Method and apparatus for continuous time representation of multiple hypothesis tracking data

• US 5,414,643 A
• Filed: 06/14/1993
• Issued: 05/09/1995
• Est. Priority Date: 06/14/1993
• Status: Expired due to Term

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Hughes Aircraft Company (Rtx Corporation)
• Inventors: Blackman, Samuel S., Dempster, Robert J.
• Primary Examiner(s): Ramirez, Ellis B.
• Application Number:
  US08/077,165
• Time in Patent Office: 694 Days
• Field of Search: 364/516, 364/514, 395/143, 358/108, 342/95
• US Class Current: 342/95
• 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
• CPC Class Codes: G01S 13/87 Combinations of radar systems, e.g. primary radar and secondary radar

Abstract

A multiple hypothesis tracking system (10) which generates a substantially continuous output to a system user. The multiple hypothesis tracking system (10) generates a primary set of tracks (12) which best represents the expected number of targets of interest in a cluster. For multiple sensor applications, a secondary set of tracks is generated having a less probability than the tracks in the primary set. A knowledge of track data is maintained from one scan to a subsequent scan such that tracks can be merged and deleted (14). A universal track file (16) is generated in which track associations from one scan to a subsequent scan are correlated such that the output of the track file remains consistent with respect to the number of tracks.

Claims

1. A method of tracking objects comprising the steps of:

generating a sequence of scans of a scene in which each scan includes a number of detections of potential targets;

establishing a set of primary tracks from a pool of potential targets, wherein a primary track includes the highest probability tracks within a particular cluster of the potential targets, each track being assigned a track number;

maintaining a running knowledge of the track numbers of previous scans which led to the assignment of the track numbers of a current scan;

eliminating redundant tracks by merging and deleting lower probability tracks;

generating a universal track file which contains a set of tracks that best represents the expected number of targets in the scene; and

outputting the universal track file to a system user.

2. The method according to claim 1 wherein the step of establishing a set of primary tracks includes determining the probability for each potential track and determining the highest probability track based on detection and false set probabilities.

3. The method according to claim 2 wherein the step of establishing a primary track includes the steps of determining which of the remaining tracks has a highest probability until the number of tracks in the cluster is equal to the number of targets in a cluster containing the tracks.

4. The method according to claim 1 wherein the step of establishing the set of primary tracks includes determining the highest probability track above a predetermined threshold which remains from the set of tracks until the track hypotheses is greater than two.

5. The method according to claim 1 wherein the step of establishing a set of primary tracks is performed for each cluster.

6. The method according to claim 1 wherein the step of generating the universal track file includes eliminating an universal track with no detections for a predetermined number of successive scans.

7. The method according to claim 1 wherein the step of generating a universal track file includes updating the universal track file after each scan by selecting the higher probability primary tracks.

8. The method according to claim 1 further comprising the step of applying last resort correlation logic for pairing primary tracks and the universal track file.

9. The method according to claim 1 further comprising the step of establishing a set of secondary tracks in which the secondary tracks have a lower probability of being targets than the set of primary tracks.

10. The method according to claim 9 wherein the step of generating the secondary tracks is based on the tracks with the highest probability of the remaining tracks after the primary tracks have been determined.

11. The method according to claim 1 wherein the step of establishing a set of primary tracks is performed at multiple sensor applications, and each set of primary tracks from each sensor is applied to a central processing system for associating common tracks from each sensor for track fusion.

12. The method according to claim 11 wherein an output from the central processing system is applied to a universal track file system for generating a universal track file for each of the sensors.

13. A multiple hypothesis tracking system for tracking multiple targets in a scene, said system comprising:

means for generating a sequence of scans of the scene in which each scan includes a number of detections of potential targets;

primary track generating means for generating a set of primary tracks from a pool of potential targets from the means for generating, wherein a primary track includes the highest probability tracks within a particular cluster of the potential targets, each primary track being assigned a track number, said primary track generating means maintaining a running knowledge of the track numbers of previous scans which led to the assignment of the track numbers of a current scan;

track merging means for eliminating redundant tracks by merging and deleting lower probability tracks;

universal track file means for generating a universal track file which contains a set of tracks that best represents the expected number of targets in a scene, and

outputting means for outputting the universal track file to a system user.

14. The multiple hypothesis tracking system according to claim 13 wherein the primary track generating means includes means for determining the probability for each potential track and determining the highest probability track based on detection and false set probabilities.

15. The multiple hypothesis tracking system according to claim 14 wherein the primary track establishing means includes means for determining which of the remaining tracks has a highest probability until the number of tracks in the cluster is equal to the number of targets in a cluster containing the tracks.

16. The multiple hypothesis tracking system according to claim 13 wherein the primary track generating means includes means for determining the highest probability track above a predetermined threshold which remains from the set of tracks until the track hypotheses is greater than two.

17. The multiple hypothesis tracking system according to claim 13 wherein the primary track generating means generates the set of primary tracks for each cluster.

18. The multiple hypothesis tracking system according to claim 13 wherein the universal track file generating means includes means for eliminating a universal track with no detections for a predetermined number of successive scans.

19. The multiple hypothesis tracking system according to claim 13 wherein the universal track file generating means includes means for obtaining the universal track file after each scan by selecting the higher probability primary tracks.

20. The multiple hypothesis tracking system according to claim 13 further comprising last resort correlation logic means for pairing primary tracks and the universal track file.

21. The multiple hypothesis tracking system according to claim 13 further comprising secondary track generating means for generating a set of secondary tracks in which the secondary tracks have a lower probability of being targets than the set of primary tracks.

22. The multiple hypothesis tracking systems according to claim 21, wherein the secondary track generating means generates the secondary tracks based on the tracks with the highest probability of the remaining tracks after the primary tracks have been determined.

23. The multiple hypothesis tracking system according to claim 13 wherein the primary track generating means establishes a set of primary tracks from multiple sensors, wherein each set of the primary tracks from each sensor is applied to a central processing system for associating common tracks from each sensor for track fusion.

24. The multiple hypothesis tracking system according to claim 23 wherein an output from the central processing system is applied to a universal track file system for generating a universal track file for each of the sensors.