Pre-crash sensing system and method for detecting and classifying objects

• US 7,069,130 B2
• Filed: 12/09/2003
• Issued: 06/27/2006
• Est. Priority Date: 12/09/2003
• Status: Active Grant

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Yopp, Wilford T.
• Primary Examiner(s): Tran, Dalena
• Application Number:
  US10/707,377
• Publication Number:
  US 20050125126A1
• Time in Patent Office: 931 Days
• Field of Search: 701/45, 701/96, 701/301, 382/103, 382/190, 382/228, 382/106, 382/224, 342/109, 342/64, 342/70, 342/71, 340/435, 340/436, 340/937, 340/903, 340/917
• US Class Current: 701/45
• CPC Class Codes: B60R 21/013 including means for detecting collisions, impending collisions or roll-over
• CPC Class Codes: B60R 21/0134 responsive to imminent contact with an obstacle , e.g. using radar systems
• CPC Class Codes: G01S 13/726 Multiple target tracking

Abstract

One embodiment of the present invention is a pre-crash sensing system (12) (“PCS system”) for a vehicle (18) for processing images within a substantially quick time. This system (12) includes one or more sensors (14) for detecting one or more objects (16) that are located external to the vehicle (18). These sensors (14) are coupled to a controller (22) for transmitting data associated with a detected object (16) to the controller (22). The controller (22) utilizes an algorithm to store this data and the object'"'"'s identity. Also, the controller (22) utilizes the algorithm to classify the object into a predetermined category when the controller (22) determines that the object (16) requires classification.

Claims

1. A pre-crash sensing system for a vehicle, comprising:

at least one sensor repeatedly detecting at least one parameter of at least one object external to the vehicle; and

a controller coupled to said at least one sensor;

said controller selectively generating an object classification list for transmission to a safety countermeasure system;

said controller comprising an object-tracking module, a process-determining module, and an object-classifying module;

said object-tracking module storing an object identification list and said object classification list;

said object identification list including a plurality of identities of previously detected objects as defined by a plurality of archived parameters;

said process-determining module receiving said object classification list, said object identification list, and said at least one parameter of said at least one object;

said process-determining module determining one of said at least one object being a previously unclassified object and said at least one object requiring an updated classification; and

said process-determining module actuating said object-classifying module to classify said at least one object into a predetermined category and update said object classification list.

2. The pre-crash sensing system as recited in claim 1 wherein at least one parameter includes at least one of a height, a width, a depth, a range, a range rate, an angle, and a visual feature.

3. The pre-crash sensing system as recited in claim 1 wherein said at least one sensor includes at least one of a visual-imaging camera and an electro-magnetic wave-ranging device.

4. The pre-crash sensing system as recited in claim 3 wherein said visual-imaging camera is selected from the group consisting of a monocular camera and a binocular camera.

5. The pre-crash sensing system as recited in claim 3 wherein said electro-magnetic wave-ranging device is selected from the group consisting of a radar device, a lidar device, and a stereo camera pair ranging device.

6. A method for operating a pre-crash sensing system for a vehicle, comprising:

utilizing at least one sensor for detecting at least one object located external to the vehicle;

utilizing a controller for producing a queue of said at least one object; and

utilizing said controller for individually determining that each of said at least one object requires that an object classification list is updated;

said object classification list being stored in said controller;

said controller coupled to said at least one sensor and intended to selectively generate an object identification list and said object classification list;

said controller determining that said at least one object was previously detected and requires an undated classification;

said object classification list for transmission to a safety countermeasure system.

7. The method as recited in claim 6 wherein utilizing said at least one sensor comprises:

utilizing at least one of a visual-imaging camera and an electro-magnetic wave-ranging device for detecting at least one of a height, a width, a depth, a range, a range rate, an angle, and a visual feature associated with said at least one object.

8. The method as recited in claim 6 wherein individually determining that each of said at least one object requires that said object classification list is updated comprises:

determining that each of said at least one object has been previously classified.

9. The method as recited in claim 6 wherein individually determining that each of said at least one object requires that said object classification list is updated comprises:

determining that said at least one object is associated with an outdated classification.

10. The method as recited in claim 6 further comprising:

transmitting said object classification list to a safety countermeasure system of the vehicle.

11. A method for operating a pre-crash sensing system for a vehicle, comprising:

utilizing at least one sensor for detecting at least one object located external to the vehicle;

utilizing a controller for producing a queue of said at least one object; and

utilizing said controller for determining one of said at least one object being a previously unclassified object and said at least one object requiring an updated classification;

said object identification list and said object classification list being stored in said controller;

said controller coupled to said at least one sensor and intended to selectively generate said object identification list and said object classification list;

said object classification list for transmission to a safety countermeasure system.

12. The method as recited in claim 11 wherein utilizing said at least one sensor comprises:

utilizing at least one of a visual-imaging camera and an electro-magnetic wave-ranging device for detecting at least one of a height, a width, a depth, a range, a range rate, an angle, and a visual feature associated with said at least one object.

13. The method as recited in claim 11 wherein individually determining that each of said at least one object requires that said object classification list is updated comprises:

determining that each of said at least one object has been previously unclassified.

14. The method as recited in claim 11 wherein individually determining that each of said at least one object requires that said object classification list is updated comprises:

determining that said at least one object is associated with an outdated classification.

15. The method as recited in claim 11 further comprising:

storing at least one of an updated object identification list and an updated object classification list in an object-tracking module within said controller at the end of an image processing cycle.

16. A pre-crash sensing system for a vehicle, comprising:

at least one sensor detecting at least one object external to the vehicle; and

a controller coupled to said at least one sensor and selectively generating an object identification list and an object classification list;

said object classification list for transmission to a safety countermeasure system;

said controller producing a queue of said at least one object;

said controller determining that each of said at least one object individually requires that said object classification list is updated;

said controller executing a classification subroutine when said at least one object is omitted from said object classification list;

said classification subroutine classifying said at least one object into at least one predetermined category for updating said object classification list;

said object classification list stored in said controller.

17. The pre-crash sensing system recited in claim 16 wherein said controller determines that each of said at least one object individually requires that said object identification list is updated.

18. The pre-crash sensing system recited in claim 16 wherein said controller determines that said object classification list requires updating one time for each of said at least one object.

19. The pre-crash sensing system recited in claim 16 wherein said at least one sensor repeatedly detects said at least one object over a series of sensing cycles and said controller updates said object classification list the first time each of said at least one object is processed by said controller.

20. The pre-crash sensing system recited in claim 16 wherein said object classification list includes at least one previously classified object and at least one predetermined category.

21. The pre-crash sensing system recited in claim 16 wherein said controller bypasses a classification subroutine when said at least one object is included in said object classification list.