SYSTEM AND METHOD FOR OCCUPANCY ESTIMATION

• US 20100299116A1
• Filed: 02/26/2008
• Published: 11/25/2010
• Est. Priority Date: 09/19/2007
• Status: Abandoned Application

• Current Assignee: UTC Fire & Security Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Original Assignee: United Technologies Corporation (Raytheon Technologies Corporation d/b/a RTX)
• Inventors: Meyn, Sean, Tomastik, Robert N.
• Application Number:
  US12/734,073
• Publication Number:
  US 20100299116A1
• Time in Patent Office: Days
• US Class Current: 703/2
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats

Abstract

An occupancy estimator calculates an occupancy estimate (x) of a region based on sensor data (z) provided by one or more sensor devices and a model-based occupancy estimate generated by an occupant traffic model (f). The occupant traffic model (f) is based on predicted movement of occupants throughout a region. The occupancy estimation system includes an occupancy estimator algorithm (20) that combines the sensor data (z) and the model-based occupancy estimate generated by the occupant traffic model (f) to generate an occupancy estimate (x) for the region.

Claims

1. A system for estimating occupancy in a region, the system comprising:

an input operably connected to receive sensor data from one or more sensor devices, including at least one or more motion detection sensors that provides an output having a first state or a second state, the first state indicating that no occupants have been detected within a particular region, the second state indicating that at least one occupant has been detected within a particular region;

an occupancy estimator operably connected to the input, wherein the occupancy estimator executes an algorithm to generate an occupancy estimate for the region based on the received sensor data and a model-based occupancy estimate generated by an occupant traffic model, wherein based on the state of the data provided by the motion detection sensor, the algorithm assigns to the data a reliability value; and

an output operably connected to the occupancy estimator to communicate the occupancy estimate generated by the occupancy estimator.

2. The system of claim 1, wherein the algorithm interprets the motion sensor data as indicating a particular region is unoccupied when the motion sensor data provided by the motion detection sensors is in the first state

3. The system of claim 2, wherein the motion sensor reliability value assigned by the algorithm indicates that the motion sensor data is reliable if the output of the motion detection sensor is in the first state and unreliable if the output of the motion detection sensor is in the second state.

4. A system for estimating occupancy in a region, the system comprising:

an input operably connected to receive sensor data from one or more sensor devices, wherein the input provided by the sensor devices provides information regarding a number of occupants moving between adjacent regions;

an occupancy estimator operably connected to the input, wherein the occupancy estimator executes an algorithm to generate a state estimate for the region based on the received sensor data and a model-based state estimate generated by an occupant traffic model, wherein the state estimate is comprised of state variables that include both occupancy estimates for the region and flow estimates associated with the number of occupants moving between adjacent regions; and

an output operably connected to the occupancy estimator to communicate the state variable generated by the occupancy estimator.

5. The system of claim 4, wherein the occupancy estimator projects the state estimate onto a range of allowable values based on one or more constraints.

6. The system of claim 5, wherein the constraints applied to the state estimate include a minimum value associated with occupancy estimates for the region, a maximum value associated with occupancy estimates for the region, a minimum value associated with the number of occupants moving between adjacent regions, a maximum value associated with the number of occupants moving between adjacent regions, or a combination thereof.

7. The system of claim 4, wherein the occupant traffic model is comprised of state equations that generate occupancy estimates based on the current number of occupants in a region and the number of occupants modeled to move between adjacent regions.

8. The system of claim 7, wherein the state equations related to the number of occupants expected to move between adjacent regions are limited by a constraint that models physically the number of occupants likely able to move between adjacent regions in a given time-step.

9. The system of claim 8, wherein the algorithm employed by the occupancy estimator is an Extended Kalman Filter that generates with respect to each state estimate a mean estimate and a covariance estimate, wherein the covariance estimate generated with respect to the flow of occupants between adjacent regions is limited by the constraint that models the number of occupants able to move between adjacent regions in a given time-step.

10. A method for estimating occupancy in a region, the method comprising:

acquiring sensor data from one or more sensor devices, wherein at least one of the sensor devices provide data indicative of a number of occupants moving between regions;

calculating model-based estimates of state variables that include model-based estimates of number of occupants located in each region as well as model-based estimates of number of occupants flowing between regions based on an occupant traffic model that predicts movement of occupants within a region; and

generating a corrected estimate associated with each state variable by combining the model-based estimate of each state variable with the acquired sensor data.

11. The method of claim 10, wherein generating the corrected estimate associated with each state variable includes:

employing an extended Kalman filter (EKF) to combine the model-based estimate and the acquired sensor data, wherein the EKF generates a mean estimate and a covariance associated with each state variable.

12. The method of claim 11, wherein generating the covariance associated with each state variable includes:

generating a predicted covariance associated with each state variable based on the occupant traffic model, noise in the occupant traffic, and a previous covariance estimated associated with each state variable;

modifying the predicted covariance associated with the state variable describing the flow of occupants between adjacent zones by applying a flow constraint value to the predicted covariance.

generating an innovation covariance associated with each state variable based on a sensor model and the modified predicted covariance associated with each state variable;

calculating a weighting parameter based on the innovation covariance and the modified predicated covariance; and

updating the covariance estimate based on the weighting parameter, the previous covariance estimate associated with each state variable, and the innovation covariance value.

13. The method of claim 12, wherein generating the mean estimate associated with each state variable includes:

calculating a measurement prediction of each state variable based on the model-based estimates of the state variables and a sensor model;

calculating an innovation based on a comparison of the measurement prediction to the acquired sensor data;

applying the weighting parameter to the innovation estimate and combining with the model-based estimate of the state variables to generate an initial mean estimate of each of the state variables; and

generating the mean estimate based on the model-based estimates, the innovation, the weighting parameter, and a projection function defined by one or more constraints that projects the mean estimate into an allowed range of values.

14. The method of claim 13, wherein the constraints used to define the range of allowable values include a minimum value associated with occupancy estimates for the region, a maximum value associated with occupancy estimates for the region, a minimum value associated with the number of occupants moving between adjacent regions, a maximum value associated with the number of occupants moving between adjacent regions, or a combination thereof.

15. The method of claim 12, wherein the sensor model takes into account attributes associated with the type of sensor providing sensor input.

16. The method of claim 16, wherein sensor input provided by a motion detection sensor is interpreted by the sensor model based on whether the motion detection sensor detects an unoccupied region or an occupied region, wherein if an unoccupied region is detected then the sensor model interprets the sensor data as reliable, and wherein if an occupied region is detected then the sensor model is interpreted as noisy.

17. A system for estimating occupancy in a region, the system comprising:

means for acquiring sensor data from one or more sensor devices, wherein at least one of the sensor devices provide data indicative of a number of occupants moving between regions;

means for calculating model-based estimates of state variables that include model-based estimates of number of occupants located in each region as well as model-based estimates of number of occupants flowing between regions based on an occupant traffic model that predicts movement of occupants within a region; and

means for generating a corrected estimate associated with each state variable by combining the model-based estimate of each state variable with the acquired sensor data.

18. The system of claim 17, further including:

means for interpreting sensor data provided by one or more motion detection sensors based on whether the output provided by the motion detection sensor indicates an unoccupied region or an occupied region, wherein the interpreting means indicates sensor data provided by the motion detection sensor to be unreliable if the output indicates a room is occupied, and indicates sensor data provided by the motion detection sensor to be reliable if the output indicates a room is unoccupied.

19. The system of claim 17, wherein the means for generating a corrected estimate further includes:

means for projecting the corrected estimate into a feasible range of values based on one or more constraints, wherein the constraints includes a minimum value associated with occupancy estimates for the region, a maximum value associated with occupancy estimates for the region, a minimum value associated with the number of occupants moving between adjacent regions, a maximum value associated with the number of occupants moving between adjacent regions, or a combination thereof

20. The system of claim 17, wherein the means for generating a corrected estimate includes a filter that generates as part of the corrected estimate a covariance estimate associated with each state variable.

21. The system of claim 20, wherein the means for generating a corrected estimate further includes constraints associated with the maximum flow of occupants between regions, wherein the covariance estimate is modified based on the constraints to reduce uncertainty associated with the covariance estimate.

22. A computer readable storage medium encoded with a machine-readable computer program code for generating thereof propagation estimates for a region, the computer readable storage medium including instructions for causing a controller to implement a method comprising:

acquiring input from one or more sensor devices, wherein at least one of the sensor devices provides data indicative of a number of occupants flowing between adjacent regions;

calculating a model-based estimate of state variables that include number of occupants in each region and number of occupants flowing between adjacent regions based on an occupant traffic model; and

generating a corrected estimate associated with each state variable by combining the model-based estimate of each state variable with the acquired sensor data, wherein the corrected estimate includes a mean and covariance associated with each state variable.

23. The computer readable media storage of claim 22, wherein the method implemented by the controller further includes:

defining a constraint to the state estimate associated with the number of occupants flowing between adjacent regions to model the maximum number of occupants allowed to flow between adjacent regions; and

modifying the covariance of the state estimates associated with the number of occupants flowing between adjacent regions based on the defined constraint.

24. The computer readable media storage of claim 22, wherein the method implemented by the controller further includes:

acquiring input from one or more sensor devices, wherein at least one of the sensor devices is a motion detection sensor that provides a binary output that indicates whether a particular region is occupied or not; and

assigning a reliability estimate to the input acquired by the motion sensor devices based on the state of the output provided by the device, wherein if the state indicates that a region is unoccupied, the assigned reliability estimate indicates that the input provided by the motion sensor device is reliable, wherein if the state indicates that a region is occupied, the assigned reliability estimate indicates that the input provided by the motion sensor device is unreliable.