Neural network based data fusion system for source localization

• US 5,537,511 A
• Filed: 10/18/1994
• Issued: 07/16/1996
• Est. Priority Date: 10/18/1994
• Status: Expired due to Fees

• Current Assignee: the united states of america as represented by the secretary of the navy
• Original Assignee: the united states of america as represented by the secretary of the navy
• Inventors: Green, Robert W., DeAngelis, Christopher M.
• Primary Examiner(s): Downs, Robert W.
• Assistant Examiner(s): KATBAB, ABDOLLAH
• Application Number:
  US08/324,641
• Time in Patent Office: 637 Days
• Field of Search: 395/22, 395/21, 382/103, 382/106, 382/107, 367/124, 367/127, 367/129, 367/907
• US Class Current: 706/24
• CPC Class Codes: G01S 13/723 by using numerical data
• CPC Class Codes: G01S 15/66 Sonar tracking systems
• CPC Class Codes: G01S 5/00 Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using active systems G01S13/00, G01S15/00, G01S17/00
• CPC Class Codes: Y10S 367/907 Coordinate determination

Abstract

A method is described for providing an estimate of the state of a moving contact. The method comprises providing a device for estimating the state of the contact, inputting information about a location of an observer platform at particular time intervals and information from at least one sensor about a position of the moving contact relative to the observer platform at each time interval into the device, transforming the inputted information into a series of geographical grids with one grid being formed for each reading of the at least one sensor; combining grids corresponding to similar time intervals into a series of consolidated grid representations; and analyzing the series of consolidated grid representations to produce an estimate of the state of the contact at a final point in time where an observation was made. The device of the present invention includes a grid stimulation block for forming the geographical grids, a fusion block for forming the consolidated grid representations, a correlation block for providing a path likelihood vector, and an estimation block for providing the desired estimate.

Claims

1. A method for providing an estimate of the state of a moving contact using a device for estimating the state of said contact, said method comprising the steps of:

inputting into said device information about a location of an observer platform at particular time intervals and information from at least one sensor about a position of a moving contact relative to said observer platform at each of said time intervals;

transforming said inputted information into a series of geographical grids with one geographical grid being formed for each reading from said at least one sensor;

combining grids corresponding to similar time intervals into a series of consolidated grid representation; and

analyzing said series of consolidated grid representations to produce an estimate of the state of said contact at a final point in time where an observation was made.

2. The method of claim 1 wherein said analyzing step comprises:

applying a constant speed and course constraint for contact motion to said series of consolidated grid representations; and

producing a path likelihood vector containing a likelihood measure for each possible contact path.

3. The method of claim 2 wherein said analyzing step further comprises:

combining said path likelihood vector with an observation location at said final point in time; and

transforming said path likelihood vector to said estimate of the state of said contact with respect to said observation location.

4. The method of claim 1 wherein said transforming step comprises:

forming sensor observation and location pairs from said inputted information;

transforming each said sensor observation and location pair into a series of continuous probability density functions with each continuous probability density function representing the likelihood of the contact'"'"'s location over a respective time plane; and

transforming each said continuous probability density function into discrete values corresponding to stimulation levels for the cells in the geographical grids.

5. The method of claim 4 wherein said combining step comprises:

providing a first series of artificial neurons; and

using said artificial neurons to combine said geographical grids at equivalent time periods into a series of fused grids where each one of said fused grids is representative of information in a particular time interval.

6. The method of claim 5 wherein said analyzing step comprises:

providing a second series of artificial neurons;

inputting information from each of said fused grids into each of said artificial neurons in said second series; and

using said second series of artificial neurons to compute a measure of the path likelihood for each possible path taken by the contact.

7. The method of claim 6 wherein said analyzing step further comprises inputting said measures of path likelihood for each possible path taken by the contact and an observation location at said final observation point in time into a means for transforming the path likelihood measures into the estimate of said contact state at said final observation point in time.

8. A system for generating an estimate of the state of a contact at a particular point in time, said system including:

means for transforming information about a location of an observer platform at particular points in time and about a relative position between said observer platform and said contact at said points in time into a series of geographical grids;

means for combining ones of said geographical grids corresponding to similar time intervals into a series of consolidated grid representations; and

means for analyzing said series of consolidated grid representations to produce an estimate of the state of said contact at a final point in time where an observation was made.

9. The system of claim 8 wherein said analyzing means comprises:

means for applying a constant speed and course constraint for contact motion to said series of consolidated grid representations;

means for producing a path likelihood vector containing a likelihood measure for each possible contact path; and

means for transforming said path likelihood vector into said estimate of said state of said contact.

10. The system of claim 9 wherein said information transforming means comprises means within a computer for transforming said information.

11. The system of claim 9 wherein said combining means comprises a first series of artificial neurons connected to said geographical grids so that each neuron in said first series receives an input from each geographical grid and generates said consolidated grid representations.

12. The system of claim 11 wherein said constant speed and course constraint applying means comprises a second series of artificial neurons connected to said series of consolidated grid representations so that each neuron in said second series receives an input from each consolidated grid representation in said series and generates an output representing the likelihood of a particular path.

13. The system of claim 8 wherein:

said system comprises a computer having an information processing engine;

said transforming means comprises a grid stimulation block within said engine;

said combining means comprises an information fusion block within said engine; and

said analyzing means comprises a correlation block and an estimation block within said engine.