Method for reducing repeat false alarm indications in vehicle impact detection systems

• US 6,853,919 B2
• Filed: 02/04/2003
• Issued: 02/08/2005
• Est. Priority Date: 02/04/2003
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: General Motors Corporation (Motors Liquidation Co. GUC Trust)
• Inventors: Kellum, Carroll C.
• Primary Examiner(s): Hernandez, Olga
• Application Number:
  US10/358,131
• Publication Number:
  US 20040153244A1
• Time in Patent Office: 735 Days
• Field of Search: 701/301, 701/1, 701/23, 701/25, 701/36, 701/96, 701/200, 340/988, 340/992, 731/78.R, 180/167, 180/168, 180/169, 180/271, 180/274, 180/286
• US Class Current: 701/301
• CPC Class Codes: B60R 2021/01197 Warning devices
• 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: G08G 1/161 Decentralised systems, e.g. inter-vehicle communication

Abstract

A method for reducing repetitive false alarm indications in a vehicle impact detection system for a host vehicle includes generating false alarm positional information corresponding to a detected object determined to be a false alarm. The false alarm is determined subsequent to an issued warning to a driver, and the false alarm positional information is stored in a database. The false alarm positional information in the database is further evaluated to determine, based on predefined removal criteria, whether any of the false alarm positional information in the database is to be removed therefrom. The false alarm positional information remaining in the database is used to inhibit the issuance of a warning to a driver of the host vehicle.

Claims

1. A method for reducing repetitive false alarm indications in a vehicle impact detection system for a host vehicle, the method comprising:

generating false alarm positional information corresponding to a detected object determined to be a false alarm, said false alarm being determined subsequent to an issued warning to a driver, and storing said false alarm positional information in a database; and

periodically evaluating said false alarm positional information in said database and determining, based on predefined removal criteria, whether any of said false alarm positional information in said database is to be removed therefrom;

wherein said false alarm positional information remaining in said database is used to inhibit the issuance of a warning to a driver of the host vehicle.

2. The method of claim 1, wherein said evaluating said false alarm positional information further comprises:

accessing false alarm objects in said database having associated positional information corresponding to a present sensing range of the impact detection system;

determining whether said accessed false alarm objects in said database are presently observed by the impact detection system and recording the results thereof, thereby updating an object history for each of said accessed false alarm objects; and

examining said updated object history for each of said accessed false alarm objects;

wherein if said updated object history for a given false alarm object satisfies a removal criteria, then removing said given false alarm object from said database.

3. The method of claim 2, further comprising storing said object history within said database.

4. The method of claim 3, wherein said object history for a given false alarm object comprises a set of observed/not observed flags stored in said database, wherein an observed flag results from the vehicle impact detection system observing the false alarm object at the position associated therewith, and a not observed flag results from the vehicle impact system not observing the false alarm object at the position associated therewith.

5. The method of claim 4, wherein said removal criteria for a given false alarm object is based on a comparison between the number of observed flags and not observed flags recorded in said database.

7. The method of claim 1, further comprising:

sensing an object in a predicted path of travel of the host vehicle;

determining the position of said sensed object with respect to a fixed coordinate system; and

comparing said position of said sensed object with stored positional information in said database;

wherein if said database contains a false alarm object having positional information corresponding to the position of said sensed object, then the issuance of a warning to a driver of the host vehicle is inhibited.

6. A method for reducing repetitive false alarm indications in a vehicle impact detection system for a host vehicle, the method comprising:

generating false alarm positional information corresponding to a detected object determined to be a false alarm, said false alarm being determined subsequent to an issued warning to a driver, and storing said false alarm positional information in a database; and

periodically evaluating said false alarm positional information in said database and determining, based on predefined removal criteria, whether any of said false alarm positional information in said database is to be removed therefrom;

wherein said false alarm positional information remaining in said database is used to inhibit the issuance of a warning to a driver of the host vehicle;

wherein said removal criteria for a given false alarm object is based on the number of times a threat assessment algorithm within the vehicle impact detection system continues to identify the false alarm object as a threat.

8. A vehicle impact detection system for a host vehicle, comprising:

an object detection system for identifying objects in a projected path of the host vehicle, said object detection system having a position tracking mechanism for associating an identified object with a reference coordinate system;

a threat assessment algorithm for evaluating the likelihood of an impact between the host vehicle and objects detected by said object detection system;

a driver warning mechanism for providing a warning to a driver, said driver warning responsive to an output of said threat assessment algorithm; and

a false alarm learning mechanism for generating and storing in a database false alarm positional information corresponding to a detected object determined to be a false alarm;

said false alarm learning mechanism periodically evaluating said false alarm positional information in said database and determining, based on predefined removal criteria, whether any of said false alarm positional information in said database is to be removed therefrom;

wherein said false alarm positional information remaining in said database is used to inhibit the issuance of a warning from said driver warning mechanism.

9. The vehicle impact detection system of claim 8, wherein said evaluating said false alarm positional information further comprises:

accessing false alarm objects in said database having associated positional information corresponding to a present sensing range of the impact detection system;

determining whether said accessed false alarm objects in said database are presently observed by the impact detection system and recording the results thereof, thereby updating an object history for each of said accessed false alarm objects; and

examining said updated object history for each of said accessed false alarm objects;

wherein if said updated object history for a given false alarm object satisfies a removal criteria, then said given false alarm object is removed from said database.

10. The vehicle impact detection system of claim 9, wherein said false alarm learning mechanism stores said object history within said database.

11. The vehicle impact detection system of claim 10, wherein said object history for a given false alarm object comprises a set of observed/not observed flags stored in said database, wherein an observed flag results from the vehicle impact detection system observing the false alarm object at the position associated therewith, and a not observed flag results from the vehicle impact system not observing the false alarm object at the position associated therewith.

12. The vehicle impact detection system of claim 11, wherein said removal criteria for a given false alarm object is based on a comparison between the number of observed flags and not observed flags recorded in said database.

14. The vehicle impact detection system of claim 8, wherein said false alarm learning mechanism compares the position of a sensed object in a predicted path of travel of the host vehicle with stored positional information in said database, and if said database contains a false alarm object having positional information corresponding to the position of said sensed object, then said false alarm learning mechanism inhibit the issuance of a warning from said driver warning mechanism.

13. A vehicle impact detection system for a host vehicle, comprising:

an object detection system for identifying objects in a projected path of the host vehicle, said object detection system having a position tracking mechanism for associating an identified object with a reference coordinate system;

a threat assessment algorithm for evaluating the likelihood of an impact between the host vehicle and objects detected by said object detection system;

a driver warning mechanism for providing a warning to a driver, said driver warning responsive to an output of said threat assessment algorithm; and

a false alarm learning mechanism for generating and storing in a database false alarm positional information corresponding to a detected object determined to be a false alarm;

said false alarm learning mechanism periodically evaluating said false alarm positional information in said database and determining, based on predefined removal criteria, whether any of said false alarm positional information in said database is to be removed therefrom;

wherein said false alarm positional information remaining in said database is used to inhibit the issuance of a warning from said driver warning mechanism;

wherein said removal criteria for a given false alarm object is based on the number of times said threat assessment algorithm continues to identify the false alarm object as a threat when encountered.