Chemical agent detection

• US 20070207059A1
• Filed: 09/06/2006
• Published: 09/06/2007
• Est. Priority Date: 05/11/2004
• Status: Abandoned Application

• Current Assignee: Science Applications International Corporation
• Original Assignee: Science Applications International Corporation
• Inventors: Ha, James, Haupt, Steven, Rese, David
• Application Number:
  US11/515,883
• Publication Number:
  US 20070207059A1
• Time in Patent Office: Days
• US Class Current: 422/67
• CPC Class Codes: G01N 27/00 Investigating or analysing materials by the use of electric, electrochemical, or magnetic means G01N3/00 – G01N25/00 take precedence; measurement or testing of electric or magnetic variables or of electric or magnetic properties of materials G01R
• CPC Class Codes: G01N 33/0036 Specially adapted to detect a particular component all the other sub-groups of G01N33/0004 take precedence
• CPC Class Codes: Y10T 436/11 Automated chemical analysis

Abstract

A chemical agent detector with a a lesser demand sensor, a greater demand sensor, an inlet; and a manifold; and methods associated therewith. The manifold is positioned between the inlet and the sensors, and includes a first intake associated with the lesser demand sensor, a second intake associated with a greater demand sensor, and at least one restrictor. The first intake is placed closer to the inlet than the second intake, and the second intake is isolated from the inlet by at least one restrictor within the manifold. The intakes and restrictor are sized and positioned such that the ratio between: the conductance of the path from the inlet to the greater demand sensor to the conductance of the path between the sensors effectively isolates the lesser demand sensor from the effects of the greater demand sensor.

Claims

1. A chemical agent detector comprising:

at least two sensors having overlapping effectiveness and providing both processed sensor data and raw sensor data;

a processor subsystem including memory, the processor subsystem in communication with the at least two sensors; and

a computer program product for operating on the processed sensor data and raw sensor data to formulate results sensor data, the computer program product stored in the memory, executable on the processor, and comprising;

an agent class identification routine operable to determine a class of agent present based at least in part on the results sensor data.

2. The chemical agent detector of claim 1 wherein:

the processed sensor data comprises an alarm state of each of the at least two sensors, and

the detector determines the class of agent present at least in part by comparing the combined alarm state of each of the at least two sensors to a table of agent classification results.

3. The chemical agent detector of claim 2 wherein:

the table of agent classification results is derived at least in part through elimination of false alarm modes as an influence on agent classification.

4. The chemical agent detector of claim 1 wherein:

the detector uses a tiered data fusion method.

5. The chemical agent detector of claim 4 wherein:

the data fusion method includes at least a first and second tier, the first tier evaluates alarms generated by the at least two sensors to determine the class of agent present and the second tier uses the raw sensor data from the at least two sensors to identify a specific agent within the class, quantify the specific agent, and track dosage of the specific agent over time.

6. The chemical agent detector of claim 1 wherein:

each of the at least two sensors includes an alarm and an alarm status of each of the at least two sensors is monitored by the agent class identification routine.

7. The chemical agent detector of claim 1, wherein each of the at least two sensors is selected from the group consisting of a surface acoustic wave (SAW) sensor and an IM cell sensor.

8. A chemical agent detector comprising:

at least two sensors having overlapping effectiveness and providing both processed sensor data and raw sensor data;

a processor for operating on the processed sensor data and the raw sensor data, wherein the processed sensor data includes alarm state data for the at least two sensors; and

a data fusion program running on the processor including at least a first and second tier, the first tier evaluating alarm state data generated by the at least two sensors to determine the class of agent present and the second tier using the raw sensor data from the at least two sensors to identify a specific agent within the class, quantify the specific agent, and track dosage of the specific agent over time.

9. The chemical agent detector of claim 8 wherein:

the detector determines the class of agent present at least in part by comparing the combined alarm state data for each of the at least two sensors to a table of agent classification results.

10. The chemical agent detector of claim 9 wherein:

the table of agent classification results is derived at least in part through elimination of false alarm modes as an influence on agent classification.

11. The chemical agent detector of claim 8, wherein each of the at least two sensors is selected from the group consisting of a surface acoustic wave (SAW) sensor and an IM cell sensor.