Decision enhancement system for a vehicle safety restraint application

• US 6,856,694 B2
• Filed: 11/07/2003
• Issued: 02/15/2005
• Est. Priority Date: 07/10/2001
• Status: Expired due to Fees

• Current Assignee: Eaton Corp. (Wisconsin) (Eaton Corporation PLC.), General Motors Corporation (Motors Liquidation Co. GUC Trust)
• Original Assignee: Eaton Corp. (Wisconsin) (Eaton Corporation PLC.)
• Inventors: Farmer, Michael E., St. John, Christopher N., Ressler, Galen E., Dell' Eva, Mark L.
• Primary Examiner(s): Johns, Andrew W.
• Assistant Examiner(s): Nakhjavan, Shervin
• Application Number:
  US10/703,957
• Publication Number:
  US 20040151344A1
• Time in Patent Office: 466 Days
• Field of Search: 382/107, 382/291, 352/39, 307/10.1, 701/36, 701/45, 701/49, 280/728.1, 348/94, 348/169, 348/143, 348/154, 348/155, 700/13, 700/56, 700/64, 700/66, 340/686.1, 340/815.58, 356/614, 731/1.79
• US Class Current: 382/107
• CPC Class Codes: B60R 21/013 including means for detecting collisions, impending collisions or roll-over
• CPC Class Codes: B60R 21/01538 for image processing, e.g. cameras or sensor arrays
• CPC Class Codes: B60R 21/0154 in combination with seat heating
• CPC Class Codes: B60R 21/01542 detecting passenger motion
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 20/59 inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands

Abstract

The disclosure describes systems and methods that pertain to interactions between a vehicle and an occupant within the vehicle. More specifically, various systems and methods for enhancing the decisions of automated vehicle applications (collectively “decision enhancement system”) are disclosed. In a safety restraint embodiment, a sensor is used to capture various sensor readings. Sensor readings are typically in the form of images. Occupant information, such as location attributes, motion attributes, and occupant category attributes can be obtained from the sensor readings. If the system concludes that the deployment of a safety restraint is potentially justified, an at-risk-zone detector can be used to determine whether or not the occupant will be too close to the deploying safety restraint at the time of deployment for the safety restraint to be safely deployed.

Claims

1. A decision enhancement system configured to influence a deployment determination of a safety restraint application for a vehicle by communicating an at-risk-zone determination to the safety restraint application, said decision enhancement system comprising:

a sensor subsystem, said sensor subsystem providing for a sensor to capture a plurality of sensor readings, said plurality of sensor readings including a first sensor reading and a second sensor reading;

a tracking subsystem, wherein said tracking subsystem provides for selectively identifying a tracked condition from a plurality of pre-defined conditions, said plurality of pre-defined conditions including a crash condition, wherein said tracked condition is selectively identified using said first sensor reading; and

a detection subsystem, wherein said detection subsystem is invoked only after said tracking subsystem selectively identifies said crash condition as said tracked condition, wherein said detection subsystem generates said at-risk-zone determination, from said second sensor reading, and wherein said second sensor reading is not captured earlier than said first sensor reading.

2. The system of claim 1, wherein said second sensor reading is not said first sensor reading, and wherein said second sensor reading is captured after said first sensor reading.

3. The system of claim 1, wherein the sensor includes a high-speed mode and a low-speed mode, wherein said sensor is configured to operate in said low-speed mode before said tracking subsystem selectively identifies said crash condition as said tracked condition, and wherein said sensor is configured to operate in said high-speed mode after said tracking subsystem selectively identifies said crash condition as said tracked condition.

4. The system of claim 3, a subset of said plurality of sensor images being a plurality of filtered sensor images, said plurality of filtered sensor images including said second sensor image, and wherein said sensor readings captured by said sensor in said high-speed mode are said plurality of filtered sensor images.

5. The system of claim 4, wherein at least two or more of said plurality of sensor images are used to selectively identify said tracked condition, and wherein at least two or more of said plurality of filtered sensor images are used to generate said at-risk-zone determination.

6. The system of claim 3, wherein said sensor in said high-speed mode captures between about 25 and 40 sensor readings per second, and wherein said sensor in said low-speed mode captures fewer sensor readings per second than said sensor in said high-speed mode.

7. The system of claim 6, wherein said sensor in said low-speed mode captures fewer than about 11 sensor readings per second.

8. The system of claim 3, further comprising a pre-defined detection window, and wherein said plurality of filtered sensor images are filtered in accordance with said pre-defined detection window.

9. The system of claim 1, further comprising a zone intrusion and a window-of-interest, wherein said detection subsystem generates said at-risk-zone determination by identifying said zone intrusion with said second sensor reading, wherein said at-risk-zone determination is generated from the portion of said second sensor reading that is within said window-of-interest.

10. The system of claim 9, wherein said window-of-interest is divided into a plurality of patches.

11. The system of claim 10, further comprising a reference image and a correlation metric, wherein said correlation heuristic is performed on said plurality of patches and said reference image to generate a correlation metric.

12. The system of claim 11, wherein said at-risk-zone determination is generated with said correlation metric.

13. The system of claim 1, said detection subsystem further including a detection window module, a correlation module, and a window-of-interest;

wherein said detection window module generates said window-of-interest with said second sensor reading, and wherein said correlation module generates a correlation metric from said window-of-interest and a reference image, wherein said detection subsystem generates said at-risk-determination using said correlation metric.

14. The system of claim 13, said detection subsystem further including a test threshold, wherein said detection subsystem generates said at-risk-determination by comparing said correlation metric with said test threshold.

15. The system of claim 1, wherein said sensor is a standard video camera, wherein said tracking subsystem selectively identifies said tracked condition by invoking a multiple-model probability-weighted heuristic, and wherein said safety restraint is an airbag.

16. The system of claim 1, further comprising a future prediction wherein said at-risk-determination is said future prediction.

17. The system of claim 1, said plurality of pre-defined conditions including a stationary condition and human motion condition.

18. The system of claim 1, further comprising a pre-crash braking condition, wherein said crash condition is said pre-crash braking condition.

19. The system of claim 1, said sensor subsystem including an infrared illuminator for providing light in low illumination environments.

20. The system of claim 1, further comprising a power supply for providing a duration of keep-alive power.

21. The system of claim 1, wherein said detection subsystem provides for:

filtering at least one sensor reading with a pre-defined window of interest;

setting said sensor to a higher speed;

dividing at least one of the filtered sensor readings into a plurality of patches;

defining a correlation metric between said plurality of patches and a reference image; and

comparing the correlation metric with a test threshold to generate said at-risk-determination.

22. A safety restraint system for a vehicle, comprising:

a sensor, a plurality of sequential sensor images, a spatial area, a computer, a plurality of occupant attributes, a current condition, a plurality of pre-defined conditions, a deployment condition, a non-deployment condition, a detection heuristic, an at-risk-zone flag value, a safety restraint deployment mechanism;

wherein said sensor is configured to capture said plurality of sequential sensor images of said spatial area;

wherein said computer provides for;

tracking said plurality of occupant attributes from said plurality of sequential images;

selectively identifying said current condition from said plurality of pre-defined occupant conditions, wherein said plurality of pre-defined conditions includes said deployment condition and said non-deployment condition;

invoking said detection heuristic after selectively identifying said deployment condition as said occupant condition;

using said detection heuristic to set said at-risk-zone flag value;

communicating said at-risk-zone flag value to said safety restraint deployment mechanism;

wherein said safety restraint deployment mechanism selectively precludes the deployment of said safety restraint when said occupant condition is said deployment condition, and when said at-risk-zone flag value is yes.

23. The system of claim 22, further comprising a video camera, a maximum shutter speed of about 50 images per second, a vehicle, a multiple-model probability-weighted tracking heuristic, and an airbag;

said sensor being said video camera with said maximum shutter speed of about 50 images per second, said vehicle being said automobile, said multiple-model probability-weighted tracking heuristic being invoked to selectively identifying said current condition, said safety restraint being said airbag, wherein said safety restraint system has only one said sensor.

24. A method for installing a decision enhancement application into a vehicle that includes a deployment mechanism for a safety restraint device, the method comprising:

defining an at-risk-zone corresponding to a location of the deployment mechanism within the vehicle;

configuring a sensor to transmit sensor images to a computer;

instructing the computer to filter out all areas within the image that are not part of a window-of-interest corresponding to the defined at-risk-zone after receiving a preliminary determination that a deployment of the safety restraint device is necessary;

programming the computer to set an at-risk-zone flag corresponding to a detection of an occupant within the at-risk-zone, wherein the at-risk-zone flag is set using at least one window-of-interest image filtered by the computer; and

placing the sensor and computer within the vehicle.

25. The method of claim 24, further comprising adapting the adapting the sensor to function in a low-speed mode to generate the preliminary determination and a high-speed mode to set the at-risk-zone flag.

26. The method of claim 24, wherein only one sensor is used to generate the preliminary determination and set the at-risk-zone flag.

27. The method of claim 24, further comprising loading a occupant tracking heuristic into said computer to generate said preliminary determination.

28. The method of claim 24, wherein setting an at-risk-zone flag includes:

dividing at least one of the window-of-interest images into a plurality of patches;

invoking a correlation heuristic to generate a correlation metric relating to the patches and a template image; and

comparing the correlation metric to a predetermined test threshold to set the at-risk-flag.