Methods for determining the direction of rotation of rotating bodies

0Associated
Cases 
0Associated
Defendants 
0Accused
Products 
20Forward
Citations 
0
Petitions 
3
Assignments
First Claim
1. A method for determining a direction of rotation of a rotating body about an axis of rotation, which comprises the steps of:
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
differentiating mathematically the measured acceleration signals x and y resulting in a differentiated x acceleration signal and a differentiated y acceleration signal;
differentiating a further time at least one of the differentiated x acceleration signal and the differentiated y acceleration signal resulting in a further differentiated x acceleration signal and/or a further differentiated y acceleration signal; and
performing at least one of;
correlating the further differentiated x acceleration signal with the differentiated y acceleration signal for obtaining a first correlation function; and
correlating the further differentiated y acceleration signal with the differentiated x acceleration signal for obtaining a second correlation function.
3 Assignments
0 Petitions
Accused Products
Abstract
A method determines a direction of rotation of a rotating body about its axis of rotation by measuring two acceleration signals x and y with an acceleration sensor. The two acceleration signals x and y are at a 90° angle to each other, are oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration are taken into account. The measured acceleration signals x and y are differentiated mathematically. The differentiated x acceleration signal is differentiated a further time and correlated with a y signal that has been differentiated once for obtaining a correlation function. The differentiated y acceleration signal can be differentiated a further time and correlated with the x signal that has been differentiated once for obtaining a second correlation function. The method determines the direction of rotation of rotating tires of a motor vehicle.
30 Citations
View as Search Results
Apparatus, system and method for configuring a tire information sensor with a transmission protocol based on vehicle trigger characteristics  
Patent #
US 10,220,660 B2
Filed 08/01/2016

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Tire pressure monitor system apparatus and method  
Patent #
US 9,676,238 B2
Filed 11/06/2015

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Apparatus and method for data transmissions in a tire pressure monitor  
Patent #
US 9,776,463 B2
Filed 08/11/2015

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

RF transmission method and apparatus in a tire pressure monitoring system  
Patent #
US 9,517,664 B2
Filed 02/20/2015

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Apparatus and method for data transmissions in a tire pressure monitor  
Patent #
US 9,259,980 B2
Filed 10/27/2014

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Pressure check tool and method of operating the same  
Patent #
US 9,446,636 B2
Filed 02/26/2014

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Tire position determination system  
Patent #
US 9,193,224 B2
Filed 12/16/2013

Current Assignee
Kabushiki Kaisha Tokai Rika Denki Seisakusho

Sponsoring Entity
Kabushiki Kaisha Tokai Rika Denki Seisakusho

Universal tire pressure monitoring sensor  
Patent #
US 8,692,661 B2
Filed 01/26/2012

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Method of preparing a universal tire pressure monitoring sensor  
Patent #
US 8,742,913 B2
Filed 01/26/2012

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Method for determining a pivoting angle of a wheel unit mounted onto a snapin inflation valve  
Patent #
US 9,891,045 B2
Filed 12/19/2011

Current Assignee
Continental Automotive France, Continental Automotive GmbH

Sponsoring Entity
Continental Automotive France, Continental Automotive GmbH

Protocol misinterpretation avoidance apparatus and method for a tire pressure monitoring system  
Patent #
US 8,502,655 B2
Filed 08/09/2011

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Protocol arrangement in a tire pressure monitoring system  
Patent #
US 8,576,060 B2
Filed 08/09/2011

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Tire pressure monitoring apparatus and method  
Patent #
US 8,742,914 B2
Filed 08/09/2011

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Apparatus and method for activating a localization process for a tire pressure monitor  
Patent #
US 9,024,743 B2
Filed 08/09/2011

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Protocol protection  
Patent #
US 8,751,092 B2
Filed 01/13/2011

Current Assignee
Continental Automotive Systems US Incorporated

Sponsoring Entity
Continental Automotive Systems US Incorporated

Method for selfcalibrating an accelerometer mounted on a wheel of a vehicle while said vehicle is moving  
Patent #
US 8,849,601 B2
Filed 02/19/2010

Current Assignee
Continental Automotive France

Sponsoring Entity
Continental Automotive France

Method for determining the wheel position in a vehicle  
Patent #
US 20090012740A1
Filed 06/11/2008

Current Assignee
Robert Bosch GmbH

Sponsoring Entity
Robert Bosch GmbH

ACCELEROMETER BASED SYSTEM FOR DETECTION OF TIRE TREAD SEPARATION AND LOOSE WHEELS  
Patent #
US 20070299573A1
Filed 06/26/2006

Current Assignee
International Truck Intellectual Property Company LLC

Sponsoring Entity
International Truck Intellectual Property Company LLC

Universal tire pressure monitor  
Patent #
US 20050104722A1
Filed 11/18/2003

Current Assignee
Sensata Technologies Incorporated

Sponsoring Entity
Sensata Technologies Incorporated

Universal tire pressure monitor  
Patent #
US 7,518,495 B2
Filed 11/18/2003

Current Assignee
Sensata Technologies Incorporated

Sponsoring Entity
Lear Corporation

Rolling motion stability control apparatus for a vehicle  
Patent #
US 7,240,971 B2
Filed 07/28/2005

Current Assignee
Advics Company Limited

Sponsoring Entity
Advics Company Limited

Sensor system  
Patent #
US 7,284,408 B2
Filed 03/22/2005

Current Assignee
DENSO Corporation

Sponsoring Entity
DENSO Corporation

Device for detecting the position of a vehicle wheel  
Patent #
US 20060044125A1
Filed 11/17/2003

Current Assignee
Continental Automotive GmbH

Sponsoring Entity
Continental Automotive GmbH

Determination of wheel sensor position using a wireless solution  
Patent #
US 20030197603A1
Filed 04/18/2002

Current Assignee
Sensata Technologies Incorporated

Sponsoring Entity


Method for sensing side impacts in a vehicle  
Patent #
US 6,917,866 B2
Filed 12/14/2001

Current Assignee
Robert Bosch GmbH

Sponsoring Entity
Robert Bosch GmbH

BASEBALL TRAINING APPARATUS  
Patent #
US 20020082125A1
Filed 12/22/2000

Current Assignee
Little Tikes Company The

Sponsoring Entity
Little Tikes Company The

Accelerometer based angular position sensor  
Patent #
US 6,873,931 B1
Filed 10/10/2000

Current Assignee
Computational Systems Incorporated

Sponsoring Entity
CSI Technology Incorporated

Method for assigning identifiers, present in signals emitted by transmitters in a tire pressure monitoring system, to the wheels where the transmitters are located  
Patent #
US 6,446,502 B1
Filed 02/22/2000

Current Assignee
Huf Hulsbeck Furst GmbH Co. KG

Sponsoring Entity
BERU AKTIENGESELLSCHAFT

Wheel information estimating apparatus  
Patent #
US 6,142,026 A
Filed 06/01/1995

Current Assignee
Ibaraki Toyota Jidosha Kabushiki Kaisha, Kabushiki Kaisha Toyota Chuo Kenkyusho

Sponsoring Entity
Ibaraki Toyota Jidosha Kabushiki Kaisha, Kabushiki Kaisha Toyota Chuo Kenkyusho

Suspension control system for enhancing response characteristics of a vehicle body  
Patent #
US 5,173,858 A
Filed 12/18/1990

Current Assignee
Mitsubishi Electric Corporation

Sponsoring Entity
Mitsubishi Electric Corporation

8 Claims
 1. A method for determining a direction of rotation of a rotating body about an axis of rotation, which comprises the steps of:
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
differentiating mathematically the measured acceleration signals x and y resulting in a differentiated x acceleration signal and a differentiated y acceleration signal;
differentiating a further time at least one of the differentiated x acceleration signal and the differentiated y acceleration signal resulting in a further differentiated x acceleration signal and/or a further differentiated y acceleration signal; and
performing at least one of;
correlating the further differentiated x acceleration signal with the differentiated y acceleration signal for obtaining a first correlation function; and
correlating the further differentiated y acceleration signal with the differentiated x acceleration signal for obtaining a second correlation function.  View Dependent Claims (2, 3, 4, 5, 6)
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
 7. A method for determining a direction of rotation of a rotating body about an axis of rotation, which comprises the steps of:
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
differentiating mathematically the measured acceleration signals x and y resulting in a differentiated x acceleration signal and a differentiated y acceleration signal;
differentiating a further time the differentiated y acceleration signal resulting in a further differentiated y acceleration signal; and
correlating the further differentiated y acceleration signal with the differentiated x acceleration signal for obtaining a correlation function.
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
 8. A method for determining a direction of rotation of a rotating body about an axis of rotation, which comprises the steps of:
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
differentiating mathematically the measured acceleration signals x and y resulting in a differentiated x acceleration signal and a differentiated y acceleration signal;
differentiating a further time the differentiated x acceleration signal resulting in a further differentiated x acceleration signal; and
correlating the further differentiated x acceleration signal with the differentiated y acceleration signal for obtaining a correlation function.
 measuring two acceleration signals x and y with at least one acceleration sensor resulting in measured acceleration signals x and y, the two acceleration signals x and y being at a 90 degree angle to each other and oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g being taken into account;
1 Specification
The invention relates to methods for determining the direction of rotation of a rotating body about its axis of rotation by measuring two acceleration signals x and y with at least one acceleration sensor.
One function of tire pressure monitoring systems is the automatic localization of the individual wheels. The function is used for the purpose of automatically determining the position of a wheel or, more specifically, the wheel electronics installed in the wheel. This enables warnings in relation to a tire that are directly linked to the corresponding wheel position to be displayed to the driver. In the localization process a distinction is made between full localization (position of all four wheels), axle localization (front/rear) and side localization (right/left).
An approach to side localization includes determining the direction of rotation of the wheel electronics, which information is transmitted to the central control device.
Based on the directionofrotation signal, the wheel position (left/right) can be determined with the aid of the direction of movement of the vehicle. Alternatively, the wheel position can be determined directly by the wheel electronics, either based on plausibility checks or on the assumption that the vehicle travels mostly in a forward direction.
An approach to directionofrotation detection in the wheel electronics is implemented with the aid of twodimensional acceleration sensors. The sensors measure acceleration signals of the specified directions x and y. The directionofrotation detection is based on the evaluation of the phase displacement of the acceleration signals of the sensors for the x and y direction, produced due to the gravitation g.
European patent EP 0 760 299 B1 discloses a tire pressure monitoring system which determines the direction of rotation of the tires and the direction of travel of the motor vehicle by a sensor on each wheel, and thereby determines the tire position. The direction of rotation of the wheel is established via what is referred to as a rolling switch that triggers a number of switches in a sequence that is dependent on the direction of rotation.
It is known from European patent EP 1 172 656 A1 (corresponding to U.S. patent disclosure No. 2002/0047715 A1) that the side localization takes place based on the phase difference between the signals of two accelerometers within the wheel electronics. A specific solution indicating how the phase difference is evaluated, i.e. how the localization result is obtained, is not given. It is also described that the centrifugal acceleration has no significant influence on the evaluation of the phase relationship.
It is also known from U.S. patent disclosure No. 2003/0197603 Al that the side localization takes place based on the phase displacement between the signals of two acceleration sensors within the wheel electronics. A specific solution indicating how the phase difference is evaluated, i.e. how exactly the localization result is obtained, is also not given here. It is described that the influence of the centrifugal acceleration is present, but is of no relevance for the evaluation.
It is known from international patent disclosure WO 2004/048131 A1 (corresponding to U.S. patent disclosure No. 2006/0044125 A1) that the side localization takes place based on the phase shift between the signals of two accelerometers mounted within the wheel electronics. A specific solution for the evaluation is not given here either. The shock sensors, as they are called, described therein are specified in terms of functionality as accelerometers.
A disadvantage of the prior art is that the phase relationship of the acceleration signals or of the corresponding sinusoidal oscillations to one another and the associated determination of the direction of rotation of the rotating body cannot be established. Furthermore, the processing of the sinusoidal signals which are overlaid with rotational speeddependent mean values due to the centrifugal forces is not solved.
SUMMARY OF THE INVENTIONIt is accordingly an object of the invention to provide methods for determining the direction of rotation of rotating bodies that overcomes the abovementioned disadvantages of the prior art methods of this general type, which determines the direction of rotation of rotating bodies, in particular for determining the direction of rotation of the wheel of a motor vehicle.
The present invention makes use of the idea of measuring two acceleration signals within a body for the purpose of directionofrotation detection of rotating bodies, e.g. tires of motor vehicles, and processing the signals further in a mathematical procedure in such a way that the direction of rotation of the body can be determined. In connection with motor vehicles the side of the vehicle on which the tires are located is finally determined from the directionofrotation information and the direction of travel that is to be determined.
With the methods for determining the direction of rotation of a rotating body, the following acceleration signals x and y are generated at the sensors in the case of a body rotating in the clockwise direction:<FORM>x(t)=A·sin (ωt)+A<sub>0 </sub></FORM><FORM>y(t)=B·sin (ωt−π/2)+B<sub>0 </sub>(trailing). </FORM>
In the case of a body rotating in the anticlockwise direction, the following acceleration signals are produced at the sensors:<FORM>x(t)=A·sin (ωt)+A<sub>0 </sub></FORM><FORM>y(t)=B·sin (ωt+π/2)+B<sub>0 </sub>(leading). </FORM>
A 0°/90° phase arrangement is present between the two signals x and y. In other words, they must be adjusted essentially at a 90° angle to each other, i.e. in the defined directions x (0°) for the horizontal movement direction and y (90°) for the vertical movement direction. The method is stable with respect to small angular deviations of, for example, 5°, such as result in particular due to installation tolerances. In general, the sensor orientation does not have to be horizontal and vertical. The method can also be applied to sensor orientations in which the complete sensor configuration is rotated and the angle of 90° between x and y is maintained.
Without in any way restricting the generality, the absolute phase of the x(t) signal is set to 0, since it is only the phase difference of the two signals x(t) and y(t) that counts. In this case the angular frequency of the rotating body is designated by ω, which frequency is in turn dependent on the rotational speed of the body. The frequency of the oscillations does not have to be determined, calculated or estimated explicitly. This is advantageous since, for example, the rotational speed of a wheel may change constantly as a function of the speed of the vehicle. The amplitude of the oscillations likewise does not have to be determined, calculated or estimated explicitly. The mean values A<sub>0 </sub>and B<sub>0 </sub>correspond to the centrifugal acceleration at the sensors and consequently depend on the rotational speed of the wheel and the precise installation angle of the sensors. As the sensors can never be mounted at exactly the same angle due to installation and manufacturing tolerances, the mean values for x(t) and y(t) are generally different. An important requirement for a solution that is relevant to realworld practice is the effective elimination of the mean values as they are typically unknown.
The mean values can be eliminated by mathematical differentiation. The acceleration signals x and y supply the following signals for the rotation in the clockwise direction:<FORM>x<sub>2</sub>(t)=A·cos (ωt) </FORM><FORM>y<sub>s</sub>(t)=B<sub>s</sub>·cos (ωt−π/2)=B<sub>s</sub>·sin(ωt) (trailing) </FORM>and in the anticlockwise direction:x<sub>s</sub>(t)=A<sub>s</sub>·cos (ωt)<FORM>y<sub>s</sub>(t)=B<sub>s</sub>·cos (ωt+π/2)=−B<sub>s</sub>·sin(ωt) (leading). </FORM>
Alternatively (where applicable, also cumulatively) it is possible to perform the measurement of the acceleration signals using what are referred to as shock sensors. Shock sensors already supply a signal that has been differentiated once. By this measure a first mathematical differentiation would be anticipated.
Following the elimination of the mean values, a distinction can be made between the two above cases. The further processing is based on the differentiation of one of the two sensor signals with a subsequent correlation with the other sensor signal. For the case that x,(t) is differentiated a further time, the following result is obtained for the rotation in the clockwise direction:<FORM>x′<sub>s</sub>(t)=−A′<sub>s</sub>·sin (ωt) </FORM><FORM>Korr1 {x′<sub>s</sub>(t), y<sub>s</sub>(t)}=∫ x′<sub>s</sub>(t)·y<sub>s</sub>(t) dt=−A′<sub>s</sub>·B<sub>s </sub>[t/2−sin(2ωt)/(4ω)]. </FORM>
In this case Korr1 {x′<sub>s</sub>(t), y<sub>s</sub>(t)} denotes the correlation of the signal x′<sub>s</sub>(t) with the signal y<sub>s</sub>(t).
And for the rotation in the anticlockwise direction analogously:<FORM>x′<sub>s</sub>(t)=−A′<sub>s</sub>·sin (ωt) </FORM><FORM>Korr1 {x′<sub>s</sub>(t), y<sub>s</sub>(t)}=∫ x′<sub>s</sub>(t)·y<sub>s</sub>(t) dt=A′<sub>s</sub>·B<sub>s </sub>[t/2−sin (2ωt)/(4ω)]. </FORM>
For long time periods t the term t/2 in the square brackets is dominant; then the rotation in the clockwise direction is differentiated from the rotation in the anticlockwise direction by the sign: negative for the clockwise direction and positive for the anticlockwise direction. With an increasing number of measured values the subsequent correlation (Korr1) leads to a more stable end value, since the signal/noise ratio is increased. The signal noise is reduced to the desired order of magnitude at the expense of the decision time or measurement frequency. This results in an extremely small computing overhead and the directionofrotation detection is performed with any quality.
An improved noise suppression is achieved by the forming of a difference between two correlation functions (Korr1; Korr2). In this case y,(t) is differentiated and correlated with x<sub>s</sub>(t). For the rotation in the clockwise direction this yields:<FORM>y′<sub>s</sub>(t)=B′<sub>s</sub>·cos (ωt) </FORM><FORM>Korr2 {x<sub>s</sub>(t), y′<sub>s</sub>(t)}=∫ x<sub>s</sub>(t)·y′<sub>s</sub>(t) dt=A<sub>s</sub>·B′<sub>s </sub>[t/2+sin(2ωt)/(4107 )]. </FORM>
For the rotation in the anticlockwise direction analogously:<FORM>y′<sub>s</sub>(t)=−B′<sub>s</sub>·cos (ωt) </FORM><FORM>Korr2 {x<sub>s</sub>(t), y′<sub>s</sub>(t)}=∫ x<sub>s</sub>(t)·y′<sub>s</sub>(t) dt=−A<sub>s</sub>·B′<sub>s </sub>[t/2+sin(2ωt)/(4ω)]. </FORM>
The combination of the two correlations (Korr1; Korr2) in the form of Korr=Korr1 {x′<sub>s</sub>(t), y<sub>s</sub>(t)}−Korr2 {x<sub>s</sub>(t), y′<sub>s</sub>(t)} then yields a better noise suppression, because an averaging between two independent noise sources (sensor x and sensor y) is performed.
The practical implementation of the abovedescribed method requires at least two measurements, ideally taken in quick succession, by the sensors in order to be able to approximate a differentiation by a difference quotient DQ<sub>x </sub>or DQ<sub>y</sub>; for x, for example:<FORM>DQ<sub>x</sub>=x′<sub>s</sub>(t)≈[x<sub>s</sub>(t)−x<sub>s</sub>(t−ΔT)]/ΔT </FORM>
where ΔT represents the time interval between two measurements. If signals x(t), y(t) of an acceleration sensor are present instead of signals x<sub>s</sub>(t), y<sub>s</sub>(t) of a shock sensor, then the signals x<sub>s</sub>(t), y<sub>s</sub>(t) must first be generated by differentiation in the abovedescribed manner, which signals are then derived in turn. In this case at least three measurements, ideally taken in quick succession, by the acceleration sensors are required in total.
When more than two measurements are performed at short time intervals, a resulting redundancy is used for noise reduction. This can be effected by conventional finite impulse response (FIR) or infinite impulse response (IIR) filtering. An alternative to this is the determination of an approximation function of the measured values, such as, for example, a regression line or a regression polynomial.
The practical calculation of the correlation can be performed by the summation of the terms x′<sub>s</sub>(t<sub>i</sub>)·y<sub>s</sub>(t<sub>i</sub>) at the times t<sub>i</sub>, i.e. the integral is approximated by a sum total. In this case the times t<sub>i </sub>can lie at arbitrary intervals and take place at arbitrary phases within a sinusoidal oscillation, provided they do not always strike a zero passage.
Furthermore an extremely small data transfer from the sensors to the central receiver is advantageous. A calibration of the free parameters is simple.
Preceding this, according to the invention a method for determining the direction of rotation of a rotating body about its axis of rotation by the measurement of two acceleration signals x and y by use of at least one acceleration sensor. The two acceleration signals x and y are at a 90° angle to each other, are oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g are taken into account. The measured acceleration signals x and y are mathematically differentiated. The differentiated x acceleration signal is differentiated a further time and correlated with a y signal that has been differentiated once for the purpose of obtaining a first correlation function.
Alternatively or cumulatively hereto, a method for determining the direction of rotation of a rotating body about its axis of rotation by the measurement of two acceleration signals x and y by at least one acceleration sensor. The two acceleration signals x and y are at a 90° angle to each other, are oriented not parallel to the axis of rotation, and both a centrifugal acceleration generated by a circular movement and a gravitational acceleration g are taken into account. The measured acceleration signals x and y are mathematically differentiated. The differentiated y acceleration signal is differentiated a further time and correlated with an x signal that has been differentiated once for the purpose of obtaining a second correlation function.
According to the invention the difference is preferably formed from the functions produced in this way.
In a preferred embodiment of the invention a differentiation is approximated using a difference quotient.
In a further preferred embodiment of the invention the integrals of the correlation functions are calculated in an approximation using a sum total.
The invention is suitable in particular for determining the direction of rotation of the rotating tire of a motor vehicle.
Other features which are considered as characteristic for the invention are set forth in the appended claims.
Although the invention is illustrated and described herein as embodied in methods for determining the direction of rotation of rotating bodies, it is nevertheless not intended to be limited to the details shown, since various modifications and structural changes may be made therein without departing from the spirit of the invention and within the scope and range of equivalents of the claims.
The construction and method of operation of the invention, however, together with additional objects and advantages thereof will be best understood from the following description of specific embodiments when read in connection with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGSFIG. 1 is a schematic representation of a rotating body with its acceleration components according to the invention;
FIG. 2 is a graph of a time curve of acceleration components for two directions of rotation, ignoring centrifugal forces;
FIG. 3 is a graph of a time curve of acceleration components for one direction of rotation, taking the centrifugal forces into account;
FIG. 4 is a graph of a time curve of the acceleration components x<sub>s</sub>(t) and y<sub>s</sub>(t); and
FIG. 5 is a graph of a time curve of two correlation functions for two different directions of rotation.
DESCRIPTION OF THE PREFERRED EMBODIMENTSReferring now to the figures of the drawing in detail and first, particularly, to FIG. 1 thereof, there is shown a schematic representation of a rotating body 1 about an axis of rotation 2 with its acceleration components in the horizontal x and the vertical y direction. Sensors 3 measure the acceleration components x and y. The acceleration signals x and y are disposed phaseshifted by 90°. In the actual embodiment the x signal is disposed parallel to the horizontal axis at 0° and the y signal is disposed parallel to the vertical axis at 90°.
Alternative arrangements are conceivable. In particular the signals x and y can be disposed at −45° and 45° respectively. This has the advantageous consequence that the centrifugal acceleration acts only partially on the sensor(s) 3 in each case and these can be implemented with a smaller measurement range.
Fundamentally, the method is stable with respect to small angular deviations such as result, for example, due to installation tolerances. Also represented is the direction of the gravitational acceleration g.
FIG. 2 shows a time curve of the acceleration components x and y for two directions of rotation 2, ignoring the centrifugal forces. Given equal amplitudes, the trailing or leading by 90° of the x or y acceleration component respectively, depending on the direction of rotation 2, is shown. Determination of the phases of the sinusoidal oscillations with respect to each other has previously been associated with a high overhead. The individual acceleration signals would have to be sampled very frequently over a period that depends on the rotational speed of the tire or the speed of the vehicle. Based thereon, a phase shift would be determined which includes an explicit determination, calculation or estimation of the signal frequency as well as of the signal amplitude. This method is very computeintensive intensive and requires a great deal of energy, which is unsuited, for example, to a batteryoperated approach such as is used in the wheel electronics. Moreover the method is susceptible with regard to swiftly changing wheel speeds such as result during rapid acceleration or deceleration of the vehicle.
FIG. 3 shows a representation of the acceleration components x and y for a direction of rotation 2, taking the centrifugal forces into account. Both acceleration components, x and y, are propagated as a sinusoidal oscillation parallel to the x axis. In this case the acceleration component in the y direction has a higher mean value than the acceleration component in the x direction and therefore oscillates about a higher zero point. The respective mean values are dependent on the respective rotational speed. Strictly speaking, therefore, the mean values are not constant. However, the changes in the mean values usually take place much more slowly than the inherent changes in the sinusoidal curve. The determination of the phases of the sinusoidal oscillations has previously been associated with the same high overhead as shown in the description referring to FIG. 2.
FIG. 4 shows the time curve of the acceleration components x<sub>s</sub>(t) and y<sub>s</sub>(t). At the points identified by circles, two measured values were recorded in quick succession for the subsequent correlation. In the simplest case only two quickly succeeding measurements of the sensors are necessary, which measurements can be repeated at arbitrary intervals.
FIG. 5 shows the shape of a correlation function (either Korr1 or Korr2) for two different directions of rotation. In this case it can be observed that the two curves diverge from each other as the number of measured values increases. Depending on the start conditions, in this case the curves can be assigned a direction of rotation starting from approximately measured value 15. Thus, a more unequivocal conclusion with regard to the direction of rotation of the rotating body is possible with an increasing number of measured values.
In connection with a motor vehicle the question of side localization is of interest; i.e. on which side a tire is positioned. The determination of whether a wheel is mounted “right” or “left” can be made based on the movement of the vehicle. When the vehicle is traveling in a forward direction, the right wheel rotates in the clockwise direction and the left wheel in the anticlockwise direction. When the vehicle is traveling in reverse, the conditions are precisely reversed. If the movement of the vehicle is unknown, it is assumed that the vehicle travels mainly in a forward direction and the percentage share of travel in reverse is small. The correlation method thus eliminates the influence of the travel in reverse.
Finally, a decision in favor of “right” or “left” is necessary on the basis of the time curve of the correlation Korr1 or Korr2 or corresponding correlation functions (Korr). For this, many methods are conceivable. It could be required, for example, that Korr1 and Korr2 or Korr must have been in the negative range for a specific percentage share, e.g. 98%, in order to decide in favor of “right”. A decision is also possible upon reaching a specific threshold value for Korr1 and Korr2 or Korr. In addition a specific minimum number of processed measured values could be required, e.g. more than 100, in order to stabilize the decision in relation to the signal noise. Furthermore it makes sense to evaluate the correlations Korr1 and Korr2 or Korr only if they are based on more than a specific number of measured quantities, e.g. 20. It is also possible even to filter the correlations (Korr1; Korr2; Korr) again in order to achieve a higher reliability. With all these approaches a balance is drawn, based on the choice of the parameters (98%, 100, 20, time constant of the filter, etc.), between the required measured data volume, and hence the time required to make a decision, and the reliability of the decision.
The present invention is therefore suitable in particular for determining the direction of rotation of the rotating tire of a motor vehicle.
This application claims the priority, under 35 U.S.C. §119, of German application DE 10 2005 018 107.4, filed Apr. 19, 2005; the prior application is herewith incorporated by reference in its entirety.