Adaptive trigger sequencing for site control automation

• US 10,215,434 B2
• Filed: 02/27/2017
• Issued: 02/26/2019
• Est. Priority Date: 11/07/2012
• Status: Active Grant

• Current Assignee: Think Automatic, LLC
• Original Assignee: Think Automatic, LLC
• Inventors: Harris, Stephen
• Primary Examiner(s): Ali, Mohammad
• Assistant Examiner(s): Azad, Md
• Application Number:
  US15/443,970
• Publication Number:
  US 20170167746A1
• Time in Patent Office: 729 Days
• Field of Search: 700275
• CPC Class Codes: F24F 11/30 for purposes related to the operation of the system, e.g. for safety or monitoring
• CPC Class Codes: F24F 11/52 Indication arrangements, e.g. displays
• CPC Class Codes: F24F 11/62 characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
• CPC Class Codes: F24F 11/65 for selecting an operating mode
• CPC Class Codes: F24F 2110/00 Control inputs relating to air properties
• CPC Class Codes: F24F 2120/20 Feedback from users
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G05B 19/048 Monitoring; Safety
• CPC Class Codes: G05B 2219/2642 Domotique, domestic, home control, automation, smart house
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 3/126 Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

Disclosed is a method and apparatus for an environmental control system in which a genetic learning algorithm creates scenes and scene triggers and in which a fitness function scores the scenes through end-user interaction.

Claims

1. A machine-teaching environmental control method comprising:

a first hub obtaining first, second, and third output configurations that all correspond to a first trigger and that are respectively associated with first, second, and third scalar scores;

presenting said first output configuration in response to a first sequential instance of said first trigger and to said scalar scores indicating that said first output configuration is ranked before said second output configuration;

presenting said second output configuration in response to a second sequential instance of said first trigger and to said scalar scores indicating that said second output configuration is ranked before said third output configuration;

presenting said third output configuration in response to a third sequential instance of said first trigger after said first and second output configuration are presented; and

presenting a modified sequence of said first, second, and third output configurations in response to a first change in said second scalar score resulting from a magnitude of a time interval between said second and third sequential instances of said first trigger either by increasing said second scalar score as a conditional response to a determination that said first magnitude of said time interval between said second and third sequential instances of said first trigger is smaller than a minimum threshold or by decreasing said second scalar score as a conditional response to a determination that said first magnitude of said time interval between said second and third sequential instances of said first trigger is larger than a maximum threshold, said maximum threshold being greater than said minimum threshold, wherein said presentation sequence is manifested in a plurality of physical control devices, wherein said plurality of physical control devices include first and second physical control devices at a first facility and operably coupled with said first hub, and wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises;

adjusting second scalar score partly based on said magnitude of said time interval between successive timestamps respectively signifying said second and third sequential instances of said first trigger and partly based on one of said timestamps falling within a particular time-of-day, said particular time-of-day being morning in combination with the remaining elements and features of the claimed invention.

2. The machine-teaching environmental control method of claim 1, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining a first set of device commands and attributes for said first physical control device;

obtaining a first device identifier assigned to said first physical control device;

obtaining a second set of device commands and attributes for said second physical control device;

obtaining a second device identifier assigned to said second physical control device;

for at least said first and second physical control devices, generating several scene candidates by performing a genetic operator on said sets of device commands so that said first, second, and third output configurations each comprise a respective one of said several scene candidates, wherein said first and second set of attributes comprise a common location at said first facility; and

developing said first, second, and third scalar scores each as a scene trigger score by performing a fitness function on said several scene candidates, wherein said fitness function comprises assigning a respective average scene trigger score for one or more scenes at a second facility correlated to respective one of said several scene candidates.

3. A machine-teaching environmental control method comprising:

a first hub obtaining first, second, and third output configurations that all correspond to a first trigger and that are respectively associated with first, second, and third scalar scores;

presenting said first output configuration in response to a first sequential instance of said first trigger and to said scalar scores indicating that said first output configuration is ranked before said second output configuration;

presenting said second output configuration in response to a second sequential instance of said first trigger and to said scalar scores indicating that said second output configuration is ranked before said third output configuration;

presenting said third output configuration in response to a third sequential instance of said first trigger after said first and second output configuration are presented; and

presenting a modified sequence of said first, second, and third output configurations in response to a first change in said second scalar score resulting from a magnitude of a time interval between said second and third sequential instances of said first trigger, wherein said presentation sequence is manifested in a plurality of physical control devices, wherein said plurality of physical control devices include first and second physical control devices at a first facility and operably coupled with said first hub, and wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises;

adjusting second scalar score partly based on said magnitude of said time interval between successive timestamps respectively signifying said second and third sequential instances of said first trigger and partly based on one of said timestamps falling within a particular time-of-day, said particular time-of-day being morning.

4. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

adjusting second scalar score partly based on said magnitude of said time interval between successive timestamps respectively signifying said second and third sequential instances of said first trigger and partly based on one of said timestamps falling within a particular time-of-day.

5. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

adjusting second scalar score partly based on said magnitude of said time interval between successive timestamps respectively signifying said second and third sequential instances of said first trigger and partly based on one of said timestamps falling within a particular day-of-the-week set.

6. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

adjusting second scalar score partly based on said magnitude of said time interval between successive timestamps respectively signifying said second and third sequential instances of said first trigger and partly based on one of said timestamps falling within a particular time-of-year.

7. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

receiving a first device report from said first physical control device;

receiving a second device report from said second physical control device; and

obtaining a list of thousands of device reports that include said first and second device reports sorted by location and by physical control device, wherein more than half of said thousands of device reports each include first event information, wherein event information that determines said magnitude of said time interval between said second and third sequential instances of said first trigger and that triggers said first change in said second scalar score includes said first event information and wherein said first scene comprises said first output configuration and wherein said second scene comprises said second output configuration.

8. The machine-teaching environmental control method of claim 3, further comprising:

receiving a first device report from said first physical control device;

receiving a second device report from said second physical control device;

sorting several device reports that include said first and second device reports by location and by physical control device, wherein said sorting by location differentiates between said first facility and a second facility, wherein said several device reports include a device report from a third physical control that is mounted at said second facility and operably coupled with a second hub at said second facility, and wherein at least said sorting is performed at one or more servers operably coupled remotely with said first and second hubs;

defining a first scene comprising a first combination of device events associated with said first trigger;

defining a second scene comprising a second combination of device events associated with said second trigger;

creating a first scene trigger group comprising multiple scenes associated with said first and second triggers, wherein said multiple scenes of said first scene trigger group include said first and second triggers; and

determining a scene trigger score for each of said multiple scenes in said first scene trigger group utilizing at least a first fitness function, wherein said first scene comprises said first output configuration and wherein said second scene comprises said second output configuration.

9. The machine-teaching environmental control method of claim 3, further comprising:

receiving a first device report from said first physical control device;

receiving a second device report from said second physical control device;

sorting several device reports that include said first and second device reports by location and by physical control device;

defining a first scene comprising a first combination of device events associated with said first trigger;

defining a second scene comprising a second combination of device events associated with said second trigger;

creating a first scene trigger group comprising multiple scenes associated with said first and second triggers, wherein said multiple scenes of said first scene trigger group include said first and second triggers; and

determining a scene trigger score for each of said multiple scenes in said first scene trigger group utilizing at least a first fitness function, wherein said first scene comprises said first output configuration and wherein said second scene comprises said second output configuration.

10. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining a first set of device commands and attributes for said first physical control device;

obtaining a first device identifier assigned to said first physical control device;

obtaining a second set of device commands and attributes for said second physical control device;

obtaining a second device identifier assigned to said second physical control device; and

for at least said first and second physical control devices, generating several scene candidates by performing a genetic operator on said sets of device commands so that said first, second, and third output configurations each comprise a respective one of said several scene candidates, wherein said first and second set of attributes comprise a common location at said first facility.

11. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining a first set of device commands and attributes for said first physical control device;

obtaining a first device identifier assigned to said first physical control device;

obtaining a second set of device commands and attributes for said second physical control device;

obtaining a second device identifier assigned to said second physical control device;

for at least said first and second physical control devices, generating several scene candidates by performing a genetic operator on said sets of device commands so that said first, second, and third output configurations each comprise a respective one of said several scene candidates, wherein said first and second set of attributes comprise a common location at said first facility; and

developing said first, second, and third scalar scores each as a scene trigger score by performing a fitness function on said several scene candidates.

12. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining a first set of device commands and attributes for said first physical control device;

obtaining a first device identifier assigned to said first physical control device;

obtaining a second set of device commands and attributes for said second physical control device;

obtaining a second device identifier assigned to said second physical control device;

for at least said first and second physical control devices, generating several scene candidates by performing a genetic operator on said sets of device commands so that said first, second, and third output configurations each comprise a respective one of said several scene candidates, wherein said first and second set of attributes comprise a common location at said first facility; and

developing said first, second, and third scalar scores each as a scene trigger score by performing a fitness function on said several scene candidates, wherein said fitness function comprises assigning a respective default scene trigger score for one or more scenes at a second facility correlated to respective one of said several scene candidates.

13. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining a first set of device commands and attributes for said first physical control device;

obtaining a first device identifier assigned to said first physical control device;

obtaining a second set of device commands and attributes for said second physical control device;

obtaining a second device identifier assigned to said second physical control device; and

for at least said first and second physical control devices, generating several scene candidates by performing a genetic operator on said sets of device commands so that said first, second, and third output configurations each comprise a respective one of said several scene candidates, wherein said first and second set of attributes comprise a common location at said first facility.

14. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining a first set of device commands and attributes for said first physical control device;

obtaining a first device identifier assigned to said first physical control device;

obtaining a second set of device commands and attributes for said second physical control device;

obtaining a second device identifier assigned to said second physical control device;

for at least said first and second physical control devices, generating several scene candidates by performing a genetic operator on said sets of device commands so that said first, second, and third output configurations each comprise a respective one of said several scene candidates, wherein said first and second set of attributes comprise a common location at said first facility; and

developing said first, second, and third scalar scores each as a scene trigger score by performing a fitness function on said several scene candidates, wherein said genetic operator comprises an arbitrary combination of said first and second sets of device commands.

15. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

downgrading said first scalar score as a conditional response to a determination that said magnitude of said time interval between said second and third sequential instances of said first trigger is larger than a maximum threshold.

16. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

improving said second scalar score as a conditional response to a determination that said magnitude of said time interval between said second and third sequential instances of said first trigger is smaller than a magnitude of said time interval between said first and second sequential instances of said first trigger, whereby said magnitude effectively determines a minimum threshold.

17. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

downgrading said first scalar score as a conditional response to a determination that said magnitude of said time interval between said first and second sequential instances of said first trigger is smaller than a minimum threshold.

18. The machine-teaching environmental control method of claim 3, wherein said presenting said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises:

obtaining said magnitude of said time interval between said second and third sequential instances of said first trigger by computing a difference between when said second sequential instance of said first trigger occurred and when said third sequential instance of said first trigger occurred;

obtaining a user-defined threshold either as a minimum threshold or as a maximum threshold, wherein said user-defined threshold is within an order of magnitude of one minute; and

detecting said first change in said second scalar score as a conditional response to a comparison between said magnitude of said time interval between said second and third sequential instances of said first trigger.

19. A machine-teaching environmental control system comprising:

transistor-based circuitry configured to obtain first, second, and third output configurations that all correspond to a first trigger and that are respectively associated with first, second, and third scalar scores;

transistor-based circuitry configured to present said first output configuration in response to a first sequential instance of said first trigger and to said scalar scores indicating that said first output configuration is ranked before said second output configuration;

transistor-based circuitry configured to present said second output configuration in response to a second sequential instance of said first trigger and to said scalar scores indicating that said second output configuration is ranked before said third output configuration;

transistor-based circuitry configured to present said third output configuration in response to a third sequential instance of said first trigger after said first and second output configuration are presented; and

transistor-based circuitry configured to present a modified sequence of said first, second, and third output configurations in response to a first change in said second scalar score resulting from a magnitude of a time interval between said second and third sequential instances of said first trigger, wherein said presentation sequence is manifested in a plurality of physical control devices, wherein said plurality of physical control devices include first and second physical control devices at a first facility and operably coupled with said first hub, and wherein said transistor-based circuitry configured to present said modified sequence of said first, second, and third output configurations in response to said first change in said second scalar score resulting from said magnitude of said time interval between said second and third sequential instances of said first trigger comprises;

transistor-based circuitry configured to adjust second scalar score partly based on said magnitude of said time interval between successive timestamps respectively signifying said second and third sequential instances of said first trigger and partly based on one of said timestamps falling within a particular time-of-day, said particular time-of-day being morning.