RFID systems and methods for probabalistic location determination

• US 20070143072A1
• Filed: 12/20/2005
• Published: 06/21/2007
• Est. Priority Date: 12/20/2005
• Status: Active Grant

• Current Assignee: Pitney Bowes Incorporated (Böwe Systec AG)
• Original Assignee: Pitney Bowes Incorporated (Böwe Systec AG)
• Inventors: Campagna, Matthew
• Granted Patent: US 7,388,494 B2
• Time in Patent Office: Days
• US Class Current: 702/181
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/08 Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
• CPC Class Codes: G08B 31/00 Predictive alarm systems characterised by extrapolation or other computation using updated historic data

Abstract

Systems and methods for probabilistic determination of real world events in a sensory environment such as a predicting location in a responsive environment are described. A responsive environment system includes a set of sensors for making probabilistic observations of RFID sensor tags. The system also includes a control system employing a set of possible real world events and statistical processing system for predicting a particular real world event state based upon the sensor observations. In one configuration, a Hidden Markov Model is used for the statistical processing system and may be updated based upon the prediction of the model. The responsive environment will then launch a response based upon the predicted real world event state.

Claims

1. A method for determining a real world event in a sensory environment comprising:

receiving a series of sensory inputs;

using a present state of a statistical model to compute a most likely sequence of real world events that would produce the received series of sensory inputs;

updating the statistical model to an updated state using the computed most likely sequence of real world events; and

outputting the computed most likely sequence of real world events.

2. The method of claim 1, further comprising:

determining an appropriate response of the sensory environment based upon the computed most likely sequence of real world events.

3. The method of claim 2, further comprising:

initiating the appropriate response.

4. The method of claim 3, wherein:

the statistical model is a Hidden Markov Model.

5. The method of claim 4, wherein:

the sensory environment includes an RFID responsive environment having a plurality of RFID tags and a plurality of RFID sensors.

6. The method of claim 5, wherein:

the series of sensory inputs includes RFID sensor data.

7. The method of claim 6, wherein:

the RFID sensor data includes an indication that at least one of the RFID tags is present in a field of view of at least one of the RFID sensors; and

the response comprises launching an application on a computer.

8. The method of claim 7, wherein:

the statistical model is updated to the updated state after ten sensory inputs.

9. The method of claim 8, further comprising:

determining if the response includes an immediate response event; and

if the response includes an immediate response event, initiating the immediate response event before updating the statistical model.

10. The method of claim 6, wherein:

the sequence of real world events includes an item placed on a surface.

11. The method of claim 6, wherein:

the sequence of real world events includes a person entering a room.

12. The method of claim 6, wherein:

the sequence of real world events includes a person passing by a room.

13. The method of claim 4, wherein:

the Hidden Markov Model utilizes a forward-backward procedure.

14. The method of claim 4, wherein:

the Hidden Markov Model utilizes at least one definite zero transition coefficient to optimize the model.

15. The method of claim 4, wherein:

the Hidden Markov Model utilizes a controlled initial condition probability.