Attribute-based container selection for inventory

• US 10,315,231 B1
• Filed: 09/07/2017
• Issued: 06/11/2019
• Est. Priority Date: 09/07/2017
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Brazeau, Jeremiah David
• Primary Examiner(s): Cumbess, Yulanda R
• Application Number:
  US15/698,510
• Time in Patent Office: 642 Days
• Field of Search: None
• CPC Class Codes: B07C 3/008 Means for collecting objects, e.g. containers for sorted mail items pile receivers in general B65H31/00
• CPC Class Codes: B07C 5/00 Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches sorting by hand only B07C7/00; separating solids from solids by sieving, screening, or sifting or by using gas currents or other separating by dry methods applicable to bulk material B07B
• CPC Class Codes: B07C 5/38 Collecting or arranging articles in groups

Abstract

A method of placing inventory can include accessing a data store to retrieve item attribute data of an inventory item to be placed, comparing the item attribute data of the inventory item to attribute data corresponding to each respective inventory item in a collection of potential destination containers, and selecting a particular destination container for storage based on that comparison in order to maximize the relative distinctiveness of items stored together. The storage method results in pseudo-random storage of inventory in containers where each item is more readily identifiable from each other item in the same container, particularly by automated means, and can be used in conjunction with an automated or partially automated inventory storage and retrieval system.

Claims

1. An inventory system, comprising:

one or more sensors configured to sense at least one physical attribute of an incoming inventory item;

a first destination container containing a first selection of inventory items;

at least one processor; and

at least one tangible memory device storing non-transitory instructions executable by the at least one processor to cause the at least one processor to;

obtain incoming item attribute data for the incoming inventory item, via the one or more sensors, indicative of the at least one physical attribute of the incoming inventory item;

retrieve first destination container attribute data indicative of the at least one physical attribute for each inventory item of the first selection of inventory items contained in the first destination container; and

assess, based on the incoming item attribute data and the first destination container attribute data, whether the incoming inventory item is detectably distinguishable with respect to the first selection of inventory items to enable automated selection of the incoming inventory item from among the first selection of inventory items in the first destination container if the incoming inventory item is stored together with the first selection of inventory items.

2. The system of claim 1, further comprising a second destination container, wherein the executable instructions are further configured to cause the at least one processor to:

retrieve second destination container attribute data indicative of the at least one physical attribute for each of inventory items contained in the second destination container;

generate first cumulative distinctness values for the first destination container based on assessing relative uniqueness of the incoming item attribute data with respect to the first destination container attribute container data;

generate second cumulative distinctness values for the second destination container based on assessing relative uniqueness of the incoming item attribute data with respect to the second destination container attribute container data; and

select the first destination container for placement of the incoming inventory item in response to the first cumulative distinctness value being greater than the second cumulative distinctness value.

3. The system of claim 1, further comprising a second destination container, wherein the executable constructions are further configured to cause the at least one processor to:

retrieve second destination container attribute data indicative of the at least one physical attribute for each of inventory items contained in the second destination container;

assess whether the incoming item attribute data is sufficiently distinguishable relative to the second destination container attribute data to enable automated selection of the incoming inventory item from the second destination container based on the incoming item attribute data if the incoming inventory item is contained in the second destination container; and

in response to determining that (a) the incoming attribute data is sufficiently distinguishable relative to the first destination container data to enable the automated selection of the incoming item from the first destination container, (b) the incoming attribute data is sufficiently distinguishable relative to the second destination container data to enable the automated selection of the incoming item from the second destination container, and (c) the first container contains more items than the second container;

select the first destination container for placement of the incoming inventory item.

4. The system of claim 1, further comprising a robotic manipulator operable to transfer the inventory item into the first destination container;

and

wherein the executable constructions are further configured to cause the at least one processor to;

select the first destination container for placement of the incoming inventory in response to the incoming item attribute data being sufficiently distinguishable relative to the first container destination data to enable the automated selection of the incoming item from the first destination container; and

cause the robotic manipulator to transfer the inventory item into the first destination container in response to the selection of the first container for placement of the inventory item.

5. The inventory system of claim 1, wherein the first selection of inventory items in the first destination container comprise one or more inventory items that are detectably distinguishable from each other.

6. The inventory system of claim 1, wherein the instructions are further configured to cause the at least one processor to assess whether the incoming inventory item is detectably distinguishable with respect to the first selection of inventory items by:

determining a distinctness value for a combination of the incoming inventory item and the first selection of inventory items; and

comparing the distinctness value to a threshold.

7. A method of placing inventory, the method comprising:

accessing a data store to retrieve item attribute data for an incoming inventory item, the item attribute data indicative of at least one physical attribute of the incoming inventory item;

accessing attribute data indicative of the at least one physical attribute for each inventory item stored in a first destination container;

accessing attribute data indicative of the at least one physical attribute for each inventory item stored in a second destination container;

comparing the item attribute data for the incoming inventory item to the attribute data for the inventory items stored in the first destination container;

comparing the item attribute data for the incoming inventory item to the attribute data for the inventory items stored in the second destination container; and

selecting the first destination container for placement of the inventory item by determining, based in part on the item attribute data for the incoming inventory item and the attribute data for the inventory items stored in the first and second destination containers, that the incoming inventory item is detectably distinguishable with respect to the inventory items stored in the first destination container to enable automated selection of the incoming inventory item from among the inventory items stored in the first destination container.

8. The method of claim 7, further comprising:

generating instructions to cause a robotic manipulator to deposit the incoming inventory item in the first destination container in response to selecting the first destination container for placement of the incoming inventory item.

9. The method of claim 7, further comprising:

determining first destination container distinctness values for the incoming inventory item, each of the first destination container distinctiveness values being indicative of the distinctness of the incoming inventory items relative to a respective one of the inventory items stored in the first destination container;

determining a first cumulative distinctness value for the incoming inventory item relative to the inventory items stored in the first destination container based on the first destination container distinctness values;

determining second destination container distinctness values for the incoming inventory item, each of the second destination container distinctiveness values being indicative of the distinctness of the incoming inventory items relative to a respective one of the inventory items stored in the second destination container;

determining a second cumulative distinctness value for the incoming inventory item relative to the inventory items stored in the second destination container based on the second destination container distinctness values;

comparing the first cumulative distinctness value to the second cumulative distinctness value; and

selecting the first destination container for placement of the incoming inventory item in response to the first cumulative distinctness value being greater than the second cumulative distinctness value.

10. The method of claim 7, further comprising:

generating a cumulative distinctness value for the first set of contained inventory attribute data based on the comparing;

determining that the cumulative distinctness value of the first set of contained inventory attribute data passes a threshold; and

selecting the first destination container for placement of the inventory item based in part on cumulative distinctness value passing the threshold.

11. The method of claim 7, wherein the at least one physical attribute of the inventory item comprises at least two distinct physical attributes corresponding, respectively, to at least two different types of measurements selected from:

item size, item color, item reflectivity, item shape, item text content, item text density, item compressibility, item weight, item density, item spectra, item thermal properties, or item acoustic density.

12. The method of claim 7, further comprising:

accessing the data store to retrieve item attribute data for a plurality of incoming inventory items that includes the incoming inventory item, the item attribute data indicative of at least one physical attribute of each of the incoming inventory items;

determining a plurality of item distinctness values for each contained inventory item with respect to each incoming inventory item by comparing the item attribute data for each incoming inventory item to the contained inventory attribute data corresponding to each contained inventory item;

determining respective cumulative distinctness values for each incoming inventory item with respect to the first destination container; and

selecting the incoming inventory item for placement in the first destination container based in part on the cumulative distinctness value corresponding to the first destination container and incoming inventory item.

13. The method of claim 7, further comprising:

accessing the data store to retrieve item attribute data for a plurality of incoming inventory items that includes the incoming inventory item, the item attribute data indicative of at least one physical attribute of each of the incoming inventory items;

determining a plurality of item distinctness values for each contained inventory item with respect to each incoming inventory item by comparing the item attribute data for each incoming inventory item to the contained inventory attribute data corresponding to each contained inventory item;

determining respective cumulative distinctness values for each combination of the first and second destination containers with each incoming inventory item;

determining a maximum distinctness arrangement by determining a total distinctness value associated with each permutation of the combinations of first and second destination containers with each incoming inventory item; and

selecting the first container for placement of the incoming inventory item based in part on the maximum distinctness arrangement.

14. The method of claim 7, wherein the item attribute data comprises a compilation of data representing multiple physical attributes, wherein the compilation is nonvarying when the inventory item is sensed from different orientations.

15. The method of claim 7, further comprising:

receiving a positive identifier corresponding to the inventory item; and

retrieving the item attribute data of the incoming inventory item from the data store based on the positive identifier.

16. The method of claim 7, further comprising:

determining second destination container distinctness values for the incoming inventory item, each of the second destination container distinctiveness values being indicative of the distinctness of the incoming inventory items relative to a respective one of the inventory items stored in the second destination container;

excluding the second destination container as a candidate for placement of the inventory item based in part on a determination that at least one destination container distinctness value corresponding to the inventory items stored in the second destination container falls below a threshold.

17. An inventory system, comprising:

a first destination container

a second destination container; and

a management component comprising a processor and a tangible memory device storing non-transitory instructions executable by the processor to cause the processor to, at least;

obtain incoming item attribute data for an incoming inventory item indicative of at least two distinct physical attributes of the incoming inventory item;

access attribute data indicative of the at least one physical attribute for each inventory item stored in a first destination container;

access attribute data indicative of the at least one physical attribute for each inventory item stored in a second destination container;

determine first destination container distinctness values for the incoming inventory item, each of the first destination container distinctiveness values being indicative of the distinctness of the incoming inventory items relative to a respective one of the inventory items stored in the first destination container;

determine a first cumulative distinctness value for the incoming inventory item relative to the inventory items stored in the first destination container based on the first destination container distinctness values;

determine second destination container distinctness values for the incoming inventory item, each of the second destination container distinctiveness values being indicative of the distinctness of the incoming inventory items relative to a respective one of the inventory items stored in the second destination container;

determine a second cumulative distinctness value for the incoming inventory item relative to the inventory items stored in the second destination container based on the second destination container distinctness values;

select one of the first and second destination containers for placement of the inventory item based in part on one or more of the first and second cumulative distinctness values.

18. The inventory system of claim 17, further comprising a robotic manipulator configured to move inventory items, and wherein the executable instructions are further configured to cause the robotic manipulator to place the inventory item into the first destination container based on the destination container being selected for placement of the inventory item.

19. The inventory system of claim 17, wherein the executable instructions are further configured to cause the processor to:

exclude one of the first and second destination containers as a candidate for placement of the inventory item based on one of the item distinctness values corresponding to one of the first and second destination containers.

20. The inventory system of claim 19, wherein the executable instructions are further configured to cause the processor to:

after excluding the one of the first and second destination containers as a candidate, select a remaining destination container for placement of the inventory item by selecting the remaining potential destination container associated with a highest cumulative distinctness value.

21. The inventory system of claim 17, wherein the executable instructions are further configured to cause the processor to:

exclude one of the first or second destination containers as a candidate for placement of the inventory item based on the first or second cumulative distinctness value falling below a threshold.

22. The inventory system of claim 17, further comprising a sensor assembly configured to sense the at least two distinct attributes of the inventory item;

and wherein the executable instructions are further configured to cause the processor to generate the item attribute data based on the sensed at least two distinct attributes.