shipment quantity.

[0049] In some situations, units 14 from a first supply pallet can be added to a second supply pallet. For example, each of the first and second supply pallets can have thereon less than the shipment quantity. However, the first and second supply pallets can, in combination, have at least the shipment quantity'. For example, an order can require 50 units; the first supply pallet can have 35 units thereon, and the second supply pallet can have 25 units thereon. Accordingly, by moving 15 units from the second pallet to the first pallet, the system 10 can form a pallet with the order number. Accordingly, the computing device 1001 and / or a separate controller 42 can be configured to cause the conveyor to move a first supply pallet to the loading area and a second supply pallet to the unloading area. The computing device 1001 can then cause the at least one robot to consolidate a plurality' of units by moving at least one unit from the second pallet at the unloading area to the first pallet at the loadingPCT Application Attorney Docket No. 19133.0406P1 area. Optionally, the at least one controller 42 can cause the at least one robot 40 to move all units of the second pallet at the unloading area to the first pallet at the loading area. The at least one robot 40 can further move the second pallet having no units thereon from the conveyor (e.g.. to the donor staging area or a separate spent pallet area). In other aspects, at least one unit can remain on the second pallet following consolidation. In some aspects, the computing device 1001 can be configured to determine, based on the quantity of units on the first supply pallet and the quantity of units on the second pallet, whether to move units from the first pallet to the second pallet or in the reverse direction to assemble the pallet most efficiently.

[0050] In some aspects, the at least one controller 42 / computing device 1001 can be configured to determine an order in which each supply pallet from each staging area is moved to the unloading area.

[0051] In some aspects, machine learning can be used to train the vision system 30. For example, machine learning can permit the vision system 30 to determine what is and is not a unit. In this way, the vision system 30 can be configured to identify, and to determine positions of units on the supply pallet and on the donor pallet. This can permit the robot(s) 40 to orient the end effector to pick up a unit (e.g., from a supply pallet) as well as to determine a location to place the unit (e.g., on a donor pallet). Similarly, machine learning can be used to detect damaged units and or units of an incorrect size (e.g., a unit of an incorrect SKU for a particular pallet). The vision system can also be trained to read descriptions or labels on any of the faces of the unit to determine type of units, e.g., LVT, ceramic, wall panels, etc., and determine the appropriate arrangement for said unit.

[0052] In some aspects, the machine learning system can be configured to determine a number of units in a stack based on machine learning. For example, a machine learning module can receive training data for a plurality of units, the training data associated with heights and numbers of units of a plurality of stacks of a plurality of units. Using machine learning, the machine learning module can determine a height of each unit within a predetermined tolerance (e.g. + / - 0.5 units). In this way, the machine learning module permit the vision system 30 to determine a number of units in each stack for a given unit type. In some aspects, the training data can comprise numerical data, image data, video data (e.g., video clips or video streaming), and / or alphanumeric identifiers (e.g.. SKU numbers).PCT Application Attorney Docket No. 19133.0406P1

[0053] In some aspects, the machine learning system can use a unit size and shape determining model to recognize or confirm a particular type of unit on a pallet based on size and shape. For example, the machine learning model can be trained by a plurality of training data, each comprising data associated with size and shape of a particular unit. In this way. data from the vision system 30 can be provided to the unit size and shape determining model to identify a particular unit. Similarly, in some aspects, the machine learning system can use a unit-ty pe classification model. For example, the machine learning model can be trained by a plurality of training data, each comprising data associated with size and shape of a classification of units. In some aspects, the training data can comprise numerical data, image data, video data (e.g., video clips or video streaming), and / or alphanumeric identifiers (e.g., SKU numbers).

[0054] Additional description of training and using machine learning modules is disclosed herein.

[0055] In some aspects, to assemble a pallet configured to support stacking thereon, the pallet can be assembled into a shippable arrangement in which the quantity of units are stacked to have an equal height on a first side and an opposite second side of the pallet. Having equal heights on all sides allows for staking of the pallets themselves on top of each other. That is, the units on the top layer of each stack form a flat, balanced surface to distribute weight of a pallet stacked thereon. In these aspects, the controller can be configured to cause the at least one robot to move the number of units of the quantity’ of units on the supply pallet to the donor pallet into such a shippable arrangement. Accordingly, optionally, in these aspects, the controller(s) 42 can be configured to generate a stacking arrangement based on a number of units and dimensions of the units. Similarly, in some aspects, the controller can be configured to leave the supply pallet in a shippable arrangement.

[0056] In some aspects, the system 10 can be configured to assemble a shippable pallet having multiple SKUs thereon. In these aspects, after loading the number of units to the unloading area from a first supply pallet, the controller can be configured to cause the conveyor to move a second supply pallet having at least one unit with a stock keeping unit (SKU) that differs from an SKU of the number of units to the unloading area and cause the at least one robot to move a second number of units of the at least one unit corresponding to a second order quantity from the second supply pallet to the donor pallet.PCT Application Attorney Docket No. 19133.0406P1

[0057] The system 10 can further comprise at least one manual pallet makeup station 80. Each manual pallet makeup station can comprise a location for a human operator to interface with the units. For example, in some aspects, each manual pallet makeup station can permit the human operator to physically touch the units. In additional aspects, each manual pallet makeup station can comprise lifting tools (e.g., a scissor lift, a jack, a vacuum lifter, etc.) for handling or manipulating the units to move the units between pallets and to stack the units on the pallets.

[0058] It is contemplated that human operators can advantageously form shippable pallets in situations for which the robot(s) 40 are less effective. For example, in some exemplary, optional aspects, a human operator can more effectively handle tiles or packs of tiles (e.g., ceramic tiles), whereas the robot(s) can more effectively handle flooring materials such as luxury vinyl tile (LVT) and luxury vinyl plank (LVP). For example, in some aspects, units (e.g., packs of ceramic tiles) can be stacked in a horizontal configuration, in which an edge (e.g., a minor surface) is placed facing downw ard. This configuration can make handling of the unit by a human, or by a human operating lifting equipment, easier than by that of the robot(s) 40.

[0059] In additional aspects, the at least one manual pallet makeup station 80 can permit additional flexibility for processing volume. For example, if the robot(s) 40 are working at capacity, one or more (optionally, a plurality’ of) manual pallet makeup stations 80 can be operated to increase capacity7.

[0060] The at least one manual pallet makeup station 80 can comprise a staging area 82 that is configured to receive a supply pallet. The staging area 82 can comprise a conveyor that is configured to move the supply pallet to an operator space 84 at which the human operator can interface with the units on the supply pallet. In some aspects, the staging area 82 and the operator space 84 can be separated by a main conveyor 81. The operator space can comprise a first region 86 for receiving the supply pallet and a second region 88 for receiving the donor pallet. The human operator can move a number of units from the supply pallet to the donor pallet (or in the reverse direction) to form a shippable pallet.

[0061] Each of the first region 86 and the second region 88 can be formed from one or more conveyors. Following assembly of the shippable pallet, the one or more conveyors of the first region 86 and the second region 88 can move the shippable pallet to the mainPCT Application Attorney Docket No. 19133.0406P1 conveyor 81 for delivery to a shipping station. The one or more conveyors of the first region 86 and the second region 88 can further carry the pallet with remaining inventory away for restocking.

[0062] In still further aspects, it is contemplated that a shippable pallet can be formed by cooperation between the robot(s) 40 and a human operator. For example, the robot(s) 40 can assemble a first portion of the shippable pallet, and the human operator can assemble a second portion (in either order of the robot(s) assembling the first portion first or the human operator assembling the second portion first).

[0063] Referring to FIG. 8, following formation of the shippable pallet, the conveyor 20 can carry the shippable pallet to a shipping station 90. The system 10 can further comprise a w rapping station 70a that is configured to wrap the shippable pallet. For example, the wrapping station 70a can comprise at least one arm that is configured to hold a spool of wrapping material (e.g., polymer sheet) and move circumferentially around the shippable pallet to at least partly cover the number of units on the pallet. The conveyor 20 can carry a remainder of the supply pallet following removal of the number of units therefrom to a restocking station 92. The system 10 can further comprise a wrapping station 70b that is configured to wrap a remainder of the supply pallet following removal of the number of units therefrom. In other aspects, a single wrapping station 70 (FIG. 1) can wrap both the shippable pallets and the remainder of the supply pallets before delivering said pallets to their respective shipping stations or restocking stations.

[0064] The system 10 can comprise a conveyor network 94 for carrying pallets throughout the system (e.g., from the staging areas to the shipping station 90 or the restocking station 92). For example, the conveyor network 94 can comprise the conveyor 20 and the main conveyor 81. The conveyor network 94 can further comprise one or more additional portions. The conveyor netw ork 94 can further comprise a transitioning system between portions of the conveyor network. For example, the conveyor network 94 can comprise a pusher that is configured to push the pallet from one portion of the conveyor network to another. Said pusher can comprise, for example, at least one pneumatic or hydraulic cylinder. In exemplary' aspects, the pusher can be a conveyor pusher or push-off device as is known in the art. In other aspects, the conveyor network 94 can comprise a chain transfer 95 that is configured to move pallets transversely to (optionally, perpendicularly to) a longitudinal direction of the conveyor portion. For example, the chain transfer 95 can move palletsPCT Application Attorney Docket No. 19133.0406P1 transversely across the main conveyor 81 from the staging area 82 to the first region 86. In further aspects, the conveyor network 94 can comprise one or more pallet rotators 98 that are configured to rotate a pallet thereon (e.g., by 90 degrees). For example, a first conveyor portion 96a can intersect a second conveyor portion 96b. A pallet rotator 98 between the first conveyor portion 96a and the second conveyor portion 96b can rotate to determine an orientation of the pallet thereon. In exemplary aspects, each pallet rotator 98 can comprise a rotary conveyor section as is known in the art.

[0065] A method can comprise moving, to a first supply pallet having at least one unit thereon, by at least one robot, at least one additional unit from a second supply pallet. In some aspects, the at least one unit and the at least one additional unit can be of a same SKU and a different dye lot. For example, products such as flooring products can be formed in different dye lots, even when forming products of the same SKU. As a quality control, each dye lot can be serialized. The dye lots can be mixed in a shipment. Optionally, all units from the second supply pallet can be moved to the first supply pallet.

[0066] Systems and methods described herein can be used for handling units such as surface covering materials (e.g., flooring materials). Accordingly, the robot(s) 40 can be configured for handling the geometry and weight of surface covering materials (e.g., flooring materials). For example, surface covering materials (e.g., flooring materials), such as wooden planks, polymer planks (e.g., luxury vinyl plank), composite planks, polymer tiles (e.g., luxury vinyl tile), and the like can be provided in packs of a predetermined number of planks. Said packs can be the units that are handled by the system 10. The packs can further comprise packaging such as boxes, wrapping, or other containers. In other aspects, the units handled by the system 10 can be individual surface covering materials or individual flooring materials (e.g., planks or tiles) as disclosed herein. In additional aspects, the disclosed system 10 and methods can be used for assembling a pallet having any type of unit thereon, including, but not limited to, food, construction materials, and consumer products.Exemplary Computing Device and System

[0067] FIG. 5 shows an exemplary computing system 1000 that can be configured to control operation of various aspects of the system 10. The computing system 1000 can include a computing device 1001. In some exemplary' aspects, the computing device 1001 can be a master computing device that interfaces with various other remote computing device 1014a,b,c, such as controllers 42 (FIG. 1) of each robot 40 (FIG. 1) and one or morePCT Application Attorney Docket No. 19133.0406P1 controllers that operate the conveyor 20 (FIG. 1 ). More generally, control and processing can be centralized or distributed across multiple computing devices as desired. Although FIG. 1 illustrates a computing device 1001 in communication with a controller 42 for controlling the robot 40, it is contemplated that a single computing device can perform the function of both elements. More generally, the terms computing device and controller can be interchanged without departing from the spirit and scope of the present disclosure.

[0068] The computing device 1001 may comprise one or more processors 1003. a system memory 1012, and a bus 1013 that couples various components of the computing device 1001 including the one or more processors 1003 to the system memory 1012. In the case of multiple processors 1003, the computing device 1001 may utilize parallel computing.

[0069] The bus 1013 may comprise one or more of several possible types of bus structures, such as a memory bus, memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures.

[0070] The computing device 1001 may operate on and / or comprise a variety of computer readable media (e.g., non-transitory). Computer readable media may be any available media that is accessible by the computing device 1001 and comprises, non-transitory, volatile and / or non-volatile media, removable and non-removable media. The system memory 1012 has computer readable media in the form of volatile memory, such as random access memory (RAM), and / or non-volatile memory, such as read only memory (ROM). The system memory 1012 may store data such as unit geometry data 1007 and / or program modules such as operating system 1005 and robot control software 1006 that are accessible to and / or are operated on by the one or more processors 1003.

[0071] The computing device 1001 may also comprise other removable / non-removable, volatile / non-volatile computer storage media. A mass storage device 1004 may provide nonvolatile storage of computer code, computer readable instructions, data structures, program modules, and other data for the computing device 1001. The mass storage device 1004 may be a hard disk, a removable magnetic disk, a removable optical disk, magnetic cassettes or other magnetic storage devices, flash memory' cards, CD-ROM, digital versatile disks (DVD) or other optical storage, random access memories (RAM), read only memories (ROM), electrically erasable programmable read-only memory’ (EEPROM), and the like.PCT Application Attorney Docket No. 19133.0406P1

[0072] Any number of program modules may be stored on the mass storage device 1004. An operating system 1005 and the robot control software 1006 may be stored on the mass storage device 1004. One or more of the operating system 1005 and the robot control software 1006 (or some combination thereof) may comprise program modules and the robot control software 1006. Unit geometry data 1007 may also be stored on the mass storage device 1004. The unit geometry 1007 may be stored in any of one or more databases known in the art. The databases may be centralized or distributed across multiple locations within the network 1015.

[0073] A user may enter commands and information into the computing device 1001 via an input device (not show n). Such input devices comprise, but are not limited to, a keyboard, pointing device (e.g., a computer mouse, remote control), a microphone, a joystick, a scanner, tactile input devices such as gloves, and other body coverings, motion sensor, and the like These and other input devices may be connected to the one or more processors 1003 via a human machine interface 1002 that is coupled to the bus 1013, but may be connected by other interface and bus structures, such as a parallel port, game port, an IEEE 1394 Port (also known as a Firewire port), a serial port, network adapter 1008, and / or a universal serial bus (USB).

[0074] A display 1011 may also be connected to the bus 1013 via an interface, such as a display adapter 1009. It is contemplated that the computing device 1001 may have more than one display adapter 1009 and the computing device 1001 may have more than one display 1011. Each display 1011 may be amonitor, an LCD (Liquid Crystal Display), light emitting diode (LED) display, television, smart lens, smart glass, and / or a projector. In addition to the display 1011. other output peripheral devices may comprise components such as speakers (not shown) and a printer (not shown) which may be connected to the computing device 1001 via Input / Output Interface 1010. Any step and / or result of the methods may be output (or caused to be output) in any form to an output device. Such output may be any form of visual representation, including, but not limited to, textual, graphical, animation, audio, tactile, and the like. The display 1011 and computing device 1001 may be part of one device, or separate devices.

[0075] The computing device 1001 may operate in a networked environment using logical connections to one or more remote computing devices 1014a,b,c. A remote computing device 1014a,b,c may be a personal computer, computing station (e.g..PCT Application Attorney Docket No. 19133.0406P1 workstation), portable computer (e.g., laptop, mobile phone, tablet device), smart device (e.g., smartphone, smart watch, activity tracker, smart apparel, smart accessory ), security’ and / or monitoring device, a server, a router, a network computer, a peer device, edge device or other common network node, and so on. Logical connections between the computing device 1001 and a remote computing device 1014a, b,c may be made via a network 1015, such as a local area network (LAN) and / or a general wide area network (WAN). Such network connections may be through a network adapter 1008. A network adapter 1008 may be implemented in both wired and wireless environments. Such networking environments are conventional and commonplace in dwellings, offices, enterprise-wide computer networks, intranets, and the Internet. In further exemplary aspects, it is contemplated that the computing device 1001 can be in communication with the remote computing devices 1014a, b,c through a Cloud-based network.

[0076] Application programs and other executable program components such as the operating system 1005 are shown herein as discrete blocks, although it is recognized that such programs and components may reside at various times in different storage components of the computing device 1001. and are executed by the one or more processors 1003 of the computing device 1001. An implementation of the robot control software 1006 may be stored on or sent across some form of computer readable media. Any of the disclosed methods may be performed by processor-executable instructions embodied on computer readable media.Machine Learning

[0077] Turning now to FIG. 6, a system 1100 is shown. The system 1100 may be configured to use machine learning techniques to train, based on an analysis of one or more training data sets 1150A-1150B by a training module 1200, at least one machine learningbased classifier 1300 that is configured to classify optical data (e.g., image data or topographical data) as indicative of or not indicative of a particular attribute(s) of a unit (e.g., is a particular unit or is not a particular unit). The at least one machine learning-based classifier 1300 may comprise the machine learning module 1104B (e.g., a segmentation model and / or a structural data model).

[0078] The system 1100 may determine (e.g., access, receive, retrieve, etc.) the training data set 1150A. The training data set 1150A may comprise first data sets (e.g., sets of images)PCT Application Attorney Docket No. 19133.0406P1 associated with one or more units. The system 1100 may determine (e.g., access, receive, retrieve, etc.) the training data set 1150B. The training data set 1150B may comprise second data sets (e.g., sets of images) associated with one or more units. The first data sets and the second data sets may each contain one or more result datasets associated with one or more units, and each result dataset may be associated with one or more data attributes. The one or more data attributes may include the existence / absence of a unit, a size of a unit, a shape of a unit, a boundary7(e.g., edge) of a unit, and / or the like. Each result dataset may include a labeled list of results. The labels may comprise "‘attribute data" (corresponding to data that indicates a particular attribute) and “non-attribute data” (corresponding to data that does not indicate a particular attribute).

[0079] Unit image data sets may be randomly assigned to the training data set 1150B or to a testing data set. In some implementations, the assignment of data to a training data set or a testing data set may not be completely random. In this case, one or more criteria may be used during the assignment, such as ensuring that similar numbers of images are in each of the training and testing data sets. In general, any suitable method may be used to assign the data to the training or testing data sets, while ensuring that the distributions of sufficient quality and insufficient quality labels are somewhat similar in the training data set and the testing data set.

[0080] The training module 1200 may train the machine learning-based classifier 1300 by extracting a feature set from the training data set 1150A according to one or more feature selection techniques. The training module 1200 may further define the feature set obtained from the training data set 1150A by applying one or more feature selection techniques to the training data set 1150B that includes statistically significant features of positive examples (e.g., image data indicating a particular attribute(s) of a unit) and statistically significant features of negative examples (e.g., image data not indicating a particular attribute(s) of a corresponding unit). The feature set extracted from the training data set 1150A and / or the training dataset 1150B may comprise segmentation data and / or structural data as described herein. For example, the feature set may comprise features associated with data that are indicative of the one or more physical features described herein. The feature set may be derived from the segmentation data indicated by the image sets associated with one or more units and / or the structural data disclosed herein.PCT Application Attorney Docket No. 19133.0406P1

[0081] The training module 1200 may extract the feature set from the training data set 1150A and / or the training data set 1150B in a variety of ways. The training module 1200 may perform feature extraction multiple times, each time using a different feature-extraction technique. In an embodiment, the feature sets generated using the different techniques may- each be used to generate different machine learning-based classification models 1350. For example, the feature set with the highest quality metrics may be selected for use in training. The training module 1200 may use the feature set(s) to build one or more machine learningbased classification models 1350A-1350N that are configured to indicate whether or not new image sets contain or do not contain data indicating a particular attribute(s) of the corresponding unit.

[0082] The training data set 1150A and / or the training data set 1150B may be analyzed to determine any dependencies, associations, and / or correlations between extracted features and the sufficient quality / insufficient quality labels in the training data set 1150A and / or the training data set 1150B. The identified correlations may have the form of a list of features that are associated with labels for data indicating a particular attribute(s) of a corresponding unit and labels for data not indicating the particular attribute(s) of the unit. The features may be considered as variables in the machine learning context. The term '‘feature,” as used herein, may refer to any characteristic of an item of data that may be used to determine whether the item of data falls within one or more specific categories. By way of example, the features described herein may comprise the one or more data attributes. The one or more data attributes may include the existence / absence of a unit, a size of a unit, a shape of a unit, a boundary (e.g., edge) of a unit, a combination thereof, and / or the like.

[0083] A feature selection technique may comprise one or more feature selection rules. The one or more feature selection rules may comprise a data attribute and a data attribute occurrence rule. The data attribute occurrence rule may comprise determining which data attributes in the training data set 1150 A occur over a threshold number of times and identifying those data attributes that satisfy the threshold as candidate features. For example, any data attributes that appear greater than or equal to 8 times in the training data set 1150A may be considered as candidate features. Any data attributes appearing less than 8 times may be excluded from consideration as a feature. Any threshold amount may be used as needed.

[0084] A single feature selection rule may be applied to select features or multiple feature selection rules may be applied to select features. The feature selection rules may bePCT Application Attorney Docket No. 19133.0406P1 applied in a cascading fashion, with the feature selection rules being applied in a specific order and applied to the results of the previous rule. For example, the data attribute occurrence rule may be applied to the training data set 1150A to generate a first list of data attributes. A final list of candidate features may be analyzed according to additional feature selection techniques to determine one or more candidate groups (e.g., groups of data attributes). Any suitable computational technique may be used to identify the candidate feature groups using any feature selection technique such as filter, wrapper, and / or embedded methods. One or more candidate feature groups may be selected according to a filter method. Filter methods include, for example, Pearson’s correlation, linear discriminant analysis, analysis of variance (ANOVA), chi-square, combinations thereof, and the like. The selection of features according to filter methods are independent of any machine learning algorithms. Instead, features may be selected on the basis of scores in various statistical tests for their correlation with the outcome variable (e.g., images that indicate or do not indicate a particular attribute(s) of a corresponding unit).

[0085] As another example, one or more candidate feature groups may be selected according to a wrapper method. A wrapper method may be configured to use a subset of features and train a machine learning model using the subset of features. Based on the inferences that drawn from a previous model, features may be added and / or deleted from the subset. Wrapper methods include, for example, forw ard feature selection, backward feature elimination, recursive feature elimination, combinations thereof, and the like. In an embodiment, forward feature selection may be used to identify one or more candidate feature groups. Forward feature selection is an iterative method that begins with no features in the machine learning model. In each iteration, the feature which best improves the model is added until an addition of a new feature does not improve the performance of the machine learning model. In an embodiment, backward elimination may be used to identify one or more candidate feature groups. Backward elimination is an iterative method that begins with all features in the machine learning model. In each iteration, the least significant feature is removed until no improvement is observed on removal of features. Recursive feature elimination may be used to identify one or more candidate feature groups. Recursive feature elimination is a greedy optimization algorithm which aims to find the best performing feature subset. Recursive feature elimination repeatedly creates models and keeps aside the best or the worst performing feature at each iteration. Recursive feature elimination constructs thePCT Application Attorney Docket No. 19133.0406P1 next model with the features remaining until all the features are exhausted. Recursive feature elimination then ranks the features based on the order of their elimination.

[0086] As a further example, one or more candidate feature groups may be selected according to an embedded method. Embedded methods combine the qualities of filter and wrapper methods. Embedded methods include, for example, Least Absolute Shrinkage and Selection Operator (LASSO) and ridge regression which implement penalization functions to reduce overfitting. For example, LASSO regression performs LI regularization which adds a penalty equivalent to absolute value of the magnitude of coefficients and ridge regression performs L2 regularization which adds a penalty equivalent to square of the magnitude of coefficients.

[0087] After the training module 1200 has generated a feature set(s), the training module 1200 may generate a machine learning-based classification model 1350 based on the feature set(s). A machine learning-based classification model may refer to a complex mathematical model for data classification that is generated using machine-learning techniques. In one example, this machine learning-based classifier may include a map of support vectors that represent boundary features. By way of example, boundary features may be selected from, and / or represent the highest-ranked features in, a feature set.

[0088] The training module 1200 may use the feature sets extracted from the training data set 1150A and / or the training data set 1150B to build a machine learning-based classification model 1350A-1350N for each classification category (e.g., each attribute of a corresponding unit). In some examples, the machine learning-based classification models 1350A-1350N may be combined into a single machine learning-based classification model 1350. Similarly, the machine learning-based classifier 1300 may represent a single classifier containing a single or a plurality of machine learning-based classification models 1350 and / or multiple classifiers containing a single or a plurality of machine learning-based classification models 1350.

[0089] The extracted features (e.g., one or more data attributes) may be combined in a classification model trained using a machine learning approach such as discriminant analysis; decision tree; a nearest neighbor (NN) algorithm (e.g., k-NN models, replicator NN models, etc.); statistical algorithm (e.g., Bayesian networks, etc.); clustering algorithm (e.g., k-means, mean-shift, etc.); neural networks (e.g., reservoir networks, artificial neural networks, etc.);PCT Application Attorney Docket No. 19133.0406P1 support vector machines (SVMs); logistic regression algorithms; linear regression algorithms; Markov models or chains; principal component analysis (PCA) (e.g., for linear models); multi-layer perceptron (MLP) ANNs (e.g., for non-linear models); replicating reservoir networks (e.g., for non-linear models, typically for time series); random forest classification; a combination thereof and / or the like. The resulting machine learning-based classifier 1300 may comprise a decision rule or a mapping for each candidate data attribute to assign one or more data to a class (e.g., indicating or not indicating a particular attribute(s) of a corresponding unit).

[0090] The candidate data attributes and the machine learning-based classifier 1300 may be used to predict a label (e.g., indicating or not indicating a particular attribute(s) of a corresponding unit) for results in the testing data set (e.g., in a portion of image sets associated with the second plurality of images). In one example, the prediction for each result in the testing data set includes a confidence level that corresponds to a likelihood or a probability that the corresponding data indicates or does not indicate a particular attribute(s) of a corresponding unit. The confidence level may be a value between zero and one, and it may represent a likelihood that the corresponding data belongs to a particular class. In one example, when there are two statuses (e.g., indicating or not indicating a particular attribute(s) of a corresponding unit), the confidence level may correspond to a value p, which refers to a likelihood that a particular data belongs to the first status (e.g., indicating the particular attribute(s)). In this case, the value 1-p may refer to a likelihood that the particular data belongs to the second status (e.g., not indicating the particular attribute(s)). In general, multiple confidence levels may be provided for each element of data and for each candidate data attribute when there are more than two statuses. A top performing candidate data attribute may be determined by comparing the result obtained for each data with the known sufficient quality / insufficient quality status for each corresponding images associated with a unit in the testing data set (e.g., by comparing the result obtained for each data with the labeled image data sets of the second portion of the images associated with the second plurality of images). In general, the top performing candidate data attribute for a particular attribute(s) of the corresponding set of image data associated with a unit will have results that closely match the known indicating / not indicating statuses.

[0091] The top performing data attribute may be used to predict the indication / not indication of data of a new set of images associated with a unit. For example, a new set ofPCT Application Attorney Docket No. 19133.0406P1 image data associated with a unit may be determined / received. The new set of images associated with a unit may be provided to the machine learning-based classifier 1300 which may, based on the top performing data attribute for the particular attribute(s) of the corresponding unit, classify the data of the new set of images associated with a unit as indicating or not indicating the particular attribute(s).

[0092] As noted above, the application may provide an indication of one or more user edits made to any of the attributes indicated by the segmentation mask / overlay (or any created or deleted attributes) to the server 1104. For example, the user may edit any of the attributes indicated by the segmentation mask / overlay by dragging some of its points to desired positions via mouse movements in order to optimally delineate depictions of boundaries of the attribute(s). As another example, the user may draw or redraw parts of the segmentation mask / overlay via a mouse. Other input devices or methods of obtaining user commands may also be used. The one or more user edits may be used by the machine learning module 1104B to optimize the segmentation model and / or the structural data model. For example, the training module 1200 may extract one or more features from output image data containing one or more user edits as discussed above. The training module 1200 may use the one or more features to retrain the machine learning-based classifier 1300 and thereby continually improve results provided by the machine learning-based classifier 1300.

[0093] Turning now to FIG. 7, a flowchart illustrating an example training method 1400 is shown. The method 1400 may be used for generating the machine learning-based classifier 1300 using the training module 1200. The training module 1200 can implement supervised, unsupervised, and / or semi-supervised (e.g., reinforcement based) machine learning-based classification models 1350. The method 1400 illustrated in FIG. 7 is an example of a supervised learning method; variations of this example of training method are discussed below, however, other training methods can be analogously implemented to train unsupervised and / or semi-supervised machine learning models.

[0094] The training method 1400 may determine (e.g., access, receive, retrieve, etc.) first set of image data associated with a plurality of units (e g., including images of one or a plurality of units on a pallet) and second set of image data associated with the plurality of units (e.g., including images of one or a plurality of units on a pallet) at step 1410. The first set of image data and the second set of image data may each contain one or more result datasets associated with one or more units, and each result dataset may be associated with onePCT Application Attorney Docket No. 19133.0406P1 or more data attributes. The one or more data attributes may include the existence / absence of a unit, a size of a unit, a shape of a unit, a boundary (e.g., edge) of a unit, a combination thereof, and / or the like. Each result dataset may include a labeled list of results. The labels may comprise "‘attribute data" and “non-attribute data.”

[0095] The training method 1400 may generate, at step 1420, a training data set and a testing data set. The training data set and the testing data set may be generated by randomly assigning labeled results from the image data associated with a plurality of images of units to either the training data set or the testing data set. In some implementations, the assignment of labeled results as training or test samples may not be completely random. In an embodiment, only the labeled results for a specific unit type and / or class (e.g., unit of a particular SKU) may be used to generate the training data set and the testing data set. In an embodiment, a majority of the labeled results for the specific unit type and / or class may be used to generate the training data set. For example, 75% of the labeled results for the specific unit type and / or class may be used to generate the training data set and 25% may be used to generate the testing data set.

[0096] The training method 1400 may determine (e.g., extract, select, etc ), at step 1430, one or more features that can be used by, for example, a classifier to differentiate among different classifications (e.g., “attribute data” vs. “non-attribute data.”). The one or more features may comprise a set of one or more data attributes. The one or more data attributes may include the existence / absence of a unit, a size of a unit, a shape of a unit, a boundary (e.g., edge) of a unit, a combination thereof, and / or the like. In an embodiment, the training method 1400 may determine a set of features from the images associated with the first plurality of images of units. In another embodiment, the training method 1400 may determine a set of features from the images associated with the second plurality of images of units. In a further embodiment, a set of features may be determined from labeled results from images of a unit type and / or class different than the images of a unit type and / or class associated with the labeled results of the training data set and the testing data set. In other words, labeled results from images of the different unit type and / or class may be used for feature determination, rather than for training a machine learning model. The training data set may be used in conjunction with the labeled results from images of the different unit ty pe and / or class to determine the one or more features. The labeled results from the images of differentPCT Application Attorney Docket No. 19133.0406P1 unit type and / or class may be used to determine an initial set of features, which may be further reduced using the training data set.

[0097] The training method 1400 may train one or more machine learning models using the one or more features at step 1440. In one embodiment, the machine learning models may be trained using supervised learning. In another embodiment, other machine learning techniques may be employed, including unsupervised learning and semi-supervised. The machine learning models trained at 1440 may be selected based on different criteria depending on the problem to be solved and / or data available in the training data set. For example, machine learning classifiers can suffer from different degrees of bias. Accordingly, more than one machine learning model can be trained at 1440, and then optimized, improved, and cross-validated at step 1450.

[0098] The training method 1400 may select one or more machine learning models to build a predictive model at 1460 (e.g., the at least one machine learning-based classifier 1300). The predictive model may be evaluated using the testing data set. The predictive model may analyze the testing data set and generate classification values and / or predicted values at step 1470. Classification and / or prediction values may be evaluated at step 1480 to determine whether such values have achieved a desired accuracy level.

[0099] Performance of the predictive model described herein may be evaluated in a number of ways based on a number of true positives, false positives, true negatives, and / or false negatives classifications of data in data sets of images of units. For example, the false positives of the predictive model may refer to a number of times the predictive model incorrectly classified data as indicative of a particular attribute that in reality did not indicate the particular attribute. Conversely, the false negatives of the machine learning model(s) may refer to a number of times the predictive model classified one or more data of a data set associated with a unit as not indicating a particular attribute when, in fact, the one or more data did indicate the particular attribute. True negatives and true positives may refer to a number of times the predictive model correctly classified one or more data of image sets associated with a plurality of images of units as having sufficient indicating of a particular attribute or not indicating the particular attribute. Related to these measurements are the concepts of recall and precision. Generally, recall refers to a ratio of true positives to a sum of true positives and false negatives, which quantifies a sensitivity of the predictive model. Similarly, precision refers to a ratio of true positives to a sum of true positives and falsePCT Application Attorney Docket No. 19133.0406P1 positives. Further, the predictive model may be evaluated based on a level of mean error and a level of mean percentage error. Once a desired accuracy level of the predictive model is reached, the training phase ends and the predictive model may be output at step 1490.However, when the desired accuracy level is not reached a subsequent iteration of the method 1400 may be performed starting at step 1410 with variations such as, for example, considering a larger collection of image sets associated with a plurality of images of units.EXEMPLARY ASPECTS

[0100] In view of the described products, systems, and methods and variations thereof, herein below are described certain more particularly described aspects of the invention. These particularly recited aspects should not however be interpreted to have any limiting effect on any different claims containing different or more general teachings described herein, or that the ‘'particular” aspects are somehow limited in some way other than the inherent meanings of the language literally used therein.

[0101] Aspect 1: A system comprising:a conveyor comprising:a main conveyor comprising an unloading area and a loading area;at least one staging area configured to receive a supply pallet and move the supply pallet to the main conveyor; anda vision system configured to determine a quantity of units on the supply pallet; at least one robot; andat least one controller configured to cause the at least one robot to:position a donor pallet on the main conveyor in the loading area; and move a number of units of the quantity of units corresponding to an order quantity on the supply pallet to the donor pallet.

[0102] Aspect 2: The system of aspect 1, further comprising a donor staging area, wherein the at least one robot comprises:a first robot that is configured to move the donor pallet from the donor staging area to position the donor pallet on the main conveyor in the loading area; anda second robot configured to move the number of units of the quantity of units on the supply pallet to the donor pallet.PCT Application Attorney Docket No. 19133.0406P1

[0103] Aspect 3: The system of aspect 1 or aspect 2, further comprising a slip sheet staging area, wherein the second robot is configured to move a slip sheet from the slip sheet staging area to the donor pallet.

[0104] Aspect 4: The system of any one of the preceding aspects, wherein the vision system is configured to determine the quantity of units on the supply pallet by detecting a height of each stack of at least one stack of units on the supply pallet.

[0105] Aspect 5: The system of aspect 4, wherein the vision system is configured to determine the quantity of units on the supply pallet by using a known unit vertical dimension associated with the supply pallet, and calculating the quantity of units based on the known unit vertical dimension and the detected height of each stack of the at least one stack of units on the supply pallet.

[0106] Aspect 6: The system of any one of the preceding aspects, wherein the conveyor defines a vision station that is spaced from the unloading area, wherein the vision system is configured to determine the quantity of units on the supply pallet at the vision station.

[0107] Aspect 7: The system of aspect 6, wherein conveyor is configured to stop the supply pallet at the vision station prior to delivering the supply pallet to the unloading area.

[0108] Aspect 8: The system of any one of the preceding aspects, wherein the at least one controller is configured to:determine, based on the quantity of units determined by the vision system and a shipment quantity, a reverse shipment condition; andin response to determining the reverse shipment condition, cause the at least one robot to move a difference between the quantity of units determined by the vision system and the shipment quantity to the donor pallet.

[0109] Aspect 9: The system of any one of the preceding aspects, wherein the at least one controller is configured to cause the at least one robot to position the donor pallet on the main conveyor in the loading area and move the number of units of the quantity of units on the supply pallet to the donor pallet without feedback from a vision device coupled to the at least one robot.PCT Application Attorney Docket No. 19133.0406P1

[0110] Aspect 10: The system of any one of the preceding aspects, further comprising a wrapping station that is configured to wrap a remainder of the supply pallet following removal of the number of units therefrom.

[0111] Aspect 11: The system of any one of the preceding aspects, wherein the least one staging area comprises a plurality of staging areas, wherein the at least one controller is configured to determine an order in which each supply pallet from each staging area is moved to the unloading area.

[0112] Aspect 12: The system of any one of the preceding aspects, wherein the at least one controller is configured to:cause the conveyor to move a first supply pallet to the loading area and a second supply pallet to the unloading area;cause the at least one robot to consolidate a plurality of units by moving at least one unit from the second pallet at the unloading area to the first pallet at the loading area.

[0113] Aspect 13: The system of aspect 12. wherein the at least one controller is configured to cause the at least one robot to move all units of the second pallet at the unloading area to the first pallet at the loading area.

[0114] Aspect 14: The system of any one of the preceding aspects, wherein the at least one controller is configured to cause the at least one robot to move based on machine learning.

[0115] Aspect 15: The system of any one of the preceding aspects, wherein the at least one controller is configured to cause the at least one robot to move the number of units of the quantity of units on the supply pallet to the donor pallet into a shippable arrangement in which the quantity of units are stacked to have an equal height on a first side and an opposite second side of the pallet.

[0116] Aspect 16: The system of any one of the preceding aspects, wherein the vision system comprises at least one robotic scanner configured to scan opposed sides of the supply pallet.PCT Application Attorney Docket No. 19133.0406P1

[0117] Aspect 17: The system of any one of the preceding aspects, wherein the vision system comprises an optical scanner that is configured to capture an identifier indicative of a stock keeping unit (SKU) of the supply pallet.

[0118] Aspect 18: The system of aspect 17, further comprising the supply pallet, wherein the identifier is a barcode or a QR code.

[0119] Aspect 19: The system of any one of the preceding aspects, wherein the vision system is configured to generate a condition based on whether all units on the supply pallet are of a single stock keeping unit (SKU).

[0120] Aspect 20: The system of any one of the preceding aspects, wherein the at least one controller is configured to generate a condition based on whether all units on the supply pallet are of a single stock keeping unit (SKU).

[0121] Aspect 21: The system of any one of the preceding aspects, wherein the at least one controller is configured to:cause the conveyor to move a second supply pallet having at least one unit with a stock keeping unit (SKU) that differs from an SKU of the number of units to the unloading area; andcause the at least one robot to move a second number of units of the at least one unit corresponding to a second order quantity from the second supply pallet to the donor pallet.

[0122] Aspect 22: A method of using the system of any one of the preceding aspects, wherein the supply pallet is a first supply pallet, wherein the method comprises:moving, by the at least one robot, at least one additional unit from a second supply pallet to the supply pallet.

[0123] Aspect 23: The method of aspect 22, wherein the at least one unit and the at least one additional unit are of a same SKU and a different dye lot.

[0124] Aspect 24: The method of aspect 22, wherein moving at least one additional unit from the second supply pallet comprises moving all units from the second supply pallet to the first supply pallet.PCT Application Attorney Docket No. 19133.0406P1

[0125] Aspect 25: The system of claim 1, further comprising at least one manual pallet makeup station, wherein each manual pallet makeup station of the at least one manual pallet makeup station comprises:a staging area that is configured to receive a second supply pallet; andan operator space that is configured to permit a human operator to move a second number of units from the second supply pallet to a second donor pallet

[0126] Aspect 26: The system of claim 25, further comprising a main conveyor, wherein the staging area and the operator space are on opposite sides of the main conveyor, wherein each manual pallet makeup station of the at least one manual pallet makeup station further comprises:a first region that is configured to receive the second supply pallet so that units on the first conveyor are accessible to the human operator in the operator space; anda transverse conveyor that is configured to move the second supply pallet from the staging area to the first region.

[0127] Aspect 27: A method comprising:moving, to a first supply pallet having at least one unit thereon, by at least one robot, at least one additional unit from a second supply pallet.

[0128] Aspect 28: The method of aspect 27, wherein the at least one unit and the at least one additional unit are of a same SKU and a different dye lot.

[0129] Aspect 29: The method of aspect 27, wherein moving at least one additional unit from the second supply pallet comprises moving all units from the second supply pallet to the first supply pallet.

[0130] Although several embodiments of the invention have been disclosed in the foregoing specification, it is understood by those skilled in the art that many modifications and other embodiments of the invention will come to mind to which the invention pertains, having the benefit of the teaching presented in the foregoing description and associated drawings. It is thus understood that the invention is not limited to the specific embodiments disclosed hereinabove, and that many modifications and other embodiments are intended to be included within the scope of the appended claims. Moreover, although specific terms are employed herein, as well as in the claims which follow, they are used only in a generic andPCT Application Attorney Docket No. 19133.0406P1 descriptive sense, and not for the purposes of limiting the described invention, nor the claims which follow.

Claims

1. PCT Application Attorney Docket No. 19133.0406P1 CLAIMSWhat is claimed is:

1. A system comprising:a conveyor comprising:a main conveyor comprising an unloading area and a loading area;at least one staging area configured to receive a supply pallet and move the supply pallet to the main conveyor; anda vision system configured to determine a quantity of units on the supply pallet;at least one robot; andat least one controller configured to cause the at least one robot to:position a donor pallet on the main conveyor in the loading area; and move a number of units of the quantity of units corresponding to an order quantity on the supply pallet to the donor pallet.

2. The system of claim 1, further comprising a donor staging area, wherein the at least one robot comprises:a first robot that is configured to move the donor pallet from the donor staging area to position the donor pallet on the main conveyor in the loading area; anda second robot configured to move the number of units of the quantity of units on the supply pallet to the donor pallet.

3. The system of claim 1, further comprising a slip sheet staging area, wherein the second robot is configured to move a slip sheet from the slip sheet staging area to the donor pallet.

4. The system of claim 1, wherein the vision system is configured to determine the quantity of units on the supply pallet by detecting a height of each stack of at least one stack of units on the supply pallet.

5. The system of claim 4, wherein the vision system is configured to determine the quantity of units on the supply pallet by using a known unit vertical dimension associated with the supply pallet, and calculating the quantity of units based on the known unit vertical dimension and the detected height of each stack of the at least one stack of units on the supply pallet.PCT Application Attorney Docket No. 19133.0406P1 6. The system of claim 1, wherein the conveyor defines a vision station that is spaced from the unloading area, wherein the vision system is configured to determine the quantity of units on the supply pallet at the vision station.

7. The system of claim 6, wherein conveyor is configured to stop the supply pallet at the vision station prior to delivering the supply pallet to the unloading area.

8. The system of claim 1, wherein the at least one controller is configured to:determine, based on the quantity of units determined by the vision system and a shipment quantity, a reverse shipment condition; andin response to determining the reverse shipment condition, cause the at least one robot to move a difference between the quantity of units determined by the vision system and the shipment quantity to the donor pallet.

9. The system of claim 1, wherein the at least one controller is configured to cause the at least one robot to position the donor pallet on the main conveyor in the loading area and move the number of units of the quantity of units on the supply pallet to the donor pallet without feedback from a vision device coupled to the at least one robot.

10. The system of claim 1, further comprising a wrapping station that is configured to wrap a remainder of the supply pallet following removal of the number of units therefrom.

11. The system of claim 1, wherein the least one staging area comprises a plurality of staging areas, wherein the at least one controller is configured to determine an order in which each supply pallet from each staging area is moved to the unloading area.

12. The system of claim 1, wherein the at least one controller is configured to:cause the conveyor to move a first supply pallet to the loading area and a second supply pallet to the unloading area;cause the at least one robot to consolidate a plurality of units by moving at least one unit from the second pallet at the unloading area to the first pallet at the loading area.PCT Application Attorney Docket No. 19133.0406P1 13. The system of claim 12, wherein the at least one controller is configured to cause the at least one robot to move all units of the second pallet at the unloading area to the first pallet at the loading area.

14. The system of claim 1, wherein the at least one controller is configured to cause the at least one robot to move based on machine learning.

15. The system of claim 1, wherein the at least one controller is configured to cause the at least one robot to move the number of units of the quantity of units on the supply pallet to the donor pallet into a shippable arrangement in which the quantity of units are stacked to have an equal height on a first side and an opposite second side of the pallet.

16. The system of claim 1, wherein the vision system comprises at least one robotic scanner configured to scan opposed sides of the supply pallet.

17. The system of claim 1, wherein the vision system comprises an optical scanner that is configured to capture an identifier indicative of a stock keeping unit (SKU) of the supply pallet.

18. The system of claim 17, further comprising the supply pallet, wherein the identifier is a barcode or a QR code.

19. The system of claim 1, wherein the vision system is configured to generate a condition based on whether all units on the supply pallet are of a single stock keeping unit (SKU).

20. The system of claim 1, wherein the at least one controller is configured to generate a condition based on whether all units on the supply pallet are of a single stock keeping unit (SKU).

21. The system of claim 1, wherein the at least one controller is configured to:cause the conveyor to move a second supply pallet having at least one unit with a stock keeping unit (SKU) that differs from an SKU of the number of units to the unloading area; and cause the at least one robot to move a second number of units of the at least one unit corresponding to a second order quantity from the second supply pallet to the donor pallet.PCT Application Attorney Docket No. 19133.0406P1 22. The system of claim 1, further comprising at least one manual pallet makeup station, wherein each manual pallet makeup station of the at least one manual pallet makeup station comprises:a staging area that is configured to receive a second supply pallet; andan operator space that is configured to permit a human operator to move a second number of units from the second supply pallet to a second donor pallet23. The system of claim 22, further comprising a main conveyor, wherein the staging area and the operator space are on opposite sides of the main conveyor, wherein each manual pallet makeup station of the at least one manual pallet makeup station further comprises:a first region that is configured to receive the second supply pallet so that units on the first conveyor are accessible to the human operator in the operator space; anda transverse conveyor that is configured to move the second supply pallet from the staging area to the first region.

24. A method of using the system of any one of the preceding aspects, wherein the supply pallet is a first supply pallet, wherein the method comprises:moving, by the at least one robot, at least one additional unit from a second supply pallet to the supply pallet.

25. The method of claim 24, wherein the at least one unit and the at least one additional unit are of a same SKU and a different dye lot.

26. The method of claim 24, wherein moving at least one additional unit from the second supply pallet comprises moving all units from the second supply pallet to the first supply pallet.

Images