FR3147977A1 - Method for controlling a motor vehicle comprising several axles equipped with independent powertrains, and corresponding motor vehicle - Google Patents

Abstract

The invention relates to a method for controlling a motor vehicle comprising a first powertrain having a reaction time greater than that of a second, the method comprising steps of: - acquiring a general request relating to a total torque that the powertrains must transmit to the wheels, - distributing said general request into specific instructions each assigned to one of the powertrains, and - controlling the powertrains according to said specific instructions. According to the invention, in the distribution step, when the specific instruction assigned to the first powertrain varies, it is provided to: - determine the share (CNF) of the specific instruction not carried out by the first powertrain due to its reaction time, - increase the specific instruction assigned to the second powertrain by at least part of said share. Figure for the abstract: Fig.5

Description

Method for controlling a motor vehicle comprising several axles equipped with independent powertrains, and corresponding motor vehicle Technical field of the invention

The present invention relates generally to the control of motor vehicle powertrains.

It relates more particularly to a motor vehicle comprising several independent axles and each equipped with a powertrain and rolling elements (typically wheels), a first of the powertrains having a reaction time greater than that of a second of the powertrains.

The invention relates in practice to a method of piloting such a vehicle implemented by an on-board computer, comprising steps of:
- acquisition of a general query relating to a total torque that the powertrains must transmit to the rolling elements,
- distribution of said general request into specific instructions each assigned to one of the powertrains, and
- control of the powertrains according to said specific instructions.

State of the art

A motor vehicle typically has at least two axles and at least two wheels per axle. Typically, a car has four wheels distributed across two axles.

Such a car can have two or four-wheel drive.

In the case of four-wheel drive cars, it was common to use a drive shaft between the two axles in order to couple them to a single internal combustion engine.

Now, with the advent of hybrid or electric vehicles, another solution is to decouple the two axles and provide a powertrain on each axle.

In a hybrid vehicle, one of these powertrains includes an electric machine while the other includes at least an internal combustion engine. This internal combustion engine can sometimes also be associated with an electric traction or pleasure machine, making it itself a hybrid.

In an electric vehicle, both powertrains include an electric traction machine. In this case, it is possible to use electric machines with different properties, but these often share common storage batteries for both machines.

In all cases, the difficulty lies in controlling the different powertrains in order to ensure good vehicle stability and adequate distribution of torque (engine or braking) between the two axles.

In practice, a control law is then used to ensure this distribution of torque between the different traction machines used.

However, such a control law is not optimal, particularly from the point of view of the responsiveness of the powertrains.

Indeed, the natural reactivity of an internal combustion engine is much lower than that of an electric traction machine. In addition, the electric machine is connected to the wheels in direct drive without an intermediate sliding member while the internal combustion engine requires a clutch system to transmit its torque to the gearbox. This clutch system is also a source of delay in the reactivity of torque applied to the wheel since part of the engine torque produced is dissipated in sliding in the clutch to adjust the engine and wheel speeds.

Presentation of the invention

In order to overcome the aforementioned drawback of the state of the art, the present invention proposes to take advantage of the different properties of the traction machines to increase the responsiveness of the vehicle.

More particularly, the invention proposes a control method as defined in the introduction, in which, at the distribution stage, when the particular instruction assigned to the first powertrain (the one with the longest reaction time) varies, it is provided to:
- determine the portion of the specific instruction not carried out by the first power unit due to its reaction time, and
- increase the specific instruction assigned to the second powertrain by at least part of said share.

Thus, thanks to the invention, when a first of the traction machines is slower than a second, as soon as the torque that this first traction machine must exert varies significantly, it is possible to compensate for the torque that this first machine is not directly able to exert due to its significant reaction time by using the second traction machine (which is faster).

Other advantageous and non-limiting characteristics of the method according to the invention, taken individually or in all technically possible combinations, are as follows:
- the specific instruction assigned to the second powertrain is increased by the entirety of said share;
- said share is determined based on an inertia of the first powertrain;
- said inertia has a predetermined fixed value;
- said inertia has a variable value, determined according to configuration data of the first powertrain;
- said share is determined based on an acceleration undergone by a moving part of the first powertrain;
- additional steps are planned for:
¤ acquisition of a maximum traction potential and/or a minimum deceleration potential that at least one of the powertrains can transmit to the rolling elements,
¤ selection of a driving mode from several predetermined driving modes,
¤ determination of a preventive value for the first of the specific instructions, making it possible to avoid a priori the sliding of the rolling elements of the motor vehicle on the ground,
¤ prioritization of a rolling element traction criterion or a powertrain energy consumption criterion, based at least on the selected driving mode,
¤ definition of a range of values that can be used for the first specific instruction depending on the preventive value if the traction criterion is prioritized, and depending on said maximum traction potential and/or minimum deceleration potential if the energy consumption criterion is prioritized, and
¤ definition of a range of values usable for the first specific instruction depending on the driving mode, the preventive value and said maximum traction potential and/or said minimum deceleration potential, and
wherein the particular instructions are determined so that the value of the first particular instruction is included in said range of values;
- the said share is added to the preventive value.

The invention also proposes a motor vehicle as defined in the introduction, the computer of which is programmed to implement a control method as mentioned above.

Preferably, only the first powertrain has a mechanical clutch.

Of course, the various features, variants and embodiments of the invention may be combined with each other in various combinations to the extent that they are not incompatible or mutually exclusive.

Detailed description of the invention

The description which follows with reference to the attached drawings, given as non-limiting examples, will make it clear what the invention consists of and how it can be implemented.

On the attached drawings:

is a schematic top view of a motor vehicle according to the invention;

is a schematic view of the different stages of a process for controlling the powertrains of the motor vehicle of the ;

is a first diagram illustrating conditions for determining which strategy to favor for implementing the process represented in the ;

is a second diagram illustrating a method of calculating a preventive torque useful for implementing the process shown in the ;

is a third diagram illustrating an example of correction of the preventive torque of the ;

is a fourth diagram illustrating the calculation of a curative torque useful for the implementation of the process represented in the ;

is a fifth diagram illustrating another method of calculating a preventive torque useful for implementing the process shown in the ;

is a sixth diagram illustrating the determination of a range of usable torques useful for implementing the method shown in the ;

is a seventh diagram illustrating the determination of the torque instructions to be transmitted to the powertrains of the motor vehicle of the .

On the , a motor vehicle 1 is shown.

This could be any type of rolling vehicle (car, truck, bus, etc.).

This motor vehicle 1 typically comprises a chassis and rolling elements (wheels, tracks, etc.) distributed over several axles.

Very generally, this motor vehicle 1 could have any number of axles greater than or equal to two.

We would then call "front axle" any assembly of rolling elements and axle placed in front of the center of gravity CG of the motor vehicle 1 (considering the direction of travel of the vehicle in forward gear) and "rear axle" any assembly placed behind this center of gravity CG.

In the following, we will only consider motorized trains, i.e. trains whose rolling elements are capable of being driven into motion by a traction machine. We will also consider that the trains are all decoupled, which means that no mechanical transmission shaft connects one of the axles considered to the other.

Each powered axle train is equipped with its own powertrain (which does not prevent these groups from using the same energy source).

Typically, such a powertrain could be purely thermal, or purely electric or hybrid.

A thermal powertrain is understood to mean an assembly comprising at least one internal combustion engine coupled to a drive train comprising, for example, a gearbox (automatic or manual or e-tech type, etc.), possibly a clutch, and a differential.

An electric powertrain is understood to mean an assembly comprising at least one electric machine (which may have a motor function and, preferably also, an alternator function) coupled to a possible speed reducer and a differential.

A hybrid powertrain is understood to mean an assembly comprising at least one internal combustion engine and at least one electric machine, coupled to a drive train as mentioned above.

It should be noted that in the following, the term “traction machine” can designate both an internal combustion engine and an electric machine.

In the example shown in the and which will be considered more precisely in the following for simplification, the motor vehicle 1 comprises exactly two motorized axles 10, 20, including a front axle 10 equipped with two front wheels 12 and a front powertrain 11, and a rear axle 20 equipped with two rear wheels 22 and a rear powertrain 21.

As an illustration, the front powertrain 11 may be of the hybrid type (with a thermal engine coupled to a complementary electric traction machine) while the rear one may be electric. The source of electric current may be either a high-voltage battery (at 400 or 800 V for example), or a low-voltage battery (at 48 V for example). In all cases, it should be noted that here, the front powertrain 11 remains the one with the greatest cumulative torque capacity and the greatest autonomy since it is hybrid. It will therefore be considered the main train. The rear powertrain 21 is lower in deliverable torque and its autonomy depends directly on the size of the storage battery. It will therefore be considered the secondary train.

Alternatively, the front powertrain 11 could be of the thermal type, while the rear powertrain could be electric, with a low-voltage battery (48V) of low capacity. Still alternatively, the front powertrain 11 could include an electric machine powered by a higher-voltage battery (400V for example) and of greater capacity than the low-voltage battery. In these two variants, the rear powertrain 21 could then play the role of alternator-starter and motor for short periods. The front axle would therefore be the main axle.

More generally, the main train will be the one whose powertrain has the greatest torque capacity and the greatest autonomy (all tanks full, i.e. when the electricity and/or fuel storage means are full).

If the front axle 10 is equipped with an internal combustion engine, a gearbox and a clutch, the transition from zero torque to non-zero torque requires time. In other words, the reaction time of the front powertrain is greater than that of the rear powertrain. More generally, the transition from one torque value to another is slow, so that this front axle 10 can be described as “slow”. On the contrary, the rear axle 20 can be described as a “fast” axle.

Of course, the configuration illustrated on the is given as an example only, and alternatively the vehicle could be configured differently.

The motor vehicle 1 also includes bodywork and glazing elements which, with the chassis, delimit a passenger compartment capable of accommodating at least one occupant.

In this cabin, the vehicle has a driving position equipped with human-machine interfaces.

Typically, the vehicle has a steering wheel.

It also includes means for selecting one driving mode from several. These means are manual and may, for example, take the form of a wheel 8, or a touch screen controlled to display a corresponding menu.

Here, these means allow you to select driving modes from the following:
- economic,
- comfort,
- sporty,
- snow,
- degraded land,
- mud,
- sand,
- off road,
- off-road.

Of course, as an alternative or in addition, other driving modes could have been used.

Alternatively, the driving mode could be selected not manually, but automatically (e.g. based on the vehicle's geolocation, driver behavior, weather, images of the environment, etc.).

The vehicle also includes an electronic and/or computer unit for data processing. This will be a computer 9 which includes a processor, a memory and various input and output interfaces.

Thanks to its input interfaces, the calculator 9 is adapted to receive data from the thumbwheel 8 and from sensors 7 embedded in the vehicle, making it possible, for example, to measure the speed of the motor vehicle 1, the steering angle of the steering wheel or the steered wheels, the longitudinal or lateral acceleration undergone by the vehicle, etc.

Thanks to its output interfaces, the calculator 9 is adapted to control the different powertrains 11, 21.

Thanks to its memory, the computer 9 stores a computer application, consisting of computer programs comprising instructions whose execution by the processor allows the computer to implement the method described below.

Before describing in detail the process implemented by the calculator 9, we can define some concepts which will be useful in the rest of this process.

The “traction” of the vehicle will be defined as the capacity of the axles to ensure the acceleration or braking of the motor vehicle 1 without their tires slipping on the ground.

The “energy efficiency” of the vehicle will be defined as the performance of all the vehicle’s powertrains. In practice, prioritizing energy efficiency will mean optimizing the control of the powertrains in order to reduce their consumption (in terms of cost or consumption) or the quantity of pollutant emissions into the atmosphere.

The “stability” of the vehicle will be defined as the tendency of the vehicle not to skid. This stability is compromised when a yaw moment appears around the vertical axis passing through the center of gravity CG of the motor vehicle 1.

This yaw moment is generally induced by a loss of lateral guidance at the contact zone between the ground and the wheels 12, 22. Indeed, if the losses of lateral guidance are sufficient on the front and/or rear axles, then the relative difference in guidance between the front and the rear will induce this rotational movement around the center of gravity CG of the vehicle. This movement may be difficult to control for the driver of the motor vehicle 1, particularly if it is the rear axle that loses the most lateral guidance (this is then referred to as an oversteering vehicle).

In the process described below, the stability of the car will be treated curatively to react immediately to any loss of traction on the front or rear axle in order to maintain an overall vehicle behavior of the understeer type (the front axle losing grip more easily than the rear axle). Of course, as a variant, the opposite could be preferred (for reasons of sports or agile vehicle typing).

This logic of managing the relative slip between the axles therefore makes it possible to control the stability of the vehicle by quickly correcting the torque applied to the rear axle. As will be described in detail below, this correction will in practice be carried out using a feedback loop called an “inter-axle slip control loop”.

The “reaction time” of a powertrain will quantify the latency of this group to exert a given torque, starting from a significantly different torque. Typically, this reaction time can be quantified by the time required to go from a zero torque to a given non-zero torque (typically 500Nm).

On the , the process implemented by the computer 9 to properly distribute the power between the axles is schematically represented, according to the conditions encountered by the vehicle and the instructions of the driver of this vehicle.

This process is implemented recursively, that is, in a loop, at regular time steps. In each loop, the computer thus executes eleven main steps.

In the example considered here, this method will be implemented to involve the rear axle as accurately as possible in the traction of the motor vehicle 1. Indeed, it is considered here that the rear powertrain 21 is equipped with a low-voltage battery, so that it cannot be used continuously but must be used only when the need arises. In this context, to simplify, the rear axle will be mainly used to guarantee the stability and traction of the vehicle when necessary.

Here, another aspect of the process will consist of determining a preventive torque for the rear axle to avoid a priori any loss of traction and to guarantee stability in a preventive (and not curative) manner.

The steps of this process are as follows.

During a first step F1, the computer 9 acquires a general energy request requested by the driver to move the motor vehicle 1. This general request allows the computer to determine whether the vehicle must accelerate or brake.

In the following, this query will be quantified in terms of torque at the axles. Of course, as a variant, it could be quantified differently (for example in terms of power, in terms of torque at the output shafts of the traction machines, etc.).

To determine this requested torque, the calculator can rely on different data such as for example the position of the brake and accelerator pedals, the pressure exerted on these pedals, the set speed entered at the vehicle's cruise control, etc.

In any case, at this stage, the calculator here determines the total torque C _T required and to be distributed between the front 10 and rear 20 axles of the motor vehicle 1. It will be noted here that this total torque C _T will be positive when the wheels must tow the motor vehicle 1, and negative when the wheels must brake the vehicle.

During a second step F2 (then a third step F3), the computer 9 acquires the minimum negative and maximum positive values of the torque that each powertrain can generate at the front axle 10 (then at the rear axle 20). It will be noted that the minimum value will correspond to the maximum torque (in negative) that the powertrain can generate to brake the motor vehicle 1.

These values here depend mainly on the performance of the powertrains 11, 21 coupled to these axles. They also depend on the SOC charge level of the low-voltage battery. They are here, for example, determined on a test bench during the design of the motor vehicle 1 and recorded in the memory of the computer in the form of tables. In the remainder of this presentation, these threshold values will be called "maximum potential traction torque C _MAX,f , C _MAX,r " and "minimum potential deceleration torque C _MIN,f , C _MIN,r ".

It will be noted that the indices f and r used above and in the remainder of this presentation refer respectively to the front 10 and rear 20 axles.

During a fourth step F4, the calculator 9 determines a “curative torque C _C ” to be applied to the axles.

This curative torque is the one that allows, when an axle slips (its wheels slide on the ground), to reduce the torque applied to this axle and to switch it in reaction towards the other axle. It is called "curative" since it is non-zero only when an axle is detected as slipping.

This curative torque Cc comes from the aforementioned inter-axle slip control loop. In other words, its value is determined according to the difference in rotational speeds between the axles. It is positive when the wheels 12 of the front axle slip and the rear axle 20 must take up an additional part of the total torque C _T , and negative when the wheels 22 of the rear axle slip and the front axle 10 must take up an additional part of the total torque C _T .

As explained above, depending on the configuration of the control loop, this curative torque will make it possible to give the motor vehicle 1 understeer or oversteer behavior.

On the , we have represented a method of calculating this curative couple C _C .

During a first sub-step, the calculator acquires the rotation speeds:
- ω _FR of the right front wheel,
- ω _FL of the left front wheel,
- ω _f of the front axle,
- ω _r of the rear axle.

Then it determines the average of the rotation speeds of the front wheels, as well as the average of the rotation speeds of the axles.

These different values give, depending on the configuration, a “reference rotation speed”. They also make it possible to calculate the speed V1 of the motor vehicle 1. Using tables stored in the computer’s memory, it is possible to deduce a target inter-axle slip rate S1 (i.e. a percentage difference between the rotation speeds of the axles).

Alternatively, there are other means of determining the speed V1 of the motor vehicle 1: it is typically possible to rely on information provided by a braking computer.

The product of the speed V1 and the target inter-axle slip value S1 then makes it possible to determine a target inter-axle slip speed Δω ₀ .

Here, if we want the vehicle to have understeer behavior, this target inter-axle slip speed Δω ₀ requires that the front axle turns faster than the rear axle.

At the same time, the calculator determines the difference Δω between the rotation speeds of the axles.

The difference between the target inter-axle slip speed Δω ₀ and this gap Δω is then used as input to a regulator 9A (here a PID type regulator) in order to determine the curative torque Cc which will ensure that, if one of the axles slips, the torque switches from this axle to the other axle.

Regarding the aforementioned "reference rotation speed", the following comments can be made.

When the traction machines have a rather low torque dynamics (typically internal combustion engines), it is possible to control the axle speeds by considering the average speeds of the wheels on each axle.

On the other hand, when the traction machines have greater torque dynamics (typically electric machines), it is more relevant to use the speeds of these traction machines directly, considering that the rotation speed of the machine represents the speed of the axle.

When using several traction machines per axle, with different dynamics and which are potentially disconnectable at will from the wheels, the solution consists in considering, if one or more traction machines with high torque dynamics are connected to the wheels, their combined speeds calculated at the differential level.

When using only one low torque traction machine (or no high torque traction machine), then the average of the axle wheel rotational speeds will be used.

Now regarding the 9A PID type regulator, the following comments can be made.

Preferably, the setting of this 9A regulator will depend on the state of the drive train, that is to say the configuration of each powertrain.

In fact, from the moment when one of the traction machines of one of the powertrains has a greater dynamic (in terms of torque) than the others and/or it is coupled to a multi-ratio reducer, it is preferable to take into account the inertial effect of this powertrain on the curative torque.

In fact, the inertia of the drive train will not be the same depending on the gear ratio engaged, depending on whether all or only some of the traction machines (electric motor and internal combustion engine) are coupled to the axles, etc.

Thus, the proportional gain of the 9A regulator can be adjusted here taking these parameters into account, based on values stored in a table memorized in the calculator 9.

This proportional gain presents here a greater steering coefficient in the case where only one of the traction machines is used (in two-wheel drive mode) than in the case where both traction machines are used (in four-wheel drive mode).

In this four-wheel drive configuration, this proportional gain will have a greater steering coefficient if the gear ratio engaged is lower.

At this stage, it should be noted that it would be possible to drive the vehicle based on the data from the four stages F1 to F4, correcting any grip problems using this curative torque Cc.

However, in the context of this presentation, we will prefer to seek to prevent any adhesion problems and to optimize the energy consumption of the powertrains as soon as possible.

For this, during a fifth stage F5 ( ), the calculator 9 performs an arbitration in order to determine whether, taking into account the configuration and situation of the vehicle, traction or energy efficiency should be prioritized.

Here, the calculator relies on the driving mode selected by the driver using the thumbwheel 8 and on data measured by the on-board sensors 7 (in particular the speed of the motor vehicle 1, the steering angle of the steering wheel or the steered wheels, the longitudinal acceleration and the lateral acceleration experienced by the vehicle).

Given this information, the computer will be able to determine whether traction should be prioritized, or whether it is possible to operate the powertrains 11, 21 in the most economical way possible.

Typically, an economy or comfort driving mode will prioritize energy optimization as much as possible while an off-road or snow driving mode will prioritize traction regardless of the vehicle speed.

Of course, a driving mode will not necessarily imply prioritizing one or the other of traction and energy optimization in all situations. Typically, an off-road or snow driving mode may prioritize energy optimization at stabilized average speed with low torque demand and reduced longitudinal and lateral accelerations.

The curative torque Cc calculated in the previous step will in all cases guarantee the traction and stability of the vehicle if ever the steering of the vehicle nevertheless generates slippage at the level of the wheels of an axle.

Pre-established arbitration rules are then preferentially defined from the design stage of the vehicle and stored in the computer memory, for example in the form of maps or in analytical form (via formal calculations of interpolation of single-criteria curves).

In any case, the aim of these arbitration rules will be to prioritize:
- the drive (i.e. the distribution of the total torque on the axles) when necessary,
- energy optimization whenever possible, even if this induces more precarious motor skills and can lead to more frequent curative interventions (i.e. to a non-zero curative torque).

On the , a table has been shown as an example illustrating simple arbitration rules that can be effectively used.

In this figure, we observe that the calculator begins by acquiring the total torque C _T (step F1), then it determines whether it is positive or negative.

In the case that interests us first, this total couple C _T is considered positive.

The first arbitration rule is based on the driving mode selected by the driver and on the value of the total torque C _T . Depending on these data, it may in fact be necessary to prioritize traction.

In practice, this first rule of arbitration consists of determining the value of a boolean ARB1.

Here, if the driving mode is off-road or snow, the value "1" is assigned to this ARB1 boolean.

Otherwise (if another driving mode is selected), the value of the total torque C _T is compared with a predetermined threshold, which preferably depends on the speed of the motor vehicle 1. If it is higher than this threshold, the value “1” is assigned to this boolean ARB1.

Otherwise, it is assigned the value "0".

As shown in the , when this boolean ARB1 is equal to 1, the motor skills are prioritized.

Otherwise, a second arbitration rule is used to determine whether the vehicle dynamics are such that traction should be prioritized.

In practice, this second arbitration rule consists of determining the value of a boolean ARB2.

To do this, the computer acquires the lateral acceleration experienced by the motor vehicle 10, then it compares this acceleration with a predetermined threshold, which preferably depends on the speed of the motor vehicle 1. If it is higher than this threshold, the value “1” is assigned to this boolean ARB2.

Otherwise, it is assigned the value "0".

As shown in the , when this boolean ARB2 is equal to 1, the motor skills are prioritized. Otherwise, it is the energy optimization that is.

In summary, traction is considered to be prioritized over energy optimization if the driving mode is off-road or snow, or if the total torque value C _T is high, or if lateral acceleration is high. In all other cases, energy optimization is prioritized.

We can now consider the case where the total torque C _T is negative. In this case, at least one of the powertrains is used to perform regenerative braking (i.e. to use the braking energy to recharge the low-voltage battery). In the case where the front powertrain 11 includes an internal combustion engine, it is understood that only the powertrain 21 can provide this regenerative braking.

A third rule of arbitration consists in determining whether we are in a case of significant braking, in which stability must be favored. Indeed, in such a case, the brakes are dimensioned so that the braking torque is mainly taken up by the front axle (with a distribution of for example 70% of the braking torque on the front axle).

This third arbitration rule then consists of determining the value of a boolean ARB3.

In practice, the calculator acquires the maximum potential traction torque C _MAX,r at the rear axle (determined during step F3) and the value of a “rear braking stability margin”.

In practice, the calculator acquires the minimum deceleration torque C _MIN,r at the rear axle (determined during step F3) and it constructs or acquires the value of a "rear braking stability margin". This rear stability margin is directly derived from braking and the stability margin used for the ABS (Anti-lock Braking System) or constructed to prevent the wheel from locking.

If their values are different, which means that we are in a configuration where regenerative braking does not allow the vehicle to be braked in the desired way, the value "1" is assigned to this ARB3 boolean. In this case, the stability of the vehicle will be prioritized.

Otherwise, it is assigned the value "0".

In this case, a fourth arbitration rule allows us to check whether the vehicle dynamics require or not to prioritize traction. Indeed, for example, braking in a straight line does not require as much priority to traction as braking in a bend.

This fourth arbitration rule consists of determining the value of a boolean ARB4, here as a function of the driving mode, the boolean ARB2 and the yaw rotation speed of the vehicle around the center of gravity CG.

The value "1" is assigned to this ARB 4 boolean if at least one of the following three conditions is met (otherwise, the value "0" is assigned to it).

A first condition is that the economical driving mode is not selected.

A second condition is that the boolean ARB2 is equal to 1 and that the economical driving mode is not chosen.

A third condition is that the yaw rate must be greater than a predetermined threshold (which may depend on the vehicle speed).

So, if the boolean ARB4 is equal to 1, the motor skills are prioritized. Otherwise, it is the energy optimization that is.

In summary, this arbitration makes it possible to determine whether motor skills or energy optimization should be prioritized. The result of this arbitration will be used in step F7 described below.

In parallel with this arbitration, as shown by the , the process continues in a sixth step F6 which consists, for the calculator 9, in determining a preventive torque C _P,r on the rear axle, allowing the traction to be optimized.

The idea is generally to ensure that the torque exerted by the front axle is not too high, so as to prevent any loss of grip in a preventive manner (so that the curative torque Cc remains as zero as possible).

This sixth step can be implemented in different ways. In the following, two different technical solutions will be considered, and in the case of the second technical solution, a refinement will also be considered.

The first technical solution is to start from the principle that the greater the mass exerted by a tire on the ground, the greater the torque that this tire will be able to exert without slipping on the ground.

In other words, this technical solution consists of distributing, when traction is preferred, the torque between the axles according to the mass that each axle exerts on the ground.

The distribution of masses between the axles depends mainly on the position of the storage battery (a notoriously very heavy object) and the type and position of the powertrains in the vehicle. It may also depend on the number of uses in the vehicle, their respective positions, the weight of the luggage, etc. The assessment of this distribution can be done in the same way as that explained below.

On the , we have shown an example of calculation of the preventive torque C _P,r on the rear axle in accordance with this first technical solution.

It can be seen that a block 91 is adapted to calculate a preliminary rear torque C _r as a function of the total torque C _T requested and the distribution of masses in the vehicle.

In order not to exceed the value of the maximum potential traction torque C _MAX,r on the rear axle, a mathematical operator selects the minimum Min between this maximum potential traction torque C _MAX,r and the preliminary rear torque C _r .

In order not to exceed the value of the maximum potential traction torque C _MAX,f on the front axle, another mathematical operator selects the maximum between this minimum Min and the difference between the total torque C _T and the maximum potential traction torque C _MAX,f on the front axle, and thus deduces the value of the preventive torque C _P,r on the rear axle.

Here, block 91 refers, for the distribution of masses, to a variable Br called “Base ratio” whose calculation is as follows:

In this equation, the variables used are defined as follows.

In this system, constants and variables are defined as follows:
M is the mass of motor vehicle 1, expressed in kilograms,
Lr is the length between the center of gravity and the front axle, expressed in meters,
Lf is the length between the center of gravity and the rear axle, expressed in meters,
L is the sum of the two aforementioned lengths and forms the wheelbase of the vehicle,
g is the acceleration of gravity,
accx is the longitudinal acceleration of motor vehicle 1,
h is the height of motor vehicle 1, expressed in meters.

It should be noted that in this model, Wfl=Wfr and Wrl=Wrr since it only takes into account the load transfer linked to longitudinal acceleration. To take into account lateral acceleration, it would have been necessary to add or subtract (depending on whether turning right or left) the expression +/- M/2.g.accy.h/l (with l the width of the car's track and accy the lateral acceleration).

The second technical solution is not based solely on the ability of the tires to exert traction or braking torque given the mass acting on them. It is also based on the front axle's traction limits.

This second solution thus allows, when traction must be prioritized, to use as much as possible the front powertrain 11 in order to save as much as possible the energy stored in the low voltage battery.

In other words, the preventive torque C _P,r is here calculated to correspond to the torque that the rear axle must exert so that the total torque demand C _T is satisfied and the front axle does not slip.

It can take more or less complex and/or formal forms. It can be determined according to:
- theoretical ellipses of adhesion of the wheels relative to the ground (we consider that the vector representing the maximum force that each wheel can apply to the ground, in the plane of the ground, varies in intensity depending on its direction according to a closed curve which forms an ellipse: Pajecka's law, Dugoff's law, etc.), and/or
- the static distribution of masses between the front axle and the rear axle, and/or
- the dynamic distribution of masses between the front axle and the rear axle as a function of the acceleration undergone by the vehicle, and/or
- the driving mode selected by the driver and/or
- the total torque C _T required by the driver.

Typically, the ECU could adjust this preventive torque according to the driver's driving. For example, it could be planned to exert a strong preventive torque C _P,r on the rear axle if the driver presses hard on the accelerator pedal and/or when the sport driving mode is selected.

In practice, here, as shown in the , this preventive couple C _P,r is calculated in the following manner.

During a first sub-step Fn1, the calculator calculates the value of an adhesion coefficient Mu of the axle considered to be the main axle (i.e. the one for which the use of the powertrain is not being limited). Here, this adhesion coefficient Mu makes it possible to quantify the adhesion of the front wheels 12 of the motor vehicle 1.

To calculate its value, various processes are possible.

Typically, we can consider that the value of this adhesion coefficient Mu depends only on the driving mode selected by the driver. In this example, its value can then be given by a table. For example, it is equal to:
1 in economy, comfort or sport mode,
0.3 in snow mode,
0.7 in sand and mud mode (if this mode is available),
0.5 in off-road mode.

Alternatively, the value of this adhesion coefficient Mu can be estimated based on data measured instantaneously. It can thus vary according to a function taking into account the dynamics of the vehicle. For example, it can be planned to reduce its value when repeated slipping of the wheels on the ground is detected.

Here, we will consider that the value of this adhesion coefficient Mu will be obtained using the first of these methods, then adjusted using the second method.

During a second sub-step Fn2, the computer calculates the mass of the front main axle (i.e. the mass of the vehicle acting on the ground via the tires of the front axle 10).

This mass can be a constant estimated during the design of the vehicle. Alternatively, it can have a variable value, for example depending on the driving mode chosen by the driver. It can be measured, or estimated by a function taking into account the dynamics of the vehicle.

This means that a continuous mass evaluation function based on an on-board sensor can be used here and/or this mass can be extrapolated based on the dynamic behaviour of the vehicle. This solution will make it possible to take into account changes in loads or the number of occupants in the vehicle.

As an example, a simple table solution consists of considering the theoretical mass of the empty car and 70 kg per detected passenger, then distributing these masses on the two axles according to the position of the passengers.

Another example is to base the mass of the front main axle solely on the driving mode selected by the driver. In this example, the idea is to consider that the mass is high when the driving mode selected is economy or comfort mode and to consider that it is low otherwise. Indeed, it is assumed here that in economy or comfort mode, the driving conditions are such that the grip of the tires on the ground is good, so that the mass of the front axle can be overestimated. Otherwise, it is better to underestimate it.

Thus, in economy or comfort mode, we consider here that the mass of the vehicle is equal to the sum of its empty mass with all tanks full and the mass of five occupants (for example five times seventy kilograms) distributed over the five seats of the vehicle.

In other modes, the mass of the vehicle is equal to the sum of its unladen mass with all fuel and the mass of two occupants (for example, two times seventy kilograms).

Alternatively, the value of the mass of the front main gear can be estimated as explained above, then adjusted according to parameters measured on the vehicle (for example according to the suspension travel).

During a third sub-step Fn3, the calculator calculates the adhesion limit C _ad of the front axle.

To do this, it relies on the results of the two sub-steps Fn1 and Fn2.

The adhesion limit is equal to the product of the mass of the main axle, the gravity constant g, the adhesion coefficient Mu and the radius of the wheels of the main axle. With regard to this radius, we can preferably consider its dynamic value (which changes according to the speed of the vehicle, to take into account the fact that a tire crushes when stationary).

In a fourth sub-step Fn4, the calculator 9 calculates the maximum torque potential available at the wheels of the front axle.

To do this, the calculator relies on the maximum potential traction torques C _MAX,f and minimum deceleration torques C _MIN,f determined in step F2 for the front main landing gear.

During a fifth sub-step Fn5, the computer then calculates the minimum value C _m between the adhesion limit C _ad of the main landing gear (from step Fn3) and the maximum traction potential C _MAX,f of the main landing gear (from step Fn4). This makes it possible to define from which torque demand on the main landing gear before the latter will be limited.

During a sixth sub-step Fn6, the calculator then subtracts this minimum value C _m from the total torque C _T requested by the customer (from step F1). The calculator thus obtains the preventive torque C _P,r on the rear axle.

Of course, if the total torque demand C _T is less than this minimum value, we force the preventive torque result to zero.

In summary, this solution makes it possible to favor the use of a main gear (here front) compared to the secondary gear (here rear), to define with more or less precision (according to the methods chosen) a theoretical grip limit depending on the driving mode chosen by the customer, and to generate use of the secondary gear without waiting for the main gear to slip.

The main advantage therefore consists of not being satisfied with curative control in reaction to an axle slip, but rather in controlling the use of the powertrain of the rear secondary axle to the bare minimum, the energy source of which, it should be remembered, is so small that its use must remain sparing.

It should be noted that the more precise the determination of the main axle slip threshold, the more it will be possible to minimize the use of the secondary axle. Although the precision is not optimal, the solution does however allow the traction to be controlled according to the driving mode chosen by the driver and thus make use on snow or off-road more demanding for the secondary axle in order to help with crossings.

In the case of slow axles (here the front axle), the torque cannot change from zero to a high value in a short instant.

Typically, when a mechanical disc clutch is used, the reaction time required to engage the clutch discs is non-zero since it is necessary to slide these discs progressively at the same speed to ultimately equalize their rotation speeds.

This problem could also arise in the absence of a clutch, with an electric machine directly coupled to one axle slower than that coupled to the other axle.

Here, the proposed refinement consists of correcting the preventive torque C _P,r on the rear axle in order to take into account this reaction time by supplementing, thanks to the rear power unit 21, the overall torque actually transmitted to the wheels of the vehicle.

For this, we could consider different solutions.

A first solution would be to use a table recorded in the calculator and allowing, taking into account the initial torque developed by the front powertrain 11 and the final torque to be achieved, to complete the torque instruction transmitted to the rear powertrain in order to compensate for the slowness of the clutch.

Here, the solution used is different.

Thus, during a step F6' illustrated on the , the calculator begins, during a first sub-step Fn10, by acquiring the inertia of the slow powertrain, namely here the front powertrain 11.

The inertias involved are those of the moving parts that the torque instruction first seeks to accelerate sufficiently to then be able to gradually transmit the torque to the rest of the drive train and to the wheels of the slow front axle.

This inertia value is generally an input data fixed by the components of the powertrain used, but a variable inertia can possibly be proposed depending on certain configurations or connections of moving parts (typically in the case where the internal combustion engine has a variable distribution system or a cylinder disconnection system).

Then, during a second sub-step Fn11, the computer calculates the acceleration of the moving parts of the front powertrain 11, which consists of calculating the derivative of the internal combustion engine speed as a function of time. This acceleration quantity ideally allows to clearly visualize the portion of the torque setpoint that will be consumed by the inertias of the internal combustion engine as a function of time.

Then, during a sub-step Fn12, the calculator 9 multiplies the results of the inertia and the acceleration to obtain the portion C _NF of torque not supplied to the front wheels 12 compared to the instruction transmitted to the internal combustion engine of the front axle 10.

This part C _NF is added to the preventive torque C _P,r on the rear axle. We thus obtain a new corrected preventive torque C _P,r .

This solution thus makes it possible to compensate for the lack of responsiveness of a slow axle compared to a fast axle, by using the fast axle to quickly satisfy the torque demand set by the driver.

In addition, this solution makes it possible to relieve the progressive torque transmission organ (typically here, the clutch) by reducing its participation during a rapid start. This allows for a better longevity of this organ since the latter will see its sliding phase reduced and therefore the wear of its interface parts limited.

During a seventh step F7, the calculator 9 determines a range of torques that can be used on the rear axle to guarantee both the stability of the vehicle and low energy consumption, taking into account the arbitration carried out in step F5.

As shown in the , this range of couples, also called “playing field”, is here defined by the calculator 9 by means of a lower bound C _inf,r and an upper bound C _sup,r .

To define these limits, the calculator uses the results of steps F4 to F6.

More precisely, compliance with the curative torque C _C will always be a priority. In practice, if a stability problem occurs (if the wheels of one of the axles slip), the curative torque C _C could form the lower limit or the upper limit depending on which axle loses grip.

The preventive torque C _P,r on the rear axle from step F6 or F6' is used if the arbitration carried out in step F5 prioritizes traction (subject to compliance with the curative torque C _C ).

In the case where energy optimization is prioritized, the maximum potential traction torque C _MAX,r and minimum deceleration torque C _MIN,r are used (subject to compliance with the curative torque C _C ).

Finally, in the event of instability during braking (ARB3=1), it is possible to rely on restrictive information from the braking system to control a margin of stability and avoid wheel lock. This makes it possible to restrict the range of usable torque in the event of a risk of rear wheel lock.

On the , we have shown in detail an example of the implementation of this step F7.

We observe that depending on the result of step F5 (arbitration), we will determine the lower and upper limits C _inf,r , C _sup,r of usable torque differently.

It should also be noted that the lower and upper limits C _inf,r , C _sup,r of usable torque will be different depending on whether the vehicle is in traction or deceleration mode.

Thus, if the total torque C _T is positive (i.e. in traction mode) and the drive is favored, the limits are defined by:

C _inf,r = Min (Max(0, C _C +C _P,r ), C _MAX,f )

C _sup,r = Min (Max(0, C _C +C _P,r ), C _MAX,f )

We note here that the limits are equal since, when traction is favored, we do not wish to leave any latitude in the choice of torque, considering that the latter is already precisely calculated in such a way as to guarantee traction and reduce the energy consumption of the vehicle.

If the total torque C _T is positive (i.e. in traction mode) and energy optimization is favored, these limits are different. They are defined here as follows:

C _inf,r = Min(Min(Max(Max(0, C _C ), C _T - C _MAX,f ), C _MAX,r ), Max1)

C _sup,r = Min (Min(C _MAX,r ), max(0,C _C + C _MAX,r ), C _T )

Since these terminals are different, the “playing field” is more open, which will allow the torques to be distributed on the axles in the most energy-efficient way, taking into account numerous parameters (is the vehicle in the city? At what speed is it traveling? ...).

We note here that the term Max1 is equal to the bound C _sup,r , which guarantees that the playing field contains at least one value.

If the total torque C _T is negative (i.e. in regeneration mode) and traction or stability is favored, these limits are defined as follows:

C _inf,r = Max(Min(C _C +C _P,r ), 0),-Brk)

C _sup,r = Max(Min(C _C +C _P,r ), 0),-Brk)

Here again, we note that these limits are equal.

It is also noted that the term Brk is equal to the rear braking stability margin, introduced in step F5. This term thus makes it possible not to exceed the maximum torque that the regenerative brake can provide before the rear wheels lock.

Finally, if the total torque C _T is negative (i.e. in regeneration mode) and energy optimization is favored, these limits are defined as follows:

C _inf,r = -Brk

C _sup,r = 0

It should be noted here that as a variant, a controlled front/rear distribution could be proposed using the curative torque so that it limits the speed difference between the two axles... the target in the regeneration phase would of course be to impose at least a speed difference to force the rear axle to roll faster than the front axle and thus avoid locking the rear wheels. In practice, in the preferred embodiment, this is not done because we rely on the ABS stability margin coming directly from braking, so that wheel lock is already avoided.

In another variant where the stability margin to avoid wheel lock is determined autonomously (independently of the ABS), the inter-axle curative could be used or mixed with a minimum speed threshold of the rear wheels.

During an eighth stage F8 illustrated on the , the calculator 9 calculates the energy-optimized torque C _OE,r on the rear axle, in compliance with the range of usable torques determined in step F7.

In fact, in step F7, a usable torque range was defined. It is now a question of choosing from this range a torque value that the rear axle must provide. This torque value will then form a specific instruction to be transmitted to the rear powertrain.

In our example, when motor skills are prioritized by the arbitration rules, this energy optimization is restricted by the playing field which only has one value.

On the other hand, when the drive is not prioritized in the arbitration function, this energy optimization is done in a more open framework and makes it possible to optimize the use of all the traction machines, whether they belong to the front or rear powertrain.

The energy sharing strategy and calculation will be performed at this stage based on Hamiltonian formulas, depending on many criteria (is the vehicle in the city? At what speed is it traveling? Is the low-voltage battery charged? ...). Since this calculation does not form the core of the present invention, it will not be described in detail here.

During a ninth step F9, the calculator 9 deduces from the total torque C _T and the energy-optimized torque C _OE,r on the rear axle an energy-optimized torque C _OE,f on the front axle (by a simple subtraction), which torque forms a specific instruction to be transmitted to the front powertrain 11.

It can be noted here that the driver's request (translated into total torque C _T ) may sometimes not be satisfied if it goes against the stability of the motor vehicle 1.

For example:
- if the total torque C _T is 2000 Nm, the maximum potential traction torque C _MAX,f of the front axle is 500 Nm and the maximum potential traction torque C _MAX,r of the rear axle is 2000 Nm, and
- if a torque distribution of 500 Nm on the front axle and 1500 Nm on the rear axle causes the rear axle to slip,
- then the curative torque (step F4) will limit the torque exerted on the rear axle to a value lower than 1500Nm, not satisfying the driver's request.

The following step F10 is implemented only if the front powertrain 11 has several traction machines. In this case, at this step, the computer 9 distributes the torque demand between these different traction machines.

Step F11 is implemented only if the rear powertrain 21 comprises several traction machines. In this case, at this step, the computer 9 distributes the torque demand between these different traction machines.

In the case described here with a front axle 10 equipped with a hybrid powertrain 11, step F10 applies while step F11 is not implemented.

Then, the specific torque instructions are transmitted to the powertrains so that the vehicle moves at the desired speed and with the required dynamics.

In summary, the process described above is very adaptable in the sense that it can be implemented regardless of the vehicle architecture (type of engines, number of axles, etc.).

It also guarantees the stability of the vehicle both preventively and curatively, since it offers the selection of a rear axle torque of 20 in a more or less wide range, depending on the driving conditions, in order to prioritize traction when necessary and at the same time guarantee an energy compromise as good as possible.

It is thus possible, under stability or motor constraints, to modulate the level of energy optimization at any time:
- either as a priority if the risk of loss of mobility or stability is low,
- either optionally if the drive and/or stability require specific use of the different powertrains.

The present invention is in no way limited to the embodiments described and shown, but those skilled in the art will be able to provide any variation in accordance with the invention.

Claims (10)

Method for controlling a motor vehicle (1) comprising a computer (9) and at least two axles (10, 20) each equipped with a powertrain (11, 21) and rolling elements such as wheels (12, 22), a first of the powertrains (11) having a reaction time greater than that of a second of the powertrains (21), the method comprising steps implemented by the computer (9) of:
- acquisition of a general request relating to a total torque (C _T ) that the powertrains (11, 21) must transmit to the rolling elements,
- distribution of said general request into specific instructions (C _OE,f , C _OE,r ) each assigned to one of the powertrain groups (11, 21), and
- control of the powertrains (11, 21) according to said specific instructions (C _OE,f , C _OE,r ),
characterized in that, at the distribution stage, when the specific setpoint (C _OE,f ) assigned to the first powertrain (11) varies, it is provided to:
- determine the share (C _NF ) of the specific instruction (C _OE,f ) not carried out by the first power unit (11) due to its reaction time,
- increase the specific instruction (C _{OE, r} ) assigned to the second powertrain (21) by at least part of said share (C _NF ). Control method according to claim 1, in which the particular instruction (C _{OE, r} ) assigned to the second powertrain (21) is increased by the entirety of said part (C _NF ). Driving method according to one of claims 1 and 2, in which said part (C _NF ) is determined as a function of an inertia of the first powertrain (11). A driving method according to claim 3, wherein said inertia has a predetermined fixed value. Driving method according to claim 3, in which said inertia has a variable value, determined as a function of configuration data of the first powertrain (11). Driving method according to one of claims 1 to 5, in which said part (C _NF ) is determined as a function of an acceleration undergone by a moving part of the first powertrain (11). Control method according to one of claims 1 to 6, in which the following steps are further provided:
- acquisition of a maximum traction potential (C _{MAX ,r} ) and/or minimum deceleration potential (C _{M IN,r} ) that at least one of the powertrains (11, 21) can transmit to the rolling elements,
- selection of a driving mode from among several predetermined driving modes, - determination of a preventive value (C _P,r ) for the first of the specific instructions (C _OE,r ), making it possible to avoid a priori the sliding of the rolling elements of the motor vehicle (1) on the ground,
- prioritization of a rolling element traction criterion or a powertrain energy consumption criterion (11, 21), depending at least on the selected driving mode,
- definition of a range of values (C _inf,r , C _sup,r ) usable for the first specific instruction (C _OE,r ) as a function of the preventive value (C _P,r ) if the traction criterion is prioritized, and as a function of said maximum traction potential (C _MAX,r ) and/or minimum deceleration potential (C _MIN,r ) if the energy consumption criterion is prioritized, and
in which the particular instructions (C _OE,f , C _OE,r ) are determined so that the value of the first particular instruction (C _OE,r ) is included in said range of values (C _inf,r , C _sup,r ). Steering method according to claim 7, in which said part (C _NF ) is added to the preventive value (C _P,r ). Motor vehicle (1) comprising at least two axles (10, 20) each equipped with a powertrain (11, 21) and rolling elements such as wheels (12, 22), characterized in that it comprises a computer (9) programmed to implement a control method in accordance with one of claims 1 to 8. Motor vehicle (1) according to claim 9, in which only the first powertrain (11) comprises a mechanical clutch.