Back to results
US 12,749,131 B2Granted
Pickup order processing
AssigneeNCR Voyix Corporation
InventorsKip Oliver Morgan, Gina Torcivia Bennett
Life of this patent
Estimating the term…
Abstract
Cameras capture images of vehicles and customers in drive-through lanes of a store. The images are processed to identify a type of vehicle, if any, and determine whether a given customer is in a proper lane associated with cars or a proper lane associated with walkups or non cars. Customers in improper lanes are instructed to move to the proper lane to place an order. The images are further processed to generate a written description of the vehicle, if any, and a written description of the customer. The written description is linked to the customer's order and is accessible to store staff for verification when collecting payment for the order and when providing the ordered items to the customer.
Claims12
A method performed by a cloud-based order service comprising at least one processor, the method comprising:
- obtaining an image of at least one drive-through lane from a camera communicatively coupled to the cloud-based order service to overcome limitations of metal sensors that are not activated by bicycles or other personal electronic vehicles (PEVs);
- determining, by the cloud-based order service, an individual is present in a first drive-through lane from the image;
- identifying, by the cloud-based order service, the individual as a pedestrian that is present in the image;
- providing, from the cloud-based order service, a drive-through lane identifier for the first drive-through lane to an order manager of a point-of-sale terminal associated with the at least one drive-through lane;
- generating, by the cloud-based order service, a first written description of a non-car vehicle when present and a second written description of the individual from the image of the individual using a trained machine-learning model (MLM) that is executed by at least one server, the trained MLM processes the image as input and produces as output a type of vehicle when the non-car vehicle is depicted in the image, the first written description when the non-car vehicle is present and the second written description to properly associate an order with the individual and enable automated real-time customer identification;
- integrating, by the cloud-based order service, the image, the first written description when present and the second written description via a network communication into order details associated with the order, wherein the order is placed by the individual through the first drive-through lane or a second drive-through lane, wherein the at least one drive-through lane comprises the first drive-through lane and the second drive-through lane;
- linking the image, the first written description, and the second written description to an order number associated with the order details within a workflow of the order manager of a point-of-sale terminal that takes the order from the individual by sending the first written description when present and the second written description to the point-of-sale terminal;
- and
- verifying at a pickup window and payment window the individual with proper ordered items based on the image, the order details, the first written description when present and the second written description.
6 more elements
A method performed by a cloud-based order service comprising at least one processor, the method comprising:
- identifying, by the cloud-based order service, a non-car vehicle present in a non-car drive-through lane of a restaurant from an image received via a network from a camera located at the non-car drive-through lane and the image taken of an area associated with the non-car drive-through lane to overcome limitations of metal sensors that are not activated by bicycles and other personal electric vehicles (PEVs);
- generating, by the cloud-based order service, a first written description of the non-car vehicle and a second written description of an individual associated with the non-car vehicle from the image using a trained machine-learning model (MLM) that is executed by at least one server, the trained MLM is trained to provide written descriptive information for bikes, scooters, PEVs, or customers who are riding the bikes, scooters, or the PEVs to properly associate an order with the individual and enable automated real-time customer identification where metal sensor technology fails;
- integrating, by the cloud-based order service, the first written description and the second written description into order details associated with the order placed by the individual in the non-car drive-through lane;
- linking the image, the first written description, and the second written description to an order number associated with the order details within a workflow of an order manager of a point-of-sale terminal that takes the order from the individual;
- and
- verifying at a pickup window and payment window the individual with proper ordered items based on the image, the order details, the first written description and the second written description.
3 more elements
A system comprising:
- at least one server that comprises at least one processor and a communication interface for network communication with a camera and a point-of-sale terminal;
- and
- the at least one processor executes instructions that cause the at least one processor to perform operations comprising:
- verifying, by the at least one processor, from at least one image received via the communication interface from the camera that an individual is in a drive-through lane for placing an order to overcome limitations of metal sensors that are not activated by bicycles and other personal electric vehicles (PEVs);
- causing the individual to receive an audible verbal instruction to move from the drive-through lane to a non-car drive-through lane based on the verifying by transmitting an instruction from the at least one processor via the communication interface;
- generating, by the at least one processor, a first written description of the individual and a second written description of a vehicle associated with the individual, when the vehicle is present, using the at least one image with a trained machine-learning model (MLM) executed by the at least one processor and that processes the at least one image as input and produces as output a type of vehicle when a vehicle is present, the first written description, and the second written description of the vehicle when the vehicle is present to properly associate the order with the individual and enable automated real-time customer identification;
- integrating the first written description and the second written description when the vehicle is present into order details associated with the order placed by the individual;
- linking the at least one image, the first written description, and the second written description when the vehicle is present, to order details associated with the order based on an order number associated with the order and providing the at least one image, the first written description, and the second written description when the vehicle is present from the at least one processor via the communication interface to the point-of-sale terminal;
- and
- verifying at a pickup window and payment window the individual with proper ordered items based on the at least one image, the order details, the first written description and the second written description when the vehicle is present;
- wherein the operations are provided and processed by the at least one processor as a cloud-based service to an order manager of the point-of-sale terminal associated with the drive-through lane.
8 more elements
Dependencies are inferred from the claim text.