Method for monitoring temperature-controlled units in a store

1. A method for monitoring temperature-controlled inventory within a store, the method comprising:

at a robotic system, during a first scan cycle:

autonomously navigating toward a temperature-controlled unit in the store;

recording a photographic image and a thermal image of the temperature-controlled unit; and

identifying a set of products stocked in the temperature-controlled unit based on features detected in the photographic image;

for each product in the set of products:

retrieving an emissivity of a surface of the product based on an identity of the product;

selecting a thermal pixel, in the thermal image, depicting the product; and

calculating a temperature of the product during the scan cycle based on the emissivity of the surface of the product and a thermal value stored in the thermal pixel; and

generating a record of temperatures of the set of products stocked in the temperature-controlled unit during the first scan cycle.

2. The method of claim 1 :

further comprising, during the first scan cycle, recording a depth image of the temperature-controlled unit; and

wherein calculating the temperature of each product in the set of products during the first scan cycle comprises, for each product in the set of products:

calculating a distance from the robotic system to the product based on the depth image; and

calculating the temperature of the product during the scan cycle further based on the distance from the robotic system to the product.

3. The method of claim 1 :

further comprising, during the first scan cycle, recording an ambient temperature, an ambient humidity, and an ambient light level proximal the temperature-controlled unit; and

wherein calculating the temperature of each product in the set of products during the first scan cycle comprises, for each product in the set of products, passing the emissivity of the surface of the product, the thermal value stored in the thermal pixel, the ambient temperature, the ambient humidity, and the ambient light level into a physics model to calculate the temperature of the product during the scan cycle.

4. The method of claim 1 :

wherein recording the photographic image and the thermal image of the temperature-controlled unit comprises recording the photographic image and the thermal image of the temperature-controlled unit comprising an open temperature-controlled unit in the store; and

further comprising:

at the robotic system, during the first scan cycle:

autonomously navigating toward an enclosed temperature-controlled unit in the store;

recording a second photographic image of the enclosed temperature-controlled unit; and

recording a set of temperature values wirelessly transmitted by a set of temperature-enabled transmitters coupled to the enclosed temperature-controlled unit;

identifying a second set of products stocked in the enclosed temperature-controlled unit based on features detected in the second photographic image;

accessing locations of the set of temperature-enabled transmitters coupled to the enclosed temperature-controlled unit;

mapping the set of temperature values received from the set of temperature-enabled transmitters to the second set of products during the first scan cycle based on:

locations of the set of temperature-enabled transmitters coupled to the enclosed temperature-controlled unit; and

locations of the set of products identified in the second photographic image; and

generating a second record of temperatures of the second set of products stocked in the enclosed temperature-controlled unit during the first scan cycle.

5. The method of claim 1 , further comprising:

deriving a temperature gradient from the thermal image;

calculating a spatial temperature variance across the temperature-controlled unit from the temperature gradient; and

in response to the spatial temperature variance across the temperature-controlled unit exceeding a threshold variance, serving a maintenance prompt to inspect the temperature-controlled unit to a computing device affiliated with an associate of the store.

6. The method of claim 1 , further comprising:

accessing a temperature specification of a first product in the set of products;

based on the record of temperatures of the set of products, detecting deviation of a first temperature of the first product during the first scan cycle from the temperature specification of the first product; and

in response to detecting deviation of the first temperature of the first product from the temperature specification:

generating an electronic notification comprising a prompt to remove the first product from the temperature-controlled unit; and

serving the electronic notification to a computing device affiliated with an associate of the store.

7. The method of claim 1 , wherein identifying the set of products stocked in the temperature-controlled unit comprises:

accessing a first list of products assigned to the temperature-controlled unit;

retrieving a first set of template images from a database of template images, each template image in the first set of template images representing visual features of a product in the first list of products;

detecting a first object in a first region of the photographic image;

extracting a first set of features from the first region of the photographic image; and

identifying the first object as a unit of a first product, in the set of products, based on correspondence between the first set of features and template features in a first template image, in the first set of template images, representing the first product.

8. A method for monitoring temperature-controlled inventory within a store, the method comprising:

at a robotic system, during a first scan cycle:

autonomously navigating toward a temperature-controlled unit in the store;

recording a photographic image of the temperature-controlled unit; and

recording a set of temperature values measured by a set of temperature sensors coupled to the temperature-controlled unit;

identifying a set of products stocked in the temperature-controlled unit based on features detected in the photographic image;

for each product in the set of products:

detecting a location of the product stocked in the temperature-controlled unit based on a position of the product detected in the photographic image; and

estimating a temperature of the product based on:

the location of the product stocked in the temperature-controlled unit;

the set of temperature values; and

locations of the set of temperature sensors coupled to the temperature-controlled unit; and

generating a record of temperatures of the set of products stocked in the temperature-controlled unit during the first scan cycle.

9. The method of claim 8 , wherein estimating the temperature of the product for each product in the set of products comprises:

calculating a temperature gradient across the temperature-controlled unit by interpolating temperatures between the set of temperature values based on locations of corresponding temperature sensors, in the set of temperature sensors, in the temperature-controlled unit; and

for each product in the set of products:

projecting a boundary of product, identified in the photographic image, onto the temperature gradient; and

extracting a temperature of the product from a region of the temperature gradient contained within the boundary of the product projected onto the temperature gradient.

10. The method of claim 8 , further comprising:

identifying the temperature-controlled unit; and

accessing locations of the set of temperature sensors, relative to the temperature-controlled unit, from a database of temperature sensor locations in the store based on an identity of the temperature-controlled unit.

11. The method of claim 8 :

further comprising, at the robotic system during the first scan cycle, broadcasting a query for temperature values from the set of temperature sensors coupled to the temperature-controlled unit; and

wherein recording the set of temperature values comprises recording the set of temperature values wirelessly transmitted by the set of temperature sensors responsive to the query.

12. The method of claim 8 :

wherein recording the set of temperature values comprises recording a set of images of the temperature-controlled unit, the set of images:

depicting a set of temperature displays coupled to the set of temperature sensors; and

comprising the photographic image; and

further comprising:

detecting the set of temperature displays in the set of images; and

extracting the set of temperature values from regions in the set of images depicting the set of temperature displays.

13. The method of claim 12 :

wherein detecting the set of temperature displays in the set of images comprises

detecting a set of optical fiducials in the set of images; and

identifying the set of temperature displays in the set of images based on the set of optical fiducials; and

further comprising extracting locations of the set of temperature sensors, relative to the temperature-controlled unit, from the set of images based on locations of the set of optical fiducials detected in the set of images.

14. A method for monitoring temperature-controlled inventory within a store, the method comprising:

accessing a photographic image of a temperature-controlled unit within the store, the photographic image captured at a first time; and

accessing a set of temperatures within the temperature-controlled unit at approximately the first time;

identifying a set of products stocked in the temperature-controlled unit at the first time based on features detected in the photographic image;

mapping the set of temperatures to the set of products based on positions of products, in the set of products, identified in the photographic image; and

generating a record of temperatures of the set of products stocked in the temperature-controlled unit at the first time.

15. The method of claim 14 :

wherein accessing the set of temperatures within the temperature-controlled unit comprises accessing the set of temperatures returned by a set of temperature sensors within the temperature-controlled unit at approximately the first time;

further comprising:

calculating a temperature gradient across the enclosed temperature-controlled unit by interpolating between the set of temperature values based on locations of corresponding temperature sensors, in the set of temperature sensors, in the enclosed temperature-controlled unit;

calculating a spatial temperature variance in the temperature gradient across the enclosed temperature-controlled unit at the first time; and

in response to the spatial temperature variance of the temperature-controlled unit exceeding a threshold variance, serving a maintenance prompt to inspect the temperature-controlled unit to a computing device affiliated with an associate of the store.

16. The method of claim 14 :

wherein accessing the set of temperatures within the temperature-controlled unit comprises accessing a thermal image of the temperature-controlled unit captured at approximately the first time, the thermal image representing temperatures across the temperature-controlled unit; and

wherein mapping the set of temperatures to the set of products comprises, for each product in the set of products:

retrieving an emissivity of a surface of the product based on an identity of the product;

selecting a thermal pixel, in the thermal image, depicting the product; and

calculating a temperature of the product during the scan cycle based on the emissivity of the surface of the product and a thermal value stored in the thermal pixel.

17. The method of claim 16 , further comprising:

dispatching a robotic system to execute a first scan cycle in the store; and

at the robotic system, during the first scan cycle:

autonomously navigating toward the temperature-controlled unit;

recording the photographic image of the temperature-controlled unit at the first time; and

recording the thermal image of the temperature-controlled unit at approximately the first time.

18. The method of claim 14 , further comprising:

dispatching a robotic system to execute a first scan cycle in the store; and

at the robotic system, during the first scan cycle:

autonomously navigating toward the temperature-controlled unit;

recording the photographic image of the temperature-controlled unit at the first time;

broadcasting a query to a set of temperature-enabled transmitters arranged in the temperature-controlled unit; and

recording the set of temperature values, within the temperature-controlled unit, received from the set of temperature-enabled transmitters at approximately the first time.

19. The method of claim 14 , further comprising:

calculating a mean temperature of the temperature-controlled unit at the first time based on the set of temperatures within the temperature-controlled unit;

accessing a temperature setting of the temperature-controlled unit at the first time; and

in response to the mean temperature of the temperature-controlled unit deviating from the temperature setting at the first time by more than a threshold difference, serving a maintenance prompt to inspect the temperature-controlled unit to a computing device affiliated with an associate of the store.

20. The method of claim 14 , further comprising:

accessing a second photographic image of the temperature-controlled unit at a second time succeeding the first time;

accessing a second set of temperatures within the temperature-controlled unit at approximately the second time;

identifying the set of products stocked in the temperature-controlled unit at the second time based on features detected in the second photographic image;

mapping the second set of temperatures to the set of products based on positions of products, in the set of products, identified in the second photographic image; and

estimating temperature exposures of products, in the set of products, between the first time and the second time by integrating from the first set of temperatures to the second set of temperatures of the set of products over a time duration from approximately the first time to approximately the second time.

21. The method of claim 20 , further comprising:

accessing a shelf life model of a first product in the set of products;

estimating a remaining shelf life of the first product based on the shelf life model and a first temperature exposure of the first product; and

in response to the remaining shelf life of the first product falling below a threshold duration:

generating an electronic notification identifying the first product and the remaining shelf life of the first product; and

serving the electronic notification to a computing device affiliated with an associate of the store.