Method and apparatus for correlating sensor detections in space and time

• US 5,128,684 A
• Filed: 02/08/1991
• Issued: 07/07/1992
• Est. Priority Date: 02/08/1991
• Status: Expired due to Fees

• Current Assignee: Westinghouse Electric Sweden AB (Brookfield Asset Management Incorporated)
• Original Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Inventors: Brown, Mark W.
• Primary Examiner(s): Barron, Jr., Gilberto
• Application Number:
  US07/652,825
• Time in Patent Office: 515 Days
• Field of Search: 342/96, 342/189, 342/195, 342/52, 342/53, 342/59, 342/95, 342/96, 364/516
• US Class Current: 342/189
• CPC Class Codes: G01S 13/726 Multiple target tracking
• CPC Class Codes: G01S 13/86 Combinations of radar systems with non-radar systems, e.g. sonar, direction finder

Abstract

Sensor detections received over time are analyzed to find relationships therebetween. For example, radar echoes, outputs from an infrared sensor, etc., may be analyzed to find target tracks. First, gross similarities are detected by mapping, e.g., angular readings into angle bins using each of a plurality of hypothesized rates of change of angle. Those sensor detections which are found to correlate with respect to angular measurement are subjected to similar mapping into range bins using rate of change of range. The angle and range bins are compared to determine which represent target tracks. The resulting target tracks may be used to revise the parameters and repeat the same process. For example, if one of the "target" tracks detected on the first pass has an angle of 30°, an angle rate of 2° per minute, a range of 2 kilometers and a range rate of 3 kilometers/second, this target track might be evaluated to determine whether it is a cluster of targets using angle rates varying from 1° per minute to 3° per minute in steps of 1/4 degree per minute to map the sensor detections into angle bins ranging from 27° to 33° in steps of 0.2° with similar mapping in a region of range space.

Claims

1. A method of correlating sensor detections in space and time, comprising the steps of:

(a) defining at least one relationship between space and time;

(b) mapping the sensor detections in space and time into a first dimension in space using the at least one relationship between space and time to produce first mapped detections; and

(c) comparing the mapped detections with first predetermined criteria to determine whether at least one group of the sensor detections correlate with the at least one relationship.

2. A method as recited in claim 1,

wherein the at least one relationship defined in step (a) includes at least one relationship between time and the first dimension in space and between time and a second dimension in space,

wherein said mapping in step (b) is performed using the at least one relationship between time and the first dimension in space, and

wherein said method further comprises the steps of:

(d) mapping the sensor detections in space and time into the second dimension to produce second mapped detections using the at least one relationship between time and the second dimension in space for each group of the sensor detections determined in step (c) to be correlated with the at least one relationship between time and the first dimension; and

(e) comparing the second mapped detections with second predetermined criteria to determine whether at least one subset of the at least one group of the sensor detections correlate with the at least one relationship between time and the second dimension.

3. A method as recited in claim 2,

wherein the at least one relationship defined in step (a) includes a plurality of relationships between time and each of the first and second dimensions,

wherein said mapping and comparing in steps (b) and (c) are repeated in sequence for each of the relationships between time and the first dimension in space, and

wherein said mapping and comparing in steps (d) and (e) are repeated in sequence for each of the relationships between time and the second dimension in space after said mapping and said comparing in steps (b) and (c) have been completed for all of the relationships between time and the first dimension in space.

4. A method as recited in claim 3, wherein the sensor detections are obtained by at least one sensor located at a reference point,

wherein the first dimension is an angle between a reference line passing through the reference point and a line passing through each sensor detection and the reference point, and

wherein the second dimension is distance from the reference point.

5. A method as recited in claim 4, wherein the relationships defined in step (a) are rate of change of angle and rate of change of distance.

6. A method as recited in claim 5, wherein the angle is azimuth from the reference point.

7. A method as recited in claim 5, wherein the angle is elevation from the reference point.

8. A method as recited in claim 5,

further comprising the step of (f) defining bins, regularly spaced within a region of interest in space, and

wherein steps (b) and (d) each comprises the step of translating the sensor detections at different times into the bins of angle and distance, using the rates of change of angle and distance, respectively.

9. A method as recited in claim 8,

further comprising the step of (g) assigning a score to each bin in the region of interest in space in dependence upon position of the mapped sensor detections within the bin, and

wherein said comparing in each of steps (c) and (e) comprises comparing the score of each bin with surrounding bins and with a threshold.

10. A method as recited in claim 9, wherein said assigning in step (g) comprises the step of calculating the score for each bin according to:

##EQU2## where N=number of mapped detections in the bin, p_i =position of an ith mapped detection within the bin, BS=the size of each bin and p_i /BS has a value between 0 and 1.

11. A method as recited in claim 9, wherein said assigning in step (g) comprises the step of calculating the score for each bin by summing the number of mapped detections in the bin.

12. A method as recited in claim 8, further comprising the step of (h) identifying a current location for correlated sensor detections determined in step (e) to correspond to one of the bins of distance.

13. A method as recited in claim 8, wherein said translating in steps (b) and (d) uses a linear relationship between rates of change of angle and distance, respectively.

14. A method as recited in claim 8, further comprising the step of (g) converting linear hypotheses in cartesian coordinates into the relationships defining rates of change of angle and distance.

15. A method as recited in claim 8, further comprising the step of (g) repeating steps (b)-(e), after repeating steps (a) and (f) to revise the relationships and after revising the first and second predetermined criteria, to evaluate whether the sensor detections determined in step (e) to be correlated can be resolved into a larger number of groups than produced by an immediately preceding execution of step (e).

16. A method as recited in claim 5, wherein said mapping in steps (b) and (d) uses a linear relationship defining rates of change of angle and distance, respectively.

17. A method as recited in claim 5, further comprising the step of (g) converting linear hypotheses in cartesian coordinates into the relationships defining rates of change of angle and distance.

18. A method as recited in claim 3, wherein the sensor detections are obtained by at least one sensor located at a reference point,

wherein the relationships defined in step (a) include rates of change of angle and distance,

wherein said method further comprises the step of (f) defining bins, regularly spaced within a region of interest in space, and

wherein said mapping in steps (b) and (d) each comprises the step of translating the sensor detections at different times into the bins in angle and distance, using the rates of change of angle and distance, respectively.

19. A method as recited in claim 2, further comprising the steps of:

(f) revising the first and second predetermined criteria; and

(g) repeating steps (b)-(e), after said revising in step (f) and after repeating step (a) to revise the relationships, to evaluate whether the sensor detections detected as being correlated in step (e) can be resolved into a larger number of groups than produced by an immediately preceding execution of step (e).

20. A method as recited in claim 1,

wherein the at least one relationship in the first dimension defined in step (a) includes a plurality of relationships in the first dimension, and

wherein said mapping and comparing in steps (b) and (c) are repeated in sequence for each of the relationships between time and the first dimension in space.

21. A method as recited in claim 20,

further comprising the step of (d) defining bins, regularly spaced along the first dimension in a region of interest in space, and

wherein said mapping in step (b) comprises the step of translating the sensor detections at different times into the bins in the first dimension, using the relationship between time and the first dimension.

22. A method as recited in claim 21, further comprising the step of (e) identifying a current location for correlated sensor detections determined in step (c) to correspond to one of the bins in the first dimension.

23. A method as recited in claim 1, further comprising the step of:

(d) revising the first predetermined criteria; and

(e) repeating steps (b) and (c), after said revising in step (d) and after repeating step (a) to revise the at least one relationship, to evaluate whether the sensor detections detected as being correlated in step (c) can be resolved into a larger number of groups than produced by an immediately preceding execution of step (c).

24. An apparatus for correlating relationships between sensor detections in space and time, comprising:

control input means for defining at least one relationship between space and time;

processing means for mapping the sensor detections in space and time into a first dimension in space using the at least one relationship between space and time to produce mapped detections; and

output means for comparing the mapped detections least one group of the sensor detections correlate with the at least one relationship and for outputting an indication of the sensor detections found to correlate.

25. An apparatus for correlating relationships between sensor detections in space and time, comprising:

sensing devices to generate sensor detections in space and time;

a buffer, operatively connected to said sensing devices, to store the sensor detections;

a processor, operatively connected to said buffer, to scan said buffer for the sensor detections, to map the sensor detections into a first dimension in space using at least one relationship between space and time to produce mapped detections for each relationship and to compare the mapped detections with predetermined criteria to determine whether at least one group of the sensor detections correlate with the at least one relationship; and

an output device for displaying a current location corresponding to each group of the sensor detections determined by said processor to be correlated with the at least one relationship.