Method and device for event detection utilizing data from a multiplicity of sensor sources

• US 6,525,658 B2
• Filed: 06/11/2001
• Issued: 02/25/2003
• Est. Priority Date: 06/11/2001
• Status: Expired due to Term

• Current Assignee: ENSCO, Incorporated (Teris LLC)
• Original Assignee: ENSCO, Incorporated (Teris LLC)
• Inventors: McGarvey, Matthew W., Streetman, Steven S.
• Primary Examiner(s): POPE, DARYL C
• Application Number:
  US09/877,023
• Publication Number:
  US 20020196140A1
• Time in Patent Office: 624 Days
• Field of Search: 340/522, 340/506, 340/517, 340/521, 340/523, 340/526, 340/527
• US Class Current: 340/522
• CPC Class Codes: G08B 29/186 Fuzzy logic; neural networks

Abstract

A method and apparatus for event detection utilizing data from a multiplicity of sensors is provided. In a first step, actual detections from a plurality of sensors identified with predetermined sensor sequences, each indicative of an event, are compared with the predetermined sensor sequence to determine whether the times between the actual detections match the times allocated between detections for any predetermined sensor sequence. If a match occurs, the event indicated by the matching predetermined sensor sequence is provided. If no match occurs, a second step is initiated wherein the actual detections are compared to a predetermined script file which defines criteria for a plurality of events. If this criteria is matched, the event for which the criteria is provided is indicated.

Claims

1. A method utilizing data from a multiplicity of sensor sources for detecting the occurrence of one of a plurality of different possible events which may occur within an area of interest which includes:

deploying a plurality of sensor sources to provide sensor detections for an area of interest;

choosing from said plurality of deployed sensor sources a unique group of sensor sources for each of said plurality of events to provide a specific sensor detection sequence associated with each event;

identifying allowable time intervals between each sensor detection from sensor sources in each sensor detection sequence which will occur for said sensor detection sequence to indicate the occurrence of the event associated therewith;

operating upon the receipt of a sensor detection from a first sensor source to identify all sensor detection sequences which include said first sensor source;

operating upon receipt of one or more subsequent sensor detections from one or more additional sensor sources to compare the actual time interval between a sensor detections with the allowable time intervals between sensor detections in each identified sensor sequence containing said first sensor source; and

when all of said actual time intervals fall within said allowable time intervals for one of said identified sensor sequences, identifying the event indicated by said one identified sensor sequence.

2. The method of claim 1 wherein the identified allowable time intervals between sensor detections from sensor sources in each identified sensor detection sequence each include a primary time difference combined with an error time.

3. The method of claim 1 which includes deploying a plurality of sensor sources of different types to provide different types of sensor detections.

4. The method of claim 1 which includes identifying one or more objects defined by a plurality of identified events;

identifying criteria for each identified object,

determining whether an event indicated by an identified sensor detection sequence is included in an identified object,

comparing all identified objects containing the event to the identifying criteria, and

providing an object identification when an identified object containing the event matches the identifying criteria.

5. The method of claim 4 wherein the identified allowable time intervals between sensor detections from sensor sources in each identified sensor detection sequence each include a primary time difference combined with an error time.

6. The method of claim 5 which includes deploying a plurality of sensor sources of different types to provide different types of sensor detections.

7. The method of claim 1 which includes providing a script file including a plurality of events which defines criteria for a number of different detection classifications for event identification, said criteria including a coarse time gate for each event identified in the script file and a configurable set of event parameters for each detection classification,

operating when the receipt times of the sensor detections from said first and additional sensor sources do not match the allowable time intervals for an identified sensor detection sequence to compare said first and additional sensor source detections with said script file,

and identifying an event from said script file when the sensor source detections match the criteria for a detection classification.

8. The method of claim 7 which includes identifying one or more objects defined by a plurality of identified events,

identifying criteria for each identified object,

determining whether an identified event is included in an identified object,

comparing all identified objects containing the identified event to the identifying criteria, and

providing an object identification when an identified object containing the event matches the identifying criteria.

9. The method of claim 8 wherein the identified allowable time intervals between sensor detections from sensor sources in each identified sensor detection sequence each include a primary time difference combined with an error time.

10. The method of claim 9 which includes deploying a plurality of sensor sources of different types to provide different types of sensor detections.

11. A device for detecting the occurrence of one of a plurality of different possible events which may occur within an area of interest comprising:

a plurality of sensors each operative when activated to provide a sensor detection signal; and

a processor unit connected to receive said sensor detection signals and operating to store data relating to a plurality of different detection sequences, each of which is associated with one of said plurality of different possible events, data for each detection sequence including the identity of each sensor included in a detection sequence and allowable time intervals between successive sensor detection signals from sensors in each detection sequence;

said processor unit operating in response to a first received sensor detection signal from a sensor to identify all detection sequences which include said sensor and to subsequent received detection signals to determine whether actual time intervals occurring between detection signals fall within the allowable time intervals stored for one of said identified detection sequences.

12. The device of claim 11 wherein said sensors include sensors of different types which provide sensor signals in response to different types of sensor detections.

13. The device of claim 12 wherein said processor unit provides an identification of the occurrence of an event when the actual time intervals occurring between detection signals fall within the allowable time intervals stored for an identified detection sequence associated with the event identified.