Method for determining interface conditions between tire and ground

Abstract

The invention relates to a method for determining an interface condition between a tyre and a ground surface in a motor vehicle, in particular for determining the occurrence of a slipping phenomenon, said method comprising: determining a plurality of indicators (RST, REST, GRP, GPRT, GPRD, IRR, CTR) of the interface condition between the tyre and the ground surface,-determining, on the basis of the dynamic balance of the motor vehicle, the interface condition between the tyre and the ground surface; calculating respective values of said plurality of indicators (RST, REST, GRP, GPRT, GPRD, IRR, CTR) of interface conditions between the tire and the ground, determining, for each calculated value of said plurality of indicators of interface conditions between the tire and the ground, a probability value of occurrence of a slip phenomenon (PAq (RES), PAq (REST), PAq (GRP), PAq (GRPT), PAq (IRR), PAq (CTR), PAq (GPRD)), determining, for each probability value of occurrence of a slip phenomenon, a weight (K1, K2, K3, K4, K5, K6, K7), and determining a weighted probability value (PAq (RES) * K1, PAq (REST) * K2, PAq (GRP) * K3, PAq (GRPT) * K4, PAq (CTR) * K6, PAq (GPRD) * K7) for each probability value of the occurrence of the slip phenomenon by applying each weight (K1, K2, K3, K4, K5, K6, K7) to the respective probability values (PAq (RES), PAq (REST), PAq (GRP), PAq (GRPT), PAq (IRR), PAq (CTR), PAq (GPRD)) of the occurrence of the slip phenomenon,-determining, for each probability value of the occurrence of the slip phenomenon, a weighted probability value (PAq (RES) * K1, PAq (REST) * K2, PAq (GRP) *

Description

Technical Field

[0001] The present invention relates to a diagnostic method and system for a motor vehicle. In particular, the present invention has been developed with reference to the diagnosis of the interface conditions between a tire and the ground during vehicle movement. Background Art

[0002] There are known a variety of methods and systems for determining the interface conditions between a tire and the ground in a motor vehicle, the vast majority of which are used to control the operation of a system for driving and / or stabilizing a motor vehicle, or for the operation of a system for autonomous or semi-autonomous driving.

[0003] However, all the information obtained from the implementation of such methods and such systems is not sufficiently effective in predicting and contrasting some phenomena such as skidding caused by specific interface conditions between a tire and the ground. In other words, the known methods do not provide any information that helps to predict or contrast such phenomena.

[0004] This drawback may even endanger the effectiveness of the most advanced anti-skid systems (in this regard, the applicant is the holder of various national patent applications, such as 102021000011108, 102021000011111, 102021000011117 or 102014902296915), because it is not possible to control the anti-skid system in this way to obtain specific - and ultimately more effective - intervention measures for the skidding conditions that the vehicle has to cope with, nor is it possible to prepare the anti-skid system for intervention in the case where the conditions encountered by the vehicle indicate a very high probability of skidding.

[0005] On the other hand, for the known methods and systems, it is not feasible to diagnose the interface conditions between a tire and the ground for the purpose of controlling an anti-skid system without resorting to other sensors or devices that are not usually present on a vehicle, and it is difficult to implement due to its cost.

[0006] Finally, the currently known diagnostic methods are based on deductive logic, which is essentially Boolean logic or, in any case, highly discretized. On the one hand, if this logic is sufficiently robust to perturbations of the input data, on the other hand, they are prone to a critical loss of information level, which may endanger the final result.

[0007] Object of the Invention

[0008] The present invention aims to solve the foregoing technical problems. Specifically, the present invention aims to provide a method for determining the interface conditions between a tire and the ground, which method is capable of managing the operation of an anti-slip system without requiring the contribution of other sensors or devices other than those normally present on a motor vehicle. Another object of the present invention is to implement such a method in a robust manner with respect to perturbations of the input data without suffering a loss of information level. Summary of the Invention

[0009] The object of the present invention is achieved by a method having the features described in the appended claims, which claims form part of the technical disclosure provided herein in relation to the present invention. Brief Description of the Drawings

[0010] The present invention will now be described with reference to the accompanying drawings, which are provided by way of non-limiting example only, in which:

[0011] - Figure 1 shows a block diagram of a method according to the present invention,

[0012] - Figures 2 to 6 shows a block diagram related to a preferred embodiment of a first element of the method according to the present invention,

[0013] - Figure 7 and FIG. 8 shows a block diagram related to a preferred embodiment of a second element of the method according to the present invention,

[0014] - Figure 9 and Figure 10 shows a block diagram related to a preferred embodiment of a third element of the method according to the present invention,

[0015] - Figures 11 to 15 shows a logic diagram representing the method according to the present invention. Detailed Description of the Invention

[0016] Figure 1 The reference numeral 1 in the drawings generally denotes a block diagram of a method for determining the interface conditions between a tire and the ground in a motor vehicle according to an embodiment of the present invention, particularly for determining the occurrence of a slipping phenomenon.

[0017] In terms of a complete functional diagram, this means that various embodiments may include Figure 1 and Figure 15 one or more of the functional blocks shown in, and the method according to the present invention is based on a set of input data 2, a real-time calculation stage 4, an intermediate output data set 6, an analysis stage 8 (real-time), and a final output data set 10.

[0018] The functional definitions within each of the aforementioned stages or sets can vary according to the computational requirements (or resources) and / or according to the control requirements for implementing the method in real-time computational stages.

[0019] In embodiments where the strictest computational and / or control requirements must be met, the overall structure is as Figure 1 shown, where the real-time computational stage 4 includes a first computational module 12, a second computational module 14, and a third computational module 16. The first computational module 12 is configured to operate on a database of data of the vehicle powertrain. The second computational model 14 is configured to operate based on general dynamic data of the vehicle. The third computational model 16 is configured to operate based on dynamic data of each wheel of the vehicle. According to the present invention, the computational modules 14 and 16 are optional, i.e., they can be provided to determine a further level of information and further output data of the method according to the present invention, while the module 12 is generally provided in all embodiments because it enables determination of the proximity of the slip phenomenon of the interface condition between the tire and the ground, even if it cannot rely on the further level of information obtained from the modules 14 and 16.

[0020] Figure 2 A block diagram of the computational module 12 is shown, which will be denoted as the "powertrain module" hereinafter for simplicity. In the method according to the present invention, the powertrain module 12 is capable of establishing the proximity to the slip condition by estimating the lift force acting on each wheel of the vehicle and by comparing this lift force with a threshold force value that would cause the vehicle to lift off the ground, i.e., the separation of the contact between the tire and the ground.

[0021] Figure 2 The functional blocks denoted by the reference numerals 18, 20, 22, 24 in

[0022] schematically represent the method steps for performing the aforementioned determination.

[0022] Specifically, according to the present invention, the powertrain module is configured to process the input data set 2 (including data and parameters that are typically available on the CAN network and that do not require auxiliary sensors or devices in addition to those that are typically present on the vehicle), and specifically the subset related to the powertrain, in such a way that:

[0023] - Determine the reference longitudinal acceleration a XPTMDL ( Figure 3 and Figure 4 )—block 18

[0024] - Measure the actual longitudinal acceleration of the vehicle a XCAN

[0025] - Calculate the reference longitudinal acceleration a XPTMDL and the actual longitudinal acceleration a XCANDifference between—Frames 20 and 22

[0026] - Determine an additional resistance value (having a hydrodynamic nature in skidding) at the interface between the tyre and the ground based on said difference, and determine a lift force at the interface between the tyre and the ground based on said additional resistance,

[0027] - Determine a threshold force for the tyre to lift off the ground,

[0028] - Compare said lift force with said threshold force and determine the proximity of the interface conditions between the tyre and the ground to the skidding conditions—Frame 24.

[0029] Reference will now be made to Figures 3 to 6 Describe each of the foregoing steps in detail.

[0030] Figure 3 The wheel W of a motor vehicle is schematically shown facing a water film WF on the ground G, which provides a theoretical reference for the method according to the invention. A point P1 on the tread surface of the wheel W is subjected to the impact of a resultant hydrodynamic force, which includes a component F D of hydrodynamic resistance and a component F L (depending on the value of the hydrodynamic resistance component F D ) of lift.

[0031] Assume that all components of the hydrodynamic resistance act only on one axis (for example, the front axle of the vehicle), and the total resistance component acting on the front axle can be expressed as

[0032] F D,Faxle = m(a XPTMDL – a XCAN )

[0033] That is, it is a function of the difference between the reference longitudinal acceleration a XPTMDL and the actual longitudinal acceleration a XCAN .

[0034] Reference Figure 4 , the global equation of the longitudinal dynamic equilibrium of the vehicle can be expressed in the following form

[0035]

[0036] Wherein:

[0037] is the longitudinal acceleration of the vehicle

[0038] ∑F x_i,j is the sum of the longitudinal forces acting on the right / left wheels (i) of the front / rear axles (j)

[0039] AeroRes is the resultant aerodynamic drag acting on the vehicle

[0040] HydroRes is the synthetic hydrodynamic resistance (due to the interaction of the tyre with the water film WF).

[0041] Fslope is the drag or pulling force due to the ground slope (uphill or downhill). In the adopted sign convention, Fslope is positive when it is a drag force and negative when it is a pulling force.

[0042] On this premise, the unknown component HydroRes can be determined by the difference from the reference case, in which there is no water at the interface between the tyre and the ground, essentially by subtracting the following two equations:

[0043] ma XPTMDL = ∑F x_i,j – AeroRes - Fslope (reference case)

[0044] ma XCAN = ∑F x_i,j – AeroRes – HydroRes - Fslope (actual case with water film WF)

[0045] Therefore:

[0046] m(a XPTMDL – a XCAN ) = HydroRes = F D,Faxle

[0047] Therefore, referring Figure 5 , the vehicle's CAN network provides multiple operating and dynamic parameters, including:

[0048] - Gear engagement

[0049] - Engine speed [rpm]

[0050] - Torque output by the engine [Nm]

[0051] - Braking torque [Nm]

[0052] - Steering angle [°]

[0053] - Lateral acceleration [m / s 2

[0054] - Pitch angle [°]

[0055] It will be observed that such data can come from any data network of the vehicle and not necessarily from the CAN network. Therefore, whenever this description refers to using data present on the CAN network, it means that data can be retrieved from the CAN network or any information network of the vehicle.

[0056] ​Therefore, by using this data, the value of the reference longitudinal acceleration a of the vehicle can be calculated in real time XPTMDL (block 18), and compared with the acceleration a XCAN . The acceleration a XCAN is another piece of data available on the CAN network (or any other data network of the vehicle) to determine the hydrodynamic resistance F D,Faxle (HydroRes) due to the water film WF (block 22).

[0057] Subsequently ( Figure 6 ), by using the value F D,Faxle , the lift component F L can be determined in order to compare it with the threshold force (lift) value, which is necessary to lift the tire off the ground and strongly depends on the static values of the vehicle, such as the weight distribution (distance from the center of gravity) between the axles and the front and rear wheelbases. The relationship between the values F D,Faxle and F L is determined during the calibration process of the method and the associated calculation model.

[0058] Some notes on the calculation:

[0059] In the calculation of the reference longitudinal acceleration, it is preferable to adopt some simplified assumptions because the various parameters considered in the dynamic equilibrium equation (which can be obtained from the Figure 4 charts) may change during driving. For example, the mass of the vehicle and the rolling resistance of individual tires may change as the vehicle moves forward. Generally speaking, the parameters affecting the vehicle can be calibrated during driving. In addition, sensor fusion logic can be implemented to minimize the impact of the physical changes of these factors.

[0060] For example, the mass of the vehicle can be updated in real time and / or at each start-up based on the calculation of the acceleration during low-speed maneuvers. For example, when the engine is started, the first maneuver - almost certainly performed at low speed (leaving the garage or parking lot) - can be used to detect the vehicle acceleration and estimate the mass of the vehicle when it starts again, because the mass may be different from the last known value, for example due to a larger number of passengers and / or more fuel or luggage on board.

[0061] Regarding the calculation of the threshold force value for the tire to lift off the ground, the processing burden is usually likely to be low because under various conditions, the change in vehicle mass may not significantly affect the calculation of the threshold strength value, and thus the weight distribution between the axles can be considered reasonably constant (or at any rate, sufficiently constant for the calculation requirements), the same as the values of the front and rear wheelbases (which are fundamentally dependent on the position of the center of mass). Of course, a more accurate and dynamic mapping of the vehicle's center of mass position and the evolution of the vehicle mass itself leads to a more accurate estimate, which can be adopted as needed and especially when the situation requires it.

[0062] In summary, the following list is a summary of the input data and output data characterizing the preferred embodiment of the powertrain module 12.

[0063] Direct input data

[0064] - Wheel speeds (speeds of the left front wheel, right front wheel, left rear wheel, and right rear wheel) [rpm] or [rad / s]

[0065] - Vehicle forward speed [m / s]

[0066] - Gear engagement [-]

[0067] - Engine speed [rpm] or [rad / s]

[0068] - Driving torque [Nm]

[0069] - Steering angle [°]

[0070] - Braking torque [Nm]

[0071] Indirect input data

[0072] - Lateral resultant force F of the interfacial force between the tire and the ground y [N] - from module 14 (if present), otherwise estimated based on data on the CAN network;

[0073] Required parameters

[0074] - Tire rolling radius [m] - for each wheel;

[0075] - Transmission ratio T_ratio between the engine and the wheels (for front-wheel drive or rear-wheel drive vehicles), or the transmission ratio between the engine and the wheels, derived from the torque distribution between the front and rear axles (for four-wheel drive vehicles);

[0076] - Mass moment of inertia of the engine [kg·m 2

[0077] - Mass moment of inertia of the wheels [kg·m​2 — for each wheel

[0078] - Vehicle mass [kg]

[0079] - Longitudinal aerodynamic drag coefficient C x [-]

[0080] - Frontal area of the vehicle [m 2

[0081] Output data

[0082] - Degree of approaching slip condition [-] (box 24)

[0083] - Indication regarding the type of terrain (box 26).

[0084] Reference Figure 7 、Figure 8、 Figure 9 、 Figure 10 , the block diagrams and operational logics of the computing modules 14 and 15 will now be described. The computing modules 14 and 15 respectively correspond to the vehicle dynamics module (14) and the longitudinal dynamics module of the wheels (16). Such computing modules implement the mapping of the vehicle driving conditions in parallel with module 12, and this mapping is reliable under all conditions. In other words, each computing module 12, 14, 16 has a reliability range that does not cover the entire domain of the vehicle driving conditions but covers its subset. The reliability ranges of domains 12, 14, 16 have overlapping regions and non-overlapping regions. The overlapping regions can be used as a means for consistency control of the determinations performed by each module, and the module that provides the most reliable results in the non-overlapping regions can be used as a reference for vehicle control. The reliability range is affected by the quality of the sensors on the vehicle. The more precise the sensors (e.g., for autonomous driving), the wider the reliability range. In this regard, in embodiments where only module 12 is present or the integration of module 12 with modules 14 and 16 is not provided in the aforementioned manner, if necessary, the control of the vehicle and the on-vehicle anti-slip system is performed using a more cautious intervention logic based only on the calculation results of module 12 in order to mitigate the effects at the boundaries of the reliability range.

[0085] Reference Figure 7 and Figure 8, the vehicle dynamics module 14 again uses the complex information present on the CAN network (or another data network of the vehicle) as a set of input data, in the same manner as module 12.

[0086] Five main operations are performed by module 14, as follows:

[0087] - Obtain data from the CAN network (or from another data network of the vehicle),

[0088] - Process the longitudinal dynamics (box 141) and lateral dynamics (box 142) of the vehicle​

[0089] - Evolution of grip based on vehicle dynamics analysis (block 143),

[0090] - Overall analysis of contact with the ground (block 144),

[0091] - Analysis of terrain regularity based on vibrations along the vehicle z-axis (block 145)

[0092] - Definition of terrain conditions based on the instantaneous value of grip and its evolution over time (block 146).

[0093] Module 14 is configured to process data on the CAN network (or another data network of the vehicle, e.g., from the vehicle inertial platform) to obtain the following estimates:

[0094] - Lateral grip (block 28)

[0095] - Longitudinal grip (block 28)

[0096] - Terrain regularity (block 32)

[0097] - Sideslip (block 30)

[0098] This helps to evaluate the overall grip condition of the vehicle and to make some preliminary assessments of the distribution of forces exchanged at the interface with the ground between the four tires. It should be noted that this assessment is independent of the variables considered in the powertrain module 12 and, therefore, as previously mentioned, module 14 can provide a different perspective and different mapping of the vehicle dynamic state.

[0099] The following list summarizes the input data and output data characterizing the preferred embodiment of the vehicle dynamics module 14.

[0100] Direct input data

[0101] - Steering angle δ [°]

[0102] - Longitudinal acceleration [m / s 2

[0103] - Lateral acceleration [m / s 2

[0104] - Yaw rate r [° / s] or [rad / s]

[0105] - Vertical acceleration [m / s 2

[0106] Indirect input data

[0107] ​​​- None

[0108] Required parameters

[0109] - Vehicle mass m [kg]

[0110] - Mass moment of inertia (polar moment of inertia) I of the vehicle z [kg·m 2

[0111] - Position of the vehicle's center of mass (defined by l f and l r (front track and rear track) and h g (height of the center of mass above the ground))

[0112] - Longitudinal aerodynamic drag coefficient C x [-]

[0113] - Vertical aerodynamic drag coefficient C z [-] (usually very small; it can usually play a role in calculating the vertical forces acting on a motor vehicle and ultimately affect the vertical loads acting on the wheels)

[0114] Output data

[0115] - Lateral grip force [N]

[0116] - Longitudinal grip force [N]

[0117] - Drift angle [°]

[0118] - Terrain regularity [-]

[0119] - Slip

[0120] Under normal grip conditions, the mass variations due to the use of the vehicle are calculated by analyzing data on wheel torque and vehicle acceleration. Regarding the polar moment about axis z, due to its low sensitivity to variations, the provided value can be referred to without the need to update it during driving. If necessary, the value of the polar moment of inertia about axis z can be updated based on the vehicle's mass, which is essentially the only component involved in the calculation of the moment of inertia that may vary during driving. Specifically, an increase or decrease in mass results in an increase or decrease in the polar moment of inertia, respectively. In this regard, the same considerations as for updating the vehicle mass value can be adopted: it can be updated by detecting the acceleration during several reference maneuvers (such as low-speed maneuvers), and it can be updated based on the general dynamic equilibrium equations of the vehicle, which take into account fixed and known parameters (such as front track and rear track) and the values available on the inertial platform.

[0121] Regarding the lateral dynamics of the vehicle, the preferred theoretical premise corresponds to Figure 7A ​The "bicycle" model shown (of course, other computational models are possible and thus the bicycle model must be considered an example). The theoretical reference for calculating the longitudinal dynamics of the vehicle in module 14 is as Figure 7B shown.

[0122] In a preferred embodiment, module 14 operates based on data retrieved from an inertial platform of the vehicle, which provides acceleration components along the x-axis ( longitudinal), along the y-axis ( lateral), and rotational acceleration (or since it corresponds to the time derivative of the yaw rate / yaw velocity ω z which is also denoted as r):

[0123] Given the known mass (m), the wheelbase of the center of mass (l f and l r —front and rear wheelbases) and the polar moment of inertia I of the vehicle z (excluding approximations due to use, as described above), from simple balance to lateral translation and rotation, by decomposing the forces on each wheel along x and y, and by estimating the longitudinal force distribution between the front and rear wheelbases as a function of the vertical forces during acceleration or braking (load transfer), the following forces can be determined with reference to the Figure 7A "bicycle" model:

[0124] Fxf: Longitudinal force on the front axle

[0125] Fxr: Longitudinal force on the rear axle

[0126] Fyf: Lateral force on the front axle

[0127] Fyr: Lateral force on the rear axle.

[0128] Furthermore, given data on load transfer due to rolling, the distribution of such forces between the right and left sides (i.e., on each wheel) can be determined, and these data can likewise be obtained from the inertial platform.

[0129] Given the vertical force Fz (due to mass, aerodynamic load, and longitudinal load transfer due to pitch — which is also known from the inertial platform and depends on the value of Iy (polar moment of inertia about axis y) and ω y (pitch rate), see Figure 7B ), the vertical forces Fzf and Fzr acting on the front and rear axles can be determined, and ultimately the coefficient of friction μ on each wheel of the vehicle.

[0130] By analyzing the differences in the friction coefficients between the front and rear axles, it is possible to infer whether there are skidding conditions, as this phenomenon inherently affects the front wheels (in other low-grip conditions, such as driving on ice, the front and rear friction coefficients should be similar or equal).

[0131] The determination of the forces Fxf, Fxr, Fyf, and Fyr specifically originates from a set of dynamic equilibrium equations that are well-known in the literature and are as follows:

[0132] (General longitudinal and lateral equilibrium)

[0133]

[0134] (Equilibrium during lateral translation)

[0135]

[0136] (Equilibrium during rotation, bicycle model)

[0137]

[0138] (Equilibrium during translation along the x-axis)

[0139] Fxf + Fxr = Fx

[0140] where Fxf, Fxr = functions of (Fzf, Fzr)

[0141] By dividing the lateral grip forces Fyf, Fyr by the vertical forces acting on the axles (depending on the weight distribution on the vehicle), the lateral grip coefficients for each axle and each direction can be estimated

[0142] μfx = Fxf / Fzf (longitudinal grip coefficient on the front axle)

[0143] -μfy = Fyf / Fzf (lateral grip coefficient on the front axle)

[0144] -μrx = Fxr / Fzr (longitudinal grip coefficient on the rear axle)

[0145] -μry = Fyr / Fzr (lateral grip coefficient on the rear axle).

[0146] In the presence of a steering angle δ, the grip coefficients on the front axle are calculated by decomposing the forces Fyf and Fxf along the steering direction, i.e., by recalculating the longitudinal Fxf(δ) and lateral Fyf(δ) components with respect to the middle plane of the steered wheel, resulting in

[0147] Fyf(δ) = Fyf · sen(δ) + Fxf · cos(δ)

[0148] Fxf(δ) = Fxf·cos(δ) - Fyf·sen(δ)

[0149] As already observed in reference module 12, some simplifying assumptions have been made in the calculations because the various parameters involved in the dynamic equilibrium equations may vary during driving, and the dynamic equilibrium equations can be written with reference to the diagrams of Figure 7A 、 Figure 7B For example, the position of the center of mass may vary during driving, and in general, it is necessary to cyclically update the values of the parameters that affect the dynamic equilibrium of the vehicle. Regarding the discretization of the various types of terrain, it is not necessary to accurately know these values.

[0150] Of course, if the computational burden is not a problem and if the output data can be continuously determined, it is necessary to update these vehicle parameters, which may vary during driving according to one or more models available in the literature and currently used for the electronic control of vehicle dynamics.

[0151] The balance of the dynamic equilibrium can be set in order to define the average value of the force exerted by each tire on the ground, which, of course, is obtained by averaging along the weight distribution of the vehicle and depends only on the input values from the inertial platform.

[0152] Once the grip coefficient of each wheel has been determined, it is analyzed according to the Figure 8A and the diagram shown in Figure 8B. In the case of Figure 8A , the calculation corresponds to the analysis of the absolute value of the grip coefficients of the four wheels. In the case of Figure 8B, the calculation corresponds to the analysis of the difference between the grip of the front axle and the rear axle. This also allows for a first analysis of the ground conditions at the same time. For example, if a longitudinal grip coefficient of 1.1 is detected, it is reasonable to rule out the presence of ice on the ground. On the other hand, if the vehicle is about to face a skidding phenomenon, the value calculated for the longitudinal grip coefficient may be confused with the presence of ice on the road. In this regard, Figure 8A the quantitative analysis summarized in is combined with the logical analysis according to Figure 8B. The aim of this analysis is to determine the difference in the longitudinal grip coefficients between the front axle and the rear axle. The theoretical premise directly derives from the physics of the skidding phenomenon: the skidding phenomenon essentially affects the front axle of the vehicle, while the rear axle is only slightly affected because almost all of the water film is wiped away by the passage of the front axle. Due to this consideration, the various types of terrain can be distinguished from each other, and the presence of one of these terrains can also be distinguished from the skidding conditions.

[0153] In addition, the variation of the friction coefficient with speed can be analyzed. Most terrain conditions (dirt roads, asphalt roads, snow, ice) do not cause a significant change in friction with the driving speed. On the contrary, skidding results in a friction similar to that of a wet asphalt road until the skidding speed is reached (i.e., the speed at which the lift is higher than the aforementioned threshold, thus generating the lift of the axle), at which point the friction suddenly drops.

[0154] The last function of this module is to calculate the terrain irregularity, which may also help to distinguish terrain conditions. In this case, the frequency recorded along the z-axis is used as a reference.

[0155] An irregular terrain will exhibit a large signal variance around its average value, while a more regular terrain will exhibit more constant values. The same concept can indicate high-frequency terrain irregularities (dirt, gravel) or low-frequency irregularities (depressions, bumps).

[0156] The irregularity can be measured either directly (acceleration along the z-axis) or indirectly (the component of z included in the x and y measurements).

[0157] Module 16 (or "wheel module") has the function of repeatedly evaluating the grip coefficient of the driving wheels, with the same purpose as module 14 (i.e., distinguishing different terrains according to the grip coefficient), but it performs calculations by means of other parameters available on the vehicle data network (CAN network or other network) in order to improve the reliability level in critical situations.

[0158] Reference Figure 9 and Figure 9A , the wheel module 16 again uses the complex information available on the CAN network (or another network of the vehicle) as a set of input data, in exactly the same way as module 12 and module 15.

[0159] Module 16 performs five main operations, as follows:

[0160] - Obtain data from the CAN network (or from another network of the vehicle)

[0161] - Determine the slip of each wheel

[0162] - Calculate the dynamic balance and longitudinal grip / friction coefficient of each driving wheel

[0163] - Analyze out-of-control

[0164] - Analyze terrain regularity

[0165] - Define terrain conditions.

[0166] Therefore, module 16 is configured to process the data on the CAN network (or another network of the vehicle), thereby obtaining:

[0167] - Slip-based indication that defines the condition of grip loss potentially caused by terrain with a low grip coefficient, in order to send the identification of the cause of out-of-control to other modules 12, 14

[0168] - Indication of the forces acting on the dynamic balance of each wheel: The grip / friction coefficient is calculated based on the powertrain

[0169] - Indication of terrain regularity.

[0170] In summary, and in order to partially anticipate further discussion, the following list summarizes the input and output data characterizing the preferred embodiment of the wheel module 16.

[0171] Direct input data

[0172] Set 1

[0173] - Wheel speeds (front left, front right, rear left, rear right) [rpm] or [rad / s]

[0174] - Vehicle driving speed [m / s]

[0175] Set 2

[0176] - Gear engagement [-]

[0177] - Wheel speeds (front left, front right, rear left, rear right) [rpm] or [rad / s]

[0178] - Driving torque [Nm]

[0179] - Braking torque [Nm]

[0180] Set 3

[0181] Longitudinal acceleration [m / s 2

[0182] Lateral acceleration [m / s 2

[0183] Indirect input data

[0184] - None

[0185] Required parameters

[0186] - Rolling radius of the wheel [m] or [mm];

[0187] - Transmission ratio T_ratio between the engine and the wheels (for front-wheel drive or rear-wheel drive vehicles), or the transmission ratio between the engine and the wheels, derived from the torque distribution ratio between the front axle and the rear axle (for four-wheel drive vehicles);

[0188] ​​- The mass moment of inertia I of the wheel zW [kgm 2 ;

[0189] Output data

[0190] - Longitudinal grip force [N];

[0191] - Longitudinal slip [-];

[0192] - Indication of terrain regularity [-].

[0193] The function of the wheel module 16 is to cooperate with the vehicle dynamics module 14 in calculations related to the longitudinal dynamics of the vehicle, so as to expand the effective range of both, and integrate it with module 12 (which essentially involves longitudinal dynamics) in order to expand the global effective range of the method according to the present invention.

[0194] For example, in the case of a strong braking action, it is difficult to model the behavior of the vehicle brakes. Therefore, the calculation of the grip / friction coefficient based on the analysis of the powertrain unit becomes inconsistent, while the calculation model based on the vehicle inertia platform (such as implemented by means of module 14) shows greater effectiveness.

[0195] On the other hand, there may be a situation where the dynamic balance model of the wheel is very consistent and accurate, because the force distribution on each wheel is not estimated but directly calculated.

[0196] Reference Figure 9C , the slip is calculated based on the input data of set 1 and the models known in the literature

[0197] Slip ij (S ij ) = [(ω ij ·R ij ) / v ij - 1]

[0198] Where:

[0199] ω ij = rotational speed of the right / left wheel (i) of the front / rear axle (j)

[0200] R ij = rolling radius of the right / left wheel (i) of the front / rear axle (j)

[0201] v w _ ij = longitudinal speed of the right / left wheel (i) of the front / rear axle (j), which is equal to the sum of the product of the longitudinal speed v of the vehicle and the yaw rate r and the curvature radius D of the trajectory at the right / left wheel (i) of the front / rear axle (j) ij of the product.

[0202] Figure 9B shows a longitudinal dynamic balance model of the wheel, which participates in the calculation process of the wheel module 16 based on the value of module 12. Regarding Figure 9C , it strictly refers to module 16 for the values used to describe the longitudinal dynamics of the vehicle. The values describing the lateral dynamics are basically related to module 14, such as the value Fy_ij that can be obtained from the values of the lateral grip forces Fyf and Fyr.

[0203] More specifically, the data (set 2) on the vehicle's CAN network (or another network) is used to determine the value of the longitudinal force transmitted by each wheel to the ground.

[0204] The dynamic balance equation is very simple:

[0205] F x_ij = M eng_ij – M brk_ij – I w_ij ω ij ’

[0206] Where:

[0207] F x_ij is the longitudinal force transmitted by the right / left wheel (i) of the front / rear axle (j) to the ground

[0208] M eng_ij is the driving torque acting on the right / left wheel (i) of the front / rear axle (j)

[0209] M brk_ij is the braking torque acting on the right / left wheel (i) of the front / rear axle (j)

[0210] I w_ij is the mass moment of inertia of the right / left wheel (i) of the front / rear axle (j)

[0211] ω ij ' is the angular acceleration value of the right / left wheel (i) of the front / rear axle (j).

[0212] This results in obtaining the friction / grip coefficient μ x_ij of each wheel, which is defined as the ratio F x_ij / F z_ij where F z_ij is the vertical load acting on the right / left wheel (i) of the front / rear axle (j), which can be known from the values of set 3, and this value can estimate the values of longitudinal load transfer and lateral load transfer.

[0213] The wheel dynamics balance equation (F x_ij , μ x_ij ) and the wheel slip (S ij) The output data is used as the input data for the following analysis, as Figure 10 shown.

[0214] In this case, the problem of characterizing the terrain regularity is also based on a process similar to that of module 14. In this case, the reference measurement on which the variance is calculated is not the acceleration along the z-axis, but the wheel speed compared to the vehicle speed. More specifically, for each wheel, the instantaneous angular velocity is detected and compared with the theoretical / expected value in the absence of slip, which is derived from the instantaneous forward speed of the vehicle. In the absence of slip, the detected angular velocity and the theoretical / expected angular velocity are consistent or approximately consistent (e.g., due to signal noise or small instantaneous variations). In the case of slip, a variance of the instantaneous angular velocity with respect to the theoretical / expected angular velocity can be observed. A high variance indicates an irregular terrain because a loose or irregular terrain causes wheel slip due to the highly variable interface conditions between the tire and the ground. In addition, a very irregular terrain causes the movement of the suspension, which results in a slight forward movement and / or backward movement of the wheel relative to the theoretical conditions, thus causing the slip phenomenon and increasing the aforementioned variance.

[0215] Therefore, the type of terrain on which the vehicle is traveling can be characterized. With the typical performance of on-vehicle sensors, the terrain can be distinguished according to the high / medium / low grip / friction of the terrain, wet terrain / snow / ice. In addition, due to the change frequency of slip, an irregular terrain (e.g., dirt / gravel / depressions / bulges / drainage ditches) can be detected.

[0216] Regarding the dynamic balance of the vehicle, a set of output data similar to the vehicle longitudinal dynamics obtained in module 14 can be extracted. As mentioned above, having similar output data sets from different calculation modules can improve the reliability of the system and expand its effective range.

[0217] In addition, the wheel module 16 operates based on the longitudinal component of the wheel, while module 14 is configured to process the forces transmitted to the ground laterally and longitudinally.

[0218] Generally speaking, the combined computational implementation of modules 12, 14, and 16 results in the availability of three sets of output data (referring again to Figure 1 , set 10)

[0219] i) Terrain type

[0220] ii) Grip condition

[0221] iii) Information on the condition approaching slip

[0222] In data sets i)-iii), sets i) and iii) are of the discrete type, while set ii) can be of the continuous type (updating in real time such vehicle parameters that may vary during driving) or of the discrete type. In order to be able to combine the results of the three computing modules 12, 14, 16, logical simplification is preferably employed, which reduces each of sets i)-iii) to a set of the discrete type.

[0223] Reference Figure 11 , the following simplification is performed.

[0224] Output data set of module 12

[0225] PWTMDL_Drag_(box 24): It corresponds to the additional drag force F determined by means of the method according to the invention. D value.

[0226] PWTMDL_Drag_Type(box 26): The value of the additional drag force F determined by means of the method according to the invention. D value is also used for a preliminary estimate of the terrain type on which the vehicle is driving. The classification includes two levels:

[0227] - Level 0: The value of the additional drag force is constant with respect to the vehicle speed. It indicates loose terrain (dirt, gravel) or dry asphalt pavement.

[0228] - Level 1: The value of the additional drag force increases as the vehicle speed increases. This may be a sign of slipping.

[0229] Output data sets of modules 14 and 16

[0230] VEHMDL_GripLevel(box 144, box 28 - module 14, lateral grip), WHEMDL_LongitudinalGrip(box 144, box 28 - module 14, longitudinal grip; box 162, box 32 - module 16, longitudinal grip): The values of the longitudinal and lateral grip forces calculated by modules 14 and 16, which correspond to the values of the grip forces Fxf, Fxr, Fyf, Fyr, and with reference to the respective wheels, correspond to Figure 9B and Figure 9C the values Fx,ij shown in, where i = 1 (front), 2 (rear), j = 1 (left), 2 (right), are combined into a continuous resultant force, which defines an indicator of the grip force applied to the ground.

[0231] Furthermore, the foregoing description clearly shows that the vehicle dynamics module 14 calculates values for longitudinal and lateral grip, and thus it is also suitable for calculating the level of longitudinal grip. Therefore, the calculation of lateral grip is provided by module 14, while the calculation of longitudinal grip is provided by modules 14 and 16. Depending on the reliability of the two signals output separately from each module (depending on different driving conditions), it is possible to select to what extent to rely on the readings of the former or the latter. Only after a reliability analysis can these values be combined.

[0232] Likewise, in this case, the grip values determined by the method according to the invention are also used for a first estimate of the type of terrain on which the vehicle is traveling.

[0233] The classification includes two levels (VEHMDL_GripType):

[0234] - Level 0: The grip value is constant relative to the vehicle speed

[0235] - Level 1: The grip value varies with the vehicle speed.

[0236] In other words, if the detected grip value is even lower than the slip speed (e.g., 55 km / h for smooth tires), then a first determination of low grip on snow and ice may be made, rather than a determination tending towards slip conditions.

[0237] VEHMDL_SideSlip (block 145, block 30 - module 14, sideslip), WHEMDL_Longitudinal Slip (block 164, block 34 - module 16, longitudinal slip): They indicate the measured values of sideslip (tire sideslip / vehicle drift angle) and tire longitudinal slip calculated by modules 14 and 16.

[0238] On the other hand, VEHMDL_RoadRegularityLevel and WHEMDL_RoadRegularity Level (blocks 32, 36) indicate terrain regularity. Furthermore, these signals can be used to identify asphalt road conditions.

[0239] Then the output signals from the estimator ( Figure 11 ) are combined, thereby obtaining the output of the following indicators of the interaction between the tire and the ground (it can be observed that the contributions of models 12, 14, 16 to the following output information definitions are indicated by the corresponding symbols in brackets in Figures 1 - 10 the relevant diagram):

[0240] - An indicator RES of the forward resistance provided by the terrain; the value F D,Faxle (or generally F D ) described with respect to module 12 is a preferred example of the indicator RES

[0241] - Indicator REST of the type of forward resistance provided by the ground,

[0242] - Indicator GRP of the grip formed on the ground; the combined forces of the longitudinal grip Fxf, Fxr and the lateral grip Fyf, Fyr are preferred examples of the indicator GRP,

[0243] - Indicator GRPT of the dependence of grip on forward speed,

[0244] - Indicator GRPD of the grip distribution between the front and rear axles,

[0245] - Indicator IRR of terrain regularity; the calculation of the variance of the acceleration along the vehicle's vertical axis z is a preferred example of the indicator IRR of terrain regularity;

[0246] - Indicator CTR of out-of-control; information including measurements of sideslip (tire sideslip / vehicle drift angle) and tire longitudinal slip calculated by modules 14 and 16 is a preferred example of data that can define the indicator CTR.

[0247] Reference Figure 15 For these indicators, the following holds:

[0248] - The indicator of forward resistance has continuously variable values (i.e., continuous values), preferably included between the values 0 (minimum) and 1 (maximum), which correspond to normalized values

[0249] - The indicator REST of the type of forward resistance has discrete values (i.e., values that change in a discrete manner) representing constant resistance (0) or resistance depending on forward speed (1) respectively,

[0250] - The indicator GRP of the grip applied to the ground has continuously variable values, preferably included between the values 0 (minimum) and 1 (maximum), which correspond to normalized values,

[0251] - The indicator GRPT of the dependence of grip on forward speed has discrete values (i.e., values that change in a discrete manner) representing constant grip (0) or grip depending on forward speed (1) respectively,

[0252] - The indicator GRPD of the grip distribution between the front and rear axles has continuously variable values. Specifically, the (normalized) continuously variable data is included between a minimum value (0) associated with the same grip conditions on the front and rear axles and a maximum value (1) associated with a higher grip condition on the front axle relative to the rear axle,

[0253] - The regularity indicator IRR has continuously variable values, including between the values 0 (minimum) and 1 (maximum), which correspond to normalized values.

[0254] - The out-of-control indicator CTR has continuously variable values, including between the values 0 (minimum) and 1 (maximum), which correspond to normalized values.

[0255] Regarding the regularity indicator IRR, it can be implemented as a direct indicator of regularity, i.e., having a minimum value (0) in the case of high irregularity and a maximum value (1) in the case of high regularity, or conversely implemented as an indicator of irregularity, i.e., having a minimum value (0) in the case of high regularity and a maximum value (1) in the case of high irregularity.

[0256] Thus, the method according to the present invention contemplates using the aforementioned output data as input data for a calculation logic that can identify the terrain type. In this regard, it must be taken into account that the method according to the present invention has hitherto aimed to use traditional vehicle dynamics equations in order to obtain all the information characterizing the contact between the ground and the tire.

[0257] At this stage of the method, the information is used to "fill" the corresponding "data container", which represents the range of interface conditions occurring in the interaction of the tire with different types of terrain, including terrain or road surfaces whose conditions may cause skidding events.

[0258] Therefore, an indication of the terrain type is obtained based on probability calculations.

[0259] Figure 13 and Figure 14 An example of this part of the method according to the present invention is shown. Starting from the output item of the data item RES calculated based on the data item PTMDL_Drag (by module 12) and normalized to values included between 0 and 1, let us assume that the value of the indicator RES is 0.97.

[0260] Then, let us consider two "data containers" intended to specify the correspondence with two different terrains (in this example, the terrain where skidding occurs and the terrain of dry asphalt).

[0261] According to the physical reality of the behavior of the wheel on dry asphalt and on a water film that causes skidding, if RES has a high value (e.g., 0.97, considering the normalization between 0 and 1), it is less likely that the vehicle is traveling on dry asphalt.

[0262] Therefore, as Figure 13As shown, the probability PDa(RES) of the vehicle traveling on a dry asphalt road surface is very low, the normalized resistance data item RES is 0.97, and specifically, PDa(RES) = 0.1. Therefore, starting from the normalized resistance data item RES with a high value, a low probability is obtained. (Symbol: P = probability function; Da = dry asphalt road surface; (Res) = independent variable of the probability function)

[0263] For the indicator RES, the probability function PDa(RES) of traveling on a dry asphalt road surface can be imagined to have a behavior similar to that shown by the curve Da in Figure 12 while the curve Aq represents the process of the probability function of the skidding phenomenon PAq(RES).

[0264] Figure 12 The example of... comes from the assumption that the process of the probability function is linear, but conceptually, it may exhibit any behavior corresponding to the specific relationship between the tire and the ground (asphalt road surface, skidding, ice surface).

[0265] As a contrary example, always referring to Figure 13 , if the normalized resistance value RES is low, for example, as equal to 0.2 in the lower branch of the graph in Figure 13 , then the probability PDa(RES) will have a relatively high value; in other words, a low value of the forward resistance is very likely to indicate traveling on a dry asphalt road surface.

[0266] Figure 14 Supplementary examples are shown, that is, if the value of the normalized resistance RES is 0.97 and 0.2, the probability PAq(RES) is calculated. (Symbol: P = probability function; Aq = skidding; (Res) = independent variable of the probability function).

[0267] In this case, the first value will cause the probability function to have a very high value (a high level of forward resistance may indicate traveling on a water film where skidding occurs or may occur), while the second value will make the probability function have a very low value (a low level of forward resistance is less likely to indicate traveling on a water film where skidding appears or may appear).

[0268] After obtaining the probability that the specific value is associated with the specific terrain, the value is multiplied by a weight K1, where the weight K1 is a function of the indicators REST; GRP; IRR; RES, GRPT, CTR, GRPD, which takes into account the nominal weight of the value on the specific terrain (skid resistance is a major value and thus it will have a relatively high weight K1) and the reliability of the value in this specific case (therefore, based on REST; GRP; IRR; RES; GRPT; CTR; GRPD). For example, if the indicator GRP shows a low value (since it is an indicator of the grip formed on the ground, which represents the instantaneous grip used by the vehicle), the vehicle will be in a stationary state (constant speed). In this case, the powertrain calculation module 12 becomes very reliable. Since the indicator RES is derived from the calculation module 12, the indicator RES must obtain a greater dependence.

[0269] Therefore, assuming that K1 is defined by a value between 0 and 1, this value will be close to the maximum value and thus it will be close to 1 (or the limit 1).

[0270] The same may be true for K5, which "weights" the result of the probability function P(IRR). Also in this case, under the condition of constant speed, the variance of the accelerometer and wheel speed on the vehicle obtains reliability and is associated with a higher weight. Generally speaking, based on the quality of the input and applied calculation model, the coefficients K1 - K7 have the function of increasing the robustness of the estimation of the final result. Regarding reliability, it is envisaged that a higher weight is assigned to the probability values of skidding phenomena occurring above the first threshold (indicating that they are credible), and a lower weight is assigned to the probability values of the occurrence of skidding events below the second threshold (indicating that they are not very credible).

[0271] For example, when the (normalized) continuous grip value GRP is low, the normalized resistance value RES is more reliable, and thus the value of K1 will further increase.

[0272] The result of this operation is to calculate the contribution of each indicator RST, REST, GRP, GPRT, GPRD, IRR, CTR in validating the inference of a given type of terrain.

[0273] For each input item of the data RES, REST, GRP, GRPT, IRR, CTR, GPRD and for each possible terrain configuration, this logical sequence ( Figure 15 ) can be repeated, and the calculated probability levels determine the "winner" among the various inferences made, that is, the most likely terrain type. In this regard, Figure 15Shows the slip probability functions PAq(RES), PAq(REST), PAq(GRP), PAq(GRPT), PAq(IRR), PAq(CTR), PAq(GPRD) associated with each of the input data RES, REST, GRP, GRPT, IRR, CTR, GPRD, the corresponding weights K1, K2, K3, K4, K5, K6, K7, and the weighted probability functions PAq(RES)*K1, PAq(REST)*K2, PAq(GRP)*K3, PAq(GRPT)*K4, PAq(IRR)*K5, PAq(CTR)*K6, PAq(GPRD)*K7, which define the respective contributions combined in the calculation of the final slip probability PAq_OUT.

[0274] Finally, for the latter, a threshold can be defined that enables the final output data to be presented to the user / driver.

[0275] If a strategy of eliminating false alarms is preferred, a very high threshold (e.g., 0.9) can be fixed. If a strategy of tolerating false alarms but having no risk of missed alarms is preferred, the value of the activation threshold can be reduced (e.g.) to 0.6.

[0276] Therefore, the output data obtainable by the method according to the invention includes the following (which is related to the inference of terrain conditions):

[0277] - Slip

[0278] - Partial slip

[0279] - Snow

[0280] - Ice

[0281] - Dirt road

[0282] - Depressions, bumps, drains (generally, concentrated irregularities of the road surface)

[0283] - Paved road

[0284] - Dry asphalt road surface.

[0285] Definition of the probability rules for each terrain, which are determined based on the physics of various phenomena, these phenomena being studied a priori based on the theoretical formalization of the interaction between the tire and the ground, and by using the empirical data detected on various terrains, the relevant probability functions can be made more accurate. The method according to the invention represents a clear improvement over the estimation methods based only on Boolean logic and discrete states. The latter computational model, although envisaging a discrete sampling of the level and thus offering the advantage of robustness against perturbations caused by variations in vehicle parameters (such as mass or tire pressure), is on the other hand also affected by a loss of information levels that are potentially useful for the final determination. As previously mentioned, the method according to the invention makes use of the so-called "sensor fusion" technique, which maximizes the available information level and the robustness against perturbations.

[0286] Of course, without departing from the scope of the invention as defined by the appended claims, the details of the embodiments and the examples can vary sufficiently with respect to what has been previously described and shown.

Claims

1. A method for determining the interface conditions between a tire and the ground in a motor vehicle, particularly for determining the occurrence of a slip phenomenon, the method comprises: determining a plurality of indicators (RST, REST, GRP, GPRT, GPRD, IRR, CTR) of the interface conditions between the tire and the ground, - calculating, based on the dynamic balance of the motor vehicle, the respective values of the plurality of indicators (RST, REST, GRP, GPRT, GPRD, IRR, CTR) of the interface conditions between the tire and the ground, - for each calculated value of the plurality of indicators of the interface conditions between the tire and the ground, determining a probability value (PAq(RES), PAq(REST), PAq(GRP), PAq(GRPT), PAq(IRR), PAq(CTR), PAq(GPRD)) of the occurrence of a slip phenomenon, - determining a weight (K1, K2, K3, K4, K5, K6, K7) for each probability value of the occurrence of a slip phenomenon, and determining a weighted probability value (PAq(RES)*K1, PAq(REST)*K2, PAq(GRP)*K3, PAq(GRPT)*K4, PAq(IRR)*K5, PAq(CTR)*K6, PAq(GPRD)*K7) for each probability value of the occurrence of a slip phenomenon by applying each weight (K1, K2, K3, K4, K5, K6, K7) to the corresponding probability value (PAq(RES), PAq(REST), PAq(GRP), PAq(GRPT), PAq(IRR), PAq(CTR), PAq(GPRD)) of the occurrence of a slip phenomenon, - determining a final probability value (Paq_OUT) of the occurrence of a slip phenomenon by combining the weighted probability values (PAq(RES)*K1, PAq(REST)*K2, PAq(GRP)*K3, PAq(GRPT)*K4, PAq(IRR)*K5, PAq(CTR)*K6, PAq(GPRD)*K7).

2. The method according to claim 1, further comprising emitting a signal of the occurrence of a slip phenomenon when the final probability value (PAq_OUT) of the occurrence of a slip phenomenon is higher than a predetermined threshold.

3. The method according to claim 1, wherein, the determining a weight (K1, K2, K3, K4, K5, K6, K7) for each probability value of the occurrence of a slip phenomenon includes determining the weight for each probability value of the occurrence of a slip phenomenon and making the weight a function of each probability value of the occurrence of a slip phenomenon.

4. The method according to claim 1, wherein, Determining the final probability value (PAq_OUT) of the occurrence of a skidding phenomenon by combining the weighted probability values (PAq(RES)*K1, PAq(REST)*K2, PAq(GRP)*K3, PAq(GRPT)*K4, PAq(IRR)*K5, PAq(CTR)*K6, PAq(GPRD)*K7) includes performing a summation operation on the weighted probability values.

5. The method according to claim 1, wherein, Calculating the corresponding values of the indicators (RST, REST, GRP, GPRT, GPRD, IRR, CTR) of the interface conditions between the tire and the ground further includes performing a normalization operation on them within a reference value range, and the reference value range preferably includes values between 0 and 1.

6. The method according to any one of the preceding claims, wherein, Determining the weights (K1, K2, K3, K4, K5, K6, K7) for each probability value (PAq(RES), PAq(REST), PAq(GRP), PAq(GRPT), PAq(IRR), PAq(CTR), PAq(GPRD)) of the occurrence of a skidding phenomenon includes assigning higher weights to the probability values of the occurrence of a skidding phenomenon above a first threshold and lower weights to the probability values of the occurrence of a skidding phenomenon below a second threshold.

7. The method according to any one of the preceding claims, wherein, The multiple indicators (RST, REST, GRP, GPRT, GPRD, IRR, CTR) of the interface conditions between the tire and the ground include: - An indicator (RES) of the forward resistance provided by the ground, - An indicator (REST) of the type of forward resistance provided by the ground, - An indicator (GRP) of the grip formed on the ground, - An indicator (GRPT) of the dependence of the grip on the forward speed, - An indicator (GRPD) of the grip distribution between the front axle and the rear axle, - An indicator (IRR) of the terrain regularity, - An indicator (CTR) of out-of-control.

8. The method according to claim 7, wherein: - The indicator (RES) of the forward resistance has continuous values, - The indicator (REST) of the type of forward resistance provided by the ground has discrete values representing constant resistance (0) or resistance depending on the forward speed (1) respectively, - The indicator (GRP) of the grip formed on the ground has continuously variable values, - The indicator (GRPT) of the dependence of the grip on the forward speed has discrete values representing constant grip (0) or grip depending on the forward speed (1) respectively, - The indicator (GRPD) of the grip distribution between the front axle and the rear axle has continuously variable values, especially continuously variable values between a minimum value associated with the same grip condition between the front axle and the rear axle and a maximum value associated with a higher grip condition of the front axle relative to the rear axle, - The indicator (IRR) of the ground regularity has continuously variable values, - The indicator (CTR) of out-of-control has continuously variable values.

9. The method according to claim 7 or claim 8, comprising: - Determine the reference longitudinal acceleration (a XPTMDL ) of the vehicle - Measure the actual longitudinal acceleration (a XCAN ) of the vehicle - calculating the difference between the reference longitudinal acceleration and the actual longitudinal acceleration, - Determine an additional resistance (F D , PWTMDL_Drag_Level) at the interface between the tire and the ground based on the difference, and determine a lift force (F D ) at the interface between the tire and the ground based on the additional resistance (F L ), the additional resistance defining an indicator (RES) of the forward resistance - determining a threshold force at which the tire lifts off the ground, - comparing the lift force with the threshold force and determining the proximity of the interface condition between the tire and the ground to the slip condition.

10. The method according to claim 9, wherein, The determination of the additional resistance (F D , PWTMDL_Drag_Level) includes determining that the additional resistance has a dependence on the forward speed (PWTMDL_Drag_Type) of the motor vehicle to define the indicator (REST) of the type of forward resistance provided by the ground.

11. The method according to claim 7 or claim 8, further comprising: - determining the longitudinal grip of the vehicle - determining the lateral grip of the vehicle - defining an indicator (GRP) of the grip formed on the ground based on the combined force of the longitudinal grip and the lateral grip (WHEMDL_LongGrip_Level, VEHMDL_LatGrip_Level).

12. The method according to claim 7 or claim 8, further comprising: - determining a vehicle sideslip indicator (VEHMDL_SideSlip_Level), - determining the longitudinal slip of the vehicle tires (WHEMDL_Slip_Level), - determining an indicator (CTR) of out-of-control based on the vehicle sideslip indicator and the longitudinal slip.

13. The method according to claim 7 or claim 8, comprising determining an indicator (IRR) of the ground regularity by computing the variance of the vertical acceleration of the motor vehicle.

14. The method according to claim 6, further comprising activating an intervention of an anti-slip system on the vehicle.

Images