Method for the detection of a change in the rolling radius of a tire, and associated monitoring

Abstract

A method for detecting a change in the rolling radius of a tire and a method for monitoring the state of the tire are based on the analysis of signals representing an actual speed of a vehicle on which the tyre is fitted and a speed of rotation of the wheel bearing the vehicle.

Claims

1. A method for determining a change in a rolling radius of a tire fitted on a vehicle, the method comprising the steps: obtaining a first signal representing an actual speed of the vehicle for a first period of time by obtaining geolocation data of the vehicle; obtaining a second signal representing a speed of rotation of the vehicle wheel bearing the tire for the same period of time by reading data transmitted on a CAN bus of the vehicle; selecting and processing data constituting the first and the second signals so as to make them comparable, wherein the step of selecting and processing data comprises: eliminating data representing a speed greater than a predetermined speed threshold; deriving a signal; and eliminating data of the derived signal representing an acceleration less, in absolute terms, than a predetermined acceleration threshold; after the selecting and processing step, calculating a ratio of the first signal and the second signal in the form of a set of data; determining a set of statistical indicators representative of the set of data; and determining a change in the rolling radius of the tire based on the statistical indicators in order to monitor the wear of the tire.

2. The method according to claim 1, wherein the statistical indicators are selected from a number of points, a minimum, a maximum, an average, a standard deviation, a median and a quartile.

3. The method according to claim 1, wherein the step of selecting and processing the data comprises smoothing the signals using a filter kernel.

4. The method according to claim 1, wherein the predetermined speed threshold is equal to 50 km/h.

5. The method according to claim 1, wherein the predetermined acceleration threshold is equal to 0.05 m.Math.s.sup.−2.

6. A method for monitoring a state of a tire comprising the steps: implementing the method according to claim 1 more than once; saving the statistical indicators of the sets of data in a memory; determining a development of the statistical indicators; and determining a need for tire maintenance depending on the development.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) Further objectives and advantages of the invention will become clearly apparent from the following description of a preferred, but non-limiting, embodiment, illustrated by the following figures, in which:

(2) FIG. 1 shows a block diagram of a method for determining a change in rolling radius according to the invention, and

(3) FIGS. 2a and 2b show examples of GPS and CAN signals obtained while a method for determining a change in rolling radius according to the invention is being implemented.

(4) FIG. 3 shows a distribution of statistical data obtained while a method for monitoring the state of a tyre according to the invention is being implemented.

DESCRIPTION OF THE BEST EMBODIMENT OF THE INVENTION

(5) The different steps in a method for determining a change in rolling radius according to the invention will now be described with the aid of FIG. 1 and of FIGS. 2a and 2b.

(6) In block 1, a first signal representing the actual speed of the vehicle is determined, for example via GPS, and a second signal representing the theoretical speed of the wheels is determined, for example by reading the data on the CAN bus. It is emphasized here that, according to the constructors, the information available on the CAN bus differs; thus, it is sometimes not possible to obtain an individual wheel speed, but only an axle speed. In this case, a method according to the invention will make it possible to obtain information about all the tyres fitted on this axle, without being able to distinguish between tyres.

(7) Once these data have been obtained, a step of extracting the relevant data is carried out, in block 11. This extraction consists first of all in selecting the data of the first signal for a predetermined period of time. When the signal has an insufficient number of points over this period, a method according to the invention is then interrupted (block 15). In one example, a period of time equal to one minute will be chosen, and the minimum number of points will be determined to be 50, representing an average of 0.8 points per second.

(8) The data of the second signal are then selected for the same period of time. If the CAN signal does not comprise any points, a method according to the invention is interrupted (block 15).

(9) Once the data have been extracted, the signals are smoothed by a Gaussian kernel. First of all, in block 12, the GPS signal is filtered in the following form:

(10) At each instant t.sub.i of the GPS speed signal (i=1 . . . n), the smoothed value is calculated as an average of the other points:

(11) ${\hat{v}}_{\mathrm{GPS}}\left(t_i\right)=\frac{\underset{j=1}{\overset{n}{.Math.}}K\left(t_i-t_j\right).Math.v_{\mathrm{GPS}}\left(t_j\right)}{\underset{j=1}{\overset{n}{.Math.}}K\left(t_i-t_j\right)},$
Where

(12) $K\left(\mathrm{dt}\right)=e^{-\frac{{\mathrm{dt}}^2}{2{\sigma }^2}}.$
In matrix form, it is possible to simply write:

(13) ${\hat{v}}_{\mathrm{GPS}}=\frac{K_{\mathrm{GPS}}.Math.v_{\mathrm{GPS}}}{K_{\mathrm{GPS}}.Math.1},$
where K.sub.GPS| is the matrix defined by k.sub.i,j=K(t.sub.i−t.sub.j)

(14) The CAN signal is then smoothed by a method similar to the GPS signal, by changing the parameter of scale of the filter, and it is resampled over the scale of time of the GPS signal, in order to obtain comparable data. Specifically, a GPS signal is generally at a frequency of 1 Hz, while a CAN signal is generally at a frequency of between 50 and 100 Hz.

(15) Thus, if the instants of the CAN signal are denoted {tilde over (t)}.sub.1, . . . , {tilde over (m)}, the following calculation is made for each instant t.sub.i of the GPS signal:

(16) ${\hat{v}}_{\mathrm{CAN}}\left(t_i\right)=\frac{\underset{j=1}{\overset{m}{.Math.}}K\left(t_i-{\tilde{t}}_j\right).Math.v_{\mathrm{GPS}}\left({\tilde{t}}_j\right)}{\underset{j=1}{\overset{n}{.Math.}}K\left(t_i-{\tilde{t}}_j\right)}=\frac{K_{\mathrm{CAN}}.Math.v_{\mathrm{CAN}}}{K_{\mathrm{CAN}}.Math.1},$
Where this time, K.sub.CAN is a rectangular matrix (n×m) defined by k.sub.i,j=K(t.sub.i−{tilde over (t)}.sub.j).

(17) Moreover, a derivation of the CAN signal is also carried out, in order to obtain a signal representative of the acceleration of the wheels.

(18) ${\hat{v}}_{\mathrm{CAN}}^{\prime }\left(t_i\right)=\frac{\partial }{\partial t_i}\left(\frac{K_{\mathrm{CAN}}.Math.v_{\mathrm{CAN}}}{K_{\mathrm{CAN}}.Math.1}\right)=\frac{\left(K_{\mathrm{CAN}}^{\prime }.Math.v_{\mathrm{CAN}}\right)\left(K_{\mathrm{CAN}}.Math.1\right)-\left(K_{\mathrm{CAN}}.Math.v_{\mathrm{CAN}}\right)\left(K_{\mathrm{CAN}}^{\prime }.Math.1\right)}{{\left(K_{\mathrm{CAN}}.Math.1\right)}^2},$
where K.sub.CAN′ is the derived matrix defined by k.sub.i,j=K′(t.sub.i−{tilde over (t)}.sub.j).

(19) FIG. 2a shows the CAN and GPS signals before and after this smoothing step. Thus, the upper graph shows four curves. The two top curves represent the unsmoothed GPS speed and the smoothed GPS speed, and the two other curves represent the unsmoothed CAN speed and the smoothed CAN speed. The lower graph shows the smoothed CAN acceleration.

(20) Once the signals have been smoothed, the signals are filtered, in block 14. To this end, the instants at which the GPS speed is greater than a predetermined speed threshold, and the instants at which the acceleration derived from the CAN speed is less, in absolute terms, than a predetermined threshold, are selected.

(21) This filtering step is shown in FIG. 2b. Thus, the upper graph shows an example of smoothed GPS speed and, from among the set of points forming this signal, only those situated above 13.88 m.Math.s.sup.−1, that is to say those situated in the grey area, are selected.

(22) In the same way, the lower graph shows an example of smoothed CAN acceleration and, from among the set of points forming this signal, only those situated, in absolute terms, below 0505 m.Math.s.sup.−2, that is to say those situated in the grey area, are selected.

(23) Next, in block 15, the ratio

(24) $\frac{{\hat{v}}_{\mathrm{GPS}}\left(t_i\right)}{{\hat{v}}_{\mathrm{CAN}}\left(t_i\right)}$
for all the remaining points is calculated, and this set of data is then summarized by several statistical indicators already mentioned in the present application.

(25) These statistical indicators can then be used directly or be saved in a memory for use over a long period, as described above. In this case, a selection will be made, from the set of ratios stored, in order to determine which ones will be used in the evaluation of the state of the tyre. Thus, the decision could be made that only the ratios relating to a number of points greater than a given threshold, for example 5, will be retained, in order to keep only the most representative ratios.

(26) FIG. 3 shows an example of the distribution of these statistical data. Thus, in this figure, the nine diagrams respectively show the ratio Actual speed/Theoretical speed for the tyres as follows:

(27) TABLE-US-00002 No. Wear Load Pressure 1 New +++ ++ 2 New ++ + 3 New + ++ 4 Average +++ + 5 Average ++ ++ 6 Average + + 7 Worn +++ ++ 8 Worn ++ + 9 Worn + ++

(28) Thus, “groups” of distribution, symbolized by the three boxes drawn on these curves, are apparent. Two things are thus observed: The first is that the load and the pressure have a negligible influence on the positioning of the distributions, since they do not mask the effect of wear. The second is that the tyres can be clearly distinguished in accordance with their state of wear, since the first box groups together the new or virtually new tyres, the second box groups together the tyres with an average level of wear, and the third box groups together the worn tyres.