Method and apparatus for joint kinematic and feature tracking using probabilistic argumentation

• US 20050001759A1
• Filed: 12/19/2003
• Published: 01/06/2005
• Est. Priority Date: 07/03/2003
• Status: Active Grant

• Current Assignee: HRL Laboratories LLC (The Boeing Co.)
• Original Assignee: HRL Laboratories LLC (The Boeing Co.)
• Inventors: Khosla, Deepak
• Granted Patent: US 7,026,979 B2
• Time in Patent Office: Days
• US Class Current: 342/90
• CPC Class Codes: G01S 13/726 Multiple target tracking

Abstract

A method, apparatus, and computer program product for joint kinematic and feature tracking are presented. Kinematic measurements and feature/class measurements regarding an object are received from a sensor. A probabilistic argumentation operation is performed on the feature/class measurements using information from a knowledge base and a track file ect to generate feature track likelihood scores regarding likely tracks for the object. Kinematic track likelihood scores are generated based on the kinematic measurements and the track file. Joint track likelihood scores are generated based on the feature track likelihood scores and the kinematic track likelihood scores. Joint track likelihood scores are used to generate a multi-frame track measurement association to determine a most likely track for the object. The track file is continually updated with the most likely track for the object, so that the most likely trajectory of the object is obtained.

Claims

1. A method for joint kinematic and feature tracking comprising acts of:

receiving kinematic measurements and feature/class measurements regarding an object from at least one sensor;

performing a probabilistic argumentation operation on the feature/class measurements using information from a rule-based knowledge base and a track file regarding the object to generate feature track likelihood scores that provide information regarding likely tracks for the object;

generating kinematic track likelihood scores based on the kinematic measurements and the track file regarding the object;

generating joint track likelihood scores based on the feature track likelihood scores and the kinematic track likelihood scores;

using the joint track likelihood scores to generate a multi-frame track measurement association to determine a most likely track for the object; and

updating the track file with the most likely track for the object;

whereby the method tracks an object and continually updates the track file for the object, so that the most likely trajectory of the object is obtained.

2. A method as set forth in claim 1, wherein the kinematic track likelihood scores are generated using a technique selected from a group consisting of a global nearest neighbor technique, multiple hypothesis tracking technique, and a log-likelihood technique.

3. A method as set forth in claim 2, further comprising an act of normalizing the feature track likelihood scores and the kinematic track likelihood scores to compute the joint track likelihood scores.

4. A method as set forth in claim 3, wherein the act of normalizing uses a degree of contradiction, k, in a Probabilistic Argumentation System used to perform the probabilistic argumentation operation as a part of an act of performing a normalizing calculation.

5. A method as set forth in claim 4, wherein the kinematic measurements are received radar signals.

6. A method as set forth in claim 5, wherein the feature/class measurements are received radar signals.

7. A method as set forth in claim 6, wherein in the act of performing a probabilistic argumentation operation, class probabilities are generated for the feature/class measurements and are used, along with the kinematic track likelihood scores in the determination of a most likely track for an object.

8. A method as set forth in claim 7, wherein the acts of generating kinematic track likelihood scores and generating joint track likelihood scores are performed substantially simultaneously, whereby kinematic and class information regarding a target may be simultaneously updated.

9. A method as set forth in claim 8, further comprising an act of outputting a representation of the most likely trajectory of an object for a user.

10. A method as set forth in claim 1, further comprising an act of normalizing the feature track likelihood scores and the kinematic track likelihood scores to allow for accurate computation of the joint track likelihood scores.

11. A method as set forth in claim 10, wherein the act of normalizing uses a degree of contradiction, k, in a Probabilistic Argumentation System used to perform the probabilistic argumentation operation as a part of an act of performing a normalizing calculation.

12. A method as set forth in claim 1, wherein the kinematic measurements are received radar signals.

13. A method as set forth in claim 1, wherein the feature/class measurements are received radar signals.

14. A method as set forth in claim 1, wherein in the act of performing a probabilistic argumentation operation, class probabilities are generated for the feature/class measurements and are used, along with the kinematic track likelihood scores in the determination of a most likely track for an object.

15. A method as set forth in claim 1, wherein the acts of generating kinematic track likelihood scores and generating joint track likelihood scores are performed substantially simultaneously, whereby kinematic and class information regarding a target may be simultaneously updated.

16. A method as set forth in claim 1, further comprising an act of outputting a representation of the most likely trajectory of an object for a user.

17. A computer program product for joint kinematic and feature tracking comprising a computer-readable medium having computer-executable instructions encoded therein for causing a computer to perform operations of:

receiving kinematic measurements and feature/class measurements regarding an object from at least one sensor;

performing a probabilistic argumentation operation on the feature/class measurements using information from a rule-based knowledge base and a track file regarding the object to generate feature track likelihood scores that provide information regarding likely tracks for the object;

generating kinematic track likelihood scores based on the kinematic measurements and the track file regarding the object;

generating joint track likelihood scores based on the feature track likelihood scores and the kinematic track likelihood scores;

using the joint track likelihood scores to generate a multi-frame track measurement association to determine a most likely track for the object; and

updating the track file with the most likely track for the object;

whereby the computer program product tracks an object and continually updates the track file for the object, so that the most likely trajectory of the object is obtained.

18. A computer program product as set forth in claim 17, wherein the kinematic track likelihood scores are generated using a technique selected from a group consisting of a global nearest neighbor technique, multiple hypothesis tracking technique, and a log-likelihood technique.

19. A computer program product as set forth in claim 18, further comprising computer-executable instructions for normalizing the feature track likelihood scores and the kinematic track likelihood scores to compute the joint track likelihood scores.

20. A computer program product as set forth in claim 19, wherein the computer-executable instructions provide that the act of normalizing uses a degree of contradiction, k, in a Probabilistic Argumentation System used to perform the probabilistic argumentation operation as a part of an act of performing a normalizing calculation.

21. A computer program product as set forth in claim 20, wherein the kinematic measurements are received radar signals.

22. A computer program product as set forth in claim 21, wherein the feature/class measurements are received radar signals.

23. A computer program product as set forth in claim 22, wherein the computer-executable instructions provide that in the operation of performing a probabilistic argumentation operation, class probabilities are generated for the feature/class measurements and are used, along with the kinematic track likelihood scores in the determination of a most likely track for an object.

24. A computer program product as set forth in claim 23, wherein the computer-executable instructions provide that the operations of generating kinematic track likelihood scores and generating joint track likelihood scores are performed substantially simultaneously, whereby kinematic and class information regarding a target may be simultaneously updated.

25. A computer program product as set forth in claim 24, further comprising computer-executable instructions for performing an operation of outputting a representation of the most likely trajectory of an object for a user.

26. A computer program product as set forth in claim 17, further comprising computer-executable instructions for normalizing the feature track likelihood scores and the kinematic track likelihood scores to allow for accurate computation of the joint track likelihood scores.

27. A computer program product as set forth in claim 26, wherein the computer-executable instructions provide that the act of normalizing uses a degree of contradiction, k, in a Probabilistic Argumentation System used to perform the probabilistic argumentation operation as a part of an act of performing a normalizing calculation.

28. A computer program product as set forth in claim 17, wherein the kinematic measurements are received radar signals.

29. A computer program product as set forth in claim 17, wherein the feature/class measurements are received radar signals.

30. A computer program product as set forth in claim 17, wherein the computer-executable instructions provide that in the operation of performing a probabilistic argumentation operation, class probabilities are generated for the feature/class measurements and are used, along with the kinematic track likelihood scores in the determination of a most likely track for an object.

31. A computer program product as set forth in claim 17, wherein the computer-executable instructions provide that the operations of generating kinematic track likelihood scores and generating joint track likelihood scores are performed substantially simultaneously, whereby kinematic and class information regarding a target may be simultaneously updated.

32. A computer program product as set forth in claim 17, further comprising computer-executable instructions for performing an operation of outputting a representation of the most likely trajectory of an object for a user.

33. An apparatus for joint kinematic and feature tracking, the apparatus comprising a computer system, the computer system including:

an input for receiving kinematic measurements and feature/class measurements regarding an object from at least one sensor;

a probabilistic argumentation module for performing a probabilistic argumentation operation on the feature/class measurements using information from a rule-based knowledge base and a track file regarding the object to generate feature track likelihood scores that provide information regarding likely tracks for the object;

a kinematic track likelihood score generator for generating kinematic track likelihood scores based on the kinematic measurements and the track file regarding the object;

a joint track likelihood score generator for generating joint track likelihood scores based on the feature track likelihood scores and the kinematic track likelihood scores;

a track measurement association generator for using the joint track likelihood scores to generate a multi-frame track measurement association to determine a most likely track for the object; and

a track file updating module for updating the track file with the most likely track for the object;

whereby the method tracks an object and continually updates the track file for the object, so that the most likely trajectory of the object is obtained.

34. An apparatus as set forth in claim 33, wherein the kinematic track likelihood scores are generated using a technique selected from a group consisting of a global nearest neighbor technique, multiple hypothesis tracking technique, and a log-likelihood technique.

35. An apparatus as set forth in claim 34, further comprising a normalizing module for normalizing the feature track likelihood scores and the kinematic track likelihood scores to compute the joint track likelihood scores.

36. An apparatus as set forth in claim 35, wherein the normalizing module uses a degree of contradiction, k, in a Probabilistic Argumentation System used to perform the probabilistic argumentation operation as a part of an act of performing a normalizing calculation.

37. An apparatus as set forth in claim 36, wherein the kinematic measurements are received radar signals.

38. An apparatus as set forth in claim 37, wherein the feature/class measurements are received radar signals.

39. An apparatus as set forth in claim 38, wherein in the probabilistic argumentation module, class probabilities are generated for the feature/class measurements and are used, along with the kinematic track likelihood scores in the determination of a most likely track for an object.

40. An apparatus as set forth in claim 39, wherein the kinematic track likelihood score generator and joint track likelihood score generator operate substantially simultaneously, whereby kinematic and class information regarding a target may be simultaneously updated.

41. An apparatus as set forth in claim 40, further comprising an output for outputting a representation of the most likely trajectory of an object for a user.

42. An apparatus as set forth in claim 33, further comprising a normalizing module for normalizing the feature track likelihood scores and the kinematic track likelihood scores to compute the joint track likelihood scores.

43. An apparatus as set forth in claim 42, wherein the normalizing module uses a degree of contradiction, k, in a Probabilistic Argumentation System used to perform the probabilistic argumentation operation as a part of an act of performing a normalizing calculation.

44. An apparatus as set forth in claim 33, wherein the kinematic measurements are received radar signals.

45. An apparatus as set forth in claim 33, wherein the feature/class measurements are received radar signals.

46. An apparatus as set forth in claim 33, wherein in the probabilistic argumentation module, class probabilities are generated for the feature/class measurements and are used, along with the kinematic track likelihood scores in the determination of a most likely track for an object.

47. An apparatus as set forth in claim 33, wherein the kinematic track likelihood score generator and joint track likelihood score generator operate substantially simultaneously, whereby kinematic and class information regarding a target may be simultaneously updated.

48. An apparatus as set forth in claim 33, further comprising an output for outputting a representation of the most likely trajectory of an object for a user.