Nest Patent – Occupancy Pattern Detection, Estimation and Prediction

OCCUPANCY PATTERN DETECTION, ESTIMATION AND PREDICTION

 

Inventors:FADELL; Anthony Michael(Portola Valley, CA) ; ROGERS; Matthew Lee(Los Gatos, CA) ; ROGERS; Kipp Avery(Chicago, IL) ; ISHIHARA; Abraham K.(Palo Alto, CA) ; BEN-MENAHEM; Shahar(Mountain View, CA) ; SHARAN; Rangoli(Sunnyvale, CA)

 

Applicant:
NameCityStateCountryType

NEST LABS, INC.
Palo AltoCAUS
Assignee:NEST LABS, INC.
Palo Alto
CA
Family ID:45807659
Appl. No.:14/322724
Filed:July 2, 2014

 

Overview

 

This patent application discloses systems and methods for predicting and/or detecting occupancy of an enclosure, such as a dwelling or other building. Using an occupancy prediction engine in conjunction with data received from at least occupancy sensor installed in a dwelling, the present invention is able to perform a number of applications.

For example, applications that can benefit from accurate occupancy prediction include heating, ventilating and air conditioning (HVAC), lighting management, hot water heating and management, security, emergency response, and the management and charging of rechargeable batteries (e.g. for electric vehicles).

In general, applications that greatly benefit from occupancy prediction are those that particularly benefit from knowing or accurately estimating, in advance, when the structure will be occupied. The lead-time of the prediction will especially aid applications that have an inherent lag-time to reach a certain state. For example, heating and cooling a structure to an acceptable level has an associated lag time of several minutes to more than one hour.

Therefore it is beneficial to accurately predict ahead of time, when an occupant or occupants will be entering and/or leaving structure. Additionally, energy savings can be obtained due to predicting and/or detecting occupancy for both short term, such as intraday periods and long term, such as multi-day vacation periods, when the structure can remain unconditioned or more economically conditioned.

What is Claimed to be the Invention

 

Note: this application is still pending, and the scope of the patent protection granted, if any, may change during patent examination.

A system for predicting occupancy of an enclosure comprising:

  1. a model of occupancy patterns based in part on information regarding the enclosure and/or the expected occupants of the enclosure;
  2. a sensor configured to detect occupancy within the enclosure; and
  3. an occupancy predictor configured to predict future occupancy of the enclosure based at least in part on the model and the occupancy detected by the sensor.

A method for predicting occupancy of an enclosure comprising:

receiving a model of occupancy patterns based in part on information regarding the enclosure and/or the expected occupants of the enclosure;

receiving occupancy data from a sensor configured to detect occupancy within the enclosure, the occupancy data being indicative of the occupancy detected by the sensor; and

predicting, by a computing device, future occupancy of the enclosure based at least in part on the model and the occupancy data.

 

How it Works

 

The systems can include a prior (a priori) stochastic model of human occupancy, thermal comfort and activity patterns, based in part on information pertaining to the type, dimensions, layout and/or the expected average number of occupants of the structure (whether a home or other type of structure) and on the calendar (time of year, day of week, time of day), and also based on prevailing and forecast local weather conditions.

Such a stochastic model can have multiple parameters, which can be initially estimated from a questionnaire filled by residents and/or from accumulated statistical data for structures of type and usage, and occupant characteristics (i.e. according to household type) similar to the structure in question.

Over time, the parameters of the a priori stochastic occupancy, comfort, activity model, can be further trained using cumulative logs of sensor data acquired within the actual structure in question. For example, if the a priori model predicts the absence of occupants on Wednesdays during daytime, but occupancy sensors sense human presence on Wednesdays consistently for several weeks, the a priori behavior model can be corrected for this information.

As used herein the term “sensor” refers generally to a device or system that measures and/or registers a substance, physical phenomenon and/or physical quantity. The sensor may convert a measurement into a signal, which can be interpreted by an observer, instrument and/or system. A sensor can be implemented as a special purpose device and/or can be implemented as software running on a general-purpose computer system.

Known methods for electronic occupancy detection include acoustical detection and optical detection (including infrared light, visible, laser and radar technology). Motion detectors can process motion-sensor data, or employ cameras connected to a computer which stores and manages captured images to be viewed and analyzed later or viewed over a computer network. Examples of motion detection and sensing applications are (a) detection of unauthorized entry, (b) detection of cessation of occupancy of an area to extinguish lighting and (c) detection of a moving object which triggers a camera to record subsequent events. A motion sensor/detector is thus important for electronic security systems, as well as preventing the wasteful illumination of unoccupied space.

The occupancy prediction can be used in the actuation and/or control of an HVAC system for the enclosure or various other applications such as: home automation, home security, lighting control, and/or the charging of rechargeable batteries.

 

If you are interested in more detail related to your situation it is best to speak with an attorney.

Yuri Eliezer heads the intellectual property practice group at Founders Legal. As an entrepreneur who saw the importance of early-stage patent protection, Yuri founded SmartUp®. Clients he has served include Microsoft, Cisco, Cox, AT&T, General Electric, the Georgia Institute of Technology, and Coca-Cola.

yuri