Systems and methods for dispensing items

Abstract

The present disclosure provides systems and methods for product handling and dispensing. In one aspect, the present disclosure provides a system comprising a plurality of storage units for storing one or more trays containing one or more items, one or more distribution conveyors for lifting and dispensing at least one of the one or more items to one or more receiving units, and a shuttle with one or more tray picking mechanisms, the shuttle configured to (a) be movable relative to the plurality of storage units, and (b) (i) pick at least one tray of the one or more trays using the one or more tray picking mechanisms, and (ii) transport the at least one tray to the one or more distribution conveyors.

Description

[Technical field]

[0001] cross reference

[0001] This application claims priority to U.S. Provisional Application No. 63 / 196,527, filed June 3, 2021, and is incorporated by reference herein in its entirety for all purposes. [Background technology]

[0002]

[0002] Efficiently handling multiple items having a variety of different shapes and / or sizes can pose increasingly complex technical challenges. Items of different shapes and / or sizes may require different product handling and packaging solutions that are tailored to the specific features and characteristics of the items. Furthermore, different items may require different storage environments to ensure that the quality of such items does not deteriorate over time. Summary of the Invention

[0003]

[0003] Product handling systems can be used to facilitate the transfer of individual product items from incoming bulk forms to dedicated trays for subsequent inspection, sorting, selection, and packaging for consumption. Inspection may include inspection of the product items in the trays by electromagnetic (e.g., optical, hyperspectral) or other (e.g., physical, acoustic, gas sensing, etc.) techniques. Prior to packaging, the product items arranged in the trays may be stored on a mobile carousel that is responsible for controlling environmental factors such as temperature, humidity, illuminance, ambient gases, product-product interactions, and / or others. The movement of the product items from the transfer station of the carousel to an external staging location may be accomplished using a robot and / or a conveyor belt. The systems and methods of the present disclosure may enable consumers to quickly and at low cost select specific individual product items based on accompanying metadata (e.g., source, identifier) ​​combined with the results of the inspection (e.g., appearance).

[0004]

[0004] Various limitations associated with currently available product handling and packaging systems are recognized herein. The present disclosure provides various systems and methods for handling, tracking, and dispensing items to ensure accurate and efficient fulfillment of customer orders.

[0005]

[0005] In one aspect, the disclosure provides a product handling system comprising a tray having one or more lanes and one or more openings disposed on a bottom surface, one or more supports positioned in at least one of the one or more lanes, the one or more supports configured to support one or more items, and one or more dispensing units having one or more dispensing arms configured to (i) couple to, engage with, or lift the one or more supports and (ii) transport (a) the one or more supports, (b) one or more items on the one or more supports, or (c) both the one or more supports and the one or more items to one or more receiving areas or receiving units.

[0006] In some embodiments, the one or more dispensing arms are configured to move through the one or more openings to lift the one or more supports and one or more items on the one or more supports, and (ii) transport the one or more supports and the one or more items to the one or more receiving units. In some embodiments, the one or more supports are configured to support the one or more items as the one or more items are transported to the one or more receiving units.

[0007]

[0007] In some embodiments, the one or more distribution units comprise one or more conveyor belts. In some embodiments, the one or more conveyor belts of the one or more distribution units are configured to move in a first direction, thereby moving the movable parts of the one or more supports in a second direction to transport the one or more items towards the one or more receiving units for distribution. In some embodiments, the second direction is different from the first direction. In some embodiments, the one or more distribution arms are configured to translate towards the one or more receiving units.

[0008] In some embodiments, the one or more supports are configured to move away from the tray when the one or more distribution arms transport the one or more supports and one or more items on the one or more supports toward the one or more receiving units. In some embodiments, the one or more supports comprise an alignment block configured to couple the one or more supports to the one or more distribution arms.

[0009] In some embodiments, the one or more supports comprise a flexible element configured to support the one or more items. In some embodiments, the flexible element comprises a belt configured to extend around or along a portion of the one or more dispense arms.

[0010] In some embodiments, the system further comprises a rotating element configured to orient at least a portion of the flexible element in a predetermined direction when the flexible element moves relative to the one or more dispense arms. In some embodiments, the rotating element is integral with the one or more receiving units.

[0011]

[0011] In some embodiments, the system further comprises an engagement mechanism configured to couple the flexible element to a portion of the one or more dispensing units. In some embodiments, the engagement mechanism comprises a latch or a magnet. In some embodiments, the engagement mechanism comprises a pin or protrusion configured to engage with a portion or component of the one or more dispensing units. In some embodiments, the one or more dispensing arms comprise a beaded chain, grooved chain links, a belt, a round belt, or a flat belt.

[0012] In some embodiments, the movable portion of the one or more supports comprises a conveyor belt. In some embodiments, the one or more supports comprise one or more features that prevent the one or more items from shifting or dropping during transport. In some embodiments, the one or more features comprise one or more side walls, grooves, ridges, or protrusions.

[0013]

[0013] In some embodiments, the one or more supports are liftable or movable relative to the tray. In some embodiments, the one or more supports are configured to drop off the one or more conveyors when conveyed beyond a distal end of the one or more conveyors. In some embodiments, the one or more supports comprise a plate having a plurality of slots through which one or more distribution units may extend or move to (i) contact the one or more items and (ii) transport the one or more items to one or more receiving units or receiving areas. In some embodiments, transporting the one or more items comprises sliding the one or more items along the plate across an edge of the tray. In some embodiments, the system further comprises a weighted shovel configured to slide adjacent to the plate to push the one or more items along a surface of the plate. In some embodiments, the one or more supports comprise a plate having a plurality of slots. In some embodiments, the one or more distribution units comprise a plurality of rollers configured to protrude from the plurality of slots to contact and transport or distribute the one or more items.

[0014]

[0014] In another aspect, the present disclosure provides a system for handling and dispensing items. The system may include a support configured to hold a plurality of items in a predetermined orientation; a movable plate configured to (i) contact at least one of the plurality of items and (ii) move the plurality of items toward a bag or packaging unit to dispense one or more of the plurality of items into the bag or packaging unit; and a gating mechanism configured to selectively permit or restrict dispensing of the one or more items. In some embodiments, the support is configured to hold the plurality of items in a vertical or upright configuration. In some embodiments, the plurality of items in the vertical or upright position are oriented such that a first surface of the plurality of items contacts the support and a second surface of the plurality of items does not contact the support. In some embodiments, the surface area of ​​the first surface is smaller than the surface area of ​​the second surface. In some embodiments, the plurality of items have a thin or flat form factor. In some embodiments, the movable plate is configured to translate toward the bag or packaging unit to displace the plurality of items toward the bag or packaging unit. In some embodiments, the movable plate is configured to ratchet forward to displace the plurality of items. In some embodiments, the system may further comprise a drive unit for driving the movable plate. In some embodiments, the drive unit comprises a chain. In some embodiments, the chain comprises a beaded chain. In some embodiments, the support further comprises one or more movable side walls adjustable to vary a width of the support. In some embodiments, the one or more movable side walls are adjustable based on a width of the plurality of items. In some embodiments, the gating mechanism comprises one or more rollers or brushes. In some embodiments, the one or more rollers or brushes are configured to singulate the one or more items during dispensing of the one or more items. In some embodiments, the system may further comprise one or more dividers for dividing a first subset of the plurality of items from a second subset of the plurality of items.

[0015]

[0015] In another aspect, the present disclosure provides a system comprising one or more imaging devices configured to capture multiple data streams including at least (i) a first data stream including a plurality of images or videos of one or more items at a first location, and (ii) a second data stream including a plurality of images or videos of one or more items at a second location, and a processor configured to generate a set of filtered images or videos using the first and second data streams, wherein the set of filtered images or videos corresponds to (a) one or more features or characteristics of the one or more items, (b) one or more components or subsystems associated with the storage, handling, or distribution of the one or more items, or (c) one or more parameters associated with the storage, handling, or distribution of the one or more items.

[0016] In some embodiments, the first location includes a tray, container, or case in which the one or more items are stored before the one or more items are dispensed, and in some embodiments, the second location includes a receiving area or receiving unit configured to receive the one or more items after the one or more items have been dispensed.

[0017] In some embodiments, the first location includes a tray, container, or case in which the one or more items are stored after the one or more items are induced and before the one or more items are received or placed in a storage area or unit. In some embodiments, the second location includes a tray, container, or case in which the one or more items are stored before the one or more items are dispensed.

[0018] In some embodiments, the one or more components include a tray, a tray storage unit containing a tray, a dispensing unit, a transport unit, or a receiving unit.

[0019] In some embodiments, the one or more features or characteristics of the one or more items include a packaging type, a product type, a product shape, a product category, a product material, a packaging material, a product dimension, a product weight, or a stock keeping unit (SKU) associated with the one or more items. In some embodiments, the one or more features or characteristics of the one or more items include an optical property, the optical property including a reflectance of the item or packaging for the item. In some embodiments, the one or more features or characteristics of the one or more items include (i) a drip risk or a spoilage risk, (ii) a designation of whether the one or more items are packaged, unpackaged, food, or non-food, or (iii) a food type or non-food type designation of the one or more items.

[0020]

[0020] In some embodiments, the one or more parameters include (i) an order associated with one or more items, (ii) a dispensing order of one or more items, (iii) a condition of a receiving unit, tray, or one or more items before, during, and after a transport of one or more items from a tray or a dispensing of one or more items into or onto a receiving unit, (iv) a fault detected by the removal of one or more items from a tray or a dispensing of one or more items, or (v) a detection of one or more items in or on a receiving unit.

[0021]

[0021] In some embodiments, the first data stream includes one or more images or videos of a tray containing one or more items before the one or more items are transported or dispensed from the tray. In some embodiments, the first data stream includes one or more images or videos of the tray after the one or more items have been transported or dispensed from the tray. In some embodiments, the second data stream includes one or more images or videos of the receiving unit before the one or more items are dispensed to the receiving unit. In some embodiments, the second data stream includes one or more images or videos of the receiving unit after the one or more items have been dispensed to the receiving unit.

[0022]

[0022] In some embodiments, the filtered set of images or videos is organized in chronological order. In some embodiments, the filtered set of images or videos is generated using (i) a plurality of images or videos of the dispensing unit transporting one or more items from a tray or dispensing one or more items to a receiving unit, or (ii) a third data stream including sensor data acquired during the guiding, storing, transporting, handling, or dispensing of one or more items. In some embodiments, the processor is configured to track or detect, based on the filtered set of images, the type of dispensed items, the amount of dispensed items, the dispensed location of each item, the success rate of dispensing the items at the target location, the movement of the items during dispensing, or one or more characteristics of the one or more items. In some embodiments, the one or more characteristics include the dimensions, shape, color, size, weight, material, or material properties of the one or more items.

[0023] In some embodiments, the processor is further configured to perform object tracking to separate one or more items from background features. In some embodiments, the processor is further configured to implement or perform one or more filtering techniques, such as blob detection, edge detection, contour detection, depth masking, or determining one or more attributes of one or more items and detecting the quantity, orientation, or location of one or more items. In some embodiments, the processor is further configured to filter out background features based on a color, hue, tint, tone, or shade of the background features. In some embodiments, the processor is further configured to filter out background features by (i) converting the images from a red-green-blue (RGB) color space to a hue-saturation-value (HSV) color space, or (ii) comparing depths using one or more depth point clouds. In some embodiments, the processor is further configured to implement a Gaussian mixture algorithm for object tracking. In some embodiments, the processor is configured to determine a probability distribution for where the one or more items will be located or dispensed within or on the receiving unit based on the plurality of images, the one or more depth point clouds, and a target location for dispensing the one or more items.

[0024] In some embodiments, the one or more imaging devices comprise a depth sensor or an RGB sensor. In some embodiments, the one or more imaging devices are configured to capture one or more images or videos before, during, or after dispensing of one or more items to the receiving unit.

[0025]

[0025] In some embodiments, the processor is configured to automatically detect whether the one or more items have been dispensed successfully. In some embodiments, the processor is configured to compare the set of filtered images to the customer order to confirm the one or more items have been dispensed successfully. In some embodiments, the processor is configured to train one or more machine learning models using at least the filtered set of images to confirm dispense accuracy and detect dispense errors. In some embodiments, the processor is configured to train one or more machine learning models based on characteristics or features of the one or more dispensed items, in combination with human labeling, the characteristics or features being determined based on one or more images or videos from the first data stream or the second data stream. In some embodiments, the processor is further configured to provide at least a subset of the plurality of images or videos to an operator for order confirmation upon detecting an error or a discrepancy between two or more dispense sensors. In some embodiments, the processor is configured to train one or more machine learning models using a subset of the images.

[0026]

[0026] In some embodiments, the system may further comprise a communication module configured to provide the filtered set of images to a customer to provide live updates during order fulfillment to enable the customer to (i) check fulfillment accuracy, quality, or item damage, or (ii) provide live feedback regarding the order fulfillment process. In some embodiments, the processor is configured to build an order fulfillment history based on the filtered set of images and identify a dispensing routine that caused an error. In some embodiments, the error corresponds to a missing item, a wrong item, or an extra item. In some embodiments, the processor is configured to determine, based on the filtered set of images, one or more quality issues regarding one or more items that were damaged during dispensing. In some embodiments, the processor is configured to detect, using computer vision, one or more dispensing actions that damage one or more items based on one or more interactions between one or more items during dispensing. In some embodiments, the one or more machine learning models are configured to determine when one or more items with a high interaction risk are dispensed based on one or more characteristics or attributes of the one or more items. In some embodiments, the high risk of interaction corresponds to a risk of spillage, a risk of damaging one or more items, a risk of damaging the item packaging, or a hole in the item or item packaging.

[0027] In some embodiments, the system may further comprise an imaging unit configured to image one or more items during item guidance or item storage to assess quality of the items. In some embodiments, the processor is configured to verify quality, maturity, temperature, correct item type, quantity, location, orientation, alignment, or placement within one or more trays based on the first data stream or the second stream or both.

[0028]

[0028] In some embodiments, the system may further include an imaging unit configured to monitor movement or navigation of one or more robots and provide one or more corrective adjustments to the one or more robots based on the monitored movement or navigation.

[0029]

[0029] In some embodiments, the system may include one or more sensors for monitoring or tracking one or more items, the one or more sensors being selected from the group consisting of a pressure sensor, an optical sensor, a depth sensor, and a thermal sensor. In some embodiments, the pressure sensor is disposed (i) in or on a receiving unit configured to receive one or more items, or (ii) on a dispensing mechanism for dispensing one or more items. In some embodiments, the optical sensor is disposed (i) at an end of a dispensing arm of a dispensing mechanism for dispensing one or more items, or (ii) above a receiving unit configured to receive one or more items.

[0030]

[0030] In some embodiments, at least one of the first data stream and the second data stream includes an image or video of a tray including one or more items before the items are lifted from the tray. In some embodiments, at least one of the first data stream and the second data stream includes an image or video of one or more items after the items are lifted from the tray using a dispensing mechanism and before the one or more items are dispensed. In some embodiments, at least one of the first data stream and the second data stream includes an image or video of one or more items as the one or more items are returned to the tray after a subset of the one or more items is dispensed using a dispensing mechanism. In some embodiments, the first data stream is captured using a first imaging device of the one or more imaging devices, and the second data stream is captured using a second imaging device of the one or more imaging devices. In some embodiments, the first data stream and the second data stream are captured using a same imaging device of the one or more imaging devices. In some embodiments, the first data stream and the second data stream are captured using different imaging devices of the one or more imaging devices. In some embodiments, at least one of the first data stream or the second data stream includes one or more images or videos of the one or more items as the one or more items are dispensed. In some embodiments, the one or more characteristics include information obtained by reading or interpreting a label associated with or affixed to the one or more items. In some embodiments, the information obtained from the label includes a product code, a product name, a product weight, a product expiration date, product ingredients, and / or nutritional information for the product.

[0031] In another aspect, the disclosure provides a system including a plurality of storage units configured to hold or store a plurality of items, a plurality of receiving units configured to move along one or more adjustable paths to one or more of the plurality of storage units to receive or pick one or more of the plurality of items, and one or more processors configured to continuously modify or update operation of the automated fulfillment system based on heuristics and an optimization algorithm that optimizes one or more order fulfillment metrics based on a state of (i) the automated fulfillment system or (ii) one or more components or subsystems of the automated fulfillment system. In some embodiments, modifying or updating operation of the automated fulfillment system includes (i) modifying or updating one or more adjustable paths for at least one receiving unit of the plurality of receiving units, (ii) modifying or updating a distribution order of one or more items, or (iii) modifying or updating a sequence in which one or more orders are fulfilled.

[0032] In some embodiments, the state of the automated fulfillment system includes (i) the availability of one or more items at a storage unit, or (ii) the availability of one or more receiving units of a plurality of receiving units to receive one or more items. In some embodiments, the state of the automated fulfillment system includes the location or distribution of one or more items within or throughout the automated fulfillment system or one or more subsystems of the automated fulfillment system. In some embodiments, the state of the automated fulfillment system includes a current state or a simulated future state. In some embodiments, the simulated future state can be used to inform one or more decisions or operations related to the storage, handling, and distribution of one or more items. In some embodiments, the simulated future state can be used to (i) generate one or more predictions related to future states of future orders or additions, and (ii) adjust the operation of the automated fulfillment system based on the one or more predictions.

[0033] In some embodiments, the one or more processors are configured to determine one or more predicted future states of the automated fulfillment system based on a current state of the automated fulfillment system. In some embodiments, the one or more processors are configured to simulate a progression or rollout from a current state to a future state based on one or more deterministic or probabilistic events. In some embodiments, the one or more processors are configured to determine the current state or the simulated future state by applying one or more machine learning algorithms.

[0034] In some embodiments, the plurality of storage units comprises a movable mechanism for distributing one or more items to one or more of the plurality of receiving units. In some embodiments, the plurality of receiving units comprises a movable mechanism for removing or picking one or more items from the plurality of storage units.

[0035] In some embodiments, the one or more order fulfillment metrics include throughput, power efficiency, order fulfillment speed, packing density, or item compatibility. In some embodiments, the one or more modified or updated routes are optimized to reduce the probability of item damage during or after a dispensing routine, to avoid or mitigate the occurrence of one or more failure modes, or to address one or more customer preferences.

[0036] In some embodiments, the processor is configured to update the one or more adjustable paths when a new order is received or an existing order is modified. In some embodiments, the processor is configured to update the one or more adjustable paths when one or more additional receiving units are deployed to fulfill a new order.

[0037] In some embodiments, the processor is configured to update the one or more adjustable paths by changing an order or sequence in which one or more receiving units travel to one or more of the plurality of storage units to retrieve items of the order. In some embodiments, the processor is configured to update the one or more adjustable paths by changing an order or sequence in which one or more receiving units retrieve one or more items of the order.

[0038]

[0038] In some embodiments, the optimization algorithm is configured to determine an optimal path routing for at least a subset of the plurality of receiving units. In some embodiments, the optimization algorithm is further configured to perform a global optimization for the plurality of receiving units to resolve one or more conflicts without an optimized local solution. In some embodiments, the optimization algorithm is further configured to assign a set of nodes for one or more receiving units to cover at a time, where the nodes are selectively reserved such that other receiving units cannot use the nodes until the nodes are unreserved. In some embodiments, the optimization algorithm is further configured to predict collisions and supplement one or more human-designed nodes with one or more additional virtual nodes to enable finer-grained node reservation that allows one or more smaller receiving units to fill spaces between one or more larger receiving units. In some embodiments, the optimization algorithm is further configured to perform a graph search using a custom heuristic that penalizes occupied nodes or additional time-consuming moves. In some embodiments, the optimization algorithm is further configured to penalize different types of nodes, where the different types of nodes include edge nodes or fast lane nodes. In some embodiments, the optimization algorithm is configured to take into account the ever-changing map and recalculate the optimal path routing for multiple receiving units when a receiving unit is stuck in traffic for a given time or when one or more receiving units are stuck between another receiving unit that is trying to cover the same node. In some embodiments, the optimization algorithm is configured to generate or update a path network that includes various types of paths and waiting locations such that one or more receiving units can take only a part of a path based on their destination. In some embodiments, the optimization algorithm is configured to determine a distribution order for one or more items. In some embodiments, the optimization algorithm is configured to calculate a cost value for each item based on the mass of the item, the drop surface area, the age or expiration date of the item, or the packaging material of the item.In some embodiments, the optimization algorithm is configured to assign higher cost values ​​to objects having a larger mass over the drop surface area. In some embodiments, the optimization algorithm is configured to assign higher cost values ​​to objects having a harder packaging material using a multiplication factor. In some embodiments, the optimization algorithm is configured to determine an optimal drop order based on a cost value associated with each item. In some embodiments, the optimization algorithm is configured to direct the robot to travel to the second storage unit to retrieve one or more items of the customer order if a difference in drop cost value between a first item and a second item in the first storage unit is greater than a difference in drop cost value between the first or second item and a third item in the second storage unit. In some embodiments, the optimization algorithm is further configured to determine an optimal location for inventory storage of the one or more items.

[0039]

[0039] In some embodiments, the one or more processors are configured to allocate multiple items to different storage units to increase tolerance to individual machine failures. In some embodiments, the multiple items include the same item. In some embodiments, the one or more processors are configured to place more frequently picked items in one or more trays located (i) at or near the distribution mechanism, or (ii) in an area directly accessible by a robot or a distribution mechanism of a receiving unit to reduce the time required to pick or distribute the items. In some embodiments, the one or more processors are configured to place items that are frequently purchased together in the same tray to reduce the number of trays that need to be picked for order fulfillment.

[0040]

[0040] In some embodiments, the optimization algorithm is configured to determine optimal item locations when selecting a path and a distribution order. In some embodiments, the optimization algorithm is configured to perform a weighted cost optimization based on at least one of: (i) travel time to a distribution location; (ii) how busy the distribution location is; and (iii) the relative order in which one or more items should be dropped to minimize damage to the items during the distribution routine. In some embodiments, the optimization algorithm is configured to weight one or more costs together to determine a final cost for each item. In some embodiments, the optimization algorithm is configured to determine an item pick order based on potential condition changes corresponding to machine failure, item unavailability, or increased traffic in an area proximate to one or more storage units. In some embodiments, the optimization algorithm is configured to direct one or more robots to an order drop-off location when no unpacked items remain for a customer order.

[0041]

[0041] In some embodiments, the optimization algorithm includes a cost function. In some embodiments, the optimization algorithm is configured to (i) calculate a weighted cost value for each available tray, (ii) select a destination with the lowest cost value, (iii) select one or more items from the destination with the lowest cost value, and (iv) instruct one or more robots to retrieve one or more items from the tray or trays associated with the destination with the lowest cost value if any unpacked items remain for the particular order. In some embodiments, the optimization algorithm is trained using machine learning or reinforcement learning to learn one or more weighted cost values ​​for each cost feature. In some embodiments, the optimization algorithm is implemented using one or more predefined rules for handling the one or more items. In some embodiments, the one or more predefined rules are set at least in part based on a set of editable features or tags that are assigned to the one or more items when the one or more items are received or directed. In some embodiments, the set of editable features or tags includes a drip risk, unpackaged, chemical, food, non-food, food type, or non-food type designation. In some embodiments, the one or more predefined rules are set based on a set of editable rule operators that correspond to one or more decisions related to order fulfillment. In some embodiments, the one or more decisions include determining whether the item can be stored on top of another item or whether the item can be bagged with another item. In some embodiments, the one or more predefined rules are established based on a set of editable rule statements that combine one or more features and one or more rule operators. In some embodiments, the one or more predefined rules are adjustable to constrain product handling decisions. In some embodiments, the optimization algorithm includes a machine learning algorithm configured to generate one or more feature classifications from one or more images of one or more newly received or derived items to facilitate onboarding.

[0042]

[0042] In some embodiments, the system further comprises a plurality of robots equipped with movable mechanisms for (i) retrieving one or more items from a plurality of storage units and / or (ii) distributing one or more items to one or more of the plurality of receiving units.

[0043] In some embodiments, modifying or updating the operation of the automated fulfillment system includes modifying or updating (i) one or more item storage locations, (ii) one or more item storage locations within a tray or storage unit, (iii) one or more tray storage locations within a storage unit, (iv) the number of items stored in each tray, (v) the number of trays stored in each storage unit, (vi) the distribution of one or more items to one or more trays, (vii) the distribution of one or more items to one or more storage units, or (viii) the distribution of one or more trays to one or more storage units. In some embodiments, modifying or updating the operation of the automated fulfillment system includes grouping or co-locating items in the same tray or in the same storage unit. In some embodiments, modifying or updating the operation of the automated fulfillment system includes grouping or co-locating items such that the items are stored in one or more adjacent trays or storage units that are adjacent or proximate to one another. In some embodiments, modifying or updating the operation of the automated fulfillment system includes modifying or updating the availability of a storage unit or a receiving unit based on a failure or need for maintenance of one or more devices associated with the storage unit or the receiving unit. In some embodiments, modifying or updating the operation of the automated fulfillment system further includes rerouting orders to various storage units. In some embodiments, modifying or updating the operation of the automated fulfillment system further includes dispatching at least one other receiving unit and adjusting the movement paths of the multiple receiving units based on the planned or actual movement path of the at least one other receiving unit.

[0044] In another aspect, the disclosure provides a system comprising a plurality of storage units for storing one or more trays containing one or more items, one or more distribution conveyors for lifting and distributing at least one of the one or more items to one or more receiving units, and a shuttle with one or more tray picking mechanisms, the shuttle being movable relative to the plurality of storage units and configured to (a) pick at least one of the one or more trays using the one or more tray picking mechanisms, and b) transport the at least one tray between a first location and a second location. In some embodiments, the first location or the second location includes (i) one or more receiving units, (ii) one or more distribution conveyors, (iii) one or more storage units of the plurality of storage units, (iv) one or more aisles between the storage units, (v) an induction area for receiving one or more items, (vi) an item washing area, (vii) an item inspection area, or (viii) one or more vehicles that can be used to transport the one or more items to or from the product handling system.

[0045]

[0045] In some embodiments, the one or more distribution conveyors are located at or near at least one of the first location and the second location. In some embodiments, the shuttle is movable relative to the one or more distribution conveyors. In some embodiments, the one or more distribution conveyors are movable relative to the storage units. In some embodiments, the one or more removal mechanisms and the one or more distribution conveyors are movable relative to each other. In some embodiments, the one or more receiving units are provided adjacent or proximate to the shuttle. In some embodiments, the one or more distribution conveyors are provided adjacent or proximate to the shuttle. In some embodiments, the one or more distribution conveyors are located away from the shuttle. In some embodiments, the one or more receiving units are located away from the shuttle.

[0046]

[0046] In some embodiments, the system further comprises an elevator configured to transport one or more receiving units. In some embodiments, the elevator is configured to transport one or more receiving units vertically. In some embodiments, the shuttle is configured to move laterally. In some embodiments, the shuttle is configured to pick and transport at least one tray to the elevator. In some embodiments, the elevator is configured to transport at least one tray into or towards one or more distribution conveyors. In some embodiments, the shuttle is configured to pick and transport at least one tray vertically or horizontally to or towards one or more distribution conveyors. In some embodiments, the shuttle is configured to pick and transport at least one tray along any one or more axes in three-dimensional space. In some embodiments, the shuttle is configured to rotate about any one or more axes in three-dimensional space.

[0047]

[0047] In some embodiments, the one or more distribution conveyors are located below one or more shelves of the plurality of storage units. In some embodiments, the one or more distribution conveyors are located adjacent to the plurality of storage units. In some embodiments, the one or more distribution conveyors are located at or near the center of an aisle between two or more of the plurality of storage units. In some embodiments, the one or more distribution conveyors are located below one or more of the plurality of storage units. In some embodiments, the one or more distribution conveyors include a first distribution conveyor set provided on or coupled to a shuttle and a second distribution conveyor set below or adjacent to the plurality of storage units. In some embodiments, the one or more receiving units are configured to move relative to the plurality of storage units. In some embodiments, the one or more distribution conveyors are located away from the plurality of storage units.

[0048]

[0048] In some embodiments, the plurality of storage units comprises one or more shelves. In some embodiments, the shuttle is configured to transport the one or more shelves to a distribution unit comprising one or more distribution conveyors. In some embodiments, the distribution unit is configured to (i) pick at least one tray from the one or more shelves and (ii) distribute one or more items from the at least one tray to one or more receiving units. In some embodiments, the one or more trays comprise one or more openings. In some embodiments, the one or more distribution conveyors are configured to move through the one or more openings to lift and transport at least one item from the one or more trays to the one or more receiving units. In some embodiments, the one or more distribution conveyors comprise one or more features configured to support at least a portion of the one or more items to prevent the one or more items from shifting or dropping during transport of the one or more items.

[0049] In some embodiments, the shuttle is configured to move in at least one of a horizontal direction and a vertical direction. In some embodiments, the shuttle is configured to move along one or more lateral planes. In some embodiments, the one or more lateral planes include two or more lateral planes corresponding to different heights. In some embodiments, the shuttle is configured to move along one or more axes of movement relative to the multiple storage units.

[0050]

[0050] In some embodiments, the system further comprises a shuttle movement mechanism, hi some embodiments, the shuttle movement mechanism is configured to provide one or more axes of movement about which the shuttle moves or advances.

[0051] In some embodiments, the one or more trays include one or more openings. In some embodiments, the one or more distribution conveyors are configured to move through the one or more openings to lift and transport at least one item from the one or more trays to the one or more receiving units.

[0052]

[0052] Another aspect of the present disclosure provides a non-transitory computer-readable medium containing machine-executable code that, when executed by one or more computer processors, performs any of the methods described above or elsewhere in this specification.

[0053] Another aspect of the present disclosure provides a system comprising one or more computer processors and a computer memory coupled thereto, the computer memory including machine executable code that, when executed by the one or more computer processors, performs any of the methods described above or elsewhere herein.

[0054]

[0054] Additional aspects and advantages of the present disclosure will become readily apparent to those skilled in the art from the following detailed description, in which only exemplary embodiments of the present disclosure are shown and described. As will be understood, the present disclosure is capable of other various embodiments, and its several details can be modified in various obvious respects, all without departing from the present disclosure. Thus, the drawings and description should be regarded as illustrative in nature, and not as restrictive.

[0055] Incorporation by Reference

[0055] All publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference. To the extent that the publications and patents or patent applications incorporated by reference conflict with the disclosure contained in the specification, the specification shall supersede and / or govern such conflicting content. [Brief description of the drawings]

[0056]

[0056] The novel features of the invention are set forth with particularity in the appended claims. The features and advantages of the present invention will be better understood by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the invention are utilized, and the accompanying drawings (also referred to herein as "Figures" and "FIGS.").

[0057] [Figure 1]

[0057] Various examples of conventional storage and retrieval systems are illustrated diagrammatically. [Figure 2A-2H]

[0058] 1A-1D are schematic diagrams illustrating various examples of automated storage and retrieval systems, according to some embodiments. [Diagram 3]

[0059] 1 illustrates a schematic diagram of a cube storage and retrieval system according to some embodiments. [Figure 4]

[0060] 1 illustrates a schematic representation of a tray having one or more open walls, according to some embodiments. [Diagram 5]

[0061] 1 illustrates a schematic diagram of a mini-load storage and retrieval system, according to some embodiments. [Figure 6]

[0062] 1 illustrates a schematic diagram of a carry pick storage and retrieval system according to some embodiments. [Figure 7-9]

[0063] 1A-1D are schematic diagrams illustrating various examples of freezers with an anteroom, according to some embodiments; [Figure 10]

[0064] 1 illustrates a schematic of a Tetra vertical lift storage and retrieval system, according to some embodiments. [Figure 11]

[0065] 1A-1C are schematic diagrams illustrating one or more robots that may be used with the Tetra Vertical Lift Storage and Retrieval System, according to some embodiments. [Figure 12]

[0066] 1 illustrates a schematic diagram of a floating conveyor, according to some embodiments. [Figure 13]

[0067] 1 illustrates a schematic of an alignment block for a conveyor system. [Figure 14]

[0068] 1 illustrates a schematic of a tongue conveyor according to some embodiments. [Figure 15]

[0069] 13A-13C are schematic diagrams illustrating rollers that may be used with a tongue conveyor, according to some embodiments. [Figure 16]

[0070] 1A-1C are schematic diagrams illustrating a tongue conveyor that may engage with one or more dispense conveyor arms to dispense items, according to some embodiments. [Figure 17]

[0071] 1 illustrates a schematic diagram of a tray with a hinge, according to some embodiments; [Figure 18]

[0072] 1 illustrates a schematic diagram of a skid plate according to some embodiments. [Figure 19]

[0073] 1 illustrates a schematic diagram of a tongue spool, according to some embodiments. [Figure 20]

[0074] 1 illustrates a schematic diagram of a trapdoor conveyor according to some embodiments. [Figure 21]

[0075] 1 illustrates a schematic diagram of a fixed position belt, according to some embodiments. [Figure 22]

[0076] 13A-13C are schematic diagrams illustrating grooved dividers that may be used with a tongue conveyor, according to some embodiments. [Figure 23-24]

[0077] 1 illustrates a schematic diagram of one or more pop-up rollers, according to some embodiments. [Diagram 25]

[0078] 1 illustrates a schematic diagram of a self-contained bucket configuration, according to some embodiments. [Figure 26]

[0079] 13A-13C are schematic diagrams illustrating various features for inhibiting movement of items during dispensing, according to some embodiments. [Figure 27]

[0080] 1 illustrates a schematic diagram of an example of a link chain for transporting and / or distributing items, according to some embodiments. [Figure 28]

[0081] 1 illustrates a schematic diagram of a reverse roller chain, according to some embodiments. [Figure 29]

[0082] 13A-13C are schematic diagrams illustrating a chain arrangement for interfacing with a tongue of a conveyor according to some embodiments. [Diagram 30]

[0083] 13A-13C are schematic diagrams illustrating a forward flag that may be provided on a distribution chain, according to some embodiments; [Figure 31-32]

[0084] 1A-1D are schematic diagrams illustrating various examples of beaded chains for transporting and / or dispensing one or more items, according to some embodiments. [Diagram 33]

[0085] 1 illustrates a schematic diagram of a raft for supporting one or more items, according to some embodiments. [Figure 34-36]

[0086] 13A-13C illustrate schematic diagrams of various locations where one or more load cells may be placed, according to some embodiments. [Figure 37]

[0087] 1 illustrates a schematic diagram of one or more cameras for imaging an item, according to some embodiments; [Figure 38]

[0088] 4 illustrates a schematic diagram of weighting of distributed recognition results by modality according to some embodiments; [Figure 39]

[0089] 1 illustrates a schematic diagram of a line-following robotic sorting tool, according to some embodiments. [Diagram 40]

[0090] 1 illustrates a schematic diagram of an LFC dump truck configuration, according to some embodiments. [Diagram 41]

[0091] 1A-1C are schematic diagrams illustrating a tray rack configured for under-rack dispensing, according to some embodiments. [Diagram 42]

[0092] 1 illustrates a schematic diagram of a tray rack that can be docked to a vertical lift system, according to some embodiments. [Diagram 43]

[0093] 1 illustrates a schematic diagram of a number of line tracking bags (LFBs) that may receive one or more dispensed items, according to some embodiments. [Figure 44-45]

[0094] 1 illustrates a schematic representation of the layout of a shipping container or trailer compatible with the systems and methods of the present disclosure. [Figure 46]

[0095] 1 illustrates a schematic diagram of a double shuttle tray picker configuration, according to some embodiments. [Figure 47]

[0096] 1 illustrates a schematic diagram of a multi-temperature container for storing items, according to some embodiments. [Figure 48]

[0097] 1 illustrates a schematic of a shuttle and distribution architecture including shuttle-integrated distribution, according to some embodiments. [Figure 49]

[0098] 13A-13C are schematic diagrams illustrating receptacles located on a shuttle according to some embodiments. [Figure 50]

[0099] 1A-1D are schematic illustrations of various examples of tray supports for supporting one or more items in a tray, according to some embodiments. [Fig. 51-52]

[0100] 1 illustrates a schematic diagram of a tray with multiple lanes according to some embodiments. [Diagram 53]

[0101] 13A-13C are schematic diagrams illustrating an example of a single lane tray dispensing operation according to some embodiments. [Figure 54]

[0102] 1A-1C are schematic diagrams illustrating clip-on tray dividers according to some embodiments. [Figure 55]

[0103] 1 illustrates a schematic diagram of a tray hygiene system, according to some embodiments. [Figure 56]

[0104] 1A-1C are schematic diagrams illustrating one or more pivotable tray dividers according to some embodiments. [Figure 57]

[0105] 1 illustrates a schematic diagram of a tray imaging system for item inspection and quality control according to some embodiments; [Figure 58]

[0106] 13A-B illustrate schematic diagrams of locating and sorting items within one or more lanes of a tray, according to some embodiments; [Figure 59]

[0107] 1 illustrates a schematic diagram of a system for imaging items from below during loading of a tray, according to some embodiments; [Figure 60-61]

[0108] 1 illustrates a schematic diagram of a system and method for image capture for order building and validation, according to some embodiments; [Figure 62]

[0109] 13A-B illustrate schematic diagrams of an assessment of risk when distributing one or more items into a bag, according to some embodiments; [Figure 63]

[0110] 1A-1C are schematic diagrams illustrating one or more drawers that may be automatically opened to provide access to various trays or items, according to some embodiments; [Figure 64]

[0111] 1 illustrates a schematic diagram of a load cell that may be used to measure the weight of one or more items, according to some embodiments. [Figure 65]

[0112] 1 illustrates a schematic diagram of a hybrid guided and distributed vertical lift system, according to some embodiments. [Figure 66]

[0113] 1 illustrates a schematic diagram of the use of graph searching to assist in optimal path routing, according to some embodiments; [Figure 67]

[0114] 1 illustrates generally a method for determining a path and distribution order for one or more items using a weighted cost optimization algorithm. [Figure 68]

[0115] 13A illustrates a schematic diagram of a depth grid that may be generated for a bag, according to some embodiments; [Figure 69]

[0116] 13A-13C illustrate diagrams of adjusting the resolution of a depth grid map generated for a bag, according to some embodiments; [Figure 70]

[0117] 1 illustrates a schematic diagram of an exemplary method for determining or predicting potential damage during item dispensing, according to some embodiments; [Figure 71]

[0118] 13 illustrates a schematic representation of an implementation of a distributed target algorithm for tracking drop targets according to some embodiments; [Figure 72]

[0119] 1 illustrates a schematic diagram of a tray reloading system according to some embodiments. [Figure 73]

[0120] 1 illustrates a schematic diagram of a locker system for customers to receive their orders, according to some embodiments. [Figure 74]

[0121] 1 illustrates a schematic diagram of a floor lock system for a tray rack, according to some embodiments. [Figure 75]

[0122] 1A-1C are schematic diagrams illustrating a tray rack with a packaging mechanism to prevent items from falling or being damaged, according to some embodiments. [Figure 76]

[0123] 1 illustrates a schematic diagram of an airlock for a freezer, according to some embodiments. [Figure 77]

[0124] 1 illustrates a schematic diagram of a cooling system connected to a vertical lift system via a supply duct and a return duct, according to some embodiments. [Figure 78]

[0125] 1 illustrates a schematic diagram of a vertical lift system with additional storage according to some embodiments. [Figure 79]

[0126] 1 illustrates a schematic diagram of a refrigerated vertical lift system configured as an anteroom, according to some embodiments. [Figure 80]

[0127] 1 illustrates generally a computer system that may be programmed or otherwise configured to implement the methods provided herein. [Figure 81]

[0128] 1A-1D illustrate various examples of items that may be suitable for storage, handling, or transport using a cartridge conveyor; [Figure 82]

[0129] 1A-1C are schematic diagrams illustrating examples of items that may be vertically reoriented to achieve higher packing or storage density. [Figure 83]

[0130] 1 illustrates a schematic diagram of an example of a cartridge conveyor that may be used to handle and / or dispense items stored in a vertical orientation, according to some embodiments. [Figure 84]

[0131] 1 illustrates generally one example of a cartridge tray that may be used to handle and / or dispense items stored in a vertical orientation. [Figure 85]

[0132] 1 illustrates a schematic of a sequence of time steps in which the camera field of view can be adjusted to capture the dispensing order. [Figure 86]

[0133] 1A-1C are schematic diagrams illustrating exemplary tray down lane dividers that may be used to organize various items in a tray, according to some embodiments. [Figure 87]

[0134] 1 illustrates a schematic diagram of an exemplary system configured for automated order staging and pickup, according to some embodiments. [Figure 88]

[0135] 1 illustrates a schematic diagram of an exemplary configuration of a pickup system, according to some embodiments. [Figure 89]

[0136] 13A-13C illustrate schematic diagrams of human or robot interaction with a movable door or window of a pickup bay for retrieval of one or more items or one or more bags, according to some embodiments. [Figure 90]

[0137] 1 illustrates a schematic diagram of an example pickup bay according to some embodiments. [Figure 91]

[0138] 1 illustrates a schematic diagram of an exemplary interaction between a robot (e.g., a line tracking bag or LFB) and a pickup bay, according to some embodiments. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0058]

[0139] While various embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will occur to those skilled in the art without departing from the invention. It is understood that various alternatives to the embodiments of the invention described herein may be used.

[0059]

[0140] Whenever the terms "at least," "greater than," or "greater than or equal to" appear before the first number in a series of two or more numbers, the terms "at least," "greater than," or "greater than or equal to" apply to every number in the series. For example, 1, 2, or 3 or more is equivalent to 1 or more, 2 or more, or 3 or more.

[0060]

[0141] Whenever the terms "at most," "less than," or "or less than" appear before the first number in a series of two or more numbers, the terms "at most," "less than," or "or less than" apply to each number in the series. For example, 3, 2, or 1 or less is equivalent to 3 or less, 2 or less, or 1 or less.

[0061]

[0142] The terms "real time" (or "real-time"), as used interchangeably herein, generally refer to events (e.g., operations, processes, methods, techniques, calculations, computations, analyses, visualizations, optimizations, etc.) that are performed using recently obtained (e.g., collected or received) data. In some cases, real-time events may occur nearly instantly, or may occur within a sufficiently short period of time, such as, for example, within at least 0.0001 milliseconds (ms), 0.0005 ms, 0.001 ms, 0.005 ms, 0.01 ms, 0.05 ms, 0.1 ms, 0.5 ms, 1 ms, 5 ms, 0.01 s, 0.05 s, 0.1 s, 0.5 s, 1 s, or more. In some cases, real-time events may occur almost instantly, or in other cases over a sufficiently short period of time, for example, within up to 1 second, 0.5 seconds, 0.1 seconds, 0.05 seconds, 0.01 seconds, 5 milliseconds, 1 millisecond, 0.5 ms, 0.1 ms, 0.05 ms, 0.01 ms, 0.005 ms, 0.001 ms, 0.0005 ms, 0.0001 ms, or even less.

[0062]

[0143] The present disclosure provides systems and methods for removing one or more items from a tray and transporting or distributing the one or more items into a bag or box that can be shipped to a consumer or customer. As used herein, a customer may refer to an end consumer or an entity that buys and sells various items to end consumers or other entities (e.g., a grocery store).

[0063]

[0144] A shuttle system may include some type of automated fulfillment technology in the family of technologies associated with automated storage and retrieval systems (AS / RS). In general, there are two main ways these technologies may be used. One is to store and retrieve full cases or boxes that are palletized or sent in some way to a shipping dock. The other is to store and retrieve totes full of individual items that are picked one at a time. Typically this is for e-commerce or spare parts applications. In this latter case, the tote may typically be sent to a picker to pick the items. The picker is typically a human, but may be a pick-and-place robot.

[0064]

[0145] In some cases, a shuttle (picker) may be configured to travel through each level of the storage unit and pick up the totes. The totes may be brought to the end where there is an elevator (z) that carries the totes to ground level. The totes may then be moved by a conveyor to a human or a pick-and-place robot. In some cases, a robot may retrieve the tote from the elevator and move it to an item picker, which can reduce the complexity of the conveyor. In other cases, a shuttle may pick the tote instead of a lift and carry it to the picker. The shuttle may be configured to move vertically to perform this operation. Various examples of AS / RS systems are shown in Figure 1.

[0065]

[0146] integrated distribution

[0147] In one aspect, the present disclosure provides a product handling system with integrated distribution functions and capabilities. The product handling system may include a plurality of storage units for storing one or more trays containing one or more items, one or more distribution conveyors for lifting and distributing at least one of the one or more items to one or more receiving units, and a shuttle with one or more tray pick mechanisms, the shuttle (a) movable relative to the plurality of storage units and (b) configured to (i) pick at least one of the one or more trays using the one or more tray pick mechanisms and (ii) transport the at least one tray to the one or more distribution conveyors.

[0066]

[0148] The storage units may include one or more shelves for storing trays and / or items. The storage units may include, for example, a vertical lift system (VLS). The VLS may include a lift, platform, support, or carrier that may move laterally (e.g., vertically and / or horizontally) to access one or more shelves, trays, and / or items. In some cases, the storage units may include a lift, platform, support, or carrier that may move along any axis or plane in three-dimensional space to access one or more shelves, trays, and / or items. In some cases, one or more shelves may include one or more fixed platforms or supports. In other cases, one or more shelves may be movable using a carousel mechanism. The carousel mechanism may be configured to move one or more shelves along a closed-loop internal path within the storage unit such that the one or more shelves sequentially arrive at target areas within the storage unit. The target areas may be aligned with one or more openings within the storage unit. The target areas may be used for item guidance through the one or more openings. In some cases, the target area may be used for retrieval of one or more trays or items stored on a tray or shelf of a storage unit such that one or more items stored on the tray or shelf of a storage unit may be transported to a receiving unit and dispensed into or onto the receiving unit. The receiving unit may comprise a box, bag, or container configured to receive one or more items. In some cases, the receiving unit may comprise a platform or support on which one or more items may be placed. The receiving unit may be interchangeably referred to as a line tracked bag (LFB). The receiving unit or LFB may comprise a robot, platform, or support configured to receive one or more items. In some cases, the LFB may comprise a robot with a support configured to receive one or more items. In some cases, the LFB may comprise a robot with a bag or box configured to receive one or more items.The bag or box may be flexible or rigid. The bag or box may be collapsible or non-collapsible. In some examples, the LFB may comprise a robot with a collapsible bag on top (e.g., to receive one or more items) or on a surface onto which one or more items may be placed.

[0067]

[0149] The one or more distribution conveyors may be used to lift one or more items from the trays once the trays are positioned at or near the target area. The one or more distribution conveyors may be configured to move under the trays along one or more axes or planes to align with various areas of the trays. In some cases, the trays may include multiple grooves that accommodate one or more items. The grooves may have one or more openings or slots at the bottom of the grooves. In some cases, the distribution conveyors may be configured to move under the trays along one or more axes or planes to align with one or more openings or slots in the grooves that accommodate one or more target items (e.g., one or more items associated with a customer order). When aligned with a particular opening or slot, the distribution conveyor may be configured to move vertically relative to the trays to extend through the slots and lift one or more items from the trays. In some cases, the trays may be configured to move vertically relative to the distribution conveyor such that the distribution conveyor extends through the openings or slots and lifts one or more items from the trays. Once one or more items are picked up by the distribution conveyor, the distribution conveyor may be configured to transport the one or more items towards a receiving unit or carrier. The carrier may comprise a robot capable of moving in a number of different directions to access a number of different trays stored in one or more storage units. In some non-limiting examples, the storage units may comprise a VLS as described elsewhere herein. The carrier may be configured to receive or support one or more items picked from a tray using the distribution conveyor. In some cases, the carrier may comprise a receiving unit. In other cases, the receiving unit may be integral with the carrier. In some cases, the carrier may be configured to transport one or more items to a receiving unit that may be located remotely from the carrier. In some embodiments, at least a portion of the distribution conveyor may be configured to translate or extend to or towards the carrier or receiving unit.In some embodiments, a moving part of a distribution conveyor may be used to transport or convey one or more items to or towards a carrier or receiving unit to distribute one or more items into or onto the carrier or receiving unit. The moving part may comprise, for example, one or more conveyor belts or one or more chains.

[0068]

[0150] In some embodiments, the system may comprise a shuttle. The shuttle may comprise a carrier as described above. Alternatively, the shuttle may be a different mechanism or subsystem than the carrier as described above. The shuttle may comprise a robot configured to move relative to the plurality of storage units to access one or more trays stored in the plurality of storage units. In some cases, the shuttle may comprise one or more tray retrieval mechanisms. In some cases, the shuttle may be configured to move in at least one of a horizontal direction and a vertical direction. In some cases, the shuttle may be configured to move along one or more lateral planes. The one or more lateral planes may include two or more lateral planes corresponding to different heights. In some cases, the shuttle may be configured to move along one or more axes of movement relative to the plurality of storage units. In some embodiments, the system may further comprise a shuttle movement mechanism. The shuttle movement mechanism may be configured to provide one or more axes of movement along which the shuttle moves or advances. The shuttle movement mechanism may comprise, for example, a gantry-like system. The shuttle may be configured to (i) retrieve at least one tray of the one or more trays using the one or more tray retrieval mechanisms, and (ii) transport the at least one tray to one or more distribution conveyors. The one or more tray ejection mechanisms may comprise an extendable arm, member, or frame configured to move towards and / or away from the one or more trays. The one or more tray ejection mechanisms may comprise a hook or latch that engages with a portion of the one or more trays. The hook or latch may enable the tray ejection mechanism to move or withdraw the one or more trays so that the distribution conveyor can access the trays and items stored therein. In some cases, the shuttle may be movable relative to the one or more distribution conveyors. The one or more tray ejection mechanisms and the one or more distribution conveyors may be movable relative to each other. In some cases, the shuttle may be configured to move in one or more lateral directions.

[0069]

[0151] In some cases, the shuttle may be configured to pick and transport at least one tray to an elevator. The elevator may be configured to transport at least one tray to one or more distribution conveyors. In other cases, the shuttle may be configured to pick and transport at least one tray vertically or horizontally to or toward one or more distribution conveyors.

[0070]

[0152] In some cases, the one or more receiving units may be located adjacent or proximate to the shuttle. In some cases, the one or more distribution conveyors may be located adjacent or proximate to the shuttle. In some cases, the one or more distribution conveyors may be located remotely from the shuttle. In some cases, the one or more receiving units may be located remotely from the shuttle. In some embodiments, the one or more receiving units may be configured to move relative to the multiple storage units.

[0071]

[0153] In some embodiments, the system may further comprise an elevator configured to transport the one or more receiving units. The elevator may be configured to transport the one or more receiving units horizontally and / or vertically.

[0072]

[0154] In some cases, the one or more distribution conveyors may be located beneath one or more shelves of the plurality of storage units. In some cases, the one or more distribution conveyors may be located adjacent to the plurality of storage units. In some cases, the one or more distribution conveyors may be located away from the plurality of storage units.

[0073]

[0155] As described above, the plurality of storage units may comprise one or more shelves. The shuttle may be configured to transport the one or more shelves to a distribution unit that comprises one or more distribution conveyors. The distribution unit may be configured to (i) pick at least one tray from the one or more shelves and (ii) distribute one or more items from the at least one tray to one or more receiving units. In some cases, the one or more trays may comprise one or more openings. In such cases, the one or more distribution conveyors may be configured to move through the one or more openings to lift and transport at least one item from the one or more trays to the one or more receiving units. In some embodiments, the one or more distribution conveyors may comprise one or more features configured to support at least a portion of the one or more items to prevent the one or more items from shifting or dropping during transportation of the one or more items. The one or more features may include, for example, one or more beads or other physical components that support one or more sides of the items as the items are transported. In some cases, the one or more features may include various protrusions and / or depressions. Such protrusions and / or depressions may be spaced in a periodic or non-periodic configuration.

[0074]

[0156] In some embodiments, the dispensing conveyor may be integrated with customer receptacles attached to the shuttle inside it, as shown in Figure 2A. When filling an order, the shuttle travels to each location, pulls a tray, dispenses the items, replaces the tray, and then moves to the next tray location. This repeats until the bag is full. The shuttle then travels somewhere to remove the customer receptacle.

[0075]

[0157] In other embodiments, picking and dispensing may be performed on the shuttle, but the receptacles may be on a separate robot below (e.g., on the ground), as shown in Figure 2B. When a tray is picked, the shuttle travels to the ground and dispenses the items into the receptacles.

[0076]

[0158] In some cases, as shown in Figure 2C, the shuttle functions similarly to Figure 2B but does not move vertically: instead of descending to the ground to dispense, it moves horizontally to an end where there are receptacles on elevators that serve each level.

[0077]

[0159] In another example, as shown in Figure 2D, the shuttle does not move in z but remains on one level. The shuttle transfers the trays onto an elevator that takes them to ground level. When the elevator arrives at ground level, it passes over a distribution conveyor that distributes them into receptacles.

[0078]

[0160] In some cases, as shown in Figure 2E, distribution may be at the last row on the ground floor, but instead of an elevator carrying the trays there, a shuttle may move vertically and horizontally to transport and distribute the trays.

[0079]

[0161] In other cases, the dispense may be located under the shelf rather than under the shuttle aisle, as shown in Figure 2F. The shuttle drops a tray onto the dispense, which travels upward through the tray to dispense the items out. The dispense may be on one side or both sides.

[0080]

[0162] In some embodiments, one or more of the distributions may be located below the shelf, as shown in Figure 2G, and the shuttle aisle is also the aisle for the receptacle to travel through. In one variation, the receptacle may be able to travel through the shelf at ground level for quick access to each distribution conveyor.

[0081]

[0163] In other embodiments, the distribution conveyor may not be co-located with the storage, as shown in Figure 2H. In such cases, the shuttle may pick up one or more trays and transport them to one or more distribution units.

[0082]

[0164] Cube storage with distribution conveyor

[0165] Another AS / RS technology is cube storage, as shown in Figure 3. If a distribution conveyor is installed in this architecture, the best location is on the outside perimeter of the system. One or more picker bots (red) may move around, picking up trays from each storage column. The picker bots may move the trays to a distribution cell (blue) and drop the trays off above the distribution so that one or more items can be dispensed into a receptacle. The trays may then be removed and replaced.

[0083]

[0166] To allow for this, the tray needs to be designed more like a tote, but with one or two open walls (e.g., as shown in Figure 4). The four corners (red dots) are picking locations for the picker bot. Products will be dispensed from the open side of the tote. There may be a variety of tote heights to accommodate different SKU heights without wasting vertical storage space. The walls may just be columns at the four corners. The columns / walls may also be automatically removable and / or interchangeable to minimize the number of different tray and / or tote configurations required.

[0084]

[0167] Mini road with distribution conveyor

[0168] FIG. 5 shows a mini-load AS / RS configuration. In some cases, a distribution station may be provided along the mini-load configuration. A pick may be on a rail and move horizontally and / or vertically. The pick may then place the tray on a distribution conveyor to pick the item. In some cases, the receptacle may be outside the system so as not to interfere with the rails of the mini-loader. The receptacle may be at the end of an aisle, with the rail ending just short of the distribution. In some embodiments, there may be one or more distributions. The distribution may be at ground level or at multiple levels.

[0085]

[0169] AGV storage system with distribution conveyor

[0170] FIG. 6 illustrates another example of an AS / RS configuration. Such a configuration may include one or more automated guided vehicle (AGV) robots configured to lift shelves and carry the shelves to a picker. The AGV may be configured to move between various locations (e.g., from a first location where shelves or trays are located to a second location where a picker is located). The AGV may include a support surface on which trays containing various items or products can be placed or held for transport to various areas within a warehouse, grocery store, or item storage, handling, packaging, and distribution center. The support surface may include a horizontal support surface located on top of the AGV. The support surface may be integrated with one or more structural components of the AGV. In some cases, the AGV may also be configured to hold or support trays or containers (e.g., boxes or bags) on which multiple items and products can be placed or held. In some cases, the containers may be optional (i.e., the trays and / or items and products stored in the trays may be placed directly on the support surface of the AGV). The trays, items, and / or products may be on a support structure or surface of the AGV.

[0086]

[0171] In some embodiments, the architecture shown in FIG. 6 may also be integrated with a distribution station. The shelves may be lifted and carried forward by a line tracked tray (LFT). The LFT may comprise a robot configured to carry one or more trays. In some cases, the LFT may comprise a floor-traveling AGV. The shelves may dock with the distribution station and the trays may be pulled from the shelves to pick items from the shelves. The distribution may have a small elevator that picks from anywhere in the rack. The items may then be distributed and the trays and racks may be returned to storage.

[0087]

[0172] Freezer with front compartment underneath

[0173] One of the main challenges in designing a freezer is that condensation can cause frost and ice to accumulate, a problem that can be particularly bad on an automatic setting. Figures 7 and 8 show a side view of a vertical lift machine with a vestibule that can help reduce ice accumulation. The vertical lift machine may include multiple storage areas 710 and one or more elevators 715 for transporting items between the storage areas 710 and one or more distribution units 720. Insulation may be used to separate the various zones or areas of the vertical lift machine. One or more doors may be provided to allow the elevators 715 to move between the various zones or areas. Different zones or areas may have different storage conditions (e.g., different temperatures, pressures, humidity, etc.). The storage system shown in Figures 7 and 8 can reduce condensation accumulation in low temperature (LT) areas by creating a vestibule that is kept at a refrigerated temperature (RT). The RT areas may be colder and therefore less humid, which may reduce moisture entering the LFT when the door is opened. As used herein, the term "LT" or "low temperature" may refer to or correspond to a freezer temperature. Low temperature or freezer temperature may include a temperature range for storing one or more items that may need to be frozen to maintain freshness, quality, or shelf life. In some cases, the temperature range of the LT area may be from about -20°C to about 0°C. As used herein, the term "RT" or "refrigerated temperature" may refer to or correspond to a temperature range for storing one or more items that may need to be refrigerated to maintain freshness, quality, or shelf life. In some cases, the temperature range of the RT area may be from about 0°C to about 5°C.

[0088]

[0174] In some cases, the VLS packs may be placed in an insulated room with modular walk-in panels or other insulating structures. In some cases, the freezer may have a horizontal or other door mechanism. The door mechanism may be normally closed and may separate the RT and LT areas. In some cases, the distribution area may be located in the RT. In some cases, a single or double layer air curtain may be provided to allow one or more LFBs and / or LFTs to enter. In some cases, the freezer may be located above the refrigerated storage area, as shown in Figure 8. Figure 9 shows an alternative embodiment of the VLS pack-vessel. The system may include a freezer with a RT vestibule and a vertical door.

[0089]

[0175] Tetra VLS Configuration

[0176] 10 and 11 show alternative vertical lift system configurations in which trays can be picked from multiple sides and lowered for distribution. A vertical lift system may include multiple storage areas 710 located adjacent to one or more vertical lift elevator shafts 711. The multiple storage areas 710 may be accessible from multiple different sides via one or more vertical lift elevator shafts 711. One or more elevators 715 may move vertically up and down the vertical lift elevator shafts 711 to retrieve items from the multiple storage areas 710. The one or more elevators 715 may be able to retrieve items from multiple different storage areas 710 located on different sides of the vertical lift elevator shafts 711. In some cases, trays may be picked from one or more directions. For example, trays may be picked from a first direction using a carrier located in a first vertical lift elevator shaft or from a second direction using a carrier located in a second vertical lift elevator shaft. In any of the embodiments described herein, one or more LFBs may be configured to travel beneath the vertical lift system to receive one or more items dispensed from the vertical lift system.

[0090]

[0177] Tetra VLS-LFR Navigation

[0178] In some cases, one or more bots may drive down to access one or more distribution stations, as shown in Figure 11. In some cases, the Tetra VLS configuration may provide an LFB only highway and / or an LFT and LFB highway.

[0091]

[0179] Pick Station Design

[0180] Floating and Tongue Conveyors

[0181] In some cases, distribution conveyors may have problems handling some categories of SKUs, including soft items (tortillas) and leafy items (bundles of carrots and kale). The underlying problem is that these items cannot be picked up with two rows of pulleys on the belt; they need a perfectly supported surface underneath. But to lift off the tray with a perfectly flat surface, the flat belt needs to move through a support structure above the tray. One solution is to provide a flat surface on the tray that can be lifted with items on it and then transported into a bag / box. Such a solution can be implemented using a floating and / or tongue conveyor. A floating and / or tongue conveyor may comprise any type of support, apparatus, or device that can be used to transport, deliver, move, or convey items from a first location to a second location. As used herein, a "floating conveyor" may refer to a support on which one or more items may rest. The support may be liftable off the tray. The support may comprise a conveyor that may be used to transport one or more items towards the support or towards the receiving unit. The conveyor may be driven by another transport unit (e.g., one or more distribution conveyors described elsewhere herein). In some cases, the one or more distribution conveyors may lift the support from the tray so that the support "floats" relative to the tray. In some cases, the one or more distribution conveyors may be coupled to the support such that the conveyor of the distribution conveyor can be used to drive or manipulate the conveyor of the support. The one or more distribution conveyors may be operated or driven in a first direction to drive the conveyor of the support (e.g., in a second direction). As used herein, a "tongue conveyor" may refer to a support on which one or more items may be placed. The support may include a flexible material on which one or more items may be placed. The support and one or more items provided on the support may be transported simultaneously towards the receiving unit.In some cases, the support may be configured to extend around or along a bottom of one or more of the dispensing arms as they are transported to or towards the receiving unit, in other cases, the support may be configured to hang off an end (like a tongue) of one or more of the dispensing arms as they are transported to or towards the receiving unit.

[0092]

[0182] In one aspect, the present disclosure provides a product handling system comprising a tray having one or more lanes and one or more openings disposed on a bottom surface, one or more supports positioned in at least one of the one or more lanes, the one or more supports configured to support one or more items, and one or more dispensing units having one or more dispensing arms configured to (i) couple or engage with a portion of the one or more supports and (ii) transport the one or more supports and one or more items on the one or more supports to one or more receiving units.

[0093]

[0183] In some cases, the one or more supports may be liftable, removable, and / or movable relative to the tray. The one or more supports may be lifted and / or moved using one or more distribution conveyors. The one or more distribution conveyors may comprise one or more distribution arms with a movable mechanism for transporting or conveying the one or more items to the receiving unit. The movable mechanism may comprise, for example, one or more conveyor belts or chains. In some embodiments, the one or more distribution arms may be configured to (i) move through one or more openings to lift the one or more supports and one or more items on the one or more supports, and (ii) transport the one or more supports and one or more items to the one or more receiving units.

[0094]

[0184] In some embodiments, the one or more supports may be configured to support the one or more items as they are transported to the one or more receiving units. The one or more distribution conveyors may be configured to lift the one or more supports while the one or more items are on the one or more supports. Lifting the one or more supports may result in simultaneously lifting one or more items on the one or more supports. The one or more distribution conveyors may be configured to simultaneously transport one or more items and one or more supports to or towards one or more receiving units by (i) extending an upper portion of the one or more distribution conveyors to or towards the one or more receiving units and (ii) transporting the one or more items into or onto the receiving unit.

[0095]

[0185] In some embodiments, the one or more distribution conveyors may have one or more conveyor belts, and the one or more supports may have one or more secondary conveyor belts. The one or more conveyor belts of the distribution units may be configured to move in a first direction to cause the one or more secondary conveyor belts of the one or more supports to move in a second direction to transport one or more items towards one or more receiving units for distribution. Movement of the conveyor belts of the distribution conveyor in a first direction (e.g., clockwise or counterclockwise) may cause the secondary conveyor belts of the one or more supports to move in a second direction (e.g., counterclockwise or clockwise), thereby transporting one or more items on the supports towards one or more receiving units. In some cases, the second direction may be different from the first direction. In other cases, the second direction may be the same as the first direction.

[0096]

[0186] In some cases, the one or more supports may include an alignment mechanism configured to couple the one or more supports to the one or more distribution arms. The alignment mechanism allows the one or more supports to mate or couple with the distribution arms. When coupled, the one or more supports may move as a single unit with the one or more distribution arms. Furthermore, the conveyor of the distribution arms and the secondary conveyor of the supports may be arranged in contact with each other such that movement of the conveyor causes a reactionary movement of the secondary conveyor in a different or opposite direction. For example, when the top of the one or more distribution arms extends or advances toward the one or more receiving units, the one or more supports (and one or more items on the supports) may also advance the same distance toward the one or more receiving units. When in the extended position, the conveyor of the distribution arms may move in a first direction, which causes the secondary conveyor in contact with the conveyor of the distribution arms to move in a second direction different from the first direction.

[0097]

[0187] In some embodiments, the one or more supports may comprise a flexible element configured to support one or more items. The flexible element may act as a tongue (i.e., the flexible element may be configured to extend away from the end of the dispense arm or around the bottom of the dispense arm when conveyed towards the receiving unit). The flexible element may comprise a belt configured to extend around or along a portion of the one or more dispense arms. In some embodiments, the system may further comprise a rotating element configured to orient at least a portion of the flexible element in a predetermined direction (e.g., downward or out of the way of the dispense arm when one or more items are dispensed) as the flexible element moves relative to the one or more dispense arms. In some cases, the rotating element may be integrated with the one or more receiving units. In such cases, the one or more receiving units may be configured to position the rotating element adjacent or in close proximity to the one or more dispense arms to enable controlled routing of the flexible element in a desired direction.

[0098]

[0188] In some cases, the system may further comprise an engagement mechanism configured to couple the flexible element to a portion of the one or more dispensing units. The engagement mechanism may include, for example, a latch or a hook. In some cases, the engagement mechanism may include a pin configured to extend into and / or engage with a portion or component of the one or more dispensing units.

[0099]

[0189] In some cases, one or more of the dispense arms may include beaded or grooved chain links, which may help reduce movement of and improve stability of one or more items as they are transported to or towards the receiving unit.

[0100]

[0190] In some cases, the flexible element may include one or more features to prevent one or more items from rolling, shifting, or dropping during transport. The one or more features may include various anti-rolling mechanisms, such as, for example, grooves, ridges, or protrusions. The features may have periodic or non-periodic spacing to accommodate various item sizes, shapes, and profiles.

[0101]

[0191] Floating Conveyor

[0192] As shown in FIG. 12, the floating conveyor may include a conveyor located at each tray lane. The conveyor may not or need not have an actuator. The distribution conveyor may be configured to penetrate the tray and lift the floating conveyor. A notch in the back of the floating conveyor may lock onto the distribution arm of the distribution conveyor. The distribution conveyor may begin to move its belt backwards, which may rotate the belt of the floating conveyor forward until one or more items drop off the front of the floating conveyor into the bag / box or onto the support. The items may be fully supported on the flat belt surface of the floating conveyor as they are transported towards the bag / box or support for distribution.

[0102]

[0193] As shown in Figure 13, alignment blocks are sometimes used to allow for varying amounts of misalignment when the floating conveyor is lifted off the tray. After the conveyor is lifted, the blocks can rigidly couple the conveyor to the distribution arm allowing full control of the flat conveyor belt. The flat conveyor belt can move forward when the distribution belt moves backward and move backward when the distribution belt moves forward.

[0103]

[0194] Floating conveyors may be constructed with simple walls suspended from a four-tray frame, allowing a single tray to carry a set of conveyors of various sizes. The position of the conveyor within the tray may be controlled by interlocking with adjacent trays. One design is a single sheet metal base piece with four separate side pieces, two of which are slideable to accommodate tension and conveyor belt manufacturing variations. Other designs to reduce part count and assembly steps may include, for example, two frame pieces that slide or pivot into place using an over-center mechanism to tension the belt.

[0104]

[0195] FIG. 14 shows a tongue conveyor that may be used to distribute items. Tongue conveyors are similar to floating conveyors, with one major difference. Floating conveyors may have a continuous belt that folds back under the floating conveyor. The tongue conveyor does not rotate under itself, but drops off an edge and hangs down. Alternatively, it may follow the belt path back according to the shape of the distribution arm. Once the items are distributed, the belt may retract into place before the tray is returned to storage. In some cases, the tongue follows the belt path under the distribution arm. To prevent ellipsoidal objects from rolling over the tongue, the tongue may have a "V" or "U" shaped cross section and one or more ridges or protrusions. Alternatively, the tongue may be configured like a ladder with holes along the length of the tongue that create stable points for the ellipsoids.

[0105]

[0196] Tongue conveyor with front rollers

[0106]

[0197] In some cases, the tongue may be deflected by a roller, as shown in Figure 15. The roller may be attached to the LFB or may be part of the VLS. This may reduce the need for a mating interface between the tongue and the dispensing arm.

[0107]

[0198] Tongue conveyor engagement

[0199] As shown in FIG. 16, the tongue conveyor may engage with the distribution conveyor arm to distribute the items and then disengage from the distribution conveyor arm to return the tongue conveyor to the tray. Various engagement methods include dropping the front tongue conveyor link onto the rail of the distribution arm. The tongue may then be pushed forward and / or backward using a push element on the chain. The tongue may be engaged from two slots at the outer positions of the tongue or from one slot in the center of the tongue. In some other alternative embodiments, a pin on the tongue conveyor may be used to engage the grooved chain link by the distribution arm moving inward and / or outward to slide the slot over the pin. The tongue can be moved back and forth as needed by moving the chain link back and forth.

[0108]

[0200] Conveyor integrated into tray

[0201] In some cases, one or more conveyors may be incorporated into or integrated into a structural part or element of the tray. This may be useful for distributing soft and / or irregular items. The one or more conveyors incorporated into or integrated into the tray may comprise a floating conveyor incorporated into the tray. This may reduce capital expenditures by utilizing the tray frame as the structure of the conveyor, and may also reduce operational costs by eliminating the need to maintain separate parts (e.g., tongues) that may require an operator to clean and place on the tray. In some embodiments, the one or more conveyors incorporated into or integrated into the tray may comprise, for example, a fixed position belt, fixed tongues in a grooved divider, one or more pop-up rollers for items that are not a contamination risk, and / or one or more pop-up rollers for items that are a contamination risk.

[0109]

[0202] Tray hinge

[0203] In some scenarios, the handoff of items between the tray and conveyor during pop-through can be unpredictable, making it more difficult for current tray / conveyor systems to dispense soft or oddly shaped items. Eliminating this handoff and the need to lift items off the tray during dispensing can improve reliability.

[0110]

[0204] One way to achieve this is to design a hinge into the front wall of the tray, as shown in Figure 17. The wall remains vertical while the tray is moved, keeping the items in place. When ready to dispense, the front wall rotates 90 degrees to create a plane with a surface that supports one or more items. Items can then be pushed or pulled along this flat surface to be dispensed.

[0111]

[0205] Skid plate

[0206] Soft items that need to be supported on a flat surface may still be transported by the conveyor belt and slide on the low friction support surface. In some cases, flat plates or "skid plates" with two slots may be installed on the tray and the soft items may be placed on these plates, as shown in Figure 18. During the dispensing procedure, the plate may be lifted by the dispensing arm so that the dispensing belt protrudes above the plate to contact the soft items and slide them along the plate and over the front tray edge into the LFB.

[0112]

[0207] To facilitate this process, especially for soft items, a weighted "shovel" may be installed on the underside of the plate (blue). The shovel is configured to slide along a distribution belt just above the skid plate surface and may help push soft items along the surface of the skid plate. In some cases, the skid plate may include an array of slots and the distribution arms may be replaced with corresponding rolling actuators. To distribute the items, the tray may be lowered until the distribution wheels protrude above the top surface of the plate. A rotating wheel may be used to distribute the items from the tray.

[0113]

[0208] Tongue spool

[0209] In some cases, the dispense arm may need to engage and disengage from the tongue during each dispense procedure. Also, the tongue may need to move away from the tray to dispense an item. An alternative design is to have the tongue wrap around a pulley at the front of the dispense tray, as shown in FIG. 19. To dispense an item, the dispense arm belt touches the front pulley and rotates backwards, thereby winding up the tongue and dispensing the item from the front of the tray.

[0114]

[0210] Trapdoor Conveyor

[0211] In some cases, the dispensing systems and methods disclosed herein may require some acceleration to transfer the conveyed items from the tray to the LFB. In some cases, such acceleration may cause unstable items to move in unpredictable ways, making accurate dispensing more difficult. As shown in FIG. 20, an alternative design with zero acceleration for unstable items is to have a tray with a front pulley that can slide forward and / or backward on a slot in the tray. A flat conveyor belt may be wrapped around this front pulley and fixed to the top of the tray, allowing it to move at the bottom of the tray. To dispense items from this tray, the bottom of the belt may be moved to the left, either by pulling the belt directly or by wrapping the belt around a second fixed pulley. This causes the front pulley to move to the left, pulling the floor out from under the items so that the items can drop into the LFB.

[0115]

[0212] Fixed position belt

[0213] As shown in FIG. 21, in some cases, one or more belts with slider beds may be integral with and / or coupled to the tray. The one or more belts may not or need not exit the tray during dispensing of one or more items. The one or more belts may be provided in a fixed position and / or orientation relative to the tray. In some embodiments, the tray may be moved on the LFB to dispense one or more items. In some cases, there may be a lower front member with rollers. In some cases, multiple cross members may be provided on the inside of the belt. The belt may be driven by direct contact with the dispensing arm.

[0116]

[0214] Tongue + groove divider

[0215] FIG. 22 shows a cost-effective method for feeding a flat conveyor into a tray. The tongue may be fed into a slot in a lane divider and may be driven under the divider by a complementary chain on a distribution arm. This design may offer several advantages, including a simplified construction that does not require pulleys, bearings, and / or bushings in the tray. The tray divider may be a single piece. Additionally, the tongue fits within the tray, making connection easy. This configuration may be modified to include a drip tray under the tongue and may be designed to be symmetrical front to back.

[0117]

[0216] Pop-up roller

[0217] As shown in FIG. 23, one or more pop-up rollers may be used to transport soft or irregular items. The pop-up rollers may be sized, shaped, and / or configured to approximate a flat conveyor. In some embodiments, one or more belt conveyors may be used in place of the rollers. In some embodiments, a tray divider may have slots into which the roller axles fit. The tray may be extended out of the elevator above the bag by a tray picking mechanism. In some cases, the tray picking mechanism may be integrated into the elevator. The tray may be lowered onto a distribution unit, which may be used to lift the rollers in the slots. A distribution belt may be used to run or drive the rollers and carry one or more items over the edge for distribution into the bag or onto a support.

[0118]

[0218] Pop-up roller driven within the divider

[0219] In some cases, the distribution belt may be configured to lift and drive one or more wheels located within the divider, as shown in Figure 24. The distribution belt may be positioned so that it does not touch the yellow rollers that hold the food products, thereby reducing the possibility of cross contamination.

[0119]

[0220] Built-in bucket

[0221] Figure 25 shows a self-contained bucket configuration. This configuration may have a floating conveyor integrated into the bucket that may catch drips from the food product. One or more drive shafts may protrude from the sides of the bucket and may be engaged by a dispensing conveyor. The entire bucket may be lifted off the tray by a dispensing lift under the drive wheels.

[0120]

[0222] Chain Distribution Concept

[0223] Ellipsoidal objects (e.g., avocados or pears) on the distribution belt may roll due to the positive and negative acceleration of the distribution belt, or may lean against each other such that after the previous object is dispensed, the second item becomes unstable and rolls off the edge of the conveyor.

[0121]

[0224] As shown in FIG. 26, one or more features or "bumps" (shown in red) may be used to provide hard stops that may prevent items from rolling and / or separate items from one another so that they do not lean against one another. Some exemplary methods for creating these features on a sliding conveyor line include (i) thermoplastic pieces with ridges connected by a single plastic bend and containing reinforcing cords for additional support, or (ii) a chain with a central link that includes one or more features (e.g., grooves, ridges, protrusions, indentations, etc.). In some embodiments, the chain may be designed to slide on the dispense arm by sliding within a slot or gripping the outside of the slot.

[0122]

[0225] Chain Concept

[0226] Some chain designs may be used to redirect the force vector to point through a support arm (resulting in no moment). Chain designs may utilize a lever arm to provide the reaction force, which may be longer than the lever arm on which the item is placed.

[0123]

[0227] A link chain may be used in conjunction with the system of the present disclosure, as shown in FIG. 27. The links may wrap around the outer edge of the arm, which brings the force vector closer to that dictated by the arm. The length and spacing of the chain irregularities may be adjusted by inserting different combinations of links. Such a configuration may not require lubrication and may require fewer rollers. The chain may slide on a metal support arm. Such a configuration may be easy to clean by spraying with a cleaning solution. The arm and link geometry does not have no nooks to trap dirt. The chain may be driven by a custom sprocket that interfaces with an extension on the pivot pin.

[0124]

[0228] In some cases, reverse roller chains may be used, as shown in Figure 28. Reverse roller chains may have one or more link plates located in the center of the chain and one or more pins extending outwardly from the chain. The pins may be captured in a groove in which the chain rides.

[0125]

[0229] In some cases, one or more concave link plates may be used. This may allow for back bending and may provide a smoother continuous contact surface. The concave link plates may allow moments to be transferred through the pivot joint to the pin.

[0126]

[0230] In some cases, one or more concave and convex pin plates may be used. Such plates may be spaced further apart to allow for back bending. Convex and convex pin plates may allow moments to be transferred directly through the plate to the pins.

[0127]

[0231] FIG. 29 shows a chain configuration that may be engaged with the tang. The chain may pass through one or more holes in the tang. The chain may be used to capture the tang as it travels around the sprocket. At the bottom, the chain releases the tang so it can run in a groove in the arm.

[0128]

[0232] Chain distribution "forward flag"

[0233] In some cases, the leading edge radius of the dispense arm may fit into the tray but leave a space at the front of the tray for items to fall out of dispense upon pop-through. One solution is to place a leading spacer on the tray that fills this space and effectively shortens the lane. Another solution is to add a "leading flag" to the dispense chain, as shown in Figure 30. The leading flag may comprise a chain link that extends away from the front of the dispense arm to lift objects in this space without compromising storage space.

[0129]

[0234] Chain Distribution Concept - Beaded Chain

[0235] Another way to create positive features to control or limit item movement during dispensing is to use beaded or ball chains to transport items, as shown in FIG. 31. In some cases, continuous strands with larger spaced features may provide a predetermined or desired spacing between items to prevent them from rolling during transport. Beaded chains may be easier to clean than typical roller chains. They can be manufactured without cavities or mechanical joints to pinch items or collect debris. They can also slide along a continuous track without the need for rollers to similarly capture debris. Within one chain, the spacing, size, shape, and / or material of the beads can be varied to better hold different types of items. Different areas of the chain can be used depending on the items being dispensed.

[0130]

[0236] Tongue Conveyor Engagement - Bead Chain

[0237] The bead chain can also be used to control a tongue conveyor, as shown in Figure 32. As the tongue conveyor lowers onto the dispense arm, the fingers on the tongue may wrap around a portion of the bead chain. A pin on the tongue may drop into a slot in the dispense arm to capture the tongue. The tongue may be pulled forward by driving the bead chain, causing one or more items to be dispensed from the tongue. The tongue may become captured in the slot rather than falling off the dispense arm. When the dispensing operation is complete, the tongue can be retracted by driving the bead chain in the opposite direction.

[0131]

[0238] Raft capture on LFB

[0239] As shown in Figure 33, individual rafts can be used in a tray to support soft items. In some cases, individual rafts may be dispensed into the LFB along with the items they are supporting. This can be disadvantageous as it may require the rafts to be removed from inside the bag of customer's items, a time consuming process that is done manually. Furthermore, the rafts occupy volume that compromises space for other items and puts other items at risk of damage. To address these scenarios, the rafts can be captured using collection bins located in front of the VLS or in front of the LFB.

[0132]

[0240] In some embodiments, the raft (dashed line) may be transported off the front of the distribution conveyor arm until it tilts. The tilted raft may get caught in the front slot of the LFB while items are being tilted into the bag. The distribution arm may extend until the raft aligns with the LFB slot. The raft may then slide into the LFB raft bucket. This process allows all rafts to be automatically removed at the same time and prevents the rafts from interacting with items in the bag.

[0133]

[0241] Cartridge Conveyor

[0242] In some embodiments, one or more cartridge conveyors may be used to handle, convey, and / or transport items towards a target location. The target location may include a bag, package, or structure or platform in or on which one or more items may be placed. The one or more cartridge conveyors may be configured to hold one or more items in a predetermined or desired orientation during storage, handling, conveyance, or transport of the one or more items to the target location. In some cases, items handled or dispensed using a cartridge conveyor may be placed, stored, or held on a support. The cartridge conveyor may be configured to dispense items from a support on which the one or more items are placed, stored, or held.

[0134]

[0243] FIG. 81 shows various examples of items that may be suitable for handling, transporting, or dispensing using a cartridge conveyor. These items may include small items with thin or flat form factors, such as gum, tic tacs, chocolate bars, spice or sauce packs, instant noodles, cans, jars, bags of potato chips, boxes of cereal, etc. Such items may typically be stored in a low density configuration when placed on a standard tray or support because they must be stored or placed in a flat orientation on the tray or support, and may not be reliably dispensed when stored in the tray or may not be stationary when the tray is moved or transported. Furthermore, such items may be difficult or difficult to load or unload from the tray. In some cases, storing or placing such items on a tray may result in wasted vertical storage space (e.g., the top of the tray and buffer may be at least 50 mm for products that are often less than 10 mm high when stored flat).

[0135]

[0244] In some cases, items may be stored more densely when stored vertically. The use of vertical space may be further optimized by adjusting the vertical spacing between various supports or trays each holding a plurality of items that may be vertically oriented. The various supports or trays may be arranged such that a first support or tray holding a first plurality of items in a vertical configuration is positioned above a second support or tray holding a second plurality of items in a vertical configuration. The first support or tray may be positioned at a different vertical height than the second support or tray. The vertical gap between the first support or tray and the second support or tray may be sized or dimensioned to accommodate storage of a plurality of items arranged or held in a vertical or substantially vertical configuration in the second support or tray.

[0136]

[0245] FIG. 82 shows additional examples of items that may be reoriented vertically to achieve higher packing or storage density depending on the form factor of the item. In some embodiments, items of a particular form factor (e.g., flat and / or thin) may be reoriented and stored vertically, which may free up additional storage space within the tray or support for additional items of the same or different types. In some cases, the additional storage space may allow for the use of multiple lanes within the tray or support to store or hold various items, and a vertical lift system may be used to retrieve items associated with customer orders from multiple different lanes or trays. Storing items vertically may also reduce the number of trays required to store items, reducing capital investments and allowing for more frequent dispensing of items from the same tray while maximizing the number of products that can be stored on a single tray. Storing items in a vertical orientation can also allow a user or operator to load more items or products onto the same tray at one time, resulting in more efficient loading of trays, directing larger quantities of items onto trays or supports (potentially reducing or eliminating the need for stockrooms), and allowing operators to order more items at one time less frequently, making purchasing and receiving items more efficient.

[0137]

[0246] FIG. 83 shows an example of a cartridge conveyor that may be used to handle and / or dispense items stored in a vertical orientation. The items may be stored in or on a tray or support. In some cases, a beaded chain may be used to drive a rigid backplate forward toward the dispensing direction or location. In some cases, a beaded chain may be used to ratchet the backplate forward or backward in a controlled incremental manner. In some cases, a rigid backplate may be used to hold the items in a stable upright position by providing or exerting a force to hold the items in a stable upright position. The items may be held in a stable upright position using a rigid backplate positioned at one end of the item and a gate or brush positioned at the other end of the item. The rigid backplate and the gate or brush may be configured to collectively hold the items in a stable upright position.

[0138]

[0247] In some cases, the items may be placed on a support configured to fit within the tray. The support may be supported by a pin that may be lifted by a side action of a beaded chain. The beaded chain may be configured to move freely in a clockwise or counterclockwise direction.

[0139]

[0248] In some cases, the cartridge conveyor may include spring-loaded roller or brush gates that may physically restrict or allow the movement or dispensing of items from the cartridge conveyor. In some cases, the conveyor may include side walls that may move inward or outward to accommodate a range of different product or item widths.

[0140]

[0249] The cartridge conveyors described herein may enable quick loading of items, possibly at or near the edge of a support or tray. The cartridge conveyors may be configured to provide, enable, or otherwise facilitate more reliable singulation of items, prevent items from rolling off in front, and direct items into a vertical drop instead of dispensing items by swing / tumble drop. The cartridge conveyors described herein may be stable for hub-to-edge tray delivery. In some cases, items may be retained on a support or tray for extended periods of time without the need to shuffle or rearrange the items. The cartridge conveyor may not shuffle items or may not need to shuffle items, and items may be retained on a support or tray in the configuration or order in which they were originally loaded until the items are dispensed and the cartridge is emptied. In some cases, the logistical savings from infrequent reloading of items (due to the higher packing density achievable when items are vertically oriented for storage) may outweigh the potential downsides associated with long-term retention of items and the empty space penalty associated with retaining items.

[0141]

[0250] FIG. 84 shows an example of a cartridge tray that may be used to handle and / or dispense items stored in a vertical orientation. The cartridge tray may comprise a dedicated or customized tray for dispensing cartridges using a cartridge conveyor as described and referenced elsewhere herein. In some cases, the cartridge tray may be configured to extend over a bag. In some cases, the cartridge tray may comprise one or more standard lane supports, one or more lane dividers, and one or more brush sides and / or brush gates configured to selectively restrict or allow movement of items over the edges of the tray and provide friction for item singulation. In some cases, the cartridge tray may be manipulated using a bead chain (e.g., using the ratchet mechanism described above) configured to drive a rigid plate back and forth.

[0142]

[0251] Load Cell Distribution Recognition Concept

[0252] Various methods can be implemented to detect when an item has left the dispense conveyor. One method is to use load cells at various positions on the dispense arm to determine when there is a weight change on the dispense arm. Load cells can also be used to determine if an item has accidentally fallen off the conveyor to the side or back.

[0143]

[0253] In some cases, the load cells are located on the dispense arms. This configuration can be used to detect changes in tip mass. In some cases, the load cells are located under each dispense arm. This configuration can be used to detect the weight of the entire string of items on the conveyor and the weight changes of the tips. In some cases, the load cells are located under the rails along which the dispense arms travel. This configuration can be used to detect the weight of the entire string of items on the conveyor and the weight changes of the tips.

[0144]

[0254] Moment Sensor

[0255] In some cases, a strain gauge may be provided on the compression side of the dispense arm to detect the moment of the cantilevered portion of the arm, as shown in Figure 34. Items may move along the length of the conveyor, causing changes in compression. Such changes in compression may be detected using a strain gauge or other type of moment sensor.

[0145]

[0256] Under each arm

[0257] In some cases, a load cell may be located under each dispense arm, as shown in Figure 35. Each arm may measure the total weight or mass of the items on the dispense arm. In some cases, a drop may be identified by measuring or detecting the load leaving the arm.

[0146]

[0258] Under the rail where the arm moves

[0259] In some cases, the load cell may be located under the entire arm and rail assembly, as shown in Figure 36. A total weight may be measured for items arriving at the assembly. Drops may be detected as items are dispensed from the assembly.

[0147]

[0260] Distributed Recognition Concepts in Computer Vision

[0261] As described elsewhere herein, a load cell may be used to detect whether an item has been successfully dispensed into an LFB bag. In some embodiments, computer vision techniques may also be used to detect whether an item has been successfully dispensed into an LFB bag.

[0148]

[0262] As shown in FIG. 37, there may be one or more overhead cameras pointing straight down towards the floor, allowing a clear view inside the LFB. In some cases, the one or more overhead cameras may be angled relative to the LFB. In any of the embodiments described herein, the position and / or orientation of the overhead cameras may be adjusted (e.g., manually or automatically using one or more motors or actuators) to capture images or video from an optimal position or angle of the LFB or any items within the LFB. Depending on the camera's field of view, imaging may occur when the LFB is positioned directly below or near the camera. The LFB may also be moving.

[0149]

[0263] During each dispensing routine, one of the cameras may capture a "pre-dose image" before the items are dispensed into the bag, followed by a "post-dose image" after the dispense has occurred. The system may include a processor configured to detect and / or identify one or more items in the "pre-dose image" and / or the "post-dose image" using computer vision techniques. The "pre-dose image" and / or the "post-dose image" may be taken using the same camera or using different cameras at different dispensing bays around the facility. The "pre-dose image" and / or the "post-dose image" may be used for order building, order confirmation, and tracking errors in the dispensing routine, as described in more detail below. In some cases, a "post-dose image" for a first dispensing operation at a first dispensing station may be used as a "pre-dose image" for a second dispensing operation at a second dispensing station.

[0150]

[0264] As described above, computer vision techniques can then be used to take both the image and additional relevant data as input to assess whether dispensing was successful. Some cameras (e.g., depth cameras, TOF sensors, RGB-D sensors) can provide both RGB images and depth point cloud data, and both of these modalities can be analyzed to determine successful dispensing.

[0151]

[0265] In some cases, classical computer vision techniques can be applied to the RGB stream. For example, object tracking techniques such as MOG ("Mixture of Gaussians") can be used to separate newly present items from the background. In some cases, blob detection, edge detection, and contour detection techniques can then be used in combination with known item attributes (e.g. size, shape, etc.) to detect item quantity and location.

[0152]

[0266] In some cases, items may also be filtered from a background bag (which may have a known color, such as green) based on the known hue of the item. This can be accomplished by converting the image from one color space to another (e.g., from the RGB color space to the HSV color space).

[0153]

[0267] In some cases, machine learning object detection and segmentation techniques can be used to analyze the RGB stream. A simple differential analysis of the depth stream between two images can also be performed. Various levels of decimation of the depth point cloud data can be used to balance resolution and processing time.

[0154]

[0268] In some embodiments, the RGB analysis, depth analysis, and target locations provided for dispensing can all be combined and weighted to generate a prediction of where the item landed in the bag. This can also be represented by a probability distribution of where the item will be in the bag. As shown in Figure 38, the dispensing recognition results can be weighted by imaging modality. The dispensing recognition results of each modality can be aggregated to generate a composite belief or prediction of where the item landed.

[0155]

[0269] Customer bag and shipping design

[0270] Line Follower Robot (LFR) Sorting Tool

[0271] As shown in Figure 39, a conveyor on a line following robot (LFR) can be used to sort items from the center of the tray lane. One or more items can be brought to the LFB for the next item to be bagged. Other items can then be brought back. The LFR conveyor may be on its own LFR, and the LFB may pick up items mid-sort. Alternatively, the conveyor may be on the LFB where the desired items are dispensed.

[0156]

[0272] Tilt Line Tracking Cart (LFC)

[0273] As described elsewhere herein, rafts may be used in conjunction with the systems of the present disclosure. Rafts may include pieces of material on which items are placed. Rafts may be placed within trays. Rafts may be used to transport and distribute odd or uniquely shaped items that cannot be successfully lifted using a distribution conveyor.

[0157]

[0274] In some cases, one or more Line Tracked Carts (LFCs) may be used. The LFC may have a bristle platform on top that allows the dispensing arm to penetrate so that items can rest on top of the bristle. LFCs may be used for items that are too large to fit into a bag or that cannot be turned around.

[0158]

[0275] As shown in Figure 40, an LFC may be modified so that it can tilt at an angle to slide an object down. This may be useful in a pickup area. For example, an LFC may bring a pack of paper towels forward and then tilt so that the paper towels slide down onto a table or shelf. An LFC with tilt functionality may also be used to transfer items from the LFC to the LFB. For example, two bots may travel next to each other and the LFC may be configured to move or guide one or more items into the LFB.

[0159]

[0276] Another possible configuration is that the LFC bristles may be placed on the bag so as to cover a portion of the opening of the bag. A small, fragile item (e.g. a ripe peach) may be placed on the LFC bristles. The bristles may then be lowered to the bottom of the bag and tilted so that only the item falls a very short height into the bag.

[0160]

[0277] Tray rack with distribution means underneath

[0278] As shown in Figure 41, tray racks may be designed to allow the LFB to go underneath and distribute items into the LFB. This creates a "flex" storage configuration that provides space under the overhang that may allow a larger rack to go in for replenishment. During peak times, this space can be used for static storage.

[0161]

[0279] Tray Rack Docking

[0280] The tray rack may be docked to the VLS so that the floor-running robot does not need to be perfectly aligned with the VLS, as shown in Figure 42. Handover can be accomplished with a kinematic constraint system using tooling balls, cups, cones, and / or wedges.

[0162]

[0281] LFT racks for delivering bulk goods for pick-up

[0282] In some cases, an LFT may dock to a VLS and a tray containing bulk items may be placed on the VLS. The LFT may be configured to move to receive the items, which are removed from the tray at that location. Alternatively, a dispensing mechanism may be used to dispense items to the bottom shelf of the LFT. The LFT may then move to receive the items, which are removed from the LFT at that location.

[0163]

[0283] Multiple LFB Decks

[0284] As shown in Figure 43, in a shuttle system, multiple LFBs can be provided to receive one or more dispensed items. Multiple LFBs can be distributed across multiple levels, decks, or floors. This can increase throughput and system density. In some cases, a tray picker shuttle can enter a row of dispense units horizontally and drop onto one or more dispense units.

[0164]

[0285] Small Format System

[0286] In some cases, the disclosed systems and methods may be implemented for facilities that are at least about 1,000 square feet (SQFT) and capable of processing tens of thousands of orders per week or more. In some cases, the disclosed systems and methods may be implemented for facilities that are at least about 100 SQFT, 200 SQFT, 300 SQFT, 400 SQFT, 500 SQFT, 600 SQFT, 700 SQFT, 800 SQFT, 900 SQFT, 1000 SQFT, 2000 SQFT, 3000 SQFT, 4000 SQFT, 5000 SQFT, or more. Some conventional storage and retrieval systems may not be compatible or suitable for storing and retrieving items in operating environments with small physical footprints. In such operating environments, smaller AS / RS form factors may be required. The disclosed systems and methods may be adapted and implemented for smaller operating environments while still maintaining high throughput and efficiency.

[0165]

[0287] In some embodiments, shuttle or mini-load technology may be used as the basis for the storage and retrieval system. The shuttle may fit better in facilities with low ceiling heights because it does not need to be tall and can be as long as the space requires. Such a system may have one or more features or characteristics of the system shown in Figures 2F and 2G, or may be a miniature version of the system shown in Figures 2F and 2G. The system may have a more specialized configuration, such as a shipping container or truck trailer with automation built into it.

[0166]

[0288] Container / Trailer Layout

[0289] As shown in Figure 44, a shipping container or trailer may be built and tested entirely in a low-cost manufacturing area and then transported to its final location. There may be no installation or configuration required other than adding power. This can be a very simple and low-cost way to expand quickly to new locations at low cost. In some cases, the container / trailer may be designed with one aisle in the middle for the shuttle to run down. The door end may have one or more distribution conveyors installed to accept trays and distribute one or more items to receptacles. The same shuttle may also be used to reload the trays into storage.

[0167]

[0290] An alternative is to use the AGV AS / RS design as a basis, as shown in Figure 45. In this layout, shelves may be tightly packed in a container. Popular SKUs may be located at or near the front. An AGV may be configured to travel under the shelves, lifting them up and moving them to a distribution conveyor. A lift may be used to pick a tray and present it above a distribution arm which removes the items and distributes them to supports or receptacles. The trays and shelves may be returned to storage until they are needed again.

[0168]

[0291] In some cases, a storage configuration with a gantry system (similar to the mini-load configuration described elsewhere herein) may be used for the container or trailer layout. Such a storage configuration may include two gantries on each side of the aisle, with a picker robot moving along or alongside the aisle to pick from each side of the tray storage. In some embodiments, multiple gantries may be provided on each side of the aisle. The multiple gantries may include at least 2, 3, 4, 5, 6, 7, 8, 9, 10 or more gantries. The multiple gantries may be arranged in a transverse configuration along the length of the aisle. In some cases, a storage row may have multiple gantries on each side of the storage row. The multiple gantries may be arranged along the length of the aisle. In some cases, the multiple gantries on each side of the storage row may be provided in different aisles located on different sides of the storage row.

[0169]

[0292] Double Shuttle Tray Picker

[0293] As shown in FIG. 46, one way to increase throughput is to provide a shuttle that can hold two trays at the same time. This allows the shuttle to pick one tray, drop it onto the distribution conveyor, go pick a second tray while the first tray is dispensing items, and then switch trays to ensure minimum cycle time between dispenses. In some cases, the shuttle may be configured to move in y and z. In other cases, the shuttle may be configured to move only in z or y. In some cases, the shuttle may have one or more levels to hold multiple trays.

[0170]

[0294] Multi-Temperature Containers / Trailers

[0295] As shown in FIG. 47, the container may include multiple temperature zones for storing various items at different temperatures. Automation may be designed to operate in each of these different environments. The container may have a frozen storage section at the rear, a refrigerated storage section in the middle, and an ambient section at the front. Shuttles and / or tray racks may be configured to move between the different environments as and when needed. Another alternative is to design a refrigerated barrier along the length of the container and have automation operate permanently on each side.

[0171]

[0296] Shuttle with distribution

[0297] As shown in FIG. 48, variations on the shuttle and distribution architecture may include a distribution system collocated with the shuttle. The shuttle may have a large hole in the center and rails on each side. Two distribution conveyors may be mounted on the rails and may move independently in the X direction. As a tray is pulled onto the shuttle, it may penetrate a slot in the tray, pick up one or more items, and advance the conveyor forward over the edge of the shuttle, after which one or more items are distributed from the end of the conveyor. Items may be distributed to receptacles that are separate from the shuttle.

[0172]

[0298] Shuttle with distribution and bag

[0299] Instead of dispensing from a conveyor to a receptacle off the shuttle, as shown in Figure 49, the receptacle may be on the shuttle. One or more receptacles may be located on one or both sides of the shuttle. In some cases, a movable mechanism on the shuttle may be used to remove the receptacle. The receptacle may comprise a bag or box. In some cases, the bag may comprise a collapsible bag where the bottom starts at the top and drops down (similar to an LFB).

[0173]

[0300] Tray

[0301] Inverted T-tray support concept

[0302] The present disclosure provides various configurations for supporting objects within a tray that operate interchangeably with the distribution conveyors disclosed elsewhere herein. For example, as shown in FIG. 50, one configuration may include a center beam support with divider walls and pop-through slots on the outside of the support. Another configuration may include an inverted T-shape with supports on the outside and pop-through slots on the inside. An inverted T may be more preferable for smaller lane trays, allowing round or cylindrical items to be lowered into the tray for increased storage density.

[0174]

[0303] Sliding Lane

[0304] The trays disclosed herein may include a platform that can be pulled or pushed. The trays may have no gap for a pop-through conveyor. The trays may have a low friction surface that allows the "lanes" containing the items to slide back and forth along the surface of the tray. The trays may have grooves for wheels or another feature to aid in the sliding of the lanes and ensure that they remain in the correct horizontal position.

[0175]

[0305] The trays disclosed herein may include one or more lanes. The lanes may contain items and may be configured to slide back and forth on the tray. As shown in FIG. 51 and FIG. 52, the tray may include multiple different types of lanes. The tray may also have a combination of various lane types and sizes. For example, the tray may include a Type A lane for standard item shapes (e.g., items with standard dimensions, standard shapes, etc.) that slide in a deterministic manner when pressed. The tray may also include a Type B lane for oddly shaped items (e.g., items not sold in uniform sizes or items that are not easy to slide). Each lane may be configured to hold a different item. In some cases, dividers may be used to separate items within a lane. The dividers may be easily insertable, removable, and adjustable in position along the length or width of the lane. In some cases, the dividers may have hinged flaps that slip under the items and facilitate sliding the items. In some cases, the containers may be designed to be adjustable in one or more dimensions.

[0176]

[0306] The location of one or more items can be precisely defined and recorded during loading and / or guidance into the tray with the aid of the novel lane container configuration and one or more dividers. Inspection algorithms can be configured to look for consistent features on the container and / or dividers to define the location of the item, rather than attempting to detect features of the item that may vary by thousands.

[0177]

[0307] Figure 53 shows an example of a dispensing operation for a single lane tray. The tray may need to extend out from the bay for the item to reach the LFB for dropping. A multi-axis system may be used to align the tray with the lane and push and / or pull the lane to slide along the tray. For Type A lanes, items may be loaded as close to the front of the tray as possible upon induction. The dispensing distance may be the length of the item. If the lane is not full, the distance may be adjusted based on induction image information.

[0178]

[0308] Clip-on Tray Dividers

[0309] Some items become difficult to control with the standard tray design, making their dispensing action unreliable. This can be addressed by using an add-on system that quickly clips onto or off the standard lane dividers as shown in Figure 54. This configuration allows existing trays to hold a variety of items by adjusting the height or thickness of the lane dividers per lane or by adding compliant features such as bristles.

[0179]

[0310] Sterilization tray

[0311] During normal operation, it may be necessary to periodically clean the dispense arm. As shown in Figure 55, the normal tray contents can be replaced with a downward facing UV light, sponges soaked in cleaning solution, etc. The disinfectant tray (grey) may be kept in storage as normal and picked by the elevator (blue). By picking this tray and pushing it against the dispense arm (green), the belt will collide with the disinfectant device (yellow) allowing the dispense tool to be cleaned without the need for human input or intervention.

[0180]

[0312] Tray flap

[0313] One limitation of existing trays is that for small items, the outer lip of the tray may need to be large enough to support the item. This reduces the space left available for the dispense arm to penetrate the tray. With less space, the dispense arm's stance becomes narrower, increasing the chance of the item tipping sideways. As shown in FIG. 56, one solution to this problem is to allow the tray divider (shown in blue) to pivot upward and outward as the dispense arm penetrates the tray. This means that the tray lip and the dispense arm can occupy the same protruding space. In this case, the dispense arm's stance can be made wider, thus providing more stability when receiving an item.

[0181]

[0314] Software Inspection + Quality Grading

[0315] Full Tray Imaging - Applications

[0316] As shown in FIG. 57, the trays can be imaged from above to inspect the items for quality control purposes. This can be beneficial at several stages of the process. For example, when manually loading trays, images of the entire tray may be taken during and / or after loading. This can be useful for identifying mistakes during loading such as wrong items, placement, orientation, etc., inspecting the quality of the items (packaged or perishable), and / or splitting out pictures of each item to present specific items to the consumer. Tray imaging can be done when automatically loading trays after quality inspection. This can be useful for identifying mistakes during loading such as wrong items, placement, orientation, etc., detecting when a tray is in storage or at a dispensing station, identifying and / or correcting mistakes during the dispensing or storage process, identifying features to improve the dispensing process, inspecting the quality of the items (packaged or perishable), verifying that the correct tray was presented, or verifying that the number of items per lane matches inventory expectations. Images can be taken with a standard camera, video, depth camera, or other type of imaging sensor.

[0182]

[0317] In some cases, the tray may be imaged from directly above or at an angle. In the latter case, a calibration operation may be performed to align the color pixel coordinates and depth data to a coordinate system parallel to the tray plane. The calibration operation may use a combination of RGB and depth data to extract the placement, size, and / or orientation of items in the tray for reporting to other automated systems.

[0183]

[0318] Full Tray Imaging Technology

[0319] The present disclosure provides a more effective approach and solution for combining depth data with RGB color data and implementing new machine learning methods to improve the use of localization in tray packaging verification.

[0184]

[0320] Inventory metadata may include item size, packaging type, and allowed orientation for a given tray type. RGB ML models may be trained to find and identify various package types on a tray and can alert operators when packages identified on a tray lane do not match inventory data. These models can also be trained at various product resolutions, such as individual items, SKUs, product categories, shapes, etc.

[0185]

[0321] In some cases, classification can be based on packaging type (i.e., glass, long neck, bottle, or Tetra Pak carton) as opposed to precise item identification, allowing for rapid adjustments to new product lines without additional model training.

[0186]

[0322] The localization data can be improved by estimating the X and Y coordinates of each item's center point, the height and width of the bounding box, and the θ of rotation at the center. These pixel-wise dimensions can be scaled using the calibration of the depth data to estimate or determine real-world dimensions and positions in three-dimensional space. An operator can be alerted if an item appears too large, if a miscount is detected, or if the item is tilted beyond the orientation constraints associated with the item. As shown in FIG. 58, items may be individually located and classified and aligned in lanes based on the location of the item's center. In some cases, RGB and depth data may be combined to mask items, and item distances to the leading edge of the tray may be reported to the operator. Lanes with all correct item types but improper counts may be flagged as possible inventory errors. Lanes containing incorrect packaging types may be flagged as possible dispensing rollover errors. The operator may receive alerts regarding various errors or potential errors flagged by the system.

[0187]

[0323] Bottom view of tray loading

[0324] As shown in FIG. 59, when manually loading a tray, the camera can be pointed down on the top of the tray to image everything facing up. In some cases, this can leave the bottom of the item uninspected. To solve this, a camera can be provided facing up next to the tray loading station. When a human (or pick-and-place robot) removes an item from the box, the item can be moved over the camera to trigger a photo and place the item in the tray. Analysis can then be performed on both the bottom and top photos to evaluate for quality control issues. Some alternative methods may include flipping the items individually or as a group so that both sides can be imaged. Flipping may include individual flipping, where each item can be flipped one at a time, for example, using a pick-and-place robot or a gantry system with suction cups or grippers. Alternatively, flipping may include group flipping, where an empty tray can be placed on top of a full tray after the top photo. Both trays (now with items between them) can then be flipped or inverted as a single unit. Another photo can then be taken again from above, but this time the bottom of the item may be facing the camera.

[0188]

[0325] Build to order + capture images for verification

[0326] In some cases, before and after images can be taken during the dispensing of an item to the LFB. The before and after images can be used to automatically detect whether the dispensing was successful.

[0189]

[0327] Throughout the order fulfillment process, one or more pre-dispense images and / or one or more post-dispense images can be compiled and associated with the associated LFB bag and particular customer order to create a complete history of the contents of each LFB at each stage of the fulfillment process. Each stage of the fulfillment process may correspond to the dispensing of one or more items into the LFB. The image history, including such dispensed images, can be stored locally on-site and / or in the cloud.

[0190]

[0328] In some cases, the compiled image history of each bag can be used for machine learning model training. For example, using the compiled image history of each bag and associated inputs regarding the items and item quantities dispensed into the bag at each step, a machine learning model can be trained to check future dispensing accuracy and automatically detect potential accuracy errors.

[0191]

[0329] In some cases, the compiled image history of each bag can be used for individual dispensing confirmation. The compiled image history can be used for internal purposes. For example, if a dispensing sensor (optical, planar, load cell, computer vision, machine learning, etc.) indicates that there was a potential error in dispensing an individual item, or if there is a discrepancy between two or more of these sensors, the associated dispensed image can be instantly displayed to an operator (either locally at the store or remotely via the cloud) for real-time or later review, as shown in FIG. 60. In some cases, the status of every active order, bag, and / or dispense can be monitored live by the operations team on-site at each store or at a remote control location via the cloud using a similar system. This system ensures that the order fulfillment of every order is 100% accurate before the item is sent to the customer.

[0192]

[0330] In some cases, the compiled image history can be used for one or more external purposes. For example, these images can be displayed to allow live updates to the customer as the order is being fulfilled. As shown in Figure 60, as items are added to each bag of the order, the customer can review the fulfillment accuracy, quality, and item damage, allowing live feedback to affect the fulfillment process.

[0193]

[0331] Build to Order + Capture Images for Verification (cont.)

[0332] In some cases, the image capture system disclosed herein can be used for complete order verification. For example, pairwise comparison of pre- and post-dispense images can be used to evaluate the success of one or more dispensing routines. Combining these images for all dispenses for a single bag, as shown in FIG. 61, allows the entire history of the bag through its fulfillment to be examined. This enables even more valuable applications. For example, an operator (on-site or remote) can easily verify the entire contents of a bag before sending it to a customer. Without a complete order history, this would not be possible since items at the top of the bag may obscure items below. This complete image history may eliminate the need for individual human operators to remove items from the bag if one wishes to verify that the bag is 100% accurate before sending the contents to a customer. This check may be done for every order, or for a subset of orders flagged as potentially problematic by an automated sensor or operator. In some cases, if an item is missing from a bag or if there is an extra item in the bag, the exact dispensing routine that caused the error can also be easily identified and investigated. The order history may include one or more video clips from each dispensing routine, rather than the still images shown in the exemplary illustrations on these slides. Video clips may provide more information about the items as they are dispensed into the bag (e.g., how the items fall in the bag after being dispensed, how they interact with other items, etc.), which may be particularly useful for damage assessment.

[0194]

[0333] In one aspect, the present disclosure provides an order fulfillment management system that includes one or more imaging devices configured to capture (i) a first data stream including a plurality of images or videos of a tray including one or more items, and (ii) a second data stream including a plurality of images or videos of a receiving unit configured to receive the one or more items, and a processor configured to generate a filtered set of images or videos using the first data stream and the second data stream. The filtered set of images or videos may correspond to (a) one or more features or characteristics of the one or more items, (b) one or more components or subsystems associated with the storage, handling, or dispensing of the one or more items, or (c) one or more parameters associated with the storage, handling, or dispensing of the one or more items.

[0195]

[0334] In some cases, the one or more components may include a tray, a storage unit including a tray, a dispensing unit, a transport unit, or a receiving unit. In some cases, the one or more features or characteristics of the one or more items may include a product type, a product shape, a product category, a product material, a packaging material, a product dimension, a product weight, or a stock keeping unit (SKU) associated with the one or more items. In some cases, the one or more characteristics may include a dimension, a shape, a color, a size, a weight, a material, or a material characteristic of the one or more items. In some cases, the one or more parameters may include an order associated with the one or more items, a dispensing instruction for the one or more items, a state of the receiving unit before or after the one or more items are dispensed, a fault detected in the dispensing of the one or more items, or a detection in or on the receiving unit of the one or more items. In some alternative embodiments, the one or more characteristics may include information obtained by reading or interpreting a label associated with or attached to the one or more items or a package of the one or more items. In some cases, the information obtained from the label may include a product code, a product name, a product weight, a product expiration date, product ingredients, and / or product nutritional information.

[0196]

[0335] In some embodiments, the first data stream may include one or more images or videos of a tray including one or more items before the one or more items are transported or dispensed from the tray. In some embodiments, the first data stream may include one or more images or videos of a tray after the one or more items have been transported or dispensed from the tray. In some embodiments, the second data stream may include one or more images or videos of a receiving unit before the one or more items are dispensed to the receiving unit. In some embodiments, the second data stream may include one or more images or videos of a receiving unit after the one or more items have been dispensed to the receiving unit.

[0197]

[0336] In some cases, the filtered set of images or videos may be generated based on user input or selection of criteria and information based on when the user wants to filter the images or videos. In some cases, the filtered set of images or videos may be organized in chronological order. In some cases, the filtered set of images or videos may be generated using a third data stream that includes a plurality of images of a dispensing unit transporting one or more items from a tray or dispensing one or more items to a receiving unit.

[0198]

[0337] In some cases, images or videos from the first and second data streams may be filtered or sorted by VLS, LFB, SKU, product category, product shape, product material, product weight, dispense sequence, how many SKUs are already in the receiving unit, the condition of the receiving unit before and after dispensing, failures of other sensors (that are part of the dispensing process), failure modes, or any combination of the aforementioned criteria or factors. Such filtering or sorting may also be performed for a particular tray and may take into account the number of items in the lane, lane size, VLS environment, SKUs, materials, etc.

[0199]

[0338] In some cases, the processor may be configured to track or detect the type of items dispensed, the amount of items dispensed, the dispense location of each item, the success rate of dispensing the items at the target location, the movement of the items during dispensing, or one or more characteristics of one or more items based on the set of filtered images. The processor may be configured to look for errors where the wrong product is dispensed or the wrong number is provided. The processor may track where the items were supposed to land in the receiving unit. The processor may track the characteristics of the items, the success rate of dispensing reaching the target, and / or where the items end up after dispensing to train one or more machine learning models that may be implemented to utilize that information to improve the quality and performance of future dispensing routines.

[0200]

[0339] In some cases, the processor may be configured to track whether an item contains glass and instruct the product handling system to avoid or prevent other glass items from being dropped directly onto the item. The processor may also track the movement of objects that may open like the clams of a berry upon impact. In some cases, the processor may track the movement of objects with protrusions (e.g., stems of products, corners of boxes, etc.) that may damage other things. In some cases, the processor may track the orientation of the object. This may help the processor control the operation of the product handling system to allow, for example, a peach to be dropped on the flat surface of a box, while avoiding or preventing a scenario where a peach falls on a corner of an item or product.

[0201]

[0340] In some cases, the processor may be further configured to perform object tracking to separate one or more items from background features. This allows for removal of one object from other objects in the bag. In some cases, the processor may be further configured to perform blob detection, edge detection, or contour detection to determine one or more attributes of one or more items and detect the quantity, orientation, or location of one or more items. In some cases, the processor may be configured to track or determine item counts and item locations. In some cases, the processor may be further configured to filter out background features based on a color, hue, tint, tone, or shade of the background features. In some cases, the processor may be further configured to filter out background features by (i) converting the images from a red-green-blue (RGB) color space to a hue-saturation-value (HSV) color space, or (ii) comparing depths using one or more depth point clouds. In some cases, a depth map may be used to determine the item filling results. In some cases, the processor may be configured to determine a probability distribution for where one or more items will be located or dispensed within the bag based on the multiple images, the one or more depth point clouds, and target locations for dispensing the one or more items.

[0202]

[0341] In some cases, the processor may be configured to automatically detect whether the one or more items were successfully dispensed. In some cases, the processor may be configured to automatically detect whether the one or more items failed to be dispensed. In some cases, the processor may be configured to compare the set of filtered images to a customer order to confirm successful dispense of the one or more items. In some embodiments, the processor may be configured to train one or more machine learning models using at least the set of filtered images to confirm dispensing accuracy and detect dispensing errors.

[0203]

[0342] In some embodiments, the processor may be configured to train one or more machine learning models based on properties or characteristics of one or more dispensed items, which properties or characteristics are determined based on one or more images or videos from the first data stream or the second data stream. Such properties or characteristics may be, for example, ingredients or quantities. In some cases, the processor may be configured to train one or more machine learning models using a subset of the images.

[0204]

[0343] In some cases, the processor may be configured to provide at least a subset of the images or videos to an operator for order confirmation upon detecting an error or a discrepancy between two or more dispensing sensors. In such cases, the operator may manually utilize the information to resolve the issue. In some cases, the system may further comprise a communication module configured to provide the filtered set of images to a customer to provide live updates during order fulfillment to enable the customer to (i) check fulfillment accuracy, quality, or item damage, or (ii) provide live feedback regarding the order fulfillment process.

[0205]

[0344] In some cases, the processor may be configured to build an order fulfillment history based on the set of filtered images and identify a dispensing routine that caused an error. The error may indicate, for example, a missing item, an incorrect item, or an extra item. In some cases, the processor may be configured to determine, based on the set of filtered images, one or more quality issues regarding one or more items that were damaged during dispensing. The processor may be configured to use computer vision to detect one or more dispensing actions that damage one or more items based on physical interactions between various items during dispensing (e.g., a first item in or on the receiving unit and a second item dispensed in or on the receiving unit).

[0206]

[0345] In some cases, the one or more machine learning models may be configured to determine or predict when one or more items that have a high risk of interaction may be dispensed together or next to each other. Such a high risk of interaction may be determined based on one or more characteristics or attributes of the one or more items. In some cases, a high risk of interaction may correspond to a risk of spillage, a risk of damaging one or more items, a risk of damaging the item packaging, or a hole in the item or item packaging.

[0207]

[0346] In some cases, the processor may be configured to verify the correct item type, quantity, position, orientation, sequence, or arrangement in one or more trays based on the first data stream or the second stream, or both. The first data stream may correspond to one or more items that may be or have been transported from the tray, and the second data stream may correspond to one or more items that have been dispensed into or onto the receiving unit. The first and second data streams may be evaluated together, as some errors may only appear in one of the two data streams.

[0208]

[0347] In some cases, the processor may be configured to use the first data stream and / or the second data stream to track which items have been dispensed, their quantities, the dispensed targets of each item, and associated characteristics of those items including dimensions, shape, color, size, weight, stiffness, etc.

[0209]

[0348] In some embodiments, the system may further comprise an imaging unit configured to image one or more items during item guidance or item storage to assess quality of the items, hi other embodiments, the system may further comprise an imaging unit configured to monitor movement or navigation of one or more robots and provide one or more corrective adjustments to the one or more robots based on the monitored movement or navigation.

[0210]

[0349] In some embodiments, the order fulfillment management system may include one or more imaging devices configured to capture multiple data streams. The multiple data streams may include, for example, image or video data. In one example, the system may capture multiple data streams of an item or a process related to the item. In some cases, the multiple data streams may correspond to when an item is placed on a tray (e.g., to verify the accuracy of item selection and whether the correct item is placed in the correct location), when an item is stored to check quality, when an item is in a tray before being lifted and dispensed (e.g., to verify the accuracy of item selection and whether an item is correctly positioned or oriented for lifting from a tray and dispensing), when an item is being dropped or falling (e.g., to monitor the dispensing process), when an item is being placed in a bag or package, when an item is coming to be placed in a bag or package, and / or when a consumer later submits a photo of an item (e.g., if an item is damaged or an incorrect item is selected or dispensed). In some cases, the multiple data streams may include images or video of images as an item is being dispensed (e.g., when an item is being dropped into a bag or package after a dispensing operation).

[0211]

[0350] In some embodiments, one or more live image / video streams may be analyzed in real-time (e.g., using one or more algorithms) to extract information about one or more items or subsystems (e.g., subsystems for dispensing, transporting, guiding, or handling various items or products) and provide immediate feedback to an operator / equipment. For example, real-time algorithmic analysis may include tracking items in space as they are rearranged within the system (e.g., while placing items in a storage tray or while moving items during a shuffle or item sorting operation), tracking items as they move from one system to another (e.g., during distribution of items from a tray to a receiving system or during movement or transportation of trays from a storage location to a distribution system), and / or tracking items as they are moved into a storage system by a human (e.g., as an operator loads an item into a tray lane). In some cases, real-time algorithmic analysis may include real-time validation and detection of failures / errors to standard actions / operations, which may be used to provide immediate feedback to an operator or equipment.

[0212]

[0351] In some embodiments, one or more sensors can be mounted on the actuation system, which allows the sensors to be repositioned or reoriented to optimally detect items and operations, events, or systems of interest during the operation / operation of the product handling and dispensing system. For example, a stepper motor can be used to control the position of a camera mounted on a linear rail. FIG. 85 shows a series of time steps in which the camera's field of view can be adjusted to capture the dispensing procedure. In some cases, the robot can be fixed in place. The camera on the linear rail can be actuated in a desired direction (e.g., toward the robot or bag) to track the receiving system to capture and detect useful information during the dispensing procedure. In another example, the camera can be mounted along one or more axes of the robot, which can manipulate the axes to move the camera to an optimal observation position of the item, event, or subsystem.

[0213]

[0352] Remote inspection of automated quality

[0353] In some embodiments, pre- and post-dispense images can be compiled and transmitted to a remote user interface that an operator can use to provide decision making and corrective action regarding the dispensing routine. These evaluations can occur in real-time or after execution of the dispensing routine, depending on the application.

[0214]

[0354] Similar methods of streaming images of automated equipment and routines to a remote user interface can be used for product quality assessment (e.g., high speed grading tools for product quality inspection during storage induction, evaluation during storage in a VLS system, or evaluation during dispensing or after the product has been dispensed into an LFB bag), remote evaluation of tray inspection characteristics (e.g., correct item type in each lane, correct item quantity in each lane, correct item orientation in each lane, whether items are lined up correctly in the tray lanes and not riding up on lane dividers or tray edges, or whether items are loaded in the correct position and / or with the correct spacing within each lane). The system may be applied to many other applications as well, including remote evaluation of the LFB (e.g., checking that the LFB / LFT / LFC is navigated correctly on a line along the floor or via floor markers, or checking supply corrective actions to recover from collisions and collisions of the LFB / LFT / LFC while in navigation), evaluation of the condition of navigation and distribution areas such as spilled items, obstructions, path damage, evaluation of the condition of guidance automation systems and VLS storage and distribution systems, and / or remote debugging and maintenance analysis of automation systems to improve uptime and response time to failure modes.

[0215]

[0355] How to inspect images to determine quality errors

[0356] In addition to determining dispensing accuracy, dispense images can be used to assess potential quality issues such as if an item was damaged during the dispensing process. By viewing the bag's complete image history of all dispense images / videos, an operator can manually assess the risk of damage at each stage of the fulfillment process. Figure 62 shows an example of making an assessment based on which item types were dropped into other item types.

[0216]

[0357] In some cases, computer vision algorithms can be used to detect where an item lands in a bag after dispensing, and the location of these items can be tracked during the fulfillment process. These systems allow for automated detection of cases where dispensing was at high risk of causing item damage, and can then route the bag for manual review by a human operator.

[0217]

[0358] Over time, this compiled data can be used to train machine learning models to predict when dispensing has occurred involving high-risk item interactions. In some instances, traditional CV and ML techniques can also be used to detect if significant damage has occurred during dispensing. Examples include detection of spilled liquids, damaged boxes, dented cans, etc. The same techniques may be applied to imagery from other stages such as induction, VLS storage, etc.

[0218]

[0359] mechanical induction design

[0360] Manual Guidance - A Higher Throughput Concept

[0361] As shown in Figure 63, to improve throughput during manual induction, instead of manually pulling out the trays, the drawers can be opened automatically. The drawers can be used to cycle through a bottom-up tray loader system and can present themselves to the operator at a faster rate, as shown in the image. In some instances, the elevator may load an empty tray into spot C. The empty tray may move to slot D once it is empty. The tray may move from slot D to slot A, and the operator may load items into the tray located in slot A. Once finished, the operator may press a button to confirm induction, one or more photos of the tray may be taken, and the tray may be moved to slot B. From there, the elevator may lift a tray containing one or more items from slot B. In some cases, the operator may be replaced by a robot. In some cases, a robot may be used to assist the operator.

[0219]

[0362] Automatic Guidance

[0363] In some cases, an operator may guide the cases from pre-guided storage to an autoloader. The autoloader may be configured to open the boxes, remove the items from the boxes, and place the items directly onto one or more trays. In some cases, a series of conveyors, an XYZ pick gantry, or a robotic arm may be used for autoguiding.

[0220]

[0364] Load cell for data collection during tray loading

[0365] As shown in Figure 64, load cells can be used to measure the weight of the item being placed in the receptacle as the item is placed in a tray or tote. This data can be used for quality control of the process. Tracking the total weight added allows the system to check that every item manually placed matches a known weight in a recorded database of items to be added. Load cells in different locations (e.g., in the four corners) can also be used to gather data on the location of the added weight by comparing the signal from each sensor.

[0221]

[0366] Breakbeam or LiDAR for measuring height

[0367] It is important to have very high density tray storage to get better value for the cost of the automated equipment. The VLS design uses a baker's rack type design, which allows the system to place trays in the most densely packed slots available. In order to do this, it may be necessary to know the height of the objects in the tray. If an item is taller than expected, there may be a possible error that could cause the item to collide with the tray above it during insertion.

[0222]

[0368] In some cases, a break beam may be used to measure the height of objects in the tray. The tray may pass through a series of break beams that are used to determine the maximum height of an item in the tray. If the bottom two beams are triggered and the third beam is not triggered, the item is said to not be as tall as the third beam. This sensor array may be located on the elevator so that it can detect the height of items as they enter or leave the elevator. It may also spot them during manual induction so that issues can be caught before they enter the system.

[0223]

[0369] In other cases, a planar LiDAR sensor may be used to measure the height of objects in the tray. A single lidar sensor may be positioned with its plane perpendicular to the elevator path. As the tray passes the lidar face of the sensor (which may include a single sensor rotating about a beam in a single plane), a trigger may be used to calculate the height of the items in the tray. The elevator position, tray thickness, and / or tray location may be determined. In some cases, the elevator position, tray thickness, and / or tray location may be known.

[0224]

[0370] Hybrid Guidance / Distribution VLS

[0371] A manual guide assembly may be positioned behind the dispense VLS as shown in Figure 65. In such a case, one VLS may be used for both manual guidance and dispensing items using the dispense assembly.

[0225]

[0372] Light placement

[0373] In some cases, LEDs may be placed under the tray to show the operator where in the tray the item should be loaded. In other cases, an overhead projector may be used to illuminate the location of the item in the tray. This lighting can be of any general shape. The lighting can also correspond to the shape / image of the item that is expected to go in that location. This is beneficial since items can be placed in the tray in many different orientations. The projection can properly project the correct orientation of the item in the tray.

[0226]

[0374] Laser Guide

[0375] In some cases, a laser can be mounted above the tray to point to the correct placement location. This laser could be a line pointing across the lane, or a dot pointing to a specific spot. This laser can also be mounted on an XY gantry or an X / Y pivot.

[0227]

[0376] Factory Software Control

[0377] Parallel and automated fulfillment

[0378] The systems and methods disclosed herein can be implemented to achieve automated fulfillment with parallel workflows. In some cases, there may be multiple distribution stations handling respective work queues. A dimensionless cost function may be utilized to autonomously make all of the decisions necessary to ensure not only fulfillment times but also the fastest and highest quality route. The dimensionless cost function may also take into account item damage, bag packing density, and product mix.

[0228]

[0379] In some cases, the system may need to select an order for the items, and may need to monitor and evaluate the state of the bag to determine the order of items to put in. This may require complex cost functions to be calculated and evaluated in a fully autonomous manner, as every LFB needs to actively move to reach its various destinations, unlike other systems that may move all items to one location for picking.

[0229]

[0380] Algorithm for determining route routing for fulfillment

[0381] In some cases, the system may be configured to implement an algorithm to determine path routing for fulfillment. As shown in FIG. 66, the system may utilize graph search with custom heuristics to penalize occupied nodes, certain edges, and extra turns that take time. Robots may be assigned a set of nodes to cover at a time, and these nodes may be reserved for a certain period of time or during the fulfillment of a certain order so that other robots cannot use them until the nodes are unreserved. To account for the ever-changing map, robots may be configured to recalculate their path if they are stuck in traffic for a certain amount of time or if they are deadlocked with another robot trying to reach the current node. In some cases, the system may provide a custom-designed path network that includes various types of edges and queue locations, including highways and local roads. Robots may only take certain edges based on their destination. For example, robots going to another section of the map may take the highway, while robots going to a distribution location may exit the highway onto local roads.

[0230]

[0382] Algorithm for determining distribution order

[0383] In some cases, the system may be configured to implement an algorithm to determine the dispensing order. The system may be configured to calculate the cost of each item based on the item's mass or volume, drop surface area, and / or packaging material. Objects with a mass greater than the drop surface area may have a higher cost. Items with harder packaging materials, such as aluminum or steel, may also have a higher cost due to multiple factors. The drop order may then be calculated based on these absolute costs. If the difference in drop cost between two items in the same storage bay is greater than the difference in drop cost between the smaller item and an item in another bay, the robot may choose to go to the other bay first to retrieve that item.

[0231]

[0384] Algorithms for determining inventory location optimization

[0385] In some cases, the system may be configured to implement algorithms for determining and optimizing inventory locations. For example, the system may distribute copies of the same product to different areas on a map to increase resilience to individual machine failures. Within a storage bay, more frequently picked items may be placed in trays with lower slots to reduce lift travel time. Items that are frequently purchased together may be placed in the same tray to reduce the number of trays that need to be retrieved.

[0232]

[0386] Algorithms for determining the location of items among all options when selecting routes and distribution orders

[0387] As shown in FIG. 67, the system may be configured to implement a weighted cost optimization algorithm when determining the particular items and drop order for the bag robot. A cost is calculated for each feature the system cares about, such as travel time to the distribution location, how busy the distribution location is, and the relative order in which to drop this item to minimize damage within the bag. These costs may be weighted together to determine the final cost for each item. Additional cost functions may be easily added to the equation without modifying the code.

[0233]

[0388] In some cases, instead of selecting all items up front, the system may be configured to pick only items that are in the next destination in a "decide as you go" approach, allowing the system to adapt to potential future condition changes that may occur if a machine goes down, a particular item becomes unavailable, or traffic increases in a particular area of ​​the map.

[0234]

[0389] Product handling rules (data systems / algorithms)

[0390] In some cases, the system may be configured to allow a system operator to set flexible rule sets for how different products can be handled at a fulfillment center. The system may be further configured to allow the factory control software module to follow one or more rules set by the system operator.

[0235]

[0391] The system may include multiple layers. The first layer of the system may include a set of editable "features" or "tags" that may be assigned to products when they are loaded (metadata entered to enable handling and sale). Some example tags may include drip risk, unpackaged, chemical, food, non-food, meat or fish. The second layer of the system may include a set of editable "rule operators" related to decisions that the software may make when planning various processes. The rule operators may relate to which items may or may not be stored above or below other items, or which items may or may not be packed in a bag with other items. The final layer of the system may include a complete set of "rule statements" that combine features and rule operators. Rule statements may include, for example, [drip risk + meat or fish][cannot be stored above][no packaging], or [chemical][cannot be packed in a bag][food]. These rules are editable in a graphical user interface. Different facilities may follow different sets of rules (set by different partners or by the grocery store's customers). The rules may be accessed and adjusted or updated by the control software via an API to constrain product handling decisions such as determining where to place trays, how to split an order into multiple bags, etc. In some embodiments, machine learning algorithms may be trained to suggest feature classifications from images of new products to make onboarding faster.

[0236]

[0392] Track your bag status

[0393] In some cases, an algorithm can be implemented to select the optimal target for distributing items into LFB bags using three main inputs: RGB images, depth point clouds, and item attributes (dimensions, shape, weight, etc.).

[0237]

[0394] In some cases, bag status tracking may be implemented as shown in Figure 68. Bag status tracking may involve storing, updating, and retrieving important information about the status of the LFB throughout the fulfillment process, which can be used to optimize the fulfillment process in various ways. One way is by improving the distribution targeting algorithm during distribution.

[0238]

[0395] The bag status may be composed of many types of information, but important components include an RGB image associated with the bag (most recent pre-dispense and post-dispense images), the entire compiled history of images / videos, a depth grid map of the bag, the average depth of each region of the bag, the number of depth points detected in each region of the bag, the maximum depth of each region of the bag, the minimum depth of each region of the bag, the depth variance of each region of the bag, an item grid map, the location of each item already in the bag, and related information such as the item's material properties (dimensions, color, weight, size, shape, damage properties, etc.), product ID with associated metadata, or the level at which the item is located (on or below the bag). Alternatively or in addition, the item locations may be represented as probability distributions.

[0239]

[0396] As shown in FIG. 69, the resolution of the depth grid map or item grid map can be adjusted to balance accuracy and processing speed. In some cases, different resolutions may be used for the depth grid map for the average depth of different bag regions. The depth and item grid maps can be configured to provide information not only about items on the surface of the bag, but also about items deep within the bag that may be covered by items on the surface. This is important to mitigate damage to items that may be impacted during dispensing, even if they are not on the surface of the bag.

[0240]

[0397] Tracking the state of the bags during fulfillment can be used to optimize distribution targeting algorithms such as optimizing fulfillment speed and LFB routing; using the state information to reduce the number of images taken of the LFB during each distribution routine (the after image of one distribution routine can be used as the before image of the next routine); simulating and evaluating whether there is a viable target for the next item distribution to the LFB based on the current state (e.g., avoiding wasting time having to send an LFB to the VLS and then reject it); real-time selection of the best LFB for distribution of an individual item; optimizing the order in which items are distributed to bags to optimize speed (e.g., simulating the order before fulfillment begins). improving bag packing efficiency (getting as many items as possible into one bag) by optimizing the division of an order's items into various bags (e.g., through simulation of the order before fulfillment begins, or through simulation in real time as the LFB is deployed and collecting items, or by detecting if items have moved during distribution), mitigating damage during distribution by providing more appropriate distribution targets based on the known location of the current item in the bag and its damage characteristics, or other applications in addition to improved monitoring (on-site or remote) of the LFB status during fulfillment for monitoring and debugging purposes.

[0241]

[0398] Distribution sequence optimization factors / cost characteristics

[0399] Optimization factors and cost features may be related to throughput. Such factors and features may include, for example, LFB travel time to the bay, bay status (busy / free), robot traffic, and robot trajectory.

[0242]

[0400] Optimization factors and cost features may be associated with damage mitigation. Such factors and features may include item damageability "scores" based on individual item characteristics, i.e. mass, drop surface area, packaging material, sharpness, mass / drop surface area, likelihood of being damaged ("fragility score") or of causing damage ("damage score"). In some cases, damage mitigation factors and features may be mapped to an interaction matrix between items by individual product and / or by product group characteristics.

[0243]

[0401] Optimization factors and cost features may be related to bag packing efficiency and bag handleability. Such factors and features may be used to prevent items from jumping out of the bag and to determine if the bag is below a threshold weight.

[0244]

[0402] Optimization factors and cost characteristics may be related to freshness and food safety. Such factors and characteristics may be used to determine whether an item is perishable and whether it should be temperature controlled for later distribution.

[0245]

[0403] In any of the embodiments described herein, the system may provide a visual presentation of the order to the consumer. The visual presentation may correspond to one or more steps of order fulfillment or to a particular item or set of items in the order.

[0246]

[0404] algorithm

[0405] The algorithm may include one or more weighted cost algorithms. In some cases, the algorithm may use mass / surface ratios or material damage characteristics to make a decision.

[0247]

[0406] The algorithm may use deterministic planning. The system may select the optimal distribution order before the bags are deployed based on the expected state of the bays and inventory at the time of the order. In some cases, the algorithm may simulate or predict the availability of bays, future state of the bags, space for new items, and interactions between items during the drop.

[0248]

[0407] In some cases, the algorithm may be a traveling salesman algorithm. In such cases, the systems and methods may be implemented to provide a baseline plan that may be updated with other algorithms during fulfillment.

[0249]

[0408] In some cases, the algorithm may implement "decision-as-you-go" with online updates. The algorithm may adapt to state changes during fulfillment, including bag status, bay status, error handling, changes in inventory status, and / or changes in LFB traffic. The algorithm may check the cost of an upcoming single distribution and / or predict a certain number of steps in a simulated rollout.

[0250]

[0409] In some cases, the algorithm may implement one or more learning methods, which may include, for example, data mining of past orders to optimize a weighted cost algorithm, or reinforcement learning with rewards based on priority factors.

[0251]

[0410] In some cases, if one or more aspects of the algorithm or simulation fail, the algorithm may allow for human-in-the-loop and / or manual intervention, which may include end-consumer intervention during fulfillment.

[0252]

[0411] Status Information

[0412] In some cases, optimization algorithms may be configured to allow for dynamic adjustments to robot planning and routing during fulfillment. These optimization algorithms may utilize various components, characteristics, or features of the factory conditions at a given time, including, for example, inventory information (e.g., availability of items and item location, or ease of access to inventory based on the status of the bay, tray, and / or lift), storage bay information (e.g., is the bay in use, free, down for maintenance, or is there another robot scheduled to use that bay?), LFB information (e.g., current location of LFBs, planned trajectory and planned sequence of each LFB's distribution, planned trajectory to reach the next distribution, or battery levels), and / or bag information (e.g., bag condition including latest image, depth information, location of items within the bag).

[0253]

[0413] Parallel and automated fulfillment

[0414] In one aspect, the present disclosure provides an automated fulfillment system comprising: a plurality of dispensing stations with movable mechanisms for dispensing one or more items; a plurality of receiving units configured to move along one or more adjustable paths to one or more of the plurality of dispensing stations to receive one or more items dispensed from the plurality of dispensing stations; and one or more processors configured to modify or update the one or more adjustable paths for at least one of the plurality of receiving units based on an algorithm including a cost function that optimizes one or more order fulfillment metrics based on (i) the availability at the dispensing stations of the one or more items and (ii) the availability of one or more of the plurality of receiving units to receive the one or more items. The one or more modified or updated paths may be optimized to reduce the probability of item damage during or after the dispensing routine, avoid or mitigate the occurrence of one or more failure modes, or address one or more customer preferences. In some embodiments, the one or more order fulfillment metrics may include, for example, throughput, power efficiency, order fulfillment speed, packing density, or item compatibility.

[0254]

[0415] In some cases, the processor may be configured to update the one or more adjustable paths when a new order is received. In some cases, the processor may be configured to update the one or more adjustable paths when one or more additional receiving units are deployed to fulfill a new order. In some cases, the processor may be configured to update the one or more adjustable paths by changing the order or sequence in which one or more receiving units travel to one or more of the plurality of dispensing stations to retrieve items of an order. In some cases, the processor may be configured to update the one or more adjustable paths by changing the order or sequence in which one or more of the receiving units retrieve one or more items of an order.

[0255]

[0416] The processor may be configured to use a cost function algorithm that tracks various order fulfillment characteristics to determine the best combination of options for operating the system. The order in which the receiving units head to each distribution station may change over time and / or instantaneously based on new information or new orders. Furthermore, storage areas may be associated with specific distribution stations or distribution tools such that the distribution station / distribution tool only distributes items received or leaving from the specific storage area. The fulfillment system may comprise a series of pick-and-place robots with AGVs moving between them. The fulfillment system of the present disclosure may be implemented to provide a low-cost distribution solution that allows the use of multiple distribution tools to allow higher throughput. The specific distribution order of any order may be changed to obtain any value, such as increased speed / throughput, reduced damage and failure modes, satisfaction of customer control / preferences, increased packing density, etc.

[0256]

[0417] In some cases, the algorithm may be configured to determine optimal path routing for at least a subset of the plurality of receiving units. In some cases, the algorithm may be configured to assign a set of nodes to one or more receiving units to cover at a time. Nodes may be selectively reserved such that other receiving units cannot use the nodes until the reservation is released. In some cases, the algorithm may be configured to perform a graph search using custom heuristics that penalize occupied nodes and additional time-consuming moves.

[0257]

[0418] In some cases, the algorithm may be configured to take into account an ever-changing map and recalculate optimal path routing for multiple receiving units when a receiving unit is stuck in traffic for a given time or when one or more receiving units are stuck between another receiving unit trying to cover the same node. In some cases, the algorithm may be configured to generate or update a path network that includes different types of paths and waiting locations so that one or more receiving units can take only a portion of a path based on their destination.

[0258]

[0419] In some cases, the algorithm may be configured to determine a dispense order for one or more items dispensed by one or more dispense stations. In some cases, the algorithm may be configured to calculate a cost value for each item based on the mass of the item, the drop surface area, or the packaging material of the item. The algorithm may be configured to assign higher cost values ​​to objects with greater mass over the drop surface area. The algorithm may be configured to assign higher cost values ​​to objects with harder packaging materials using a multiplication factor. In some cases, the algorithm may be configured to determine an optimal drop order based on a cost value associated with each item. In some cases, the algorithm may be configured to instruct the robot to move to a second dispense station to retrieve one or more items of the customer order if a difference in drop cost value between a first item and a second item at the first dispense station is greater than a difference in drop cost value between the first or second item and a third item at the second dispense station.

[0259]

[0420] In some embodiments, the algorithm may be configured to determine an optimal location for inventory storage of one or more items. In some cases, the one or more processors may be configured to distribute multiple items across multiple different storage units to increase resistance to individual machine failures, where the multiple items include the same item. In some cases, the one or more processors may be configured to place more frequently picked items in one or more trays located on lower shelves of a storage unit to reduce lift travel time. In some cases, the one or more processors may be configured to place items that are frequently purchased together in the same tray to reduce the number of trays that need to be retrieved for order fulfillment.

[0260]

[0421] In some cases, the algorithm may be configured to determine optimal item locations when selecting a route and distribution order. In such cases, the algorithm may be configured to perform a weighted cost optimization based on at least one of: (i) travel time to a distribution location, (ii) how busy the distribution location is, and (iii) the relative order in which one or more items should be dropped to minimize damage to the items during a distribution routine. In some embodiments, the algorithm may be configured to weight one or more costs together to determine a final cost for each item.

[0261]

[0422] In some cases, the algorithm may be configured to determine the pick order of items based on potential condition changes corresponding to machine failure, item unavailability, or increased traffic in an area near one or more distribution stations. In some cases, the algorithm may be configured to direct one or more robots to an order delivery location when no unpacked items remain for a customer order. In some cases, the algorithm may be configured to (i) calculate a weighted cost value for each available tray, (ii) select a destination with the lowest cost value, (iii) select one or more items from the destination with the lowest cost value, and (iv) instruct one or more robots to retrieve one or more items from the tray or trays associated with the destination with the lowest cost value when unpacked items remain for a particular order.

[0262]

[0423] In some embodiments, the algorithm may be implemented using one or more predefined rules for handling one or more items. The one or more predefined rules may be established based at least in part on a set of editable features or tags that are assigned to one or more items when the one or more items are received or directed. The set of editable features or tags may include, for example, drip risk, unpackaged, chemical, food, non-food, food type, or non-food type designation. In some cases, the one or more predefined rules may be established based on a set of editable rule operators that correspond to one or more decisions related to order fulfillment. In some cases, the one or more decisions may include a determination of whether an item can be stored on top of another item or whether an item can be bagged with another item. In some cases, the one or more predefined rules may be established based on a set of editable rule statements that combine one or more features and one or more rule operators. In some cases, the one or more predefined rules may be adjustable to constrain product handling decisions.

[0263]

[0424] In some embodiments, the algorithm may include a machine learning algorithm configured to generate one or more feature classifications from one or more images of one or more newly received or derived items to facilitate onboarding.

[0264]

[0425] Item Quality Tracking

[0426] As shown in Figure 70, the present disclosure provides a method to inspect dispensed images to predict when item damage may occur, or to directly detect damage that has occurred. In either case, the LFB bags can be directed to an area where a human can see the severity and potential cause of the damage. A compiled history of bag dispensed images and / or videos can be used to further pinpoint where the damage occurred.

[0265]

[0427] Each time damage to an item is detected and / or confirmed by a human operator, the resulting data can be stored and tracked for further use. Data mining and machine learning algorithms can be used to uncover general trends and rules governing interactions between items that lead to damage. Big data analytics can make it possible to understand (1) the risk of damage to item type A when distributing item type A to item type B, (2) the risk of damage to item type B when distributing item type A to item type B (example diagram, bottom left), (3) the general risk to item type A when item type A is distributed, and (4) the general risk to item type B when any item is distributed.

[0266]

[0428] In the above case, an item “type” can be defined as an individual SKU, product type, product group, or material packaging group (metal, plastic, produce, etc.). Big data analytics can effectively find trends at all these “type” resolutions. These findings, in conjunction with bag condition tracking, can inform algorithms for selecting the optimal dispense location within the bag. Additionally, each dispense can be given a score for the likelihood that damage has occurred based on the above findings and the detected location if the item lands within the bag. If the risk score of a particular bag for a particular item, particular dispense, or cumulative dispenses becomes too high, the LFB can be flagged for manual review (by a human operator or remotely using image / video history) before being sent to the customer.

[0267]

[0429] Depth Camera and Distributed Target Algorithm

[0430] As shown in Figure 71, a 3D depth camera and / or an RGB camera can be used to provide targets for dispensing items into bags. In some cases, HSV filtering of the RGB image stream can be used to remove items within the bag from the bag itself. An example of the use of green hue HSV filtering is shown in the top right. Using the HSV color space allows for bag detection despite changing lighting conditions and bag texture. This is important to distinguish which corresponding depth data points refer to items and not the bag itself.

[0268]

[0431] The dispense target algorithm can also utilize a physical model of how an item will fall into the bag when dispensed. The bottom right diagram shows an example of this. That is, the algorithm looks for the best possible target, assuming that the item will land in the bag at a 90 degree rotation from where it is placed on the VLS tray. A swing collision region is also evaluated for each potential target location based on the item's dimensions to ensure that the dispensed item does not collide with items already in the bag. In some embodiments, the dispense target algorithm can also provide dispense targets for multiple items simultaneously to improve system throughput. In some cases, the machine learning model can take RGB images, depth data, and item dimension information as inputs and output the optimal drop location based on a set of labeled training data compiled from past dispenses.

[0269]

[0432] In an alternate mode of system operation, an RGB image of the LFB bag can be sent to an operator interface with an overlay of targets showing the size of the item dimensions. The operator can manually select the optimal dispense target on the image, which can be instantly fed back to the system in real time to influence the dispense routine. The operator can be on-site or remote. This system can be used for any dispense or when the automatic dispense target algorithm fails.

[0270]

[0433] Automatic tray "BreakPack" loading

[0434] Break-pack is the process of splitting a quantity delivered by a vendor (for example, a 12-pack of wine bottles) into smaller quantities for delivery to a fulfillment center. This is useful, for example, when a fulfillment center sells wine infrequently, so storing all 12 bottles in-house would be an inefficient use of space. In a hub-and-spoke fulfillment network, large volumes of all SKUs may be delivered to a hub location. Nearly homogenous trays (with few individual products) can be loaded with high efficiency.

[0271]

[0435] As shown in FIG. 72, in the tray reloading system depicted and described herein, items may be transported from the homogenous trays on the left onto a set of shuttle conveyors in the center. The shuttle conveyors may move to the tray lanes on the right and deposit the items into a mixed tray. The shuttle conveyors may be configured to feed multiple tiers of a distribution assembly (page farther back). The mixed trays may be shipped to a "spoke" facility with the exact amount of trays and / or items required by the facility. This system may also be coordinated with a "shuttle" tray pick system that allows a tray pick robot to move up and down the aisles. In some cases, this system may also be used to empty the contents of one tray into another tray so it can be returned empty to the hub facility.

[0272]

[0436] Customer Interface

[0437] Customer facing VLS and pick-up side locker system

[0438] As shown in FIG. 73, to provide buffer storage for pickup and better optimize customer interaction, the systems and methods of the present disclosure may be used to collect items of an order into multiple bags using a robot, to grab the bags from the robot and move one or more items from the bags to a tray using an automated system or subsystem, to store the tray in a buffer VLS in suitable environmental conditions, and to move the tray and / or tray rack by a conveyor or robot to a customer interface system when the customer is ready to pick up or receive the items on the tray. In some cases, the customer interface may be a locker type interface or other type of high density / access point interface.

[0273]

[0439] Storage before induction

[0440] Moving trays within a container

[0441] Induction (i.e., loading of items onto or into the trays) may occur at a central location or at the manufacturer or distributor of the items or goods. Full trays may be shipped to smaller hub stores where they are stored until purchased by a customer. Either the shuttle or the AGV AS / RS system disclosed herein may be used for both storage and transportation of one or more trays.

[0274]

[0442] Floor lock system for tray racks

[0443] When tray racks are used, as shown in FIG. 74, the racks can be driven and positioned so that their feet are inserted into a floor and / or ceiling device that locks the rack in place to minimize or prevent movement during transport. The locks may be designed so as not to interfere with a rack-lifting robot that will move underneath the trays. The locks may be activated by the rack-lifting robot when the rack is inserted and released when the rack-lifting robot is ready to pick up the rack.

[0275]

[0444] Tray rack wrapping mechanism

[0445] As shown in FIG. 75, tray racks can be designed to prevent items from falling or being damaged. One approach is to use a door at the front where the trays are inserted. The door can be configured to rotate and have stops on both sides. Another approach is to wrap the rack in a flexible or rigid material. In some cases, a blanket or plastic wrap can be used if one or more items on the rack are soft. In some cases, a wire cage or one or more walls can be used if one or more items on the rack are rigid.

[0276]

[0446] Inflatable mechanism on tray

[0447] In some cases, one or more inflatable devices may be placed over the tray to hold the items in place and protect them, in other cases the items may be wrapped in netting or other covering material.

[0277]

[0448] Storage system design

[0449] Freezer airlock

[0450] As shown in Figure 76, the storage system may have double entry doors that provide an airlock to limit moisture and temperature infiltration into the conditioned space. For low temperature storage, the dew point temperature differential to the distribution area may be the most extreme. Thus, limiting infiltration may help reduce the cooling requirements of the system, reduce power requirements, and improve storage conditions of the food. Additionally, reducing moisture infiltration may help reduce or minimize the occurrence of frost and condensation on the stored products.

[0278]

[0451] High temperature VLS

[0452] In some cases, the VLS may be adjusted for dry bulb and dew point temperatures to keep cooked foods ready to serve. For example, the VLS may be configured to maintain a temperature of at least about 60° C. The VLS may be configured to maintain temperature uniformity at a minimum rate, utilize waste heat from other cooling systems for heating, control temperature, and adjust humidity up or down depending on the items stored in the VLS.

[0279]

[0453] Duct for VLS air distribution

[0454] A cooling system may be connected to the VLS through supply and return ducts as shown in Figure 77. The duct system may be configured to distribute air throughout the trays at various levels throughout the VLS. The ducts may be configured to maintain air temperature and product uniformity by distributing the air. Distributing the air over a wide area in this manner may help control humidity.

[0280]

[0455] Add-on Storage VLS

[0456] As shown in Figure 78, a VLS may have a shaft in the middle with columns of trays at the front and back. In some cases, the storage capacity of this design may be limited since it can only scale in high ceiling environments. The solution to this is to add more VLS with no dispensing capacity to the back. The storage will be lowered to the floor. This will allow more storage per dispense, thus leveraging the dispense cost.

[0281]

[0457] In some cases, trays may be transported between two or more elevator shafts using conveyors. Such tray passing conveyors may be located anywhere along the height of the elevator shaft.

[0282]

[0458] VLS front room concept

[0459] Add-on storage may be used for refrigerated and / or frozen storage of items, as shown in Figure 79. The pass-through conveyor may have at least one door to a refrigerated VLS. The refrigerated VLS may act as an antechamber to keep condensation low and minimize cooling losses.

[0283]

[0460] Tray Down Lane Divider

[0461] Referring to FIG. 86, in some cases, one or more tray dividers may be used to organize various items in a tray. The one or more dividers may include, for example, one or more tray down lane dividers. As shown in the top panel of FIG. 86, some tray configurations may hold one product per lane, and items may not be stacked on top of each other. The bottom panel of FIG. 86 shows examples of lane dividers that may be added to any portion of a lane of a tray to create a lane section in which items may be stacked. The item stacks may be kept stable and in the same place as the tray is moved using an automated system. In some cases, a lane may have multiple dividers. In some cases, the lane dividers may have various heights. In some non-limiting embodiments, the lane dividers may include clip-on lane dividers. In some non-limiting embodiments, items in a tray with one or more dividers may be picked from above by a gripper or suction robot.

[0284]

[0462] Automated order staging and pick-up system (storage lockers)

[0463] In one aspect, the present disclosure provides a system for automated order staging and pickup. An exemplary system configured for automated order staging and pickup is shown in FIG. 87. In some cases, mobile robots, each carrying one or more filled shopping bags, can stop at one or more stations. Stationary robots at one or more stations can pick one or more bags from the mobile robots and place the bags on one or more trays. The trays may be configured to hold multiple shopping bags, possibly from different robots. In some embodiments, the trays can also hold products that are not suitable for automated pick and pack, such as very bulky items.

[0285]

[0464] In some cases, the trays can be transported by a shuttle or gantry tray transport robot. The tray transport robot can move the trays vertically and horizontally along the aisles of the storage location. In some cases, the tray transport can pick up the trays from the storage location or locker and place the trays in the storage location or locker. Some storage locations or lockers can be accessible through automatic doors. In some cases, one or more entities (e.g., humans or robots) can pick up filled shopping bags and / or other groceries from the trays stored in the locker. In some cases, there may also be automatic doors at the back of the locker location to increase safety and prevent people from reaching the work area of ​​the tray transport robot. In some cases, the storage locations or lockers can be temperature controlled (e.g., refrigerator or freezer temperature ranges).

[0286]

[0465] Automated order pickup system (robots in automated bays)

[0466] In another aspect, the present disclosure provides a system for automated order pickup. In some cases, one or more robots may be configured to run under a receipt printer before entering a pickup bay. In some cases, an automatic cut label printer may deposit one or more receipts into one or more bags. The printing station may include a printer mounted on a stand with a "chute" to direct the cut labels into the robot's bag. In some cases, multiple printers may operate in parallel to simultaneously print multiple receipts individually for delivery to one or more bags.

[0287]

[0467] In some cases, customers can pick up bags directly from the robot. The robot can enter the pickup bay from the automation system side and docks in the pickup bay. In some cases, the pickup bay has an angled plate that guides the robot to a fixed position as it moves into position. In some cases, the pickup bay has a lid that is an interface between the automation system side and the human operation side. In some cases, the pickup bay may have a tablet for people to interact with the pickup bay or to view information about order pickup. In some cases, the pickup bay may include one or more lights to further guide or inform human interaction. In some cases, the pickup bay may have recessed charging prongs to allow the robot to charge while "docked."

[0288]

[0468] In some cases, the lid of the pickup bay may be locked in place (docked) when the robot is parked and locked closed until it is powered off or stopped. In some cases, a "roadblock" mechanism may be used to prevent the robot from rolling backwards in the bay. In some cases, metal bars may be inserted into the bot's "wheel well" to reduce the likelihood of an emergency stop scenario. In some cases, the lid may be configured to automatically unlock for the customer when the robot is parked and emergency stopped. If one or more sensors detect that the robot has been manually pushed out of the pickup bay, at least one or more of the other remaining bots in the facility may be emergency stopped. In some cases, a safety PLC may be used to control this logic. In some cases, the lid sensor may be used to detect if the lid has been pried open or electronically failed, which can be used to initiate an emergency stop protocol.

[0289]

[0469] In some cases, the pickup bay lid may be configured to close automatically. In some cases, an electronic release mechanism may release and close the lid after one or more events occur. Events may include, for example, the robot detecting that a bag has been removed, removal of a bag within a predetermined time, failure to remove a bag within a predetermined time, a customer clicking a button on a tablet interface, and / or a customer pressing a lid release button. Events may be software programmable. In some cases, the lid closure may be dampened in the closing direction to prevent slamming.

[0290]

[0470] 88 illustrates an example configuration of a pickup system. The pickup system may include multiple pickup bays configured to accommodate one or more items or a bag containing one or more items. In some cases, the pickup bays may include movable doors or windows configured to provide access to the items or bags in the pickup bays. The pickup bays may allow a human or robot to interact with the movable doors or windows to retrieve the items or bags, as shown in FIG.

[0291]

[0471] FIG. 90 illustrates an example of a pickup bay in schematic form. The pickup bay may include a lid. The lid may be an auto-closing lid. In some cases, the pickup bay may include a lid sensor and a solenoid lock to limit movement of the lid relative to the body of the bay. In some cases, the pickup bay may include an embedded control printed circuit board (PCB). In some embodiments, the pickup bay may include a robot presence sensor to detect if a robot is in proximity to or docked with the bay. In some embodiments, the pickup bay may include a charging port and / or a robot auto-stop. The robot auto-stop may include components that physically or electronically limit the movement or motion of the robot (e.g., by disconnecting a circuit or wire, or by using software to trigger a locking mechanism or stop condition). In some cases, the pickup bay may include LED lighting to indicate whether the pickup bay is in use, whether a robot is docked or undocked from the pickup bay, or whether the pickup bay is ready for access or unavailable.

[0292]

[0472] FIG. 91 illustrates an exemplary interaction between a robot (e.g., a line tracked bag or LFB) and a pickup bay. The pickup bay may initially be empty. The robot may dock with the pickup bay. In some cases, a robot presence sensor may be activated. If the sensor detects the presence of the robot, the lift may lower and the bag hammer may open. The robot may then stop. In some cases, a light may change to indicate that the robot has stopped. After the robot has stopped, a solenoid latch may open, the pickup bay lid may open, and a load block may engage. When the bag is removed (e.g., by the robot or a human such as a consumer), the pickup bay lid may close. In some cases, the closure of the lid may be confirmed using a sensor (e.g., a touch sensor). The robot emergency stop may then be removed, the lift may be raised, and in some non-limiting embodiments, the bag hammer may close. The LFB may then undock from the bay and move to another location (e.g., to retrieve one or more other bags associated with the customer order, or to transport the items or trays to a storage bay or dispensing station).

[0293]

[0473] Computer Systems

[0474] In an aspect, the present disclosure provides a computer system programmed or otherwise configured to perform any of the methods of the present disclosure, e.g., the subject product handling method. FIG. 80 illustrates a computer system 8001 programmed or otherwise configured to perform the product handling method. The computer system 8001 may be configured, for example, to change or update a distribution order of one or more items, or to adjust one or more movement paths of one or more receiving units using a cost function that optimizes one or more order fulfillment metrics based on the availability of one or more items at a distribution station, the availability of one or more of a plurality of receiving units to receive one or more items, or one or more newly received customer orders. The computer system 8001 may be a user's electronic device or a computer system located remotely relative to the electronic device. The electronic device may be a mobile electronic device.

[0294]

[0475] The computer system 8001 may include a central processing unit (CPU, also referred to herein as "processor" and "computer processor") 8005, which may be a single-core or multi-core processor, or multiple processors for parallel processing. The computer system 8001 also includes memory or storage locations 8010 (e.g., random access memory, read-only memory, flash memory), an electronic storage unit 8015 (e.g., hard disk), a communication interface 8020 (e.g., network adapter) for communicating with one or more other systems, and peripheral devices 8025, such as cache, other memory, data storage devices, and / or electronic display adapters. The memory 8010, storage unit 8015, interface 8020, and peripheral devices 8025 are in communication with the CPU 8005 via a communication bus (solid lines), such as a motherboard. The storage unit 8015 may be a data storage unit (or data repository) for storing data. The computer system 8001 can be operatively connected to a computer network ("network") 8030 using the communication interface 8020. The network 8030 can be the Internet, an Internet and / or an extranet, or an intranet and / or an extranet in communication with the Internet. The network 8030 is possibly a telecommunications network and / or a data network. The network 8030 can include one or more computer servers, which can enable distributed computing, such as cloud computing. The network 8030 can possibly use the computer system 8001 to implement a peer-to-peer network, which can enable devices connected to the computer system 8001 to act as clients or servers.

[0295]

[0476] The CPU 8005 may execute a series of machine-readable instructions, which may be embodied in a program or software. The instructions may be stored in a memory location, such as the memory 8010. The instructions may be directed to the CPU 8005, which may then be programmed or otherwise configured to implement the methods of the present disclosure. Examples of operations performed by the CPU 8005 may include fetch, decode, execute, and writeback.

[0296]

[0477] The CPU 8005 may be part of a circuit, such as an integrated circuit. One or more other components of the system 8001 may be included in the circuit. In some cases, the circuit is an application specific integrated circuit (ASIC).

[0297]

[0478] The storage unit 8015 can store files such as drivers, libraries, and saved programs. The storage unit 8015 can store user data, such as user preferences and user programs. The computer system 8001 may comprise one or more additional data storage units, possibly located outside the computer system 8001 (e.g., on a remote server in communication with the computer system 8001 via an intranet or the Internet).

[0298]

[0479] The computer system 8001 can communicate with one or more remote computer systems via the network 8030. For example, the computer system 8001 can communicate with a remote computer system of a user (e.g., a product handling system operator, a product handler, a grocery store, a consumer, an end user, etc.). Examples of remote computer systems include a personal computer (e.g., a portable PC), a slate or tablet PC (e.g., an Apple® iPad, a Samsung® Gala80 Tab), a phone, a smartphone (e.g., an Apple® iPhone, an Android-enabled device, a Blackberry®), or a personal digital assistant. A user can access the computer system 8001 via the network 8030.

[0299]

[0480] The methods described herein may be implemented by machine (e.g., a computer processor) executable code stored in an electronic storage location of the computer system 8001, such as the memory 8010 or the electronic storage unit 8015. The machine executable or machine readable code may be provided in the form of software. In use, the code may be executed by the processor 8005. In some cases, the code may be retrieved from the storage unit 8015 and stored in the memory 8010 for immediate access by the processor 8005. In some cases, the electronic storage unit 8015 may be eliminated and the machine executable instructions are stored in the memory 8010.

[0300]

[0481] The code may be pre-compiled and configured for use on a machine having a processor adapted to execute the code, or it may be compiled at run time. The code may be provided in a programming language that may be selected so that the code can be executed in a pre-compiled or co-compiled manner.

[0301]

[0482] Aspects of the systems and methods provided herein, such as computer system 8001, may be embodied in programming. Various aspects of the technology may be considered as "products" or "articles of manufacture" that typically take the form of machine (or processor) executable code and / or associated data carried on or embodied in some type of machine-readable medium. The machine executable code may be stored in an electronic storage unit, such as a memory (e.g., read-only memory, random access memory, flash memory) or a hard disk. A "storage" type medium may include some or all of the tangible memory of a computer, a processor, etc., or associated modules such as various semiconductor memories, tape drives, disk drives, etc., that may provide non-transitory storage for software programming at any time. All or parts of the software may be communicated from time to time over the Internet or various other telecommunications networks. Such communication may, for example, allow the software to be loaded from one computer or processor to another, for example, from a management server or host computer to the computer platform of an application server. Thus, other types of media that may carry software elements include light, radio and electromagnetic waves used over physical interfaces between local devices, wired and optical land line networks, and various air links. Physical elements that transmit such waves, such as wired or wireless links, optical links, etc., may also be considered software-bearing media. As used herein, unless limited to non-transitory tangible "storage" media, terms such as computer or machine "readable medium" refer to any medium that participates in providing instructions to a processor for execution.

[0302]

[0483] Thus, a machine-readable medium such as a computer-executable code may take many forms, including but not limited to a tangible storage medium, a carrier wave medium, or a physical transmission medium. For example, a non-volatile storage medium such as an optical or magnetic disk, or any storage device such as any computer may be used to implement the databases shown in the figures. Volatile storage media include dynamic memory such as the main memory of such a computer platform. Tangible transmission media include coaxial cables, copper wire and optical fibers, including the wires that make up a bus within a computer system. Carrier wave transmission media may take the form of electric or electromagnetic signals, or acoustic or light waves such as those generated during radio frequency (RF) and infrared (IR) data communications. Thus, common forms of computer readable media include, for example, floppy disks, flexible disks, hard disks, magnetic tape, other magnetic media, CD-ROMs, DVDs or DVD-ROMs, other optical media, punch cards paper tape, other physical storage media with patterns of holes, RAM, ROM, PROMs and EPROMs, FLASH-EPROMs, other memory chips or cartridges, carrier waves carrying data or instructions, cables or links carrying such carrier waves, or other media from which a computer may read programming code and / or data. Many of these forms of computer readable media may be involved in carrying one or more sequences of one or more instructions to a processor for execution.

[0303]

[0484] The computer system 8001 may include or communicate with an electronic display 8035 that includes a user interface (UI) 8040 to provide a portal for a product handler, consumer, or human operator to monitor the dispensing of one or more items into a container (e.g., a bag or box for a customer), for example. The portal may be provided through an application programming interface (API). A user or entity may also interact with various elements in the portal through the UI. Examples of a UI include, but are not limited to, a graphical user interface (GUI) or a web-based user interface.

[0304]

[0485] The methods and systems of the present disclosure may be implemented by one or more algorithms. The algorithm may be implemented by software upon execution by the central processing unit 8005. For example, the algorithm may be configured to track the distribution of one or more items or the movement of one or more receiving units. In some embodiments, the algorithm may be further configured to change or update the distribution order of one or more items, or adjust one or more movement paths of one or more receiving units using a cost function that optimizes one or more order fulfillment metrics based on the availability of one or more items at a distribution station, the availability of one or more of a plurality of receiving units to receive one or more items, or one or more newly received customer orders.

[0305]

[0486] While preferred embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. The present invention is not intended to be limited by the specific examples provided within the specification. Although the present invention has been described with reference to the above specification, the description and illustration of the embodiments herein are not intended to be construed in a limiting sense. Numerous variations, changes, and substitutions will occur to those skilled in the art without departing from the present invention. Furthermore, it is to be understood that all aspects of the present invention are not limited to the specific depictions, configurations, or relative proportions set forth herein, which depend upon a variety of conditions and variables. It is to be understood that various alternatives to the embodiments of the present invention described herein may be employed in carrying out the present invention. It is therefore contemplated that the present invention will cover such alternatives, modifications, variations, or equivalents. It is intended that the following claims define the scope of the present invention, and that methods and structures within the scope of these claims, and their equivalents, are covered thereby.

Claims

1. A product handling system comprising: A first tray having a first opening and a first set of supports, The first set of supports comprises: disposed above the first opening; defining a first set of lanes within the first tray; a first tray configured to support a manufacturing unit within the first set of lanes; a first distribution unit configured to temporarily receive the first tray, the first distribution unit having a first set of distribution arms; The first set of distribution arms comprises: a first opening of the first tray extending through the first set of supports; in said first set of lanes, ascending through a first lane defined between a first pair of supports of said first set of supports to lift a first row of product units stored in said first lane between said first pair of supports above said first lane; a product handling system configured to convey the first row of product units to a first receiving area.

2. The product handling system according to claim 1, further comprising a controller, The controller: receiving a customer order specifying a set of product types; in response to the customer order specifying a first quantity of product units of a first product type to be stored in the first lane; moving the mobile robotic system to the first receiving area; lifting the first set of dispense arms through the first lane to lift the first row of product units above the first lane; and a first set of dispensing arms configured to transport a first row of the product units to the first receiving area and dispense a first quantity of the product units of the first product type into a container temporarily loaded onto the mobile robotic system located at the first receiving area.

3. In the product handling system described in claim 1, the first tray is In the first configuration, temporarily storing a first row of said product units of a first product type in said first lane, said first product type including a canned product; temporarily storing a second row of product units of a second product type in a second lane in the first set of lanes adjacent to the first lane, the second product type including boxed products; In a second configuration, temporarily storing a third row of product units of a third product type in the first lane, the third product type including unpackaged products; a fourth row of product units of a fourth product type is temporarily stored in the second lane, the fourth product type including a bagged product.

4. In the product handling system according to claim 1, the first set of dispense arms is configured to dispense a first row of product units of a first product type stored in a first lane in a first arrangement of the first trays; The distribution unit further comprises a second set of distribution arms, the second set of distribution arms comprising: a first opening of the first tray extending through the first set of supports; ascending through the first lane to lift a second row of product units above the first lane, the second row of product units of a second product type being different from the first product type and being stored in the first lane between the first pair of supports in a second configuration of the first tray; a product handling system configured to convey the second row of product units to the first receiving area.

5. In the product handling system according to claim 4, each arm in the first set of distribution arms comprises a bead chain; the first set of dispensing arms is configured to dispense product units of a first product type including a boxed product; 13. A product handling system comprising: a second set of dispensing arms including a set of rollers configured to dispense product units of a second product type including an unpackaged product.

6. In the product handling system according to claim 1, the first dispensing unit is disposed within a first dispensing station; the first set of dispensing arms is configured to dispense product units of a first product type exhibiting a first product shape; The product system further comprises: a second tray having a second opening and a second set of supports, The second set of supports comprises: located above the second opening, defining a second set of lanes within the second tray; a second tray configured to support product units within the second set of lanes; and a second distribution unit disposed within a second distribution station adjacent to the first distribution station and configured to temporarily receive the second tray, the second distribution unit having a second set of distribution arms; the second set of distribution arms a second opening of the second tray extending through the second set of supports; and ascending through a second lane defined between a second pair of supports of the second set of supports in the second set of lanes to elevate above the second lane a second row of product units of a second product type stored in the second lane between the second pair of supports, the second product type exhibiting a second product shape different from the first product shape; a product handling system configured to convey the second row of product units to a second receiving area.

7. In the product handling system according to claim 6, the first set of dispensing arms are configured to dispense product units of the first product type including boxed products; 13. A product handling system comprising: a second set of dispensing arms configured to dispense product units of the second product type including unpackaged products.

8. The product handling system according to claim 1, further comprising: A tray storage rack, A set of slots; a tray storage rack configured to temporarily store the first tray in a first slot in the set of slots; a tray routing conveyor, in response to detecting that the first tray has been assigned to the first receiving area, removing the first tray from the first slot of the tray storage rack; a tray routing conveyor configured to place the first tray into a first distribution station that houses the first distribution unit.

9. In the product handling system according to claim 1, The first set of distribution arms comprises: a first opening of the first tray extending through the first lane; expand laterally to a first separation distance corresponding to a first width of the first row of product units of a first product type stored in the first lane; through a second opening in the first tray to a second lane defined by the first set of supports in the first set of lanes; contract laterally to a second separation distance that corresponds to a second width of a second row of product units of a second product type stored in the second lane, the second product type being narrower in width than the first product type; ascending through the second lane to lift the second row of product units above the second lane; and conveying the second row of product units to the first receiving area.

10. In the product handling system according to claim 1, a first set of dispensing arms configured to dispense a first row of product units into a container temporarily installed in the first receiving area, the container being removable by a customer from a customer pickup bay.

11. The product handling system of claim 1, further comprising a mobile robot system; The mobile robot system comprises: Receive units of product of the product type specified in the customer's order; to move to the first receiving area in response to the customer order specifying a first quantity of product units of a first product type stored in the first lane; a first set of dispensing arms configured to transport a first row of product units to the first receiving area and dispense the first quantity of product units of the first product type to the mobile robotic system positioned in the first receiving area.

12. In the product handling system according to claim 1, The first set of distribution arms comprises: moving to a first position aligned with the first lane; lifting a first row of said product units of a first product type stored in said first lane above said first lane; moving the first row of product units a first distance based on a first dimension of the first product type to dispense a first product unit of the first product type into the first receiving area; moving in the first set of lanes to a second position aligned with a second lane defined between a second pair of supports in the first set of supports of the first tray; ascending through the second lane and lifting a second row of product units stored in the second lane between the second pair of supports above the second lane, wherein the second row of product units of a second product type is different from the first product type; moving the second row of product units a second distance based on a second dimension of the second product type to dispense a second product unit of the second product type into the first receiving area; The product handling system of claim 1, wherein the second dimension is greater than the first dimension and the second distance is greater than the first distance.

13. The product handling system of claim 1, further comprising a controller, The controller: receiving a customer order specifying a set of product types; For a first product type in the set of product types specified in the customer order, identifying a first tray within a set of trays containing a first row of the product units of the first product type; Obtaining a current position of the first tray; allocating the first receiving area to the first tray for dispensing product units of the first product type from the first tray; moving the mobile robotic system to the first receiving area; lifting the first set of dispense arms through the first lane to lift the first row of product units above the first lane; a first set of dispensing arms configured to transport a first row of product units to the first receiving area and dispense product units of the first product type into a container temporarily loaded into the mobile robotic system located in the first receiving area.

14. A product handling system comprising: A first tray having a first set of supports, The first set of supports comprises: defining a first set of lanes within the first tray; a first tray configured to support a manufacturing unit within the first set of lanes; a first dispensing unit configured to temporarily receive the first tray and having a first set of dispensing arms, The first set of distribution arms comprises: ascending through a first lane defined between a first pair of supports of the first set of supports toward the first set of supports in the first set of lanes to lift a first row of product units stored in the first lane between the first pair of supports above the first lane; a first distribution unit configured to convey the first row of product units to a first receiving area; A controller for controlling the first set of distribution arms: lifting a first row of product units up through the first lane and toward the first set of supports; extending toward the first receiving area such that a distal end of the first set of dispensing arms is disposed across the first receiving area; a controller configured to distribute the first row of product units to the first receiving area; A product handling system comprising:

15. In the product handling system according to claim 14, The first tray comprises: temporarily storing a first row of said product units of a first product type in said first lane; configured to temporarily store a second row of product units of a second product type in a second lane in the first set of lanes adjacent to the first lane; The second product type is different from the first product type.

16. The product handling system of claim 14, further comprising a mobile robot system; The mobile robot system comprises: receiving product units of a product type specified in a customer order and dispensed by the first dispensing unit into a container that is temporarily loaded into the mobile robotic system when located in the first receiving area; and transporting the container to a storage locker; The container may be removed from the storage locker by a customer.

17. In the product handling system according to claim 14, the first dispensing unit is disposed within a first dispensing station; the first set of dispensing arms are configured to dispense product units of a first product type exhibiting a first geometric characteristic from the first tray; The product handling system further comprises: a second tray having a second set of supports, The second set of supports comprises: defining a second set of lanes within the second tray; a second tray configured to support product units within the second set of lanes; and a second distribution unit disposed within a second distribution station adjacent to the first distribution station and configured to temporarily receive the second tray, the second distribution unit having a second set of distribution arms; the second set of distribution arms and ascending through a second lane defined between a second pair of supports of the second set of supports in the second set of lanes to lift a second row of product units stored in the second lane between the second pair of supports above the second lane, the second row of product units of a second product type exhibiting a second shape characteristic different from the first shape characteristic; a product handling system configured to convey the second row of product units to a second receiving area.

18. In the product handling system according to claim 17, The controller: receiving a customer order specifying a set of product types; in response to the customer order specifying a first quantity of product units of a first product type to be stored in the first lane; moving the mobile robotic system to the first receiving area; lifting the first set of dispense arms through the first lane to lift the first row of product units above the first lane; carrying a first row of the product units to the first receiving area with a first set of the dispensing arms and dispensing a first quantity of the product units of the first product type into a container temporarily loaded onto the mobile robotic system located at the first receiving area; In response to the customer order further specifying a second quantity of product units of a second product type to be stored in the second lane, moving the mobile robotic system to the second receiving area; raising the second set of dispensing arms through the second lane to lift the second row of product units above the second lane; and a second set of the dispensing arms configured to transport a second row of the product units to the second receiving area and dispense a second quantity of the product units of the second product type into a container temporarily loaded onto the mobile robotic system located in the second receiving area.

19. In the product handling system according to claim 14, the first tray temporarily stores a first row of the product units of a first product type in the first lane, the first product type including packaged products; In response to the controller receiving a customer order specifying a first quantity of product units of a bagged product stored in the first lane, moving the mobile robotic system to the first receiving area; lifting the first set of dispense arms through the first lane to lift the first row of product units above the first lane; and a first set of dispensing arms configured to move a first row of product units a distance based on a first amount of the product units and a dimension of the bagged product to dispense the first amount of the bagged product to the mobile robot system located in the first receiving area.

20. A product handling system comprising: A tray having a base and a support set, The set of supports comprises: are placed across the base, defining a set of lanes within said tray; a tray configured to support product units within the set of lanes; a distribution unit, The distribution unit comprises: Temporarily receiving the tray; a support member extending through the base toward the set of supports; in said set of lanes, by ascending through lanes defined between pairs of supports of said set of supports, thereby lifting the rows of product units stored in said lanes between said pairs of supports above said lanes; and conveying the row of product units to a receiving area.

Images