Method for characterizing the coefficient of friction that can be mobilized by a motor vehicle

Abstract

The invention relates to a method for characterizing the friction coefficient available to a motor vehicle, and the vehicle implementing the method. The method comprises: a step (501) of estimating a drift stiffness and a step (502) of estimating the tire caster of one or more steerable wheels of the motor vehicle; a step (503) of classifying the friction coefficient available to the motor vehicle from a finite set of friction coefficient classes, by implementing a machine learning classification algorithm trained for this purpose, taking as inputs said drift stiffness and tire caster; a step (504) of using the class of friction coefficient available to the vehicle calculated in the preceding step by a vehicle driver assistance system. Figure for the abstract: Fig. 5

Description

Title of the invention: Method for characterizing the coefficient of friction that can be mobilized by a motor vehicle technical field

[0001] The invention relates to the field of motor vehicles in the broadest sense (i.e., passenger cars, vans, trucks, sports cars, electric or otherwise, autonomous or otherwise), and more particularly to a method for characterizing the coefficient of friction available to such a motor vehicle, based on information about its lateral dynamics. The estimated coefficient of friction available can then be used by the various driver assistance systems implemented on the vehicle. Previous technique

[0002] Motor vehicles are now equipped with increasingly sophisticated advanced driver assistance systems (ADAS). These ADAS systems aim, in particular, to improve safety and driving efficiency by acting on the vehicle's various actuators.

[0003] Indeed, modern vehicles can be viewed as multi-actuator systems (or "overactuated systems") in which different actuators can be used to perform the same action, with the aim of improving comfort or safety performance. Each actuator has its operating range, in which it is optimal for achieving the best performance. For example, for lateral vehicle control, differential braking actuators are very effective when the road surface has low grip, because the self-alignment of the wheels (which characterizes the tendency of the wheels to return on their own towards the longitudinal axis of the vehicle) is low. Conversely, steering actuators are more effective when the road surface has high grip, because the self-alignment of the wheels is high.

[0004] In order to use the most suitable actuators for each situation, it is necessary for the ADAS to have as much information as possible regarding the vehicle's operation. One such piece of information is the maximum value of the tire-road friction coefficient that can be mobilized by the vehicle, that is, the representation of the maximum longitudinal or lateral force that the tires can generate on the road without exceeding the limits of adhesion. This coefficient is also called the tire friction ellipse. This mobilizable friction coefficient is a function of the road surface adhesion and intrinsic vehicle characteristics, such as the type of tires. and their condition, but also other factors such as weather conditions, tire pressure, vehicle driving dynamics, etc.

[0005] Knowledge of the mobilizable (or potential, or maximum) tire-road friction coefficient can be used, for example, to select the most suitable actuators for lateral or longitudinal vehicle control, to adjust safety distances, etc.

[0006] In the state of the art, most work aimed at estimating the friction coefficient uses two main approaches: - the so-called "direct" approaches, which require the use of additional sensors, for example acoustic sensors attached to the chassis near the wheels to measure the noise produced by the contact between the tire and the road, or vision sensors (cameras) that capture images of the road for classification purposes; - the so-called "indirect" approaches, which are based on observing the effects of tire-road adhesion on vehicle dynamics. They mainly use information relating to the longitudinal dynamics of the vehicle and wheel slippage.

[0007] Direct approaches have the disadvantage of the additional cost associated with installing dedicated sensors. It should be noted that today, vehicles are increasingly equipped with sensors that classify road surface grip for noise control or ADAS functions. However, these solutions do not take into account vehicle dynamics (i.e., how the vehicle moves on the road and how it responds to the forces and movements acting upon it) and tire condition.

[0008] Indirect approaches mitigate the problem of additional cost by using only the information available on the CAN bus (Controller Area Network, a serial system bus used in modern vehicles to allow the various vehicle controllers and sensors to exchange data). However, their response time is highly dependent on the vehicle's driving dynamics.

[0009] We know, for example, of US patent 7751961 B2, which estimates the coefficient of friction that can be mobilized by a vehicle by intentionally accelerating or decelerating the vehicle, that is to say by sending an additional instruction to the engine. This acceleration or deceleration provides the longitudinal excitations or dynamics necessary to estimate the coefficient of friction. This method is advantageous because the engine is used to generate longitudinal dynamics at a moderate level even in the absence of driver input, particularly in low-dynamic driving conditions. However, the The level of such arousal demand, if not properly controlled, can be intrusive or disruptive for the driver and passengers. This method also leads to an increased computational load on the control laws used to regulate, for example, the speed to a given reference point, particularly when adaptive cruise control (ACC) is activated, and an increase in energy consumption.

[0010] One object of the invention is therefore to define a method for characterizing the tire-road friction coefficient that can be mobilized by the motor vehicle, usable for controlling the vehicle's control laws, according to an indirect approach independent of the longitudinal dynamics of the vehicle (i.e., operating in the absence of strong acceleration or braking). Summary of the invention

[0011] To this end, the present invention describes a method for characterizing the coefficient of friction that can be mobilized by a motor vehicle. The method comprises: - a step to estimate the drift stiffness Cy of the tires of at least one steerable wheel set of the motor vehicle, - a step involving the estimation of the pneumatic tire clearance (TP) of at least one set of steering wheels on the motor vehicle, - a classification step of the friction coefficient that can be mobilized by the motor vehicle, among a finite set of classes of mobilizable friction coefficients, by implementing a machine learning classification algorithm trained for this purpose taking as inputs said drift stiffness Cy of the vehicle's tires and said tire caster tp of the vehicle, - a step of using the class of friction coefficient that can be mobilized by the motor vehicle calculated during the previous step by a motor vehicle driving assistance system.

[0012] According to various embodiments, the step of classifying the friction coefficient that can be mobilized by the motor vehicle may further include providing the machine learning classification algorithm with at least one type of data from among: - data on the nature and condition of a road surface, - meteorological data, - data relating to tires, - data relating to vehicle dynamics.

[0013] Advantageously: - the step of estimating a drift stiffness Cy may include smoothing a calculated drift stiffness value and / or - The step of estimating a pneumatic flush may include smoothing a calculated pneumatic flush value.

[0014] According to a particular embodiment, the step of estimating a drift stiffness Cy is implemented only when the tires of the steering wheel assembly of the motor vehicle considered are in a linear operating zone.

[0015] According to a particular embodiment of the invention, the method further comprises a preliminary step of validating the lateral dynamic conditions of the motor vehicle. For example, the lateral dynamic conditions of the motor vehicle are validated when the lateral dynamics of the motor vehicle are greater than its longitudinal dynamics.

[0016] According to one embodiment of the method according to the invention, the finite set of classes of friction coefficients that can be mobilized by the vehicle comprises two classes, and the machine learning classification algorithm is a logistic regression algorithm.

[0017] According to an embodiment of the method for characterizing the coefficient of friction available to a motor vehicle according to the invention, the step of estimating a drift stiffness Cy of the tires of at least one steerable wheel set of the motor vehicle includes the calculation of Cy such that: C -1^1 ^y- | dp |

[0018] with: - dFy a variation in the lateral force exerted on the tires of the wheel assembly under consideration, - d.0 the variation of the drift angle of the wheel assembly considered.

[0019] According to an embodiment of the method for characterizing the coefficient of friction available to a motor vehicle according to the invention, the step of estimating the pneumatic caster tp of the tires of at least one directional wheel set of the motor vehicle includes the calculation of rP such that: TP~ Fy

[0020] with: - T is the self-alignment torque of the considered wheel assembly, - Fy the lateral force exerted on the tires of the wheel assembly considered.

[0021] According to an embodiment of the method for characterizing the coefficient of friction that can be mobilized by a motor vehicle according to the invention, the supervised learning classification algorithm is configured to provide a probability associated with the class of the mobilizable friction coefficient, said probability being taken into account during step (504) of using the class of mobilizable friction coefficient by a vehicle driving assistance system.

[0022] The invention also relates to a motor vehicle comprising an electronic data processing device configured to implement a method for characterizing the mobilizable friction coefficient according to one of the embodiments of the invention. Brief description of the drawings

[0023] The invention will be better understood and other features, details and advantages will become clearer from the following description, given by way of non-limiting reason, and from the accompanying figures, given by way of example.

[0024] [Fig-1] Fig. 1 represents drift stiffness measurements of a tire on two classes of supports.

[0025] [Fig.2] Figure 2 shows the evolution of the lateral force exerted on a pneumatic depending on the drift angle Æ of the vehicle for two support classes.

[0026] [Fig. 3] Figure 3 represents the tire-road contact of a tire when a steering angle is applied to the wheel assembly.

[0027] [Fig.4] Fig.4 represents the distribution of pneumatic shunt values obtained during rolling tests carried out on two rolling classes.

[0028] [Fig. 5] Figure 5 schematically represents the steps of a process of characterization of the mobilizable friction coefficient of a motor vehicle according to the invention.

[0029] [Fig. 6] Fig. 6 schematically represents steps to carry out learning the classification algorithm used in the characterization process according to the invention. Description of the implementation methods

[0030] In the remainder of the description, and unless otherwise stated, the term "friction coefficient" will be used to refer to the "friction coefficient that can be mobilized by the vehicle".

[0031] The invention relates to a method for characterizing the coefficient of friction available to a motor vehicle. The purpose of the method is not to precisely estimate this coefficient of friction, but to quickly determine a class representative of its value, possibly associated with a confidence level. Any number of classes can be used, for example, two classes of values ​​(high / low coefficient of friction), or three classes of values ​​(high / medium / low coefficient of friction).

[0032] To achieve this, the method according to the invention relies on processing data relating to the lateral dynamics of the vehicle in a supervised machine learning classification algorithm. A classification algorithm is an algorithm that allows elements or data to be classified into predefined categories based on certain common characteristics. These algorithms are commonly used in fields such as image recognition, data analysis, natural language processing, document classification, etc.

[0033] Classification algorithms are used for ADAS, but primarily for identifying contextual elements of the vehicle's environment from camera images. The invention proposes using such a classification algorithm to identify the tire-road friction coefficient from data other than optical flow. One of the difficulties encountered lies in choosing the data used as inputs for these classification algorithms, which must be significant with respect to the classes being sought for the classification results to be relevant.

[0034] The method according to the invention proposes to use a supervised learning classification algorithm to determine the class of the friction coefficient that can be mobilized by the vehicle using at least two parameters representative of its lateral dynamics: - the Cy drift stiffness of the tires on a steering wheel assembly of the vehicle, and - pneumatic tire pressure monitoring (TP) of a set of steering wheels of the vehicle.

[0035] The slip resistance Cy of a tire corresponds to its ability to generate lateral forces that can be applied during driving. It relates the lateral force exerted on a tire to follow a given slip angle P relative to the longitudinal axis of the vehicle to the lateral deformations it undergoes. The higher the slip resistance, the less the tire deforms laterally in turns, and vice versa.

[0036] Figure 1 shows estimates of the slip stiffness Cy of a tire, obtained by driving a vehicle on two classes of surfaces: a class with low grip (such as snow) and a class h with high grip (for example, dry asphalt). These estimates are made by measuring the forces exerted on the tire and the vehicle's slip angle. The results are represented as a density, with the dispersion along the x-axis representing this density.

[0037] Observations from these measurements lead to the conclusion that, during lateral stress, the more the tire adheres to the road surface, the higher the drift stiffness value Cy.

[0038] Figure 2 shows the evolution of the lateral force Fy exerted on a tire as a function of the vehicle's drift angle fi, again for the two support classes and h. These curves reach a plateau, respectively Fl and corresponding to the maximum friction coefficient that can be mobilized by the vehicle on the corresponding support. The drift stiffness Cy corresponds to the absolute value of the slope of the curve relating the force Fy to the drift angle fi.

[0039] This figure highlights the fact that, for mobilizable friction coefficients F2 and F2, with F2 < 2, the drift stiffness Cy1 corresponding to F2 is less than the drift stiffness Cy2 corresponding to F2. There is therefore a link between the drift stiffness Cy and the mobilizable friction coefficient of the vehicle, which is why the invention uses this information for the characterization of the friction coefficient class.

[0040] Pneumatic trail tp (in English “pneumatic trail”) of a tire is a term used in automotive engineering to describe the distance between the center of the tire-road contact area and the point of application of the lateral force.

[0041] Figure 3 represents the tire-road contact 301 of a tire when a steering angle fi is applied to the wheel assembly, fi being the angle between the longitudinal axis of the vehicle 302 and the steering direction of the directional wheels 303. The tire-road contact has a length 304 and a width 305.

[0042] Point B represents the center of contact between the tire and the road surface. It is also the point of application of the normal force acting on the tire. Applying a steering angle results in a lateral force Fy on the tire, which deforms the tire-road contact patch. The center of contact of the force Fy is represented by point A, positioned slightly ahead of point B in the direction of vehicle travel. The distance between points A and B defines the caster angle tp. This caster angle is considered the lever arm of the self-alignment moment TAJ of the wheel.

[0043] When tp decreases towards zero, the tire enters the slip zone and loses its self-aligning power: the force Fy enters the non-linear operating zone of the tire and the self-aligning torque gradually decreases towards zero.

[0044] The self-aligning torque Ta reaches its maximum value before the force Fy. In other words, the peak of the wheel self-aligning torque occurs before the tires enter their non-linear operating zone. The tire caster rP is therefore a good representative of the coefficient of friction that can be mobilized by the vehicle, before the tire enters its non-linear operating zone.

[0045] Figure 4 represents the distribution of the pneumatic caster values ​​obtained during runs carried out on the two running classes and ^2. It can be observed that the value of the pneumatic caster tp of the wheels of a steering wheel set of the vehicle is well representative of the class of the coefficient of friction available to this vehicle.

[0046] The method according to the invention consists of using a classification algorithm to define a class of the friction coefficient that can be mobilized by the vehicle. The inputs to this classification algorithm are values ​​of the vehicle's lateral dynamics that are representative of this friction coefficient, namely at least the drift stiffness Cy of a steering wheel assembly of the vehicle and the tire caster tp of the tires of a steering wheel assembly of the vehicle (potentially the same assembly). An advantage of the invention is that these values ​​are available or directly derivable from vehicle data transmitted on the CAN bus. Therefore, implementing the invention does not require the installation of dedicated sensors.

[0047] Figure 5 schematically represents the steps of a method according to the invention for characterizing the coefficient of friction that can be mobilized by a motor vehicle.

[0048] The method includes a first step 501 of estimating a drift stiffness Cy of the tires of a steering wheel assembly of the vehicle, the drift stiffness Cy corresponding to the slope relating the lateral force exerted on the tires to the drift angle P. A positive force Fy is induced by a negative drift angle. In order to maintain a positive drift stiffness, the slope of Fy with respect to the drift angle is considered in absolute value, such that:

[0049] with: - dFy the variation of the lateral force, measured or estimated for example from the lateral acceleration, mass and yaw rate of the vehicle. The force Fy can be obtained by a simple method, by multiplying the mass of the vehicle by its lateral acceleration, or by more complex methods, for example by using a Kalman filter; - d / 3 the variation in the wheel set drift angle, measured or estimated for example from the steering wheel angle or the yaw rate.

[0050] Advantageously, the calculated drift stiffness estimate Cy can be smoothed, for example by a Kalman filter, a recursive least squares algorithm, or any other similar technique. Such smoothing avoids abrupt jumps in Cy related to point variations that are not representative of a change in adhesion. of tire-road contact, particularly when grip conditions are on the borderline of two classes.

[0051] Advantageously, step 501 of estimating the drift stiffness Cy of the vehicle's tires is only carried out when the drift angle 0 has a small value, i.e., when the tires are in their linear operating range. For example, the estimation of Cy may only be performed when the absolute value of the vehicle's drift angle A is less than 5 degrees. When this condition is not met, the drift stiffness value Cy used for implementing the invention corresponds to the last estimated value, possibly after smoothing of that value.

[0052] The method also includes a second step 502 for estimating the tire caster tp of a steering wheel assembly of the vehicle. This step can be performed before, after, or concurrently with step 501. The wheel assembly considered for calculating the tire caster can be the same as that considered for the drift stiffness, or a different wheel assembly in the case of a four-wheel steering vehicle.

[0053] The pneumatic caster rP can be estimated in different ways. One example of an implementation, which does not require taking into account the mechanical caster (which is a constant related to the chassis), consists of calculating:

[0054] with: - ^al the self-alignment torque of the wheel assembly, measured or estimated for example from information from a driver torque sensor (which measures the rotational force applied by the driver on the steering wheel of a car), the steering assistance torque (which indicates the rotational force applied by the steering wheel on the entire steering column) and a dry friction torque sensor of the front axle of the vehicle (observed on the steering column), - Fy, the lateral force applied to the tires of the wheel assembly, measured on one or more wheels or estimated for example from the lateral acceleration, mass and yaw rate of the vehicle.

[0055] Advantageously, the estimation of the tire caster rP of the vehicle's tires can be smoothed, for example by a Kalman filter, a recursive least squares algorithm or any other estimation technique, to avoid sudden jumps in tp linked to point variations not representative of a change in adhesion at the contact of the tire-road.

[0056] The method according to the invention includes a third step 503 of classifying the friction coefficient that can be mobilized by the vehicle among a finite set of classes of friction coefficients that can be mobilized, carried out by a machine learning classification algorithm previously trained for this purpose, and taking as inputs the values ​​of the lateral dynamics of the vehicle that are representative of the friction coefficient that can be mobilized, namely the drift stiffness Cy calculated during step 401 and the pneumatic caster tp calculated during step 402.

[0057] The machine learning classification algorithm used can, for example, be of the neural network, logistic regression, decision tree, etc. type. In the case of a classification of the mobilizable friction coefficient into only two classes, the machine learning classification algorithm used is advantageously a logistic regression, as it is one of the best methods for binary classification, with a smooth transition approximated by a sigmoid function between the two classes.

[0058] The use of a machine learning classification algorithm makes it possible to rely on a plurality of inputs to determine the class of the friction coefficient that can be mobilized by the vehicle. It takes as inputs at least two values ​​of the vehicle's lateral dynamics that are representative of the mobilizable friction coefficient, called primary indicators, namely the slip stiffness Cy and the tire caster tp of the tires of one or more steerable wheel assemblies of the vehicle. The wheel assembly considered is typically the front wheel assembly of the vehicle, but the invention applies identically by considering the slip stiffness Cy and / or the tire caster tp of the rear wheel assembly when it is a steerable assembly, or by supplementing the measurements taken on the front wheel assembly with measurements taken on the rear wheel assembly.

[0059] The inputs of the classification algorithm may also include secondary indicators provided that they are also representative of this friction coefficient, such as: - data on the nature and condition of the road surface, obtained for example from a camera, - meteorological data, obtained for example by a thermometer, a rain sensor, and / or a weather station, - data relating to tires, such as tire type, pressure, age and / or wear condition, - data relating to vehicle dynamics, such as speed, acceleration, yaw rate and / or slip rate.

[0060] The list of secondary indicators is not exhaustive. These secondary indicators help to improve the accuracy of the classification performed by the supervised learning classification algorithm.

[0061] The method according to the invention finally includes a fourth step 504 of using the class of available friction coefficient calculated in step 503 by a vehicle driver assistance system, for example for an ADAS system mobilizing the braking, acceleration, or steering actuators for trajectory tracking or lane centering, for adaptive ACC to adjust safety distances from the vehicle ahead, for collision detection. It may also be an engine control system that adjusts the torque setting or the maximum engine power demand according to the grip class, etc.

[0062] The classification algorithm is trained offline (i.e., outside the vehicle) and in a supervised manner. This involves providing the classification algorithm with a set of known input and output data, called training data, so that it can learn to associate the inputs with the corresponding outputs by creating a model or a mathematical function usable during the inference phase. The training data comes from measurements taken on a test vehicle or from a driving database.

[0063] Figure 6 schematically represents the steps for learning the classification algorithm used in the characterization process according to the invention. The learning process considers input / output data of the algorithm of the same type as those of the inference phase of classification step 503.

[0064] The learning process includes the following steps: - Collection of 601 input samples, including at least one measurement of the drift stiffness Cy of the tires of a steering wheel assembly of the vehicle and of the tire caster tp of the tires of a steering wheel assembly of the vehicle. The collected samples must be sufficiently numerous and representative of various application cases to allow the classification algorithm to offer satisfactory performance regardless of its implementation conditions. For example, a minimum of one million sets of measurements taken under various conditions of tire-road contact adhesion, weather conditions, tire condition, road surface condition and vehicle dynamics will be used; - association 602 of a class of friction coefficients that can be mobilized by the vehicle to each set of measurements. This association (or labeling) can be done by an operator considering groups of measurement sets, and consists of to assign a class from among the N desired output classes to each set of measurements; - Training 603 of the supervised machine learning classification algorithm. This step consists of providing it with the datasets collected during step 601 and the corresponding classes associated during step 602, which constitute the training data, so that it learns to associate the datasets with the different classes sought. - Validation of the trained classification algorithm (604) is performed by introducing new input data into the model and verifying whether the algorithm's outputs correspond to the corresponding classes. A prediction accuracy rate for each input class can then be measured. If this accuracy rate exceeds a given threshold for each class, for example, 95% accuracy, then the trained algorithm can be deployed for implementation. Otherwise, the algorithm's accuracy is insufficient, and training must be restarted with new training data.

[0065] The trained classification algorithm can be used to implement the method for characterizing the friction coefficient available to the motor vehicle according to the invention on all vehicles in the range of the vehicle used to collect the training data, as well as on comparable vehicles. Comparable means that the characteristics of the vehicles, in particular their mass, dimensions, and tire size, are of the same order of magnitude.

[0066] According to a particular embodiment of the method according to the invention, the supervised learning classification algorithm implemented in step 503 of classifying the vehicle's available friction coefficient among a finite set of available friction coefficient classes is configured to provide a probability associated with each classification. This probability provides an indication of the reliability of the determination, which can be taken into account to adapt step 504 of using the calculated available friction coefficient class by a vehicle driver assistance system.

[0067] According to a particular embodiment of the method for characterizing the coefficient of friction available to the motor vehicle according to the invention, the steps of the method are only carried out when minimum lateral dynamic conditions are met. Indeed, to be accurate, the estimates of the drift stiffness Cy and the tire caster tp of the tires of one or more steerable wheels of the motor vehicle, carried out during steps 501 and 502, require the presence of a significant lateral force Fy. Thus, this particular embodiment of the method according to the invention includes a preliminary step 505 for validating the lateral dynamic conditions of the motor vehicle. This validation can be performed in various ways, for example, using the longitudinal acceleration signals ax and lateral acceleration signals ay of the vehicle, by taking: Lateral dynamics (t) - ■fâ+aj

[0068] In this embodiment, steps 501 to 504 of the process are only carried out when Lateral Dynamics^ is greater than or equal to a predefined threshold, for example 0.5 or 0.9. Setting the threshold at 0.5 ensures that the estimates of the drift stiffness Cy and the pneumatic caster tp are carried out under conditions where the lateral dynamics of the vehicle are more stressed than the vertical dynamics.

[0069] In another embodiment, steps 501 and 502 are performed only when the vehicle is under conditions where sufficient lateral dynamics are exerted, i.e., when Lateral Dynamics(t) is greater than or equal to the predefined threshold. When this is not the case, the outputs of these two estimation steps correspond to the latest values ​​of drift stiffness Cy and tire caster tp calculated under satisfactory lateral dynamic conditions.

[0070] The invention relates to a method for characterizing the coefficient of friction available to a motor vehicle, implementing a supervised machine learning classification algorithm that takes as minimum inputs an estimate of the drift stiffness Cy and an estimate of the tire caster tp of one or more steerable wheels of the motor vehicle. This method is intended to be implemented on board a vehicle to enable it, for example at regular time intervals, to perform an estimation of this coefficient of friction available to the vehicle.

[0071] When steps 501 and 502 include smoothing of drift stiffness measurements Cy and / or pneumatic flushing tp, they can be carried out at higher frequencies than step 503 of classification, in order to increase the number of readings to converge towards very accurate measurements.

[0072] The rate at which step 504 of using the class of friction coefficient mobilizable by the motor vehicle by a driving assistance system is carried out is independent of the rate of the preceding steps.

[0073] The main advantages of the classification method according to the invention are: - it allows for the rapid detection of intervals of the mobilizable friction coefficient, based on the lateral dynamics data of the vehicle, - it can be implemented using signals available on the vehicle's CAN bus: therefore, it is not necessary to install specific sensors, - It allows for improved safety for the vehicle, whether it is an autonomous vehicle or a vehicle equipped with driver assistance systems, - It allows for the selection of the most suitable motor vehicle actuators for the context, and for better management of the driving power of these actuators, thanks to knowledge of the class of mobilizable friction coefficient.

[0074] The invention also relates to a vehicle comprising an electronic data processing device, for example a computer or a microprocessor, configured to implement the method of characterizing the coefficient of friction that can be mobilized by a motor vehicle described above, from data acquired by sensors of the vehicle.

Claims

Demands

1. A method for characterizing the friction coefficient available to a motor vehicle, the method comprising: - a step (501) of estimating the drift stiffness Cy of the tires of at least one steerable wheel set of the motor vehicle, - a step (502) of estimating the tire caster tp of the tires of at least one steerable wheel set of the motor vehicle, - a step (503) of classifying the friction coefficient available to the motor vehicle, among a finite set of classes of available friction coefficients, by implementing a machine learning classification algorithm trained for this purpose taking as inputs said drift stiffness Cy of the vehicle's tires and said tire caster tp of the vehicle,- a step (504) of using the class of friction coefficient available to the motor vehicle, calculated in the previous step by a motor vehicle driving assistance system.

2. A method for characterizing the coefficient of friction available to a motor vehicle according to claim 1, wherein the step (503) of classifying the coefficient of friction available to the motor vehicle further comprises providing the machine learning classification algorithm with at least one type of data from among: - data on the nature and condition of a road surface, - meteorological data, - data relating to tires, - data relating to vehicle dynamics.

3. A method for characterizing the coefficient of friction available to a motor vehicle according to any one of the preceding claims, wherein the step (501) of estimating a drift stiffness Cy includes smoothing a calculated drift stiffness value and / or Step (502) of estimating a pneumatic flush includes smoothing a calculated pneumatic flush value.

4. Method for characterizing the coefficient of friction that can be mobilized by a motor vehicle according to any one of the preceding claims, wherein the step (501) of estimating a drift stiffness Cy is implemented only when the tires of the steering wheel assembly of the motor vehicle in question are in a linear operating zone.

5. A method for characterizing the coefficient of friction available to a motor vehicle according to any one of the preceding claims, further comprising a preliminary step (505) of validating the lateral dynamic conditions of the motor vehicle, the steps (501, 502, 503 and 504) of estimating the drift stiffness Cy of the tires, estimating the tire caster tp of the tires, classifying the coefficient of friction available to the motor vehicle and using the class of the coefficient of friction available to the motor vehicle being carried out only when the lateral dynamic conditions are validated.

6. A method for characterizing the coefficient of friction that can be mobilized by a motor vehicle according to the preceding claim, wherein the lateral dynamic conditions of the motor vehicle are validated when the lateral dynamics of the motor vehicle are greater than its longitudinal dynamics.

7. A method for characterizing the mobilizable friction coefficient of a motor vehicle according to any one of the preceding claims, wherein the finite set of classes of mobilizable friction coefficients of the vehicle comprises two classes, and wherein the machine learning classification algorithm is a logistic regression algorithm.

8. A method for characterizing the coefficient of friction available to a motor vehicle according to any one of the preceding claims, wherein step (501) of estimating the drift stiffness Cy of the tires of at least one steerable wheel set of the motor vehicle comprises calculating Cy such that: with : - dFy a variation of a lateral force exerted on the tires of the wheel assembly considered, - dp a variation of the drift angle of the wheel assembly considered.

9. A method for characterizing the coefficient of friction available to a motor vehicle according to one of the preceding claims, wherein the step (502) of estimating a pneumatic caster tp of the tires of at least one directional wheel set of the motor vehicle includes the calculation of tp such that: TP~ F y with: - al a self-alignment torque of the wheel set considered, - Fy a lateral force exerted on the tires of the wheel set considered.

10. A method for characterizing the mobilizable friction coefficient of a motor vehicle according to any one of the preceding claims, wherein the supervised learning classification algorithm is configured to provide a probability associated with the class of the mobilizable friction coefficient, said probability being taken into account during step (504) of using the class of mobilizable friction coefficient by a vehicle driving assistance system.

11. Motor vehicle comprising an electronic data processing device configured to implement a method for characterizing the mobilizable friction coefficient according to any one of claims 1 to 10.

Images