Instant detection / diagnosis of abrupt bias fault in signals of vehicle motion sensors

• US 6,144,904 A
• Filed: 12/22/1998
• Issued: 11/07/2000
• Est. Priority Date: 12/22/1998
• Status: Expired due to Term

• Current Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Original Assignee: Ford Global Technologies, LLC (Ford Motor Company)
• Inventors: Tseng, Hongtei Eric
• Primary Examiner(s): Cuchlinski, Jr., William A.
• Assistant Examiner(s): Beaulieu, Yonel
• Application Number:
  US09/218,280
• Time in Patent Office: 686 Days
• Field of Search: 701/34, 701/63, 701/43, 701/29, 701/31, 701/32, 303/122, 303/122.05, 303/138, 303/140, 303/146, 303/147, 340/825.06, 340/825.16, 340/438, 180/170, 180/197
• US Class Current: 701/34
• CPC Class Codes: B60T 2210/22 Banked curves
• CPC Class Codes: B60T 8/171 Detecting parameters used in the regulation; Measuring values used in the regulation
• CPC Class Codes: G01P 21/00 Testing or calibrating of apparatus or devices covered by the preceding groups

Abstract

A system and method for enabling an abrupt bias fault in a vehicle motion sensor (24), such as a yaw rate sensor or a lateral acceleration sensor to be both quickly and accurately diagnosed. When a potential fault is first indicated, it is signaled as true only if the vehicle is in a stable state and vehicle vibration is ruled out as a cause of abrupt signal change. Processing of signals is conducted by microprocessor.

Claims

1. A system for detecting an abrupt bias fault in an on-board motion sensor in an automotive vehicle, the system comprising, in combination with such sensor:

signal sources that are independent of the sensor, including signal sources providing signals useful in calculating a motion that at least approximately corresponds to motion represented by a signal of the sensor; and

a processor

a) for processing the signals from the sensor and from the sources that are independent of the sensor to yield a calculation at least approximately corresponding to motion represented by the signal of the sensor;

b) the processor further processing the calculation and the signal from the sensor to calculate deviation of one from the other; and

c) the processor further processing the calculated deviation for distinguishing a frequency component indicative of such an abrupt bias fault from a frequency component that is not indicative of such a fault.

2. A system as set forth in claim 1 in which the processor, upon having distinguished said frequency component indicative of said abrupt bias fault, processes the sensor signal for noise induced by vehicle vibration to either validate or invalidate the distinguished frequency component indicative of said abrupt bias fault.

3. A system as set forth in claim 1 in which the signal sources include signal sources useful in distinguishing a stable state of the vehicle from a state other than said stable state;

and

the processor further processes a) signals from the signal sources useful in said distinguishing a stable state of the vehicle from a state other than said stable state and b) the result of the processing of the calculated deviation that distinguished said frequency component indicative of such an abrupt bias fault from a frequency component not indicative of such a fault.

4. A system as set forth in claim 3 in which the processor, upon having distinguished said frequency component indicative of said abrupt bias fault, processes the sensor signal for noise induced by vehicle vibration to either validate the distinguished frequency component as a true fault or invalidate the distinguished frequency component as a false fault, and signals the occurrence of said true fault in consequence of both fault validation by the noise processing and determination of stable vehicle state.

5. A system as set forth in claim 1 in which the processor at least approximately normalizes the calculated deviation for longitudinal vehicle velocity, processes the approximately normalized calculated deviation to obtain its value at a frequency of interest, and weights the value of the approximately normalized deviation by its value at the frequency of interest to create a residue.

6. A system as set forth in claim 5 in which the processor compares the residue to a threshold for distinguishing a residue that is indicative of said abrupt bias fault from a residue that is not indicative of said abrupt bias fault.

7. A system as set forth in claim 6 in which the processor processes the sensor signal and signals from at least some of the other signal sources a) for noise induced by vehicle vibrations and b) for distinguishing said stable state of the vehicle from a state other than said stable state, and

the processor signals a true abrupt bias fault when vibrations are below a certain vibration threshold, the vehicle state is stable, and the residue is indicative of said abrupt bias fault.

8. A method for detecting an abrupt bias fault in an on-board motion sensor in an automotive vehicle, the method comprising:

processing signals from signal sources that are independent of the sensor to calculate a motion that at least approximately corresponds to motion represented by a signal of the sensor;

calculating deviation between the sensor signal and the calculated motion; and

processing the calculated deviation for distinguishing a frequency component indicative of such an abrupt bias fault from a frequency component that is not indicative of such a fault.

9. A method as set forth in claim 8 including processing for noise induced by vehicle vibration to either validate or invalidate the distinguished frequency component indicative of said abrupt bias fault.

10. A method as set forth in claim 8 including processing signals from signal sources useful in distinguishing a stable state of the vehicle from a state other than said stable state.

11. A method as set forth in claim 10 including processing the sensor signal for noise induced by vehicle vibration to either validate the distinguished frequency component as a true fault or invalidate the distinguished frequency component as a false fault, and signaling the occurrence of said true fault in consequence of both fault validation by the noise processing and determination of stable vehicle state.

12. A method as set forth in claim 8 including at least approximately normalizing the calculated deviation for longitudinal vehicle velocity, processing the approximately normalized calculated deviation to obtain its value at a frequency of interest, and weighting the value of the approximately normalized deviation by its value at the frequency of interest to create a residue.

13. A method as set forth in claim 12 including comparing the residue to a threshold for distinguishing another residue that is indicative of said abrupt bias fault from a residue that is not indicative of said abrupt bias fault.

14. A method as set forth in claim 13 including processing the sensor signal and signals from at least some of the other signal sources a) for noise induced by vehicle vibrations and b) for distinguishing a stable state of the vehicle from a state other than said stable state, and signaling a true abrupt bias fault when vibrations are below a certain vibration threshold, the vehicle state is stable, and the residue is indicative of said abrupt bias fault.