Antiskid brake control system using kalman filtering

• US 5,918,951 A
• Filed: 05/09/1997
• Issued: 07/06/1999
• Est. Priority Date: 05/09/1997
• Status: Expired due to Fees

• Current Assignee: Goodrich Corporation (Rtx Corporation)
• Original Assignee: Goodrich Corporation (Rtx Corporation)
• Inventors: Rudd, Robert Edward III
• Primary Examiner(s): Butler, Douglas C.
• Application Number:
  US08/853,555
• Time in Patent Office: 788 Days
• Field of Search: 303/150, 303/126, 303/149, 303/148, 303/154, 303/160, 303/163, 303/164, 303/165, 303/169, 303/170, 303/171, 303/172, 303/176, 303/177, 303/199, 303/191, 303/183, 303/178 , 303/112, 188/181 A, 188/181 C, 188/181 T, 701/71, 701/76, 701/75, 701/78, 701/79, 701/80, 701/90, 244/111
• US Class Current: 303/150
• CPC Class Codes: B60T 8/1703 for aircrafts
• CPC Class Codes: B60T 8/17616 Microprocessor-based systems

Abstract

A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising a sensor for measuring a speed of the wheel and for providing an output signal indicative of the speed; and a controller, operatively coupled to the sensor, which implements a Kalman filter to estimate an amplitude and location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and which controls the amount of braking force applied to the wheel based on the amplitude and location of the peak.

Claims

1. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel and for providing an output signal indicative of the speed; and

a controller, operatively coupled to the sensor, which implements a Kalman filter to estimate an amplitude and location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and which controls the amount of braking force applied to the wheel based on the amplitude and location of the peak.

2. The system of claim 1, wherein the amplitude and location of the peak are each modeled by a system state equation implemented within the controller.

3. The system of claim 2, wherein the amplitude and location of the peak are modeled substantially as constants within the controller and slowly varying changes in the amplitude and location are accounted for by adding fictitious noise to the respective constants.

4. The system of claim 2, wherein the amplitude and location of the peak are modeled substantially as constants and rapid changes in the amplitude and location are accounted for by resetting covariances within the Kalman filter associated with the amplitude and location.

5. The system of claim 4, wherein the covariance associated with the location is reset to cover predefined limits based on an expected range of covariance.

6. The system of claim 2, wherein values of the amplitude and location of the peak in the system state equations are each initialized approximately at an upper end of respective predefined ranges.

7. The system of claim 6, wherein the vehicle is an aircraft and the wheel is a wheel of the aircraft running on the runway, and the upper ends for the location and amplitude of the peak are approximately 0.15 and approximately 0.6, respectively.

8. The system of claim 6, wherein the values of the amplitude and location of the peak are initialized whenever the value of at least one parameter defined by state equations used within the controller falls outside a predefined range.

9. The system of claim 1, wherein the controller detects a skid condition based on estimates of velocity of the vehicle and velocity of the wheel provided by the Kalman filter, and as a result of detecting a skid condition causes a reduction in the amount of braking force applied to the wheel.

10. The system of claim 9, wherein the controller is configured to cause a reapplication of braking force to the wheel upon an estimated slip ratio between the wheel and the running surface dropping to approximately 40% the location of the peak.

11. The system of claim 1, wherein the controller causes the braking force applied to the wheel to incrementally increase towards the peak of the mu-slip curve based on an adaptive proportional gain term.

12. The system of claim 11, wherein the adaptive proportional gain term is based on the amplitude and the location of the peak in the mu-slip curve.

13. The system of claim 12, wherein the adaptive proportional gain term is proportional to an estimated average slope of the mu-slip curve.

14. The system of claim 1, wherein the controller computes a slip ratio set point based on the location of the peak of the mu-slip curve and controls the amount of braking force provided to the wheel based on the slip ratio set point.

15. The system of claim 14, wherein in computing the slip ratio set point the controller introduces an excitement signal to a term based on the location of the peak of the mu-slip curve in order to increase the observability of the term by the Kalman filter.

16. The system of claim 15, wherein the excitement signal functions to increase the observability of the amplitude and location of the peak of the mu-slip curve by the Kalman filter.

17. The system of claim 15, wherein the excitement signal is a periodic waveform having an amplitude based on the location of the peak.

18. The system of claim 17, wherein the excitement signal is a sinusoid.

19. The system of claim 1, further comprising a brake actuator and brake assembly operatively coupled to the controller for providing the braking force to the wheel based on a control signal output by the controller, and wherein the controller models the brake actuator to account for a response time of the brake actuator as a nominal value with a fictitious process noise representing changes in the response of the brake actuator due to wear.

20. The system of claim 1, wherein system state equations implemented within the controller define the speed of the wheel, the speed of the vehicle, and the radius of the wheel, and during initialization of the system state equations the speed of the wheel is initialized based on the output of the sensor.

21. The system of claim 20, wherein the speed of the vehicle is initialized by multiplying the initialized speed of the wheel by a predefined nominal value of the radius of the wheel.

38. The system of claim 1, wherein a torque to pressure ratio k of a brake applying the braking force to the wheel is modeled by a system state equation implemented within the controller, the torque to pressure ratio k being modeled substantially as a constant and slowly varying changes in k being accounted for by adding fictitious noise to the constant.

39. The system of claim 38, wherein the fictitious noise comprises a component which takes into account a change in brake coefficient of friction due to heating.

40. The system of claim 38, wherein the fictitious noise comprises a component which takes into account a time rate of change of the pressure applied by the brake.

42. The system of claim 15, wherein the excitement signal is a periodic waveform having an amplitude preselected to provide a preselected degree of modulation of pressure provided by the brake to apply the braking force.

22. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel; and

a controller, operatively coupled to the sensor, which implements a state estimator to estimate an amplitude and location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on an output of the sensor, and which controls the amount of braking force applied to the wheel based on the amplitude and location of the peak.

23. The system of claim 22, wherein the state estimator comprises a state vector including elements V, ω

, μ

_p and r_p, where V represents the velocity of the vehicle, ω

represents the speed of the wheel, μ

_p represents the peak of the mu-slip curve, and r_p represents the location of the peak of the mu-slip curve.

24. The system of claim 23, wherein the state vector further includes elements P and r, where P represents brake pressure applied to the wheel and r represents the radius of the wheel.

25. The system of claim 22, wherein the mu-slip curve, represented by μ

, is generally modeled within the state estimator by the following mathematical function;

space="preserve" listing-type="equation">μ

=(μ

.sub.p 62 )(r.sub.slip /r.sub.p)/{(β

-1)+|r.sub.slip /r.sub.p |.sup.β

}

where μ

_p represents the peak of the mu-slip curve, r_slip represents the relative amount of slip between the wheel and the surface, r_p represents the location of the peak of the mu-slip curve, and β

is a parameter which controls the steepness of the mu-slip curve.

41. The system of claim 23, wherein the state vector further includes torque to pressure ratio k of a brake applying the braking force to the wheel.

26. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel; and

a controller, operatively coupled to the sensor, for estimating a location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on an output of the sensor, and for controlling the amount of braking force applied to the wheel so as to maintain a slip ratio with respect to the velocity of the wheel compared to a velocity of the vehicle at or near the location of the peak in the mu-slip curve without tending to exceed the location of the peak.

27. The brake control system of claim 26, wherein the controller controls the amount of braking force applied to the wheel based on the following equation for peak applied brake pressure P_p :

space="preserve" listing-type="equation">P.sub.p =(μ

.sub.p /k) W.sub.t r+g I(1-r.sub.p)/r!

where μ

_p represents the peak of the mu-slip curve, k represents the torque to pressure ratio between a brake and the wheel, W_t represents the weight per wheel, r represents the radius of the wheel, g represents acceleration due to gravity, I represents inertia of the brake and the wheel, and r_p represents the location of the peak of the mu-slip curve.

28. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel and for providing an output signal indicative of the speed; and

a controller, operatively coupled to the sensor, which implements a Kalman filter to estimate at least one parameter of a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and controls the amount of braking force applied to the wheel based on the estimated parameter;

wherein the controller is configured to reset a covariance associated with the at least one parameter in the Kalman filter upon detection of a skid condition between the wheel and the surface.

29. The system of claim 28, wherein the at least one parameter comprises an amplitude and location of a peak of the mu-slip curve.

47. The system of claim 28, wherein the covariance is reset based on a difference between a current estimate of the at least one parameter and a preselected nominal value for the at least one parameter.

48. The system of claim 47, wherein the at least one parameter comprises a plurality of parameters and respective covariances are independently set for each of the plurality of parameters based on a difference between a corresponding current estimate and a corresponding preselected nominal value.

30. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel and for providing an output signal indicative of the speed; and

a controller, operatively coupled to the sensor, which implements a Kalman filter to estimate at least one parameter of a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and controls the amount of braking force applied to the wheel based on the estimated parameter;

wherein the controller is configured to employ an adjustable amount of control gain in controlling the amount of braking based on the estimated parameter.

31. The system of claim 30, wherein the adjustable amount of control gain is proportional to a slope of the mu-slip curve as estimated by the Kalman filter.

32. The system of claim 31, wherein the controller provides a higher level of control when the slope of the mu-slip curve is relatively steep as compared to a lower level of control which is provided when the slope of the mu-slip curve is relatively less steep.

33. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel; and

a controller, operatively coupled to the sensor, which implements a state estimator to estimate at least one parameter of a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on an output of the sensor, and which controls the amount of braking force applied to the wheel based on the at least one parameter;

wherein the amount of braking force is determined at least in part by a predefined mathematical function comprising a periodic component for increasing an ability of the state estimator to observe the at least one parameter.

34. The system of claim 33, wherein the function comprises a first term which is proportional to a value of the at least one parameter, and a second term which comprises the periodic component multiplied by the value of the at least one parameter.

35. The system of claim 33, wherein the at least one parameter is a peak of the mu-slip curve.

36. The system of claim 35, wherein the periodic component comprises a sinusoid.

37. A method for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

measuring a speed of the wheel and for providing an output signal indicative of the speed; and

using Kalman filter techniques to estimate an amplitude and location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and controlling the amount of braking force applied to the wheel based on the amplitude and location of the peak.

43. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel; and

a controller, operatively coupled to the sensor, which implements a state estimator to estimate at least one parameter of a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on an output of the sensor, and which controls the amount of braking force applied to the wheel based on the at least one parameter;

wherein the state estimator initializes the at least one parameter by setting the at least one parameter to an initial value which is predetermined to be greater than an expected true value of the at least one parameter.

44. The system of claim 43, wherein the state estimator selects a covariance for the at least one parameter which is large enough to encompass a predefined range within which the at least one parameter is expected to exist.

45. The system of claim 44, wherein the at least one parameter is selected from a group consisting of a peak of the mu-slip curve and a location of the peak of the mu-slip curve.

46. A brake control system for controlling an amount of braking force applied to a wheel of a vehicle running on a surface, comprising:

a sensor for measuring a speed of the wheel; and

a controller, operatively coupled to the sensor, for estimating a location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on an output of the sensor, and for controlling the amount of braking force applied to the wheel based on an operating point biased towards maintaining a slip ratio with respect to the speed of the wheel compared to a velocity of the vehicle, below the location of the peak in the mu-slip curve.

49. A method for controlling rotation of a wheel of a vehicle running on a surface, comprising:

measuring a speed of the wheel with a sensor to provide an output signal indicative of the speed;

applying a state estimator to the output signal of the sensor to estimate an amplitude and location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface; and

controlling the rotation of the wheel relative to the surface based on the amplitude and location of the peak.

50. The method of claim 49, wherein the state estimator comprises a Kalman filter.

51. The method of claim 49, wherein the step of controlling the rotation comprises a step of applying a braking force to the wheel.

52. A method for controlling rotation of a wheel of a vehicle running on a surface, comprising:

measuring a speed of the wheel with a sensor to provide an output signal indicative of the speed;

estimating a location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and controlling the rotation of the wheel relative to the surface to maintain a slip ratio with respect to the speed of the wheel compared to a velocity of the vehicle at or near the location of the peak in the mu-slip curve without tending to exceed the location of the peak.

53. A method for controlling rotation of a wheel of a vehicle running on a surface, comprising:

measuring a speed of the wheel with a sensor to provide an output signal indicative of the speed;

estimating a location of a peak in a mu-slip curve representative of a coefficient of friction between the wheel and the surface based on the output signal of the sensor, and controlling the rotation of the wheel relative to the surface based on an operating point biased towards maintaining a slip ratio with respect to the speed of the wheel compared to a velocity of the vehicle, below the location of the peak in the mu-slip curve.

54. The method of claim 53, wherein the step of estimating the location of the peak comprises applying a Kalman filter to the output signal of the sensor.

55. The method of claim 53, wherein the step of controlling the rotation of the wheel comprises the step of applying a braking force to the wheel.