METHOD FOR INDIRECT TIRE PRESSURE MONITORING AND TIRE PRESSURE MONITORING SYSTEM

• US 20100211250A1
• Filed: 09/01/2008
• Published: 08/19/2010
• Est. Priority Date: 09/03/2007
• Status: Active Grant

• Current Assignee: Continental Teves AG & Co. OHG (Continental AG)
• Original Assignee: Continental Teves AG & Company HG (Continental AG)
• Inventors: Sussmann, Christian, Irth, Markus, Köbe, Andreas, Fischer, Daniel
• Granted Patent: US 8,397,559 B2
• Time in Patent Office: Days
• US Class Current: 701/29
• CPC Class Codes: B60C 23/061 by monitoring wheel speed measuring distance traversed on the ground by vehicles G01C22/00
• CPC Class Codes: B60C 23/062 Frequency spectrum analysis of wheel speed signals, e.g. using Fourier transformation

Abstract

A system and method for indirect tire pressure monitoring. A pressure loss variable (X, f, c) is determined based on an analysis of an oscillation behavior of a tire of a motor vehicle for the detection of a pressure loss in the tire of a wheel. A variable (α, α/v) representing an increase (α) of a wheel torque slip curve, or depending on an increase (α) of a wheel torque slip curve (α/v), is determined from the oscillation behavior of the tire, and the variable (α, α/v) and the pressure loss variable (X, f, c) are used for detecting the pressure loss.

Claims

1-16. -16. (canceled)

17. A method for indirect tire pressure monitoring, in which a pressure loss identification variable (X, f, c) is determined for identification of a pressure loss at the tire of the wheel on the basis of an analysis of the oscillation behavior of a wheel of a motor vehicle, wherein a characteristic variable (α

, α

/v), which represents a gradient (α

) of a wheel torque slip curve, or depends on a gradient (α

) of a wheel torque slip curve (α

/v), is determined from the oscillation behavior of the wheel, and wherein the characteristic variable (α

, α

/v) and the pressure loss identification variable (X, f, c) are used for pressure loss identification.

18. The method as claimed in claim 17, wherein the pressure loss identification variable (X, f) is corrected, and/or the pressure loss identification variable (X, f) is determined as a function of the determined characteristic variable (α

/v).

19. The method as claimed in claim 17, wherein the characteristic variable (α

/V) is determined from a ratio (A_ratio) of two dimension variables, wherein each dimension variable represents the energy content (G(f_eigen1), G(f_eigen2), E₁, E₂) of the oscillation behavior of the wheel at a frequency (f_eigen1, f_eigen2) or in a frequency range (f_eigen1−

Δ

f₁, f_eigen1+Δ

f₁;

f_eigen2−

Δ

f₂, f_eigen2+Δ

f₂).

20. The method as claimed in claim 17, wherein a frequency analysis is carried out on the oscillation behavior of the wheel, and the characteristic variable (α

/v) is determined from a frequency spectrum.

21. The method as claimed in claim 20, wherein the oscillation behavior of the wheel is an angular velocity signal (ω

) from the wheel, and the frequency spectrum is determined in particular by Fourier transformation.

22. The method as claimed in claim 20, wherein the characteristic variable (α

/v) is determined as a ratio (A_ratio)

of the amplitude values (G(f_eigen1), G(f_eigen2)) of the frequency spectrum at two predetermined or determined resonant frequencies (f_eigen1, f_eigen2), or

the energy contents of the frequency spectrum in the region of two predetermined or determined frequency ranges (f_eigen1+Δ

f₁, f_eigen1+Δ

f₁;

f_eigen2−

Δ

f₂, f_eigen2+Δ

f₂).

23. The method as claimed in claim 17, wherein the characteristic variable (α

/V) is determined from the time profile of an angular velocity signal (ω

) from the wheel.

24. The method as claimed in claim 23, wherein the characteristic variable (α

/v) is determined by filtering by at least one bandpass filter.

25. The method as claimed in claim 23, wherein the angular velocity signal (ω

) from the wheel is filtered by at least one bandpass filter, whose cut-off frequencies (f_eigen1−

Δ

f₁, f_eigen1+Δ

f₁;

f_eigen2−

Δ

f₂, f_eigen2+Δ

f₂) are located around a predetermined or determined resonant frequency (f_eigen1, f_eigen2), and wherein the characteristic variable (α

/v) is determined from the variance of the filtered signal.

26. The method as claimed in claim 25, wherein the characteristic variable (α

/v) is determined from a ratio (A_ratio) of the variances of two signals which have been filtered by bandpass filters with different cut-off frequencies.

27. The method as claimed in claim 17, wherein a model-based analysis is carried out of the oscillation behavior of the wheel, and wherein at least one parameter (J_rim, J_{radial belt}, c) of the basic model or of the basic model-based equation is determined, and wherein the characteristic variable (α

/v) is determined from the determined parameter or parameters.

28. The method as claimed in claim 27, wherein the oscillation behavior of the wheel is an angular velocity signal (ω

) from the wheel, and the at least one parameter (J_rim, J_{radial belt}, c) of the basic model or of the basic model-based equation is determined by parameter estimation.

29. The method as claimed in claim 17, wherein a frequency analysis is carried out of the oscillation behavior of the wheel, and wherein the pressure loss identification variable (X) is a resonant frequency (f_eigen) determined on the basis of the frequency analysis.

30. The method as claimed in claim 29, wherein the oscillation behavior of the wheel is an angular velocity signal (ω

) from the wheel.

31. The method as claimed in claim 29, wherein a value of the characteristic variable (α

/v) is determined for each determined frequency spectrum, and wherein frequency spectra are averaged using values of the characteristic variable (α

/v) which are located within a predetermined value range, wherein the pressure loss identification variable (f_eigen) is determined from the averaged frequency spectrum.

32. The method as claimed in claim 17, wherein a model-based analysis is carried out of the oscillation behavior of the wheel, wherein at least one parameter (J_rim, J_{radial belt}, c) of the basic model or of the basic model-based equation is determined, and wherein the pressure loss identification variable (X) is a parameter of the model (c) or is determined (f_eigen) from the parameter or parameters of the model.

33. The method as claimed in claim 32, wherein the oscillation behavior of the wheel is an angular velocity signal (ω

) from the wheel, and the at least one parameter (J_rim, J_{radial belt}, c) of the basic model or of the basic model-based equation is determined by parameter estimation.

34. The method as claimed in claim 17, wherein a decision is made as a function of the characteristic variable (α

/v) as to whether or not a pressure loss identification variable (X, f, c) is used for pressure loss monitoring.

35. The method as claimed in claim 17, wherein in a learning phase, value pairs are learnt for the characteristic variable (α

/v) and for the pressure loss identification variable (X, f, c) as comparison values, wherein currently determined value pairs are compared with these comparison values in a pressure loss monitoring phase, in order to identify tire pressure loss.

36. The method as claimed in claim 17, wherein in a learning phase, value pairs are learnt for the characteristic variable (α

/v) and for the pressure loss identification variable (X, f, c), wherein a functional relationship (g) between at least the characteristic variable (α

/v) and the pressure loss identification variable (X, f, c) is determined from the value pairs, and wherein a pressure loss identification is carried out in a pressure loss monitoring phase on the basis of a currently determined pressure loss identification variable (X, f, c) and the determined functional relationship (g).

37. The method as claimed in claim 36, wherein the pressure loss identification is carried out with a currently determined pressure loss identification variable (X, f, c) being corrected on the basis of the functional relationship (g).

38. The method as claimed in claim 17, wherein a gradient (α

) of the wheel torque slip curve is determined from the characteristic variable (α

/v), and any change in a coefficient of friction (μ

) between the roadway and the tire and/or any load change (Fz) of the vehicle are/is determined on the basis of the gradient (α

).

39. A tire pressure monitoring system, in which the wheel rotation speed (ω

) of at least one wheel, or a variable which is related to the wheel rotation speed of the wheel, is determined and is evaluated for identification of the tire pressure loss, comprising an evaluation unit in which a method as claimed in claim 17 is carried out.