Dual disimilar sensing object detection and targeting system

• US 6,834,232 B1
• Filed: 07/30/2003
• Issued: 12/21/2004
• Est. Priority Date: 07/30/2003
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Malhotra, Anita B.
• Primary Examiner(s): Beaulieu, Yonel
• Application Number:
  US10/604,547
• Time in Patent Office: 510 Days
• Field of Search: 701/300-302, 340/901, 340/903, 340/425.5, 340/435, 340/436, 342/69-72
• US Class Current: 701/301
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• CPC Class Codes: G08G 1/163 involving continuous checking

Abstract

A method of performing object detection within a collision warning and countermeasure system (10) is provided. The method includes generating an object detection signal and generating an image detection signal. Centers-of-reflection (56) and centers-of-intensity (59) are determined in response to the object detection signal and the image detection signal. The centers-of-intensity (59) are associated with the center-of-reflection (56). Differences between the centers-of-reflection (56) and the centroids (62) are determined for a plurality of frames (65) and a sensor difference signal is generated. An object (36) is classified in response to the sensor difference signal.

Claims

1. A method of performing object detection within a collision warning and countermeasure system comprising:

generating an object detection signal;

generating an image detection signal;

determining at least one center-of-reflection and at least one center-of-intensity in response to said object detection signal and said image detection signal;

associating said at least one center-of-intensity with said at least one center-of-reflection;

determining difference between said at least one center-of-reflection and said at least one center-of-intensity for a plurality of frames and generating a sensor difference signal; and

classifying at least one object in response to said sensor difference signal.

2. A method as in claim 1 wherein generating an object detection signal comprises performing an upramp sweep and a downramp sweep.

3. A method as in claim 1 wherein generating an object detection signal comprises determining range and range-rate for at least one object.

4. A method as in claim 1 wherein generating an object detection signal comprises gating and pairing targets for at least one object.

5. A method as in claim 1 wherein generating an object detection signal comprises determining relative velocity and relative direction of travel for at least one paired target.

6. A method as in claim 1 wherein determining at least one center-of-reflection comprises transforming polar coordinates of said object detection signal into Cartesian coordinates.

7. A method as in claim 1 wherein determining at least one center-of-reflection comprises:

performing a perspective transform of said world coordinates into image coordinates; and

determining said at least one center-of-reflection in response to said image coordinates.

8. A method as in claim 1 further comprising determining at least one centroid in response to said at least one center-of-intensity.

9. A method as in claim 8 wherein determining at least one centroid comprises determining a current frame region-of-interest in response to a prior frame region-of-interest.

10. A method as in claim 8 wherein determining at least one centroid comprises adjusting size of a current frame region-of-interest in response to a prior frame region-of-interest and in response to said sensor difference signal.

11. A method as in claim 8 wherein determining at least one centroid comprises:

determining whether an object is a false object or a ghost object and generating an error signal; and

determining whether to use said sensor difference signal when determining a subsequent region-of-interest in response to said error signal.

12. A method as in claim 8 further comprising:

determining a first set of vectors between said at least one center-of-reflection and said at least one centroid for a prior frame region-of-interest;

determining a second set of vectors between said at least one center-of-reflection and at least one updated centroid for a current frame region-of-interest; and

generating said sensor difference signal in response to said first set of vectors and said second set of vectors.

13. A method as in claim 12 wherein generating said sensor difference signal comprises averaging differences between said second set of vectors and said first set of vectors.

14. A method as in claim 1 wherein generating said sensor difference signal comprises accounting for errors between said at least one center-of-reflection and said at least one center-of-intensity.

15. A collision warning and countermeasure system for an automotive vehicle comprising:

at least one object detection sensor generating an object detection signal;

at least one image generating sensor generating an image detection signal; and

a controller electrically coupled to said at least one object detection sensor and said at least one image generating sensor, said controller determining at least one center-of-reflection and at least one center-of-intensity in response to said object detection signal and said image detection signal, associating said at least one center-of-intensity with said at least one center-of-reflection, determining difference between said at least one center-of-reflection and said at least one center-of-intensity for a plurality of frames and generating a sensor difference signal, and classifying at least one object in response to said sensor difference signal.

16. A system as in claim 15 wherein said object detection sensor in generating said object detection signal performs an upramp sweep and a downramp sweep.

17. A system as in claim 15 wherein said controller in determining said at least one center-of-reflection gates and pairs targets generated from said radar sensor performing an upramp sweep and a downramp sweep.

18. A system as in claim 15 wherein said controller determines at least one centroid in response to said at least one center-of-intensity and determines a set of vectors between said at least one center-of reflection and said at least one centroid.

19. A system as in claim 18 wherein said controller compares differences between said set of vectors and an average distance parameter to determine an updated average distance parameter and determines a subsequent set of regions-of-interest in response to said updated average distance parameter.

20. A method of performing object detection within a collision warning and countermeasure system comprising:

generating an object detection signal by performing an upramp sweep and a downramp sweep;

generating an image detection signal;

determining at least one center-of-reflection and at least one center-of-intensity in response to said object detection signal and said image detection signal;

associating said at least one center-of-intensity with said at least one center-of-reflection;

determining difference between said at least one center-of-reflection and said at least one center-of-intensity for a plurality of frames and generating a sensor difference signal;

determining at least one centroid in response to said center-of-intensity;

determining a set of vectors between said at least one center-of-reflection and said at least one centroid;

comparing differences between said set of vectors and an average distance parameter to determine an updated average distance parameter;

determining a subsequent set of regions-of-interest in response to said updated average distance parameter;

determining updated centroids in response to said subsequent set of regions-of-interest; and

classifying and tracking at least one object in response to said updated centroids.