Registration method for multiple sensor radar

• US 6,225,942 B1
• Filed: 07/30/1999
• Issued: 05/01/2001
• Est. Priority Date: 07/30/1999
• Status: Expired due to Term

• Current Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Original Assignee: Litton Systems Incorporated
• Inventors: Alon, Yair
• Primary Examiner(s): Lobo, Ian J.
• Application Number:
  US09/365,307
• Time in Patent Office: 641 Days
• Field of Search: 342/59, 342/36, 342/29
• US Class Current: 342/59
• CPC Class Codes: G01S 13/42 Simultaneous measurement of distance and other co-ordinates indirect measurement G01S13/46
• CPC Class Codes: G01S 13/878 Combination of several spaced transmitters or receivers of known location for determining the position of a transponder or a reflector G01S13/874 takes precedence
• CPC Class Codes: G01S 7/4004 of parts of a radar system

Abstract

The invention is a method for radar registration by determining initially unknown azimuth and range biases (errors) in a system of multiple, overlapping coverage radars. Track data from multiple radar systems corresponding to a common target are associated into track pairs. Track pair data is then used to calculate state vectors in a multi-dimensional vector space (preferably six-dimensional), with state vector components corresponding to both position and velocity information. From these state vectors an average normalized statistical distance is calculated, where the averaging is over multiple track pairs. An azimuthal bias parameter (and preferably also a range bias parameter) are then varied to minimize the average normalized statistical distance, thereby finding the best estimates of the corrections required to register the multiple radars.

Claims

1. A method for determining an initially unknown bias for more accurately registering a multiple-radar system having at least two radars which track common targets, wherein for each target said radars each report track data, derived from azimuth and range measurements of at least one of the N common targets, the method comprising the steps of:

identifying track data from the multiple-radar system corresponding to the one of the common targets;

forming state vectors from said track data, said state vectors including both directional position and directional velocity components;

calculating a statistical distance D between state vectors corresponding to the one of the common targets; and

determining a measurement bias which if applied to correct said track data, optimizes a function of said statistical distance D.

2. The method of claim 1, wherein said measurement bias comprises azimuthal bias.

3. The method of claim 2, wherein said measurement bias further comprises range bias.

4. The method of 3, wherein said function of D is determined by the step of:

averaging N statistical distances D over N track data pairs, to obtain an average value for the statistical distance D.

5. The method of claim 4 where said statistical distance D is normalized in at least one coefficient by dividing the corresponding distance coefficient, in said vector space, by an associated variance.

6. A method for determining an initially unknown bias for more accurately registering a multiple radar system having at least two radars which track N common targets, wherein for each target said radars each report track data, derived from azimuth and range measurements of the common target, the method comprising the steps of:

(a) acquiring first and second track data from first and second radars respectively, said first and second track data together comprising a track pair, each said track data representing a target'"'"'s position and velocity components at a known time;

(b) expressing said first and second track data as respective first and second state vectors both in an (1+m) dimensional vector space, each said state vector having (i) L independent coefficients proportional to L components of target position, and (ii) m independent coefficients proportional to m components of target velocity;

(c) compensating said second state vector to account for an azimuthal bias, based on an estimated azimuthal bias parameter, to obtain an adjusted second state vector; and

(d) determining a value of said estimated azimuthal bias parameter which minimizes a function of a normalized distance D, where D is defined as the distance between said first and second state vectors in said vector space.

7. The method of claim 6, further comprising the steps of:

(e) further compensating said adjusted second vector to account for a range bias, based on an estimated range bias parameter, to obtain a further adjusted second vector; and

(f) determining the value of said estimated range bias parameter which minimizes the normalized distance D with respect to said estimated range bias parameter.

8. The method of claim 7, further comprising the steps of:

performing N iterations of steps (a) through (c), to obtain N vector pairs of first and second vectors, derived from N track pairs from N common targets; and

wherein the function minimized in step (d) is an average over N of the square of D, with D calculated for each vector pair.

9. The method of claim 8, wherein said step (b) of expressing said track data as state vectors is accomplished by the steps of:

for each track pair, dividing each position component by the sum of corresponding first and second track data position component variances to obtain a normalized state vector coefficient corresponding to said position component, and

for each track pair, dividing each velocity component by the sum of corresponding first and second track data velocity component variances to obtain a normalized state vector coefficient corresponding to said velocity component;

thereby expressing said track data as vectors normalized with respect to variance.

10. The method of claim 9, wherein said track pairs in step (b) are represented in radar referenced coordinates.

11. The method of claim 10, wherein said track data are represented in three position components and three velocity components and said vector space is consequently six dimensional.

12. The method of claim 6, further comprising the step of:

before step (b), defining as first data that track data acquired from the radar which is nearest the target, and as second track data that track data acquired from the radar farther from the target.

13. The method of claim 12, further comprising the step:

reiterating the steps (a) through (f) at discrete time intervals, and

computing a weighted average of said estimated azimuthal bias based on its values calculated at different times.

14. The method of claim 13, wherein said weighted average is computed by using a Kalman estimator.