Occupancy pattern detection, estimation and prediction

Patent metadata

Field	Value
Publication number	US-9245229-B2
Application number	US-201414322724-A
Country	US
Kind code	B2
Filing date	Jul 2, 2014
Priority date	Sep 14, 2010
Publication date	Jan 26, 2016
Grant date	Jan 26, 2016

Abstract

Systems and methods are described for predicting and/or detecting occupancy of an enclosure, such as a dwelling or other building, which can be used for a number of applications. An a priori stochastic model of occupancy patterns based on information of the enclosure and/or the expected occupants of the enclosure is used to pre-seed an occupancy prediction engine. Along with data from an occupancy sensor, the occupancy prediction engine predicts future occupancy of the enclosure. Various systems and methods for detecting occupancy of an enclosure, such as a dwelling, are also described.

First claim

What is claimed is:

1. An environmental control system for predicting occupancy of an enclosure comprising:
a memory storing a model of occupancy patterns comprising a plurality of coefficients;
an occupancy sensor configured to detect occupancy within the enclosure;
a wireless communication port that receives a location of a cellular phone associated with one of more of the occupants of the enclosure;
one or more processors programmed to:
pre-seed the plurality of coefficients for the model of occupancy patterns with values that are selected based on a geographic location of the enclosure and information describing an initial occupant schedule received through a user interface;
initiate a learning phase for the model of occupancy patterns after installation of the control system in the enclosure;
combine information from the occupancy sensor with the location of the cellular phone to generate an occupancy status for the enclosure;
generate a future occupancy pattern of the enclosure, the future occupancy pattern being based at least in part on the model of occupancy patterns and the occupancy status for the enclosure;
refine the values for the plurality of coefficients using the occupancy status for the enclosure during the learning phase; and
control an environmental condition of the enclosure based on the future occupancy pattern after an end of the learning phase.

2. The system of claim 1 wherein the model is an a priori stochastic model of human occupancy.

3. The system of claim 2 wherein the a priori stochastic model is a comfort and spatial occupancy model that includes one or more statistical profiles.

4. The system of claim 2 wherein the a priori stochastic model includes behavior modeling of activity, itinerary, and/or thermal behavior.

5. The system of claim 1 wherein the model is based at least in part on information selected from the group consisting of: a type of the enclosure, geometrical data about the enclosure, structural data about the enclosure, geographic location of the enclosure, an expected type of occupant of the enclosure, an expected number of occupants of the enclosure, the relational attributes of the occupants of the enclosure, seasons of the year, days of the week, types of day, and times of day.

6. The system of claim 1 wherein the occupancy sensor is selected from a group consisting of: motion detector, powerline sensor, network traffic monitor, radio traffic monitor, microphone, infrared sensor, accelerometer, ultrasonic sensor, pressure sensor, smart utility meter, and light sensor.

7. The system of claim 1 further comprising a second occupancy sensor, wherein the one or more processors are programmed to generate the future occupancy pattern of the enclosure based at least in part on the model, the occupancy status, and the second sensor.

8. A method for predicting occupancy of an enclosure comprising:
accessing, by an environmental control system, a model of occupancy patterns comprising a plurality of coefficients;
receiving, by the environmental control system, occupancy data from an occupancy sensor configured to detect occupancy within the enclosure;
receiving, by the environmental control system, a location of a cellular phone associated with one of more of the occupants of the enclosure;
pre-seeding, by the environmental control system, the plurality of coefficients for the model of occupancy patterns with values that are selected based on a geographic location of the enclosure and information describing an initial occupant schedule received through a user interface;
initiating, by the environmental control system, a learning phase for the model of occupancy patterns after installation of the control system in the enclosure;
combining, by the environmental control system, the occupancy data from the occupancy sensor with the location of the cellular phone to generate an occupancy status for the enclosure;
generating, by the environmental control system, a future occupancy pattern of the enclosure based at least in part on the model of occupancy patterns and the occupancy status for the enclosure;
refining, by the environmental control system, the values for the plurality of coefficients using the occupancy status for the enclosure during the learning phase; and
controlling, by the environmental control system, an environmental condition of the enclosure based on the future occupancy pattern after an end of the learning phase.

9. The method of claim 8 further comprising receiving user inputted data, wherein the future occupancy pattern of the enclosure is generated further based in part on the user inputted data.

10. The method of claim 9 wherein the user inputted data includes occupancy information directly inputted by an occupant of the enclosure and/or calendar information.

11. The method of claim 9 further comprising detecting periodicities in the user inputted data, wherein the future occupancy pattern of the enclosure is generated further based in part on the detected periodicities in the user inputted data.

12. The method of claim 8 further comprising:
comparing the future occupancy pattern of the enclosure with the occupancy data from the occupancy sensor; and
updating the future occupancy pattern based at least in part on the result of the comparison.

13. The method of claim 8 wherein the occupancy sensor is one of a plurality of sensors arranged at different sub-regions of the enclosure, receiving occupancy data from the occupancy sensor includes receiving occupancy data from the plurality of sensors, and generating the future occupancy pattern of the enclosure includes predicting future occupancy of the enclosure based at least in part on the occupancy data received from the plurality of sensors.

14. The method of claim 8 wherein the future occupancy pattern is based at least in part on a maximum-likelihood approach.

15. A tangible non-transitory computer-readable storage medium having instructions stored thereon that, when executed by a computing device, cause the computing device to perform operations comprising:
accessing a model of occupancy patterns comprising a plurality of coefficients;
receiving occupancy data from an occupancy sensor configured to detect occupancy within the enclosure;
receiving a location of a cellular phone associated with one of more of the occupants of the enclosure;
pre-seeding the plurality of coefficients for the model of occupancy patterns with values that are selected based on a geographic location of the enclosure and information describing an initial occupant schedule received through a user interface;
initiating a learning phase for the model of occupancy patterns after installation of the control system in the enclosure;
combining the occupancy data from the occupancy sensor with the location of the cellular phone to generate an occupancy status for the enclosure;
generating a future occupancy pattern of the enclosure based at least in part on the model of occupancy patterns and the occupancy status for the enclosure;
refining the values for the plurality of coefficients using the occupancy status for the enclosure during the learning phase; and
controlling an environmental condition of the enclosure based on the future occupancy pattern after an end of the learning phase.

16. The storage medium of claim 15 wherein the model is based at least in part on an occupant type.

17. The storage medium of claim 16 wherein the occupant type depends on one or more occupant attributes selected from a group consisting of: age, school enrollment status,

Assignees

• Google Inc

Inventors

• Fadell Anthony Michael
• Rogers Matthew Lee
• Rogers Kipp Avery
• Ishihara Abraham K
• Ben-Menahem Shahar
• Sharan Rangoli

Classifications

• G06N5/04

  Inference or reasoning models · CPC title

• H04Q2209/40

  using a wireless architecture · CPC title

• G05B2219/2642

  Domotique, domestic, home control, automation, smart house · CPC title

• H04L12/2823

  Reporting information sensed by appliance or service execution status of appliance services in a home automation network (device-related reporting H04L43/065; arrangements in telecontrol or telemetry systems for selectively calling a substation from a main station, in which substation desired apparatus is selected for applying a control signal thereto or for obtaining measured values therefrom H04Q9/00) · CPC title

• G06N20/00

  Machine learning · CPC title