Methods, systems, and computer readable media for tracking consumer interactions with products using modular sensor units

• US 10,083,453 B2
• Filed: 12/31/2013
• Issued: 09/25/2018
• Est. Priority Date: 03/17/2011
• Status: Active Grant

• Current Assignee: Triangle Strategy Group, LLC
• Original Assignee: Triangle Strategy Group, LLC
• Inventors: Campbell, Patrick Joseph
• Primary Examiner(s): Rotaru, Octavian
• Application Number:
  US14/145,649
• Publication Number:
  US 20140114708A1
• Time in Patent Office: 1,729 Days
• Field of Search: None
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G08B 5/22 using electric transmission; using electromagnetic transmission

Abstract

The subject matter described herein includes methods, systems, and computer readable media for tracking consumer interactions with products using modular sensor units. In one system, a plurality of weight and motion sensors is configured to track consumer interactions with products. The system further includes a plurality of sensor mounts for holding the sensors in place. Each sensor mount includes signal lines for transmitting power to the sensors and for conducting signals between the sensors and a signal processor. The system further includes a plurality of conductors for interconnecting sensor mounts to a signal processor. The system further includes a plurality of connectors to allow rapid addition and removal of the sensors and their mounts as needed for a particular product planogram.

Claims

1. A system for tracking interactions with products, comprising:

a merchandising unit;

a plurality of sensors configured to detect signals that change over time in response to a plurality of product items being removed or returned from the merchandising unit;

a daisy-chain system, for incremental electrical connection of the plurality of sensors to upstream and downstream wiring segments joined by multi-conductor wires transmitting power to the plurality of sensors;

a database containing allowable and non-allowable patterns of the plurality of product items; and

a signal processor connected to the daisy-chain system configured to;

receive and log the signals detected by the plurality of sensors;

communicate with the database containing the allowable and non-allowable patterns; and

execute a pattern matching algorithm by;

identifying a pickup event, of at least one of the product items, detected as a peak in at least one of the detected signals indicating that a shopper has pressed down on the at least one of the product items prior to removing the at least one of the product items, followed by a period, detected by a lower signal level at a tail of the at least one of the detected signals, when the at least one of the product items weighs less than prior to the pickup event,

identifying a vibration event when the at least one of the product items is jolted,

generating a percentage best pattern match using a plurality of identified pickup and vibrations events for the at least one of the product items within the database,

applying an acceptance percentage threshold to the generated percentage best pattern match to classify the plurality of identified pickup events into an allowable pattern,

applying a rejection percentage threshold to the generated percentage best pattern match to classify the vibrations events into a non-allowable pattern,

identifying un-assignable events that do not meet the acceptance percentage threshold or the rejection percentage threshold, and

determining from the acceptance percentage threshold, the rejection percentage threshold, and newly discovered patterns, valid signals produced by the plurality of identified pickup events of the at least one of the product items and invalid signals produced by the vibrations events, and

upon determination of the valid signals further identifying a position and an associated facing of the at least one of the product items that was identified as being removed or returned from the merchandising unit.

2. The system of claim 1, wherein the percentage best pattern match is derived according to a vector dot product.

3. The system of claim 1, wherein the acceptance percentage threshold is used to confirm the pattern in a signal received by the signal processor as being one of the allowable patterns.

4. The system of claim 1, wherein the percentage best pattern match is at least partially implemented by a machine learning system.

5. The system of claim 1, wherein the signal processor is further configured to validate a predetermined positioning of the plurality of sensors.

6. The system of claim 1, further comprising an alarm system to alert an operator of the system to an anomalous condition that interferes with operation of the system, the alert including at least one of:

an out of stock alert, a hardware failure alert, a power failure alert, or a sensor blockage alert.

7. The system of claim 1, wherein consumer interactions tracked by the system are used to control an operation of an output device including lighting, activated signage, an audio loop, or a digital media screen.

8. The system of claim 1, further comprising a video capture system configured to record still or video images in response to the signal processor classifying a received pattern as one of the allowable patterns.

9. The system of claim 1, wherein the plurality of sensors includes weight sensors.

10. The system of claim 9, further comprising a plurality of range sensors configured to track a position and motion of a person when the person is in proximity to the merchandising unit.

11. The system of claim 9, further comprising anti-vibration mounts under the merchandising unit to reduce an amount of non-allowable patterns and false readings.

12. The system of claim 9, further comprising a plurality of proximity sensors configured to track movement of people or the product items through an area of space, the plurality of proximity sensors being positioned at a predetermined height to the movement of the people or the product items.

13. The system of claim 12, wherein the predetermined height is an average waist height of a sampled group of people.

14. A method for tracking interactions with products, comprising:

detecting, by a plurality of sensors, signals that change over time in response to a plurality of product items being removed or returned from a merchandising unit;

incrementally connecting, by a daisy-chain system, the plurality of sensors to upstream and downstream wiring segments joined by multi-conductor wires transmitting power to the plurality of sensors;

storing, by a database, allowable and non-allowable patterns of the plurality of product items;

connecting a signal processor to the daisy-chain system;

receiving and logging, by the signal processor, the detected signals by the plurality of sensors;

transmitting, by the signal processor, the detected signals to the database containing the allowable and non-allowable patterns;

executing a pattern matching algorithm by;

identifying, by the signal processor, a pickup event, of at least one of the product items, detected as a peak in at least one of the detected signals indicating that a shopper has pressed down on the at least one of the product items prior to remove the at least one of the product items, followed by a period, detected by a lower signal level at a tail of the at least one of the detected signals, when the at least one of the product items weighs less than prior to the pickup event,

identifying, by the signal processor, a vibration event when the at least one of the product items is jolted,

generating, by the signal processor, a percentage best pattern match using a plurality of identified pickup and vibrations events for the at least one of the product items within the database,

applying, by the signal processor, an acceptance percentage threshold to the generated percentage best pattern match to classify the plurality of identified pickup events into an allowable pattern,

applying, by the signal processor, a rejection percentage threshold to the generated percentage best pattern match to classify the vibrations events into a non-allowable pattern,

identifying, by the signal processor, un-assignable events that do not meet the acceptance percentage threshold or the rejection percentage threshold,

determining, by the signal processor, from the acceptance percentage threshold, the rejection percentage threshold, and newly discovered patterns, valid signals produced by the plurality of identified pickup events of the at least one of the product items and invalid signals produced by the vibrations events; and

upon determination of the valid signals further identifying, by the signal processor, a position and an associated facing of the at least one of the product items that was identified as being removed or returned from the merchandising unit.