PROCESS OF PROBABILISTIC MULTI-SOURCE MULTI-INT FUSION BENEFIT ANALYSIS

• US 20150205760A1
• Filed: 01/20/2015
• Published: 07/23/2015
• Est. Priority Date: 01/20/2014
• Status: Active Grant

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Hershey, Paul C., Deardorff, Thomas P., Guisewite, Geoffrey, Wisniewski, David J., Williams, John J.
• Granted Patent: US 10,068,177 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models

Abstract

A method of fusing sensor detection probabilities. The fusing of detection probabilities may allow a first force to detect an imminent threat from a second force, with enough time to counter the threat. The detection probabilities may include accuracy probability of one or more sensors and an available time probability of the one or more sensors. The detection probabilities allow a determination of accuracy of intelligence gathered by each of the sensors. Also, the detection probabilities allow a determination of a probable benefit of an additional platform, sensor, or processing method. The detection probabilities allow a system or mission analyst to quickly decompose a problem space and build a detailed analysis of a scenario under different conditions including technology and environmental factors.

Claims

1. A method of determining sensor detection probabilities of a system of platforms each having one or more respective sensors, the method comprising:

determining a set of detection parameters of one or more combinations of platforms and respective sensors of the system of platforms based on a target object, environmental conditions affecting detection of the target object during a first point in time of a timeline of the target object, and capabilities of each of the one or more combinations;

deriving a time limit to detect the target object, the time limit being at a second point in time after the first point in time, and based on a threat level of the target object;

updating the set of detection parameters based on the second point in time;

deriving a time to process a detection of the target object by each of the one or more combinations;

deriving an accuracy of the of each of the one or more combinations based on the target object, environmental conditions affecting detection of the target object during the first point in time, and capabilities of each of the one or more combinations; and

determining sensor detection probabilities of each sensor of the one or more combinations based on the second point in time, the time to process the detection of the target object, and the accuracy of each of the one or more combinations.

2. The method of determining sensor detection probabilities of claim 1, further comprising:

determining that one or more additional combinations of platforms and respective sensors should be added to a region based on the sensor detection probabilities.

3. The method of determining sensor detection probabilities of claim 2, further comprising:

relocating the one or more additional combinations to detect the target object.

4. The method of determining sensor detection probabilities of claim 1, further comprising:

deriving a time for detection of the target object based on a derived number of detections of each intelligence type of the one or more combinations and a plurality of other combinations, and based on a derived time to report result of each intelligence type of the one or more combinations and the plurality of other combinations.

5. The method of determining sensor detection probabilities of claim 1, further comprising:

deriving a quality of detection based on a derived number of detections of each intelligence type of the one or more combinations and a plurality of other combinations, and based on a derived quality of an intelligence product of each intelligence type of the one or more combinations and the plurality of other combinations.

6. The method of determining sensor detection probabilities of claim 1, wherein the capabilities of each of the one or more combinations are based on historical data of the one or more combinations and subject matter expert data related to the one or more combinations.

7. The method of determining sensor detection probabilities claim 1, further including:

a method of fixing one or more locations of the target object based on the sensor detection probabilities; and

a method of tracking based on the sensor detection probabilities and the one or more locations of the target object.

8. A computer network including a plurality of computers in electronic communication with one another, wherein at least one of the plurality of computers performs each step of the method of claim 1;

and

wherein the computer network is in electronic communication with the system of platforms and instructs the system of platforms to add at least one additional combination of a platform and a sensor of the system of platforms to a region, and wherein the system of platforms relocates the at least one additional combination to the region.

9. The method of determining sensor detection probabilities of claim 1, further comprising:

determining a probability of one or more of the platforms and the respective sensors will be available to detect an enemy observable based on a weighted average time.

10. The method of determining sensor detection probabilities of claim 1, further comprising:

determining whether one or more additional platforms and respective sensors of the system of platforms would have a higher probability of detecting the target object based on the time to process the detection of the target object, and an accuracy of each of the additional platforms and respective sensors.

11. The method of determining sensor detection probabilities of claim 10, wherein determining whether the one or more additional platforms and respective sensors would have a higher probability of detecting the target object is further based on an intelligence method associated with the target object.

12. The method of determining sensor detection probabilities of claim 10, wherein determining whether the one or more additional platforms and respective sensors would have a higher probability of detecting the target object is further based on fusion parameters of the one or more additional platforms and respective sensors, and based on fusion parameters of the one or more combinations.

13. The method of determining sensor detection probabilities of claim 12, wherein the fusion parameters include manual fusion parameters.

14. The method of determining sensor detection probabilities of claim 12, wherein the fusion parameters include automatic fusion parameters.

15. The method of determining sensor detection probabilities of claim 12, wherein the fusion parameters include a combination of manual and automatic fusion parameters.

16. The method of determining sensor detection probabilities of claim 1, further comprising:

deriving a probability of detection of the target object based on the target object and an intelligence type of the one or more combinations.

17. The method of determining sensor detection probabilities of claim 16, wherein deriving a probability is further based on a plurality of target objects, an intelligence type of the one or more combinations, a plurality of other combinations of platforms and sensors of the system of platforms, and based on quality metric.

18. The method of determining sensor detection probabilities of claim 1, wherein the target object includes a plurality of enemy vehicles;

wherein the target object includes a plurality of enemy weapons; and

wherein one or more of the plurality of enemy weapons is a missile.

19. The method of determining sensor detection probabilities of claim 1, wherein a plurality of the one or more combinations is relocated from a low threat level area to a high threat level area based on physical conditions including weather, temperature, time, and/or environment.

20. A method of determining sensor detection probabilities of a system of platforms each having one or more respective sensors, the method comprising:

identifying intelligence methods of a plurality of combinations of platforms and respective sensors of the system of platforms based on a plurality of target objects;

deriving probability parameters of each of the plurality of combinations;

mapping accuracy and timeliness parameters of each of the plurality of combinations to the probability parameters;

integrating fusion parameters of each of the plurality of combinations with the probability parameters, the accuracy parameters, and the timeliness parameters;

deriving a probability for each of the plurality of combinations detecting each target object of the plurality of target objects based on the integrated fusion parameters;

deriving tipped probabilities based on the probabilities of each of the plurality of combinations detecting each target object.