Activity Recognition in Multi-Entity Environments

• US 20130006899A1
• Filed: 06/29/2012
• Published: 01/03/2013
• Est. Priority Date: 07/01/2011
• Status: Active Grant

• Current Assignee: Washington State University Office Of Commercialization, WSU Research Foundation
• Original Assignee: WSU Research Foundation
• Inventors: Cook, Diane J.
• Granted Patent: US 9,460,350 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69

Abstract

A physical environment is equipped with a plurality of non-obtrusive sensors (e.g., motion sensors). As a plurality of residents perform various activities within the physical environment, sensor readings are received from one or more of the sensors. Based on the sensor readings, each of the plurality of residents is identified and locations of each of the plurality of residents are tracked.

Claims

1. A method comprising:

defining a graph including vertices and edges, wherein each vertex corresponds to a sensor in a physical environment and each edge represents a direct path between two vertices with no other vertices between the two vertices along the path;

defining a set of logical rules for identifying and tracking entities within the physical environment based on the graph;

receiving data from a particular sensor in the physical environment corresponding to a vertex in the graph; and

executing instructions on a computer processor to identify and track a location of a particular entity within the physical environment based on the data received from the particular sensor and based on the set of logical rules.

2. A method as recited in claim 1, further comprising:

Receiving data from a plurality of sensors within the physical environment over a period of time;

Analyzing the data from the plurality of sensors to identify specific residents within the smart environment;

3. A method as recited in claim 2, wherein the data from the plurality of sensors is analyzed to identify specific residents within the smart environment based on behaviometrics.

4. A method as recited in claim 3, wherein a Viterbi alborithm is used to identify a particular resident that corresponds to data from a particular sensor.

5. A method as recited in claim 3, wherein a supervised learning technique is used to learn a mapping from sensor event contexts to a corresponding resident based on training data.

6. A method as recited in claim 5, wherein the supervised learning technique comprises a naï

ve Bayes classifier.

7. A method as recited in claim 5, wherein the supervised learning technique comprises a hidden Markov model.

8. A method comprising:

defining a transition matrix for a physical environment, wherein the transition matrix represents a probability of an entity moving from a first sensor location within the physical environment to a second sensor location within the physical environment;

receiving a sensor event from a particular sensor in the physical environment;

for each entity of a plurality of entities in the physical environment;

determining a known current location of the entity; and

determining, based on the transition matrix, a probability that the entity moved from the known current location to the second sensor location; and

identifying a particular entity within the physical environment as an entity that caused the sensor event based on the particular entity having a highest probability of having moved from the known current location of the particular entity to the second sensor location.

9. A method comprising:

receiving, from a smart environment equipped with non-obtrusive sensors, a plurality of sensor readings;

executing computer-readable instructions to analyze the sensor readings to detect a plurality of residents within the smart environment; and

analyzing the sensor readings to track locations of each of the plurality of residents within the smart environment.

10. A method as recited in claim 9, further comprising analyzing the sensor readings to identify one or more specific residents of the plurality of residents within the smart environment.

11. A method as recited in claim 9, further comprising analyzing the sensor readings to identify a specific activity being performed by a particular resident of the plurality of residents within the smart environment.

12. A method as recited in claim 9, further comprising making available a context-aware service within the smart environment, wherein the service determines context based on the sensor readings.