METHOD FOR DETECTING FAULTS RELATED TO WHEELS OF A MOTOR VEHICLE IN A DRIVING SITUATION

Abstract

The subject matter of the invention is a method for detecting faults related to wheels of a motor vehicle in a driving situation, comprising: a step (S.sub.1) of automatically determining a first series of corrective steering wheel angles applied successively to force the vehicle to follow a path parallel to a first rectilinear portion of a traffic lane; a step (S.sub.2) of automatically detecting, from the corrective steering wheel angles of the first series, the presence of a fault affecting a pair of steering wheels of the vehicle; and, optionally: a step (S.sub.3) of estimating a type of fault associated with the detected fault, a step (S.sub.4) of identifying the steering wheel of the pair of wheels affected by the fault; and a step (S.sub.5) of generating a warning message for the attention of the driver of the motor vehicle, the message including the estimated type of fault and the wheel identified as having the fault.

Claims

1. A method for detecting faults related to wheels of a motor vehicle in a driving situation, comprising: automatically determining a first series of corrective steering wheel angles successively applied during a first time window of driving said motor vehicle on a first portion of a substantially rectilinear traffic lane, said corrective steering wheel angles being applied so that said vehicle follows a path parallel to said first traffic lane portion; automatically detecting the presence of a fault affecting a pair of steered wheels of the motor vehicle, from the corrective steering wheel angles of the first series; estimating a type of fault associated with the fault whose presence has been detected; identifying the steered wheel of said pair of wheels affected by said fault; and generating a warning message for the attention of the driver of the motor vehicle, said message comprising the estimated type of fault and the wheel identified as having the fault.

2. The method as claimed in claim 1, wherein the automatic detection comprises calculating an average and a standard deviation on the corrective steering wheel angles of the first series.

3. The method as claimed in claim 2, wherein the steered wheel affected by said fault is identified from the sign of the average calculated in the automatic detection step.

4. The method as claimed in claim 1, wherein estimating a type of fault comprises calculating time derivatives of the corrective steering wheel angles of said first series, the type of fault being estimated on the basis of a comparison of an average of the time derivatives with respect to zero.

5. The method as claimed in claim 1, further comprising: following a detection of the presence of a fault resulting from the automatic detection, automatically determining a second series of corrective steering wheel angles successively applied during a second time window of driving said motor vehicle on a second traffic lane portion of curved profile, said corrective steering wheel angles being applied so that said vehicle follows a path parallel to said second traffic lane portion; and calculating an average on the corrective steering wheel angles of the second series.

6. The method as claimed in claim 5, wherein the steered wheel affected by said fault is identified from the sign of the average calculated on the corrective angles of the second series.

7. The method as claimed in claim 5, wherein estimating a type of fault comprises calculating the time derivative of the corrective steering wheel angles of said second series, the type of fault being estimated on the basis of a comparison of said time derivative with respect to zero.

8. The method as claimed in claim 1, further comprising: deactivating an on-board driver assistance system capable of commanding a lateral control of the path of the motor vehicle during an activation phase of said on-board driver assistance system.

9. The method as claimed in claim 8, wherein said deactivation is triggered on the basis of the estimated type of fault.

10. The method as claimed in claim 8, wherein the corrective steering wheel angles applied are delivered by said on-board driver assistance system during the activation phase of said on-board driver assistance system.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The invention will be better understood in view of the following description given with reference to the appended figures, in which:

[0030] FIG. 1 shows a functional block diagram of a system on board a motor vehicle according to a possible implementation of a method for detecting faults related to the wheels of the vehicle according to the invention;

[0031] FIG. 2 shows an example of a path followed by a motor vehicle whose wheels have a fault;

[0032] FIG. 3 shows steps that can be implemented according to a first embodiment of a detection method according to the invention;

[0033] FIG. 4 shows steps that can be implemented according to another embodiment of a detection method according to the invention.

DESCRIPTION OF EMBODIMENT(S)

[0034] Possible modes of implementation of a method for detecting faults related to wheels of a motor vehicle according to the invention will now be described, with particular reference to FIG. 1 which represents a functional block diagram, in the nonlimiting context of a motor vehicle 1 equipped with a driver assistance system 2 capable of commanding the lateral control of the path of the motor vehicle.

[0035] To keep the motor vehicle 1 in the traffic lane which it takes, the lateral-control driver assistance system 2 conventionally has one or more sensors 20 whose detections will make it possible to determine certain characteristics of this traffic lane, such as the width of the traffic lane, its orientation relative to the vehicle and its curvature, in order to estimate and follow a model of the traffic lane (typically the path formed by following the center of the traffic lane). The sensors used are, for example, cameras, LIDAR sensors, radars or else virtual sensors, or more generally any type of sensor capable of detecting marking lines on the ground. An example of a virtual sensor would be software that would read data acquired by other sensors of the vehicle, which would determine on this basis the position of objects such as other cars or safety rails, and which would deduce therefrom the position of the traffic lane relative to the vehicle. From the information delivered by the sensor or sensors 20, an electronic module 21 of the lateral control system 2 estimates the marking lines on the ground, possibly by data fusion in cases where several sensors capable of detecting marking lines on the ground are used. The system 2 also has a certain amount of information, grouped under the reference 22, such as the speed and/or the acceleration of the motor vehicle 1, and the position of the vehicle provided by an on-board navigation system, such as a GPS system, and/or an on-board high-definition map. On the basis of all of this information, an electronic lateral-control module 23 generates and transmits, to a system 3 for controlling the steering column of the motor vehicle 1, corrective steering wheel angles to be applied so that the motor vehicle 1 automatically follows a path parallel to the traffic lane, generally the center line of this traffic lane. The system 3 then directly applies the angle corrections to the steering wheel that it receives. In other words, the steering wheel of the vehicle, which controls the steered wheels of the vehicle, is automatically turned according to the corrective angle values that it receives.

[0036] The principle of the invention is based on the fact that the main symptom of a fault related to a steered wheel of a motor vehicle is that the vehicle will tend to drift to the right or to the left. This drift is all the more visible when the motor vehicle is driving on a rectilinear road portion. Under these conditions, the lateral-control assistance system 2, when it is activated, will detect these drifts and thus be caused to generate many more corrective steering wheel angles than it should have done for a vehicle whose steered wheels show no fault. This is illustrated schematically in FIG. 2 which shows several successive positions of the motor vehicle 1 driving on a rectilinear traffic lane portion 6. It is noted in this figure that the motor vehicle 1 drifts twice to the right (as illustrated by the arrows in solid lines) while it is supposed to follow the center 7 of this rectilinear traffic lane portion 6. The electronic lateral-control module 23 will in this case, at each drift detected, deliver a corrective steering wheel angle so that the motor vehicle 1 can be recentered on the road portion (as illustrated by the two arrows in dotted lines).

[0037] The method for detecting faults related to steered wheels of the motor vehicle according to the invention is based on the analysis of the corrective steering wheel angles which will make it possible, as will be detailed later, not only to automatically detect the presence of a fault while the vehicle is in a driving situation, but also to estimate a type of fault and to identify which of the two left or right steered wheels is affected by the fault.

[0038] FIG. 3 illustrates steps of a method for detecting faults related to a right and/or left wheel of a pair of steered wheels of the motor vehicle 1 according to a first possible embodiment according to the invention.

[0039] A first step S.sub.1 of the method consists in automatically determining a first series of corrective steering wheel angles which are applied successively during a first time window of driving the motor vehicle 1 while the latter is traveling on a first substantially rectilinear traffic lane portion, in order to force the motor vehicle 1 to follow a path parallel to this first traffic lane portion.

[0040] As has been seen previously, the corrective angles are here automatically determined by the electronic lateral-control module 23, and can therefore be delivered directly for processing to a module 4 for detecting faults related to the wheels (see FIG. 1). The assistance system 2 is also able, from the information delivered by the electronic module 21 and the information 22 relating to the map and to the speed and/or the acceleration of the vehicle, to identify that the corrective steering wheel angles of this first series are well determined while the motor vehicle is on a rectilinear road portion. A condition prior to triggering step S.sub.1 may be to verify that the motor vehicle 1 is driving on a substantially rectilinear road portion, for example by verifying that the radius of curvature of the road does not exceed a certain threshold of curvature, and that this road portion will remain rectilinear long enough to make it possible to collect a succession of corrective angles. It may be considered for example, to give a concrete idea, that the road portion must be substantially rectilinear over approximately 200 meters if the vehicle is moving at a speed of less than 50 km/h, or over approximately 400 meters if the vehicle is moving at a speed of between 50 and 100 km/h, so as to be able to observe the variations in the corrective steering wheel angle and the frequency of the corrections over a time window of approximately 15 seconds.

[0041] The corrective steering wheel angles of the first series are then processed by the module 4 for detecting faults related to the wheels during a step S.sub.2 for the purpose of automatically detecting the presence of a fault affecting the pair of steered wheels of the motor vehicle. To do this, the module 4 can calculate, during a step S20, the average and the standard deviation which are associated with the corrective steering wheel angles of the first series. If {SW_angle.sub.1, SW_angle.sub.2, . . . SW_angle.sub.N} is used to denote the set of N values of corrective steering wheel angles contained in the first series, then the average SW_angle and the standard deviation a are determined according to the following relationships:

[00001]$\stackrel{\_}{\mathrm{SW}\_\mathrm{angle}}=\frac{{.Math.}_{i=1}^N{\mathrm{SW}\_\mathrm{angle}}_i}{N}\sigma =\left(\sqrt{\frac{{.Math.}_{i=1}^N{\left({\mathrm{SW}}_{{\mathrm{angle}}_i}-\stackrel{\_}{S{W}_{\mathrm{angle}}}\right)}^2}{N}}\right)$

[0042] The presence of a fault is detected automatically by comparing a linear combination of the average SW_angle and the standard deviation a to the known uncertainty c of the assistance system 2. For example, as illustrated by step S21 of FIG. 1, the presence of a fault is automatically detected by the fault detection module 4 if the following relationship is fulfilled:

SW_angle±σ>ε

reflecting the fact that the corrective steering wheel angles deviate on average by a value greater than the known uncertainty of the system.

[0043] If no fault is detected at this stage, provision can be made to reprogram the automatic reiteration of steps S.sub.1 and S.sub.2 at a later date, for example one week later.

[0044] Otherwise, the method continues with a step S.sub.3 of estimating the type of fault associated with the fault whose presence has been detected in step S.sub.2, and with a step S.sub.4 of identifying the steered wheel of the pair of wheels affected by this fault. It should be noted at this stage that although step S.sub.4 is illustrated in FIG. 3 as succeeding step S.sub.3, the order of the two steps S.sub.3 and S.sub.4 can be reversed. These steps S.sub.3, S.sub.4 can alternatively be carried out in parallel.

[0045] To proceed to step S.sub.3 of estimating the type of fault, the fault detection module 4 analyzes the rate of variation of the corrective steering wheel angles determined in step S.sub.1 by calculating for example, during a calculating step S.sub.30, the time derivative dSW.sub.i at each instant of the corrective steering wheel angles of the first series, according to the relationships:

[00002]$dSW=\left\{dS{W}_1,dS{W}_2,.Math.{\mathrm{dSW}}_{N-1}\right\}{\mathrm{dSW}}_i=\frac{{\mathrm{SW}\_\mathrm{angle}}_{i+1}-{\mathrm{SW}\_\mathrm{angle}}_i}{\Delta T}$

[0046] The type of fault can then be estimated on the basis of a comparison (step S.sub.31) of an average dSW of the time derivatives dSW.sub.i with respect to zero. More precisely, if the average dSW of the time derivatives dSW.sub.i is equal to zero (corresponding to a constant rate of variation of the corrective steering wheel angles of the first series), then the fault detection module 4 deduces that the type of fault is a deflation of a tire of one of the steered wheels (left branch of the test performed in step S.sub.31). Otherwise, the fault detection module 4 deduces that the type of fault concerns an alignment problem, and more precisely a negative camber angle (wheel leaning toward the vehicle) for one of the steered wheels (right branch of the test performed in step S.sub.31).

[0047] The steered wheel affected by the fault is identified for its part from the sign of the average SW_angle calculated in step S.sub.2 of automatic detection. More precisely, if the sign of the average SW_angle is positive, then the wheel affected by the fault is the left steered wheel (right branch of the test performed in step S.sub.41). Conversely, if the sign of the average SW_angle is positive, then the wheel affected by the fault is the right steered wheel (left branch of the test performed in step S.sub.41).

[0048] At the end of steps S.sub.3 and S.sub.4, the detection module 4 can generate a warning message (step S.sub.5) for the attention of the driver of the motor vehicle, this message advantageously comprising the type of estimated fault and the wheel identified as having the fault. This message is transmitted for sound and/or visual broadcast on a user interface 5 located in the passenger compartment of the motor vehicle 1 (see FIG. 1).

[0049] The module 4 can also decide, depending on the type of fault detected, to deactivate the lateral-control assistance system 2 as indicated by the dotted command in FIG. 1. In particular, if the type of fault is a deflation of a tire for one of the steered wheels, it may be desirable to deactivate the functionalities offered by the lateral control because a deflation of the tire can falsify the calibration of the sensors 20 used by the system 2, and, consequently, automatic path control. A fault linked to the negative camber angle will, on the other hand, have less impact on the performance of the lateral-control assistance system 2, so that it is not necessary to deactivate the system 2 in this case. As a variant, the lateral control by the system 2 can be deactivated as soon as the presence of a fault linked to a wheel has been detected.

[0050] FIG. 4 illustrates steps of a method for detecting faults linked to a right and/or left wheel of a pair of steered wheels of the motor vehicle 1 according to a second possible embodiment in accordance with the invention.

[0051] The method begins with the same steps S.sub.1 and S.sub.2 as those described with reference to FIG. 3, namely the steps enabling the fault detection module 4 to detect the presence of a fault from the first series of corrective steering wheel angles collected while the motor vehicle 1 is moving on a first substantially rectilinear traffic lane portion.

[0052] The method of FIG. 4 nevertheless differs from that described in FIG. 3 in that it comprises, in the event that, at the end of step S.sub.2, the module 4 has detected the presence of a fault, a second cycle of measurements, carried out this time while the motor vehicle 1 is moving on a second curved traffic lane portion, unlike the first traffic lane portion which was substantially straight. FIG. 4 thus shows, following step S.sub.2, a new step S.sub.6 of automatic determination of a second series of corrective steering wheel angles successively applied during a second time window of driving the motor vehicle 1 while the latter is traveling on a traffic lane portion with a curved profile, the corrective steering wheel angles being applied so that said vehicle follows a path parallel to this second curved traffic lane portion.

[0053] As before, the corrective angles here are automatically determined by the electronic lateral-control module 23, and can therefore be delivered directly for processing to the module 4 for detecting faults linked to the wheels (see FIG. 1). The assistance system 2 is also able, from the information of the electronic module 21 and the information 22 relating to the map and to the speed and/or the acceleration of the vehicle, to identify that the corrective steering wheel angles of this second series are well determined while the motor vehicle is on a curved portion of road. A condition prior to triggering step S.sub.6 may be to verify that the portion of road on which the motor vehicle 1 is traveling is indeed of curved profile. It is for example possible to check that the radius of curvature of this portion of road is greater than a predefined curvature threshold, for example set at 1000 meters, and that this portion of road will remain curved long enough to make it possible to collect a succession of corrective angles.

[0054] The type of fault and the identification of the wheel having the fault are then determined in a manner similar to steps S.sub.3 and S.sub.4 described above, except that the calculations performed by the fault detection module 4 now use the corrective steering wheel angles of the second series.

[0055] More specifically, the module 4 can calculate, during a step S.sub.7, the average of the corrective steering wheel angles of the second series. If {SW_angle.sub.1, SW_angle.sub.2, . . . SW_angle.sub.P} denotes the set of P values of corrective steering wheel angles contained in the second series, then the average SW_angle is determined according to the following relationship:

[00003]$\stackrel{\_}{\mathrm{SW}\_\mathrm{angle}}=\frac{{.Math.}_{i=1}^P{\mathrm{SW}\_\mathrm{angle}}_i}{P}$

[0056] The method continues with a step S.sub.8 of estimation of the type of fault associated with the fault whose presence was detected in step S.sub.2, and by a step S.sub.9 of identification of the steered wheel of the pair of wheels affected by this fault. Here again, although step S.sub.9 is shown in FIG. 3 as succeeding step S.sub.8, the order of the two steps S.sub.8 and S.sub.9 can be reversed. These steps S.sub.8, S.sub.9 can alternatively be carried out in parallel.

[0057] These steps S.sub.8 and S.sub.9 are very similar to steps S.sub.3 and S.sub.4 of the embodiment of FIG. 3.

[0058] Thus, to proceed to step S.sub.8 for estimating the type of fault, the fault detection module 4 analyzes the rate of variation of the corrective steering wheel angles determined in step S.sub.6 by calculating for example, during a step S.sub.80 of calculation, the time derivative dSW.sub.i at each instant of the corrective steering wheel angles of the second series, according to the relationships:

[00004]$dSW=\left\{dS{W}_1,dS{W}_2,.Math.{\mathrm{dSW}}_{P-1}\right\}{\mathrm{dSW}}_i=\frac{{\mathrm{SW}\_\mathrm{angle}}_{i+1}-{\mathrm{SW}\_\mathrm{angle}}_i}{\Delta T}$

[0059] The type of fault can then be estimated on the basis of a comparison (step S.sub.81) of the average dSW of the time derivatives dSW.sub.i relative to zero. More precisely, if the average dSW of the time derivatives dSW.sub.i is equal to zero (corresponding to a constant rate of variation of the corrective steering wheel angles of the second series), then the fault detection module 4 deduces that the type of fault is a deflation of a tire of one of the steered wheels (left branch of the test performed in step S.sub.81). Otherwise, the fault detection module 4 deduces that the type of fault concerns an alignment problem, and more precisely a positive camber angle (wheel leaning toward the outside of the vehicle) for one of the steered wheels (right branch of the test performed in step S.sub.81).

[0060] The steered wheel affected by the fault is identified for its part from the sign of the average SW_angle calculated in step S.sub.7. More precisely, if the sign of the average SW_angle is positive, then the wheel affected by the fault is the left steered wheel (right branch of the test performed in step S.sub.91). Conversely, if the sign of the average SW_angle is positive, then the wheel affected by the fault is the right steered wheel (left branch of the test performed in step S.sub.91).

[0061] At the end of steps S.sub.8 and S.sub.9, there is found, as in the case of FIG. 3, step S.sub.5 during which the detection module 4 will generate a warning message for the attention of the driver of the motor vehicle, this message advantageously comprising the type of fault estimated in step S.sub.8, and the wheel identified in step S.sub.9 as having the fault, this message being transmitted for sound and/or visual broadcast on the user interface 5 located in the passenger compartment of the motor vehicle 1 (see FIG. 1).

[0062] The module 4 can also decide, once the presence of a fault has been detected, or depending on the type of fault detected, to deactivate the lateral-control assistance system 2, as indicated by the dotted command in FIG. 1. In particular, if the type of fault is a deflation of a tire for one of the steered wheels, it may be desirable to deactivate the functionalities offered by the lateral control because a deflation of the tire can falsify the calibration of the sensors 20 used by the system 2, and, consequently, automatic path control.

[0063] The two previous methods have been described as being alternative methods, comprising only steps S.sub.1 and S.sub.2 in common. In a variant that is not shown, provision can also be made to combine these two methods. For example, there can be provision, following the detection in step S.sub.2 of the method shown in FIG. 3, to also initiate in addition steps S.sub.3 and S.sub.4 and, as soon as the motor vehicle 1 drives on a traffic lane portion of curved profile, to carry out steps S6 to S9 described in relation to the method of FIG. 4.

[0064] The two methods whose steps are illustrated in FIGS. 3 and 4, or their combination, have been described in the nonlimiting case where the motor vehicle 1 is equipped with the lateral-control assistance system 2 (case of FIG. 1). The advantage in this case is twofold: on the one hand, it is possible to precisely identify a wheel having a fault in a driving situation of the vehicle by directly using measurements of corrective steering wheel angles used for the lateral control, and, on the other hand, it can be decided, for example according to the type of fault or once the presence of a fault has been detected, to deactivate the functionality linked to the lateral control.

[0065] The principles of the invention can nevertheless be applied even in the case where the motor vehicle is not equipped with such a lateral-control assistance system. In this case, there can be provision that the driver can deliberately enter, for example via the user interface 5, into a diagnostic mode when driving, and that the corrective steering wheel angles used by the fault detection module 4 are the steering wheel angles directly applied to the steering wheel of the vehicle by the driver to recenter his vehicle on a rectilinear portion of a traffic lane, and measured by a steering wheel angle sensor.