US-12617095-B1 - Dynamic allocation of robotic sortation resources

US12617095B1US 12617095 B1US12617095 B1US 12617095B1US-12617095-B1

Abstract

In various examples, systems and methods of controlling robotic sortation devices are described. In various examples, a first item associated with a first target may be determined. A first item count associated with a first buffer may be determined. A first robotic sortation device may be allocated to the first target based at least in part on the first item count. The first robotic sortation device may receive the first item. The first robotic sortation device may be controlled to place the first item in a first container associated with the first target.

Inventors

Benjamin McClosky
Di Wang

Assignees

AMAZON TECHNOLOGIES, INC.

Dates

Publication Date: 20260505
Application Date: 20221209

Claims (20)

1 . A method comprising: determining a first item associated with a first target; determining a first item count associated with a first buffer, wherein the first buffer stores items prior to loading the items onto robotic drives; determining a second item count associated with items loaded onto robotic drives; determining a first number of robotic sortation devices allocated to the first target; allocating an additional robotic sortation device to the first target based on the first item count, the second item count, and the first number of robotic sortation devices allocated to the first target; loading the first item onto a first robotic drive; sending first control instructions to the first robotic drive effective to cause the first robotic drive to take the first item to the additional robotic sortation device; receiving, by the additional robotic sortation device, the first item; determining, by the additional robotic sortation device using computer vision, that the first item is associated with the first target; and controlling the additional robotic sortation device to place the first item in a first container associated with the first target based at least in part on the allocation of the additional robotic sortation device to the first target and based on the first item being associated with the first target.
2 . The method of claim 1 , further comprising: determining a set of robotic sortation devices that are unallocated to a target; determining, for a first robotic sortation device of the set of robotic sortation devices, a first distance between the first robotic sortation device and a second robotic sortation device allocated to the first target, wherein the second robotic sortation device is a closest robotic sortation device to the first robotic sortation device among robotic sortation devices allocated to the first target; determining, for a third robotic sortation device of the set of robotic sortation devices, a second distance between the third robotic sortation device and a fourth robotic sortation device allocated to the first target, wherein the fourth robotic sortation device is the closest robotic sortation device to the third robotic sortation device among robotic sortation devices allocated to the first target; and selecting the first robotic sortation device as the additional robotic sortation device to allocate to the first target based on the first distance being greater than the second distance.
3 . The method of claim 1 , further comprising: determining that the first container has been removed from a station associated with the additional robotic sortation device; and deallocating the additional robotic sortation device from the first target.
4 . The method of claim 3 , further comprising: determining a first percentage utilization of the first container based at least in part on a number of items in the first container; and determining that the first percentage utilization is greater than or equal to a threshold percentage utilization, wherein the deallocating the additional robotic sortation device from the first target is based at least in part on the first percentage utilization being greater than or equal to the threshold percentage utilization.
5 . A method of controlling robotic sortation devices, comprising: determining a first item associated with a first target; determining a first item count associated with a first buffer; allocating a first robotic sortation device to the first target based at least in part on the first item count; receiving, by the first robotic sortation device, the first item; and controlling the first robotic sortation device to place the first item in a first container associated with the first target.
6 . The method of claim 5 , further comprising: determining a second item count associated with items loaded onto a plurality of robotic drives, wherein the first robotic sortation device is allocated to the first target further based at least in part on the second item count.
7 . The method of claim 6 , further comprising: loading the first item onto a first robotic drive of the plurality of robotic drives; and sending control instructions to the first robotic drive effective to cause the first robotic drive to navigate to the first robotic sortation device based at least in part on the allocating the first robotic sortation device to the first target.
8 . The method of claim 5 , wherein the first buffer comprises a recirculating buffer that stores overflow items that are currently unable to be processed by robotic sortation devices due to current processing load, the method further comprising: comparing the first item count to a threshold item count associated with the first buffer; and allocating the first robotic sortation device to the first target based at least in part on the first item count exceeding the threshold item count.
9 . The method of claim 5 , further comprising: determining a set of robotic sortation devices that are unallocated to a target; determining, for the first robotic sortation device of the set of robotic sortation devices, a first distance between the first robotic sortation device and a second robotic sortation device allocated to the first target, wherein the second robotic sortation device is a closest robotic sortation device to the first robotic sortation device among robotic sortation devices allocated to the first target; determining, for a third robotic sortation device of the set of robotic sortation devices, a second distance between the third robotic sortation device and a fourth robotic sortation device allocated to the first target, wherein the fourth robotic sortation device is the closest robotic sortation device to the third robotic sortation device among robotic sortation devices allocated to the first target; and selecting the first robotic sortation device to allocate to the first target based on the first distance being greater than the second distance.
10 . The method of claim 5 , further comprising: determining that the first container has been removed from a station associated with the first robotic sortation device; and deallocating the first robotic sortation device from the first target.
11 . The method of claim 10 , further comprising: determining a first percentage utilization of the first container based at least in part on a number of items in the first container; and determining that the first percentage utilization is greater than or equal to a threshold percentage utilization, wherein the deallocating the first robotic sortation device from the first target is based at least in part on the first percentage utilization being greater than or equal to the threshold percentage utilization.
12 . The method of claim 5 , further comprising: determining a first number of items associated with the first target in the first buffer; determining a second number of items associated with the first target in a laden drive buffer; and allocating, based on the first number and the second number, the first robotic sortation device to the first target.
13 . The method of claim 12 , further comprising: inputting first data into a first machine learning model, wherein the first data comprises a representation of at least the first number and the second number; generating, by the first machine learning model, output data indicating that an additional robotic sortation device is allocated to the first target; and sending control instructions to the first robotic sortation device, the control instructions designating the first container as associated with the first target.
14 . A system comprising: at least one processor; and non-transitory computer-readable memory storing instructions that, when executed by the at least one processor is effective to; determine a first item associated with a first target; determine a first item count associated with a first buffer; allocate a first robotic sortation device to the first target based at least in part on the first item count; receive, by the first robotic sortation device, the first item; and control the first robotic sortation device to place the first item in a first container associated with the first target.
15 . The system of claim 14 , the non-transitory computer-readable memory storing further instructions that, when executed by the at least one processor, are further effective to: determine a second item count associated with items loaded onto a plurality of robotic drives, wherein the first robotic sortation device is allocated to the first target further based at least in part on the second item count.
16 . The system of claim 14 , the non-transitory computer-readable memory storing further instructions that, when executed by the at least one processor, are further effective to: load the first item onto a first robotic drive of a plurality of robotic drives; and send control instructions to the first robotic drive effective to cause the first robotic drive to navigate to the first robotic sortation device based at least in part on the allocating the first robotic sortation device to the first target.
17 . The system of claim 14 , wherein the first buffer comprises a recirculating buffer that stores overflow items that are currently unable to be processed by robotic sortation devices due to current processing load, the non-transitory computer-readable memory storing further instructions that, when executed by the at least one processor, are further effective to: compare the first item count to a threshold item count associated with the first buffer; and allocate the first robotic sortation device to the first target based at least in part on the first item count exceeding the threshold item count.
18 . The system of claim 14 , the non-transitory computer-readable memory storing further instructions that, when executed by the at least one processor, are further effective to: determine a set of robotic sortation devices that are unallocated to a target; determine, for the first robotic sortation device of the set of robotic sortation devices, a first distance between the first robotic sortation device and a second robotic sortation device allocated to the first target, wherein the second robotic sortation device is a closest robotic sortation device to the first robotic sortation device among robotic sortation devices allocated to the first target; determine, for a third robotic sortation device of the set of robotic sortation devices, a second distance between the third robotic sortation device and a fourth robotic sortation device allocated to the first target, wherein the fourth robotic sortation device is the closest robotic sortation device to the third robotic sortation device among robotic sortation devices allocated to the first target; and selecting the first robotic sortation device to allocate to the first target based on the first distance being greater than the second distance.
19 . The system of claim 14 , the non-transitory computer-readable memory storing further instructions that, when executed by the at least one processor, are further effective to: determine that the first container has been removed from a station associated with the first robotic sortation device; and deallocate the first robotic sortation device from the first target.
20 . The system of claim 19 , the non-transitory computer-readable memory storing further instructions that, when executed by the at least one processor, are further effective to: determine a first percentage utilization of the first container based at least in part on a number of items in the first container; and determine that the first percentage utilization is greater than or equal to a threshold percentage utilization, wherein the deallocating the first robotic sortation device from the first target is based at least in part on the first percentage utilization being greater than or equal to the threshold percentage utilization.

Description

BACKGROUND Robotic drive systems can be used to move items and/or containers of items from one location to another. Robotic arms can be used to sort items by removing an item from one location and placing the item in a different, target location. Systems of such robots can be designed and deployed to provide desired tasks such as item processing and sortation. BRIEF DESCRIPTION OF DRAWINGS FIG. 1 is a diagram of an example robotic sortation system, according to various aspects of the present disclosure. FIG. 2 is an overhead view of an example sortation field, in accordance with various aspects of the present disclosure. FIG. 3 is a block diagram illustrating examples inputs and outputs of the dynamic allocation controller described herein, in accordance with various aspects of the present disclosure. FIG. 4A is an example image depicting laden robotic drives within a sortation facility, in accordance with various aspects of the present disclosure. FIG. 4B is an example image of one of the laden robotic drives of FIG. 4A, in accordance with an example aspect of the present disclosure. FIG. 4C is a diagram of a robotic sortation device that sorts items into target containers, in accordance with various embodiments described herein. FIG. 4D depicts an overhead example of a container build station in accordance with various examples of the present disclosure. FIG. 5 is an example computing device architecture that may be used in accordance with various techniques described herein. FIG. 6 is a diagram illustrating an example system for sending and providing data that may be used in accordance with the present disclosure. FIG. 7 is a flowchart describing an example process for dynamic allocation of robotic sortation resources, according to various aspects of the present disclosure. DETAILED DESCRIPTION In the following description, reference is made to the accompanying drawings that illustrate several example embodiments of the present invention. It is understood that other examples may be utilized and various operational changes may be made without departing from the scope of the present disclosure. The following detailed description is not to be taken in a limiting sense, and the scope of the embodiments of the present invention is defined only by the claims of the issued patent. In various examples, robotic sortation systems may be used in large scale sorting environments in which a large number of items are received, processed, sorted, and sent out. For example, large-scale delivery and inventory management systems may use fulfillment centers and other large warehouses that may serve as part of the supply chain and can serve as a hub for logistics and processes used to get items from third party sellers to the purchaser. In some cases, after items in such facilities are packaged and addressed for delivery, the packages are sorted into carts and/or other containers, with the packages in a single container being destined for the same delivery location. In some examples, such containers may be said to be “destination pure” as all items within the container may be destined for the same downstream location (whether such location is an intermediate location, such as a warehouse or last-mile delivery center, or the terminal location (e.g., a delivery address of a purchaser)). It should be noted that sortation may be used in other contexts beyond commercial delivery systems and in general may be used whenever heterogeneous items are to be sorted along some dimension of sortation (e.g., type, class, size, delivery destination, color, functionality, etc.). Accordingly, although many of the examples described herein use a package delivery example, the various dynamic allocation techniques used to control robotic resources may instead be deployed in other sortation contexts. A robotic sortation system is described herein in which a highly-variable stream of inbound packages (or other items) may be processed, transported from induct stations to eject stations by laden drive robots, and then sorted at the eject stations by robotic sortation devices (e.g., robotic “arms”) into destination pure containers. The number of physical containers that may be stationed at a robotic sortation device station may be limited due to spatial constraints (e.g., by physical constraints and/or by a reach of a stationary robotic arm). Since each such container is intended to be destination pure such that all items are bound for the same downstream location, the robotic arm is controlled to place packages that are bound for the same location in the same target container. Additionally, the laden drive robot devices are also controlled to provide packages to eject stations with containers that correspond to a target location for that package (e.g., where a robotic sortation device is sorting packages into a container bound for the target). Systems and techniques are described herein for a dynamic allocation controller that dynamically allocates