Vehicle rollover sensing using extended kalman filter

• US 6,002,974 A
• Filed: 02/06/1998
• Issued: 12/14/1999
• Est. Priority Date: 02/06/1998
• Status: Expired due to Term

• Current Assignee: Delphi Technologies, Inc. (BorgWarner Incorporated)
• Original Assignee: Delco Electronics Corporation (Motors Liquidation Co. GUC Trust)
• Inventors: Schiffmann, Jan Konried
• Primary Examiner(s): Cuchlinski, Jr., William A.
• Assistant Examiner(s): Hernandez, Olga
• Application Number:
  US09/019,878
• Time in Patent Office: 676 Days
• Field of Search: 701/36, 701/110, 701/45, 701/38, 340/440, 280/756, 180/282
• US Class Current: 701/36
• CPC Class Codes: B60R 2021/01075 between the airbag control unit and pyrotechnic fuses or equivalent actuators
• CPC Class Codes: B60R 2021/01306 monitoring vehicle inclination
• CPC Class Codes: B60R 2021/01313 monitoring the vehicle steering system or the dynamic control system
• CPC Class Codes: B60R 2021/01325 Vertical acceleration
• CPC Class Codes: B60R 2021/01327 Angular velocity or angular acceleration
• CPC Class Codes: B60R 21/01 Electrical circuits for triggering passive safety arrangements, e.g. airbags, safety belt tighteners, in case of vehicle accidents or impending vehicle accidents electrical circuits for transmission of signals between steering wheel and the vehicle itself B60R16/027; for electrically actuating belt retractor locking means B60R22/343
• CPC Class Codes: B60R 21/013 including means for detecting collisions, impending collisions or roll-over
• CPC Class Codes: B60R 21/01336 using filtering
• CPC Class Codes: B60R 22/343 with electrically actuated locking means B60R22/3405 takes precedence
• CPC Class Codes: B60R 22/46 Reels with means to tension the belt in an emergency by forced winding upelectrical circuits for triggering safety arrangements B60R21/01 ; tensioners using reels only guiding the belt during normal use, e.g. splitted reels B60R22/195

Abstract

A rollover sensing apparatus and method are provided for predicting rollover or pitchover conditions of a vehicle. The apparatus and method sense angular roll rate, angular pitch rate, longitudinal acceleration, lateral acceleration, and vertical acceleration of the vehicle and produce output signals indicative thereof. A non-linear filter, such as an extended Kalman filter, estimates a current roll angle and a current pitch angle as a function of the sensed signals. A predictor predicts a future roll angle as a function of the estimated current roll angle and the sensed roll rate, and further predicts a future pitch angle as a function of the estimated current pitch angle and the sensed pitch rate. The predicted roll and pitch attitude angles are compared to a threshold value and a vehicle overturn condition signal is output to indicate a predicted vehicle rollover or pitchover condition.

Claims

1. A rollover sensing apparatus for predicting an overturn condition for a vehicle, comprising:

an angular roll rate sensor for sensing roll rate of a vehicle and producing an output signal indicative thereof;

an angular pitch rate sensor for sensing pitch rate of the vehicle and producing an output signal indicative thereof;

a longitudinal accelerometer for measuring longitudinal acceleration of the vehicle and producing an output signal indicative thereof;

a lateral accelerometer for measuring lateral acceleration of the vehicle and producing an output signal indicative thereof;

a vertical accelerometer for measuring vertical acceleration of the vehicle and producing an output signal indicative thereof;

a non-linear filter for receiving said signals from said angular roll rate and pitch rate sensors, and said longitudinal, lateral, and vertical accelerometers, and for estimating a roll angle and a pitch angle as a function of said received signals;

a predictor for predicting a future roll angle as a function of said estimated roll angle and said sensed roll rate, and for predicting a future pitch angle as a function of said estimated pitch angle and said sensed pitch rate;

a comparator for comparing each of said predicted roll and pitch attitude angles to a threshold value; and

an output for deploying a vehicle overturn condition output based on said comparison.

2. The apparatus as defined in claim 1, wherein said non-linear filter comprises an extended Kalman filter.

3. The apparatus as defined in claim 1, further comprising a first integrator for receiving the estimated roll angle from the filter at a first time update rate and further receiving the sensed roll rate at a second time update rate which is faster than said first time update rate, such that said estimated roll angle periodically establishes an estimated current roll angle at its first time update rate, while said sensed roll rate is integrated to establish estimated current roll angle updates between filter updates.

4. The apparatus as defined in claim 3, further comprising a first integrator for receiving the estimated pitch angle from the filter at a first time update rate and further receiving the sensed pitch rate at a second time update rate which is faster than said first time update rate, such that said estimated pitch angle periodically establishes an estimated current pitch angle at its first time update rate, while said sensed pitch rate is integrated to establish estimated current pitch angle updates between filter updates.

5. The apparatus as defined in claim 1, wherein said predictor further comprises a Taylor series-based predictor for providing said prediction of roll and pitch angles.

6. The apparatus as defined in claim 1, wherein said non-linear filter further estimates bias in said input signals and further subtracts said bias from said received signals.

7. The apparatus as defined in claim 1, wherein said predictor predicts said future roll angle further as a function of an estimated roll acceleration and predicts said future pitch angle further as a function of an estimated pitch acceleration.

8. An apparatus for estimating a current attitude angle of a vehicle comprising:

an angular roll rate sensor for measuring roll rate of a vehicle and producing an output signal indicative thereof;

an angular pitch rate sensor for measuring pitch rate of the vehicle and producing an output signal indicative thereof;

a longitudinal accelerometer for measuring longitudinal acceleration of the vehicle and producing an output signal indicative thereof;

a lateral accelerometer for measuring lateral acceleration of the vehicle and producing an output signal indicative thereof;

a vertical accelerometer for measuring vertical acceleration of the vehicle and producing an output signal indicative thereof; and

a non-linear filter for receiving signals from said angular roll rate and pitch rate sensors and said longitudinal, lateral, and vertical accelerometers, and said non-linear filter estimating a roll angle and a pitch angle based on said received signals.

9. The apparatus as defined in claim 8, wherein said non-linear filter comprises an extended Kalman filter.

10. The apparatus as defined in claim 8, further comprising a first integrator for receiving the estimated roll angle from the filter at a first time update rate and further receiving the measured roll rate at a second time update rate which is faster than said first time update rate, such that said estimated roll angle periodically establishes an estimated current roll angle at its first time update rate, while said measured roll rate is integrated to establish estimated current roll angle updates between filter updates.

11. The apparatus as defined in claim 10, further comprising a first integrator for receiving the estimated pitch angle from the filter at a first time update rate and further receiving the measured pitch rate at a second time update rate which is faster than said first time update rate, such that said estimated pitch angle periodically establishes an estimated current pitch angle at its first time update rate, while said measured pitch rate is integrated to establish estimated current pitch angle updates between filter updates.

12. A method for predicting an overturn condition of a vehicle, said method comprising the steps of:

sensing angular roll rate of a vehicle and producing an output signal indicative thereof;

sensing angular pitch rate of the vehicle and producing an output signal indicative thereof;

sensing longitudinal acceleration of the vehicle and producing an output signal indicative thereof;

sensing lateral acceleration of the vehicle and producing an output signal indication thereof;

sensing vertical acceleration of the vehicle and producing an output signal indicative thereof;

estimating a roll angle and a pitch angle with a non-linear filter as a function of said sensed angular roll and pitch rates and said sensed longitudinal, lateral, and vertical accelerations;

predicting a roll angle based on said estimated roll angle and said sensed roll rate;

predicting a pitch angle based on said estimated pitch angle and said sensed pitch rate;

comparing said predicted roll angle to a roll threshold value;

comparing said predicted pitch angle to a pitch threshold value; and

deploying a vehicle overturn condition signal based on either of said comparisons.

13. The method as defined in claim 12, wherein said step of estimating said roll angle and said pitch angle includes extended Kalman filtering.

14. The method as defined in claim 12, further comprising the steps of updating said estimated roll angle at a first time update rate, integrating said measured roll rate at a second time update rate which is faster than said first time update rate, and updating said estimated roll angle as a function of said integrated roll rate to establish estimated current roll angle updates between said filter updates.

15. The method as defined in claim 14, further comprising the steps of updating said estimated pitch angle at a first time update rate, integrating said measured pitch rate at a second time update rate which is faster than said first time update rate, and updating said estimated pitch angle as a function of said integrated pitch rate to establish estimated current pitch angle updates between filter updates.

16. The method as defined in claim 12, wherein said step of predicting a roll angle comprises performing a Taylor series-based prediction.

17. The method as defined in claim 12, further comprising the step of estimating bias of said input signals and further subtracting said bias from current received signals.

18. The method as defined in claim 12, further comprising the step of estimating a current roll acceleration and predicting said roll angle further as a function of said estimated current roll acceleration.

19. A method of estimating a current attitude angle of a vehicle, said method comprising the steps of:

sensing angular roll rate of a vehicle and producing an output signal indicative thereof;

sensing angular pitch rate of the vehicle and producing an output signal indicative thereof;

sensing longitudinal acceleration of the vehicle and producing an output signal indicative thereof;

sensing lateral acceleration of the vehicle and producing an output signal indicative thereof;

sensing vertical acceleration of the vehicle and producing an output signal indicative thereof; and

estimating a roll angle and a pitch angle with a non-linear filter as a function of said sensed angular roll and pitch rates and said sensed longitudinal, lateral, and vertical accelerations.

20. The method as defined in claim 19, wherein said step of estimating said roll angle and said pitch angle includes extended Kalman filtering.

21. The method as defined in claim 19, further comprising the steps of updating said estimated roll angle at a first time update rate, integrating said measured roll rate at a second time update rate which is faster than said first time update rate, and updating said estimated roll angle as a function of said integrated roll rate to establish estimated current roll angle updates between filter updates.

22. The method as defined in claim 19, further comprising the steps of updating said estimated pitch angle at a first time update rate, integrating said measured pitch rate at a second time update rate which is faster than said first time update rate, and updating said estimated pitch angle as a function of said integrated pitch rate to establish estimated current pitch angle updates between filter updates.