System and method for estimating velocity using reliability indexed sensor fusion

• US 6,816,804 B1
• Filed: 06/04/2003
• Issued: 11/09/2004
• Est. Priority Date: 06/04/2003
• Status: Expired due to Fees

• Current Assignee: Visteon Global Technologies Incorporated (Visteon Corporation)
• Original Assignee: Visteon Global Technologies Incorporated (Visteon Corporation)
• Inventors: Lee, Hyeongcheol
• Primary Examiner(s): Bui, Bryan
• Application Number:
  US10/454,658
• Time in Patent Office: 524 Days
• Field of Search: 702/141, 702/142, 702/145, 702/147, 702/148, 702/150, 702/151, 702/191, 701/1, 701/36, 701/45, 701/72, 701/213, 303/140, 303/146-147
• US Class Current: 702/148
• CPC Class Codes: B60T 2250/04 Vehicle reference speed; Vehicle body speed
• CPC Class Codes: B60T 8/172 Determining control parameters used in the regulation, e.g. by calculations involving measured or detected parameters B60T8/17551 takes precedence
• CPC Class Codes: B60W 2050/0013 Optimal controllers
• CPC Class Codes: B60W 40/105 Speed

Abstract

A method and system for vehicle velocity estimation. The method and system effectively combine vehicle kinetic and kinematic models to determine the noise covariance or reliability of vehicle sensor signals, and vehicle approximated velocity. The vehicle approximated velocity, sensor signals, and noise covariances are used in conjunction with a Kalman filter to determine the estimated velocity of the vehicle. The noise covariance is calculated from the relationship between various sensor signals and calculated values by determining if the sensor signal and calculated values correspond to a vehicle operating state.

Claims

1. A system for dynamically estimating longitudinal and lateral velocities of a vehicle comprising:

a vehicle sensor producing a signal indicative of a vehicle state; and

a processor communicating with said vehicle sensor and configured to receive said signal, said processor further configured to;

calculate an approximated velocity of the vehicle;

determine a noise covariance associated with said signal and said approximated velocity; and

estimate the velocity of said vehicle using said signal, said approximated velocity and said noise covariance.

2. The system of claim 1 wherein said vehicle sensor is a wheel speed sensor and said signal is an angular wheel speed signal, said angular wheel speed signal being communicated to said processor.

3. The system of claim 2 wherein said approximated velocity includes an approximated longitudinal velocity and an approximated lateral velocity.

4. The system of claim 3 further including an accelerometer producing a longitudinal acceleration signal and a lateral acceleration signal, said lateral and longitudinal acceleration signals being communicated to said processor for use in determining said noise covariance and said estimated velocity.

5. The system of claim 4 further including a yaw rate sensor producing a yaw rate signal and a steering wheel angle sensor producing a steering wheel angle signal, said yaw rate signal and said steering wheel angle signals being communicated to said processor for use in calculating said approximated lateral velocity and determining said noise covariance.

6. The system of claim 5 further including a brake sensor producing a brake position signal and a throttle sensor producing a throttle position signal, said brake position signal and said throttle position signal being communicated to said processor for use in determining said noise covariance.

7. The system of claim 6 further including a four wheel drive engagement sensor producing a four wheel drive engagement signal, said four wheel drive engagement signal being communicated to said processor for use in determining said noise covariance.

8. The system of claim 1 wherein said estimated velocity includes a lateral estimated velocity and a longitudinal estimated velocity, and said Kalman filter module uses a Kalman filter framework to determine said lateral and longitudinal estimated velocities.

9. The system of claim 8 wherein said estimated velocities are defined by:

10. A system for dynamically estimating longitudinal and lateral velocities of a vehicle comprising:

a wheel speed sensor producing an angular wheel speed signal indicative of a vehicle state;

an accelerometer producing a longitudinal acceleration signal and a lateral acceleration signal;

a yaw rate sensor producing a yaw rate signal;

a steering wheel angle sensor producing a steering wheel angle signal;

a brake sensor producing a brake position signal;

a throttle sensor producing a throttle position signal; and

a processor communicating with said wheel speed sensor and said accelerometer and wherein said processor is configured to receive said wheel speed signal, said lateral acceleration signal and said longitudinal acceleration signal, said processor further configured to;

calculate an approximated longitudinal velocity;

calculate an approximated lateral velocity of the vehicle using said yaw rate signal and said steering wheel angle signal;

determine a noise covariance associated with said signal and said approximated velocity using said longitudinal acceleration signal, said lateral acceleration signal, said yaw rate signal, said brake position signal, said throttle position signal and said steering wheel angle signal, said noise covariance including a longitudinal system noise covariance, a lateral system noise covariance, a longitudinal measurement noise covariance, and a lateral measurement noise covariance, and wherein said longitudinal system noise covariance is inversely related to the reliability of said longitudinal acceleration signal, said lateral system noise covariance is inversely related to the reliability of said lateral acceleration signal, said longitudinal measurement noise covariance is inversely related to the reliability of said approximated longitudinal velocity, and said lateral measurement noise covariance is inversely related to the reliability of said approximated lateral velocity; and

estimate the velocity of said vehicle using said signal, said approximated velocity and said noise covariance.

11. The system of claim 10 wherein said longitudinal acceleration signal has a high reliability when at least one of said throttle position signal and said brake position signal indicate a vehicle operating state consistent with said longitudinal acceleration signal.

12. The system of claim 10 wherein said lateral acceleration signal has a high reliability when at least one of said steering wheel angle signal and said yaw rate signal indicate a vehicle operating state consistent with said lateral acceleration signal.

13. The system of claim 10 wherein said longitudinal approximated velocity has a high reliability when at least one of said throttle position signal and said brake position signal indicate a vehicle operating state consistent with said longitudinal approximated velocity.

14. The system of claim 10 wherein said lateral approximated velocity has a high reliability when at least one of said steering wheel angle signal and said yaw rate signal indicate a vehicle operating state consistent with said longitudinal approximated velocity.

15. The system of claim 10 wherein

said longitudinal system noise covariance Q_x is defined by:

16. A method of estimating a velocity of a vehicle comprising:

receiving sensor signals indicative of a vehicle travel state;

calculating an approximated velocity using said sensor signals;

determining a noise covariance associated with said sensor signals and said approximated velocity;

estimating a velocity of the vehicle with a Kalman filter framework, said Kalman filter framework using said noise covariance, said approximated velocity, and said sensor signals.

17. The method of claim 16 wherein said sensor signals include a wheel speed sensor signal, a lateral acceleration signal and a longitudinal acceleration signal.

18. A method of estimating a velocity of a vehicle comprising:

receiving sensor signals indicative of a vehicle travel state, including a wheel speed sensor signal, a lateral acceleration signal and a longitudinal acceleration signal;

calculating an approximated lateral velocity and an approximated longitudinal velocity using said sensor signals;

determining a noise covariance associated with said sensor signals, said longitudinal acceleration signal, said lateral acceleration signal said lateral approximated velocity and said longitudinal approximated velocity; and

estimating a velocity of the vehicle with a Kalman filter framework, said Kalman filter framework using said noise covariance, said approximated velocity, and said sensor signals.

19. The method of claim 18 wherein said approximated lateral velocity, said approximated longitudinal velocity, said lateral acceleration signal and said longitudinal acceleration signal each have an associated reliability, said associated reliability being inversely related to said noise covariance.

20. A system for estimating a velocity of a vehicle, said vehicle having a driven wheel, a throttle, a brake pedal, and a steering wheel, wherein each of said throttle, said brake pedal, and said steering wheel have a rest position, said system comprising:

an accelerometer producing an acceleration signal;

a wheel speed sensor producing a wheel speed signal;

a yaw rate sensor producing a yaw rate signal;

a steering wheel angle sensor producing a steering wheel angle signal;

a brake position sensor producing a brake pedal position signal;

a throttle position sensor producing a throttle position signal;

a processor communicating with said sensors and configured to receive said signals;

a velocity calculation module receiving said wheel speed sensor signal and calculating an approximated velocity of the vehicle;

a noise covariance module receiving said acceleration signal and at least one other of said signals, said noise covariance module calculating a noise covariance associated with said approximated velocity and said acceleration signal, said acceleration signal and said approximated velocity each having an associated reliability, wherein said noise covariance is inversely related to said reliability; and

a Kalman filter module estimating a velocity of the vehicle, said Kalman filter module including a Kalman filter framework configured to receive said noise covariance, said approximated velocity and said acceleration signal.