Method for controlling the torque of a wheel motor of a vehicle and associated torque controller

Abstract

The invention relates to a method for controlling the torque of at least one wheel motor (10) of a vehicle (1000), the method comprising: determining an actual slip of at least one vehicle wheel driven by the wheel motor, which is implemented without the torque controller obtaining a datum representative of the orientation of the vehicle; and controlling the torque of said wheel motor, based on the determined actual wheel slip.

Description

METHOD FOR CONTROLLING THE TORQUE OF A VEHICLE WHEEL MOTOR AND ASSOCIATED TORQUE CONTROLLER Description Technical Field

[0001] The present invention falls within the general field of vehicle drive systems, where at least some of the wheels are equipped with "wheel motors." More particularly, it relates to torque control of one or more wheel motors, for example, powered by an electrical source. It also relates to a torque controller for one or more wheel motors, as well as a vehicle in which such a controller is installed. Previous technique

[0002] Motor vehicles generally consist of axles that carry wheels and support the vehicle body relative to the wheels. These axles are driven in rotation by an engine located away from the wheels, via a transmission mechanism.

[0003] Other vehicle drive solutions exist, including systems called "wheel motors." A wheel motor is an assembly in which the motor is mounted on the wheel and provides either drive or braking for that wheel. In this way, the torque applied to the wheel is directly controlled by its motor. This technological solution notably eliminates the need for a wheel drive differential.

[0004] However, current torque control solutions generate significant costs related to the use of vehicle steering angle sensors, which are due in particular to the environment in which these sensors are installed - these sensors being regularly subjected to shocks, vibrations, and / or the presence of dust - but also to the constraints in terms of expected reliability on this type of sensor. Description of the invention

[0005] The present invention aims to remedy all or part of the drawbacks of the prior art, in particular those set out above, by proposing a solution which improves the traction and grip of the vehicle, through an optimal allocation and distribution of power to the wheel motors of this vehicle.

[0006] To this end, and according to a first aspect, the invention relates to a method for controlling the torque of a wheel motor of a vehicle, the method being implemented by a torque controller mounted within the vehicle and comprising:

[0007] - a determination of an effective slip of at least one wheel of the vehicle (1000) driven by said wheel motor, implemented without obtaining, by the torque controller, a data representative of a vehicle orientation; and,;

[0008] - a control of the torque of said wheel motor, depending on the determination of an effective slip of said wheel.

[0009] In this application, "vehicle orientation data" refers to an angle measurement taken on the vehicle's steering system that reflects a change in the vehicle's trajectory relative to a straight path. This orientation data corresponds, for example, to the angle of the steering wheel(s) relative to the chassis, or to the angle of the steering wheels of a first steering axle relative to a second steering axle.

[0010] In general, the steps of a process are not considered to be linked to a notion of temporal succession.

[0011] In particular modes of implementation, the method of torque control of a wheel motor may further include one or more of the following characteristics, taken individually or in all technically possible combinations.

[0012] In specific implementation modes, the wheel motor is an electric wheel motor.

[0013] In particular modes of implementation, the vehicle comprises a plurality of wheels equipped with a wheel motor, and the controller is then configured to control the torque of each of the wheel motors of the plurality of wheels.

[0014] Thus, in this particular mode of implementation, the torque controller is configured to act centrally on the different wheel motors of the vehicle, and considers, to do so, the data associated with the different wheel motors of the vehicle to allocate the torque of a specific wheel motor.

[0015] In particular embodiments, the method further includes determining a value representative of the effective slip of each of the wheels as a function of a steering radius of the vehicle, and determining said steering radius, as a function of data representative of the speeds of the wheels of the plurality, and as a function of at least one dimension of said vehicle.

[0016] Thus, in accordance with this particular method of implementation, the turning radius of the vehicle is not obtained by analyzing data from a steering angle sensor installed on the steering wheel of the vehicle or more generally linked to a steering control of the vehicle or on a joint of the vehicle or on a pivot point of the vehicle, but determined from a processing of data from the wheels and / or the wheel motors that drive them.

[0017] In particular modes of implementation, at least one dimension of said vehicle includes a radius value of a wheel of the plurality, at least one track value and / or at least one wheelbase value of the vehicle.

[0018] Generally speaking, a vehicle's track width is defined as the distance between the two wheels on the same axle. More precisely, the track width corresponds to the distance between the centers of the contact patches of the wheels on the same axle, measured when the vehicle is empty (i.e., unloaded). A single vehicle can have front and rear track widths of different lengths.

[0019] The wheelbase is defined as the distance between the outermost axle centers of a vehicle. When articulated, a vehicle can have several wheelbase values, for example, one value corresponding to the front part of the vehicle, and a second value corresponding to the rear part.

[0020] In particular embodiments, the method further includes determining a representative value of the effective slip of each of the wheels as a function of a steering radius of the vehicle, and obtaining said steering radius from an association table linking a steering radius to at least one dimension of said vehicle.

[0021] In certain implementation modes, this association table is determined by machine learning.

[0022] In particular modes of implementation, the representative value of the effective slip is further determined as a function of the rotational speeds of the wheel motors and a reference speed of the wheels of the plurality.

[0023] In particular modes of implementation, the method further includes a determination, for each of the wheel motors of the vehicle, of a limiting torque as a function of the representative value of the effective slip of the wheel driven by said wheel motor, and as a function of a representative value of an optimal slip of said wheel.

[0024] In particular modes of implementation, the process further includes an allocation of torque to each wheel motor of the vehicle, according to an overall torque setpoint and said limit torques.

[0025] In particular modes of implementation, the overall torque setpoint is determined as a function of a vehicle speed setpoint and an estimated vehicle speed, the method further including an estimation of the vehicle speed as a function of the turning radius and the reference speed of the wheels of the plurality.

[0026] In particular modes of implementation, the vehicle comprises a plurality of wheels equipped with a wheel motor, and the controller is configured to control the torque of a single wheel motor, without obtaining representative data on the rotational speeds of the other wheel motors of the vehicle.

[0027] Alternatively, the controller can be common to one axle or common to all wheel motors of the machine, and be configured to control the torque of each wheel motor without obtaining data representative of the rotational speeds of the other wheel motors of the vehicle.

[0028] In particular modes of implementation, the determination of effective slip includes a determination that the rotational speed of the controlled wheel motor increases significantly over a predetermined period, while the linear acceleration of said vehicle is less than a threshold value over that same period.

[0029] In specific implementation modes, the linear acceleration of said vehicle is determined by an inertial measurement unit equipping said vehicle.

[0030] In specific implementation modes, an inertial measurement unit is integrated into one wheel motor or all wheel motors of the vehicle.

[0031] In specific implementation modes, the control of the torque of a particular wheel motor without obtaining representative data of the rotational speeds of the other wheel motors of the vehicle is implemented by taking into account an estimate of the slope on which the vehicle is located; this slope estimation can in particular be carried out by an inertial measurement unit equipping the vehicle.

[0032] In particular modes of implementation, the control of the torque of a particular wheel motor, without obtaining representative data of the rotational speeds of the other wheel motors of the vehicle, is implemented according to an estimate of a slip / skip angle, a vehicle load, the attitude or attitude of the vehicle, the yaw rate, the temperature of the wheel motors and / or the electrical consumption of the wheel motors.

[0033] According to a second aspect, the invention relates to a computer program comprising instructions for implementing a torque control method according to the invention, when said program is executed by a processor.

[0034] This program can use any programming language, and be in the form of source code, object code, or code somewhere between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0035] According to a third aspect, the invention relates to a computer-readable recording medium on which the computer program according to the invention is recorded.

[0036] The information or recording medium can be any entity or device capable of storing the program. For example, the medium may include a storage means, such as a ROM, for example a CD-ROM or a microelectronic circuit ROM, or a magnetic recording means, for example a hard drive.

[0037] On the other hand, the information or recording medium can be a transmissible medium such as an electrical or optical signal, which can be transmitted via an electrical or optical cable, by radio, or by other means. The program according to the invention can, in particular, be uploaded to a network such as the Internet.

[0038] Alternatively, the information or recording medium may be an integrated circuit in which the program is incorporated, the circuit being adapted to execute or to be used in the execution of the process in question.

[0039] According to a fourth aspect, the invention relates to a torque controller for at least one wheel motor of a vehicle, the controller being configured to implement a torque control method according to the invention.

[0040] According to a fifth aspect, the invention relates to a vehicle comprising a plurality of wheel motors and in which a torque controller according to the invention is mounted.

[0041] According to a sixth aspect, the invention relates to a wheel in which a torque controller according to the invention is embedded. Brief description of the drawings

[0042] Other features and advantages of the present invention will become apparent from the description below, with reference to the accompanying drawings, which illustrate an example of an embodiment without being limiting in any way. In the figures:

[0043] [Fig.1] Figure 1 is a front view representation of a vehicle in which a torque controller is mounted, according to an example of implementation of the invention;

[0044] [Fig.2A] Figure 2A is a top-view representation of an articulated vehicle in a turn;

[0045] [Fig.2B] Figure 2B is a top-view representation of a non-articulated vehicle in a turn;

[0046] [Fig.3] Figure 3 represents modules embedded in a torque controller, according to a particular implementation method of the invention;

[0047] [Fig.4] Figure 4 represents modules embedded in a torque controller, according to a particular implementation method of the invention;

[0048] [Fig.5] Figure 5 schematically represents an example of the hardware architecture of a torque controller;

[0049] [Fig.6] Figure 6 represents, in the form of a flowchart, a particular method of implementing an overall method of controlling the torque of at least one wheel motor, for example executed by the torque controller of Figure 3;

[0050] [Fig.7] Figure 7 represents, in the form of a flowchart, a particular method of implementing step S160 of determining a limit torque of at least one wheel motor of a vehicle 1000;

[0051] [Fig.8] Figure 8 represents, in the form of a flowchart, a particular method of implementing an overall method of controlling the torque of a wheel motor, for example executed by the torque controller of Figure 4;

[0052] [Fig.9] Figure 9 is a top view representation of an articulated vehicle in a turn, and the notations considered for determining its turning radius;

[0053] [Fig.10] Figure 10 is a top view representation of a three-wheeled vehicle in a turn, and the notations considered for determining its turning radius;

[0054] [Fig.11] Figure 11 is a top view representation of the notations considered for determining the turning radius and effective wheel slip of an articulated vehicle having several wheelbase values ​​and several track values;

[0055] [Fig. 12] Figure 12 is a representation of a drive device, according to a particular embodiment of the invention; and,

[0056] [Fig.13] Figure 13 represents modules embedded in a torque controller, according to a particular implementation method of the invention. Description of the implementation methods

[0057] Figure 1 is a front view representation of a vehicle in which a torque controller is mounted, according to an example of implementation of the invention.

[0058] In general, a "vehicle" within the meaning of the invention corresponds to any motorized land vehicle, that is to say any vehicle capable of traveling on the ground or on rails, and which can be powered by a mechanical force generated by one or more motors.

[0059] In specific modes of implementation, the vehicle is a wheel loader, a multi-function tool carrier, a forklift, a tandem roller, a compact mini track loader or an autonomous agricultural machine.

[0060] As illustrated in Figure 1, vehicle 1000 is, in this example, an articulated loader. A loader's defining characteristic is its ability to quickly transport or move a large quantity of material, particularly during earthmoving operations. To achieve this, loader 1000 is equipped with, among other things, a loader bucket 300, a stabilizer (also called a "counterweight") 500, and a cab 400.

[0061] In this example, the vehicle 1000 is also equipped with four wheels 100. A drive unit is associated with at least one front wheel and at least one rear wheel. Preferably, each wheel 100 of the vehicle 1000 is equipped with a drive unit 10, and in this case, the vehicle 1000 comprises four drive units, forming a four-wheel drive transmission for the vehicle 1000.

[0062] It should be noted that the number of wheels (100) equipping vehicle 1000 does not constitute a limiting factor of the invention. The following developments can indeed be easily generalized by a person skilled in the art to cases where the vehicle is equipped with a number of wheels other than four.

[0063] Each drive device 10 includes, in particular, a motor, typically an electric motor. It should be noted, however, that there are no limitations on the power source supplying the motor. The following developments can indeed be easily adapted by a person skilled in the art to cases where the motor is of the hydraulic type, for example, with axial pistons.

[0064] Figure 12 is a representation of a drive device, according to a particular embodiment of the invention.

[0065] As illustrated in Figure 12, the drive unit 10 comprises an electric motor 11, a parking brake 12, a gearbox 13, and a variable speed drive 14 for the electric motor. This drive unit also includes an inertial measurement unit (IMU) 15 connected to the variable speed drive 14, as well as a sensor 16 for the rotational speed of the electric motor 10. An inertial measurement unit (IMU) is a set of sensors that measure linear accelerations and rotational velocities, allowing the motion of a point of interest to be estimated in its six degrees of freedom: three translational degrees modeled by the linear velocity vector of the point, and three rotational degrees modeled by the instantaneous rotation vector of the frame of reference. An inertial measurement unit therefore has six sensors: three accelerometers and three gyroscopes.

[0066] The reducer 13 includes a housing which is, for example, fixed to the chassis frame of the vehicle 1000. The reducer 13 also includes an output hub, rotatably mounted relative to the housing. The output hub of the drive device 10 carries a wheel 100, and said drive device 10 supports the body of the vehicle 1000 relative to said wheel 100.

[0067] Alternatively, the wheel motor has a rotating housing, and the drive device 10 therefore does not include an output hub.

[0068] Returning to the description in Figure 1, the vehicle 1000 further includes a torque controller 200 of at least one of the wheel motors whose functionalities are described in more detail with reference to Figures 3 to 11.

[0069] In one particular implementation mode, a centralized architecture is considered, and the torque controller is embedded in the vehicle body or in its engine. In this particular case, the torque controller is, for example, integrated into the vehicle's electronic control unit (ECU). Alternatively, but still considering a centralized architecture, the torque controller is integrated into one of the vehicle's wheels, known as the "master wheel." In this case, all traction management is performed by this master wheel. Alternatively, but still considering a centralized architecture, the torque controller is integrated into each wheel equipped with a wheel motor, but only one wheel, known as the "master wheel," performs the control at any given time.

[0070] The wheel motors are then connected, via one or more data buses, to the torque controller 200, the functionalities of which are described in more detail with reference to figures 3 to 11.

[0071] According to a particular implementation, the data buses are CAN type buses (acronym for "Controller Area Network"), for example conforming to the ISO 11898 standard. The use of CAN type buses is advantageous since it allows several electronic devices to be connected to the same cable, thus avoiding the use of dedicated cables for the transport of each piece of information.

[0072] Alternatively, a distributed architecture is considered, in which each wheel incorporates a torque controller configured to determine the torque to be applied by that wheel's in-wheel motor. In this case, all traction management is handled by a so-called "plug and play" (or ready-to-use) wheel. More precisely, traction management is performed autonomously by the wheel(s), requiring nothing more than a power source and a torque or speed command. It is therefore important to note that when such a distributed architecture is considered, it can be beneficial to send a slip signal from the wheel to a higher-level vehicle controller, for example, to alert the system that the wheel is slipping.

[0073] Figure 2A is a top-view representation of an articulated vehicle in a turn, in which the invention can be implemented. As illustrated in Figure 2A, the vehicle 1000 comprises four wheels 100 FL 100 FR 100 RL 100 RR and features a central articulation point connecting the front (FP) and rear (RP) sections of the chassis. Thus, to make a turn, the front (FP) section pivots on the horizontal plane relative to the rear section. The steering angle therefore depends directly on the articulation angle. This feature is advantageous because it allows articulated vehicles (such as articulated loaders) to be used for tasks requiring high maneuverability.

[0074] Figure 2B is a top-view representation of a non-articulated vehicle in a turn, in which the invention can be implemented. As illustrated in Figure 2B, to make a turn, it is the wheels 100 FL 100 FR 100 RL 100 RR of the non-articulated vehicle which pivot, for example around an axis located at each end of an axle and perpendicular to that axle.

[0075] Figure 3 shows modules embedded in a torque controller, according to a particular embodiment of the invention. In this particular embodiment, the architecture is centralized, and the torque controller 200 according to the invention is then configured to control all the wheel motors of the vehicle, taking into account, for a given wheel motor, the information from said wheel motor, but also the data from the other wheel motors of the vehicle.

[0076] As illustrated in Figure 3, the 200 torque controller includes a MOD_OBS module, also called the "observation module," which includes several sub-modules:

[0077] - a MOD_WS module for determining the vehicle's wheel speeds, also configured to determine a speed V REF called "reference speed";

[0078] - a MOD_TURN module for estimating the turning radius R of the vehicle, as a function of the speed V REF reference;

[0079] - a MOD_VS module for estimating velocity V VEC of the vehicle, depending on the speed V REF reference and turning radius R;

[0080] - a MOD_SR module for estimating the slip rate k i=FR FL 100 of each wheel i=FR FL , depending on the rotational speeds ω FL , ω FR , ω RL , ω RR vehicle wheel motors, reference speed, and turning radius R.

[0081] The 200 torque controller also includes:

[0082] - a CTRL_SR controller to determine, for each of the vehicle's wheel motors, a limit torque TL FL TL FR TL RL TL RR depending on the effective slip k i=FR FL of the wheel driven by the wheel motor, and as a function of the representative value of optimal slip k SET of this wheel;

[0083] - a CTRL_SD controller for determining an overall torque setpoint TG, based on a speed setpoint V SET and the estimated speed V VEC of the vehicle; and,

[0084] - a MOD_TQ module for torque allocation based on TL limit torques FL TL FR TL RL TL RR and the overall torque setpoint TG. This MOD_TQ module is further configured to transmit a torque setpoint T to each wheel motor of the vehicle. FL , T FR, T RL , T RR to be applied by these wheel motors.

[0085] Figure 4 shows modules embedded in a torque controller, according to another specific embodiment of the invention. Unlike the specific embodiment illustrated in Figure 3, which considered a centralized architecture, the embodiment in Figure 4 considers a distributed architecture. The torque controller 200 is then configured to control the torque of a single wheel motor, and, for example, embedded within that wheel.

[0086] As illustrated in Figure 4, this 200 torque controller includes a MOD_OBS module, also known as the "observation module," which notably includes:

[0087] - a MOD_AL sub-module for obtaining inertial data. This inertial data is determined, for example, by an inertial measurement unit connected to this MOD_AL data acquisition module, such as the inertial measurement unit 15 described with reference to Figure 12; and,

[0088] - a MOD_VR sub-module for obtaining the rotational speed of the wheel motor. This rotational speed is, for example, determined by a speed sensor connected to this MOD_AL obtaining module, such as sensor 16 described with reference to Figure 12.

[0089] This 200 torque controller also includes:

[0090] - a MOD_DET module for calculating an estimated linear velocity, and for determining that the evolution of this estimated linear velocity over a predetermined period is greater than a first threshold value, while the linear acceleration is less than a second threshold value during this same period; and,

[0091] - a MOD_CTRL module for determining a torque T t which must be applied by the wheel motor it controls.

[0092] Figure 5 schematically represents an example of the hardware architecture of a 200 torque controller.

[0093] As illustrated in Figure 5, the Torque Controller 200 has the hardware architecture of a computer. Thus, the Torque Controller 200 includes, in particular, a processor 1, RAM 2, ROM 3 and non-volatile memory 4. It also has means of communication 5.

[0094] The read-only memory 3 of the torque controller 200 constitutes a storage medium according to the invention, readable by the processor 1, on which a computer program PROG according to the invention is stored, comprising instructions for executing steps of the torque control process according to the invention. The PROG program defines functional modules of the torque controller 200, which rely on or control the hardware elements 1 to 5 of the torque controller mentioned above. These functional modules are illustrated in Figure 3 or Figure 4, without limitation, and are described in more detail below with reference to different implementation methods.

[0095] In the implementation modes described below, the communication means 5 enable the torque controller 200 to transmit instructions to the wheel motors regarding the target torque value. To this end, the communication means 5 include a wired or wireless communication interface capable of implementing any suitable communication protocol.

[0096] Figure 6 represents, in flowchart form, a particular method of implementing a global method of controlling the torque of at least one wheel motor, for example executed by the torque controller 200 of Figure 3. This torque controller 200 is configured to control, in a centralized manner, the torques of all the wheel motors of a vehicle 1000 comprising four wheels 100 FL, 100 FR, 100 RL, 100 RR all equipped with wheel motors.

[0097] The overall torque control process includes a first step S100 of obtaining the effective rotational speeds a> FL , a> FR , a> RL , a> RR vehicle wheel motors. This S100 step is for example implemented by the MOD_WS sub-module of the MOD_OBS observation module described with reference to Figure 3. According to a particular implementation, the rotation speeds are determined by rotation sensors, such as sensor 16 illustrated in Figure 12, installed on the wheel motors, and transmitted to said MOD_WS module.

[0098] The overall control process further includes a step S110 for determining the estimated speed V i=FL,FR,RL,RR of the vehicle's wheels, depending on the effective rotational speeds ωi=FL, FR, RL, RR of the wheel motors, and the radius RI=FL, FR, RL, RR of said wheels. In general, the estimated speed V t is expressed such that = ω i x R iWithout slippage, the speed The estimated and actual speeds have identical values. However, in the case of slippage, only the estimated speed conforms to this expression.

[0099] During this same step S110, a speed called "reference speed" V REF is determined. Assuming that at least one wheel of the vehicle is in contact with the ground, the wheel with the lowest slip rate then becomes the one providing the reference speed. When a positive torque (or "drive torque") is applied by the wheel motors, the reference speed V REF is expressed as V REF = MIN(V FL , V FL , V FL , V FL ), with MINQ being the minimum operator. Conversely, when a negative torque (or "braking torque") is applied by the wheel motors, the reference speed V REF can be expressed, for example, as V REF = MAX(V FL , V FL , V FL, V FL ), with MAXQ being the maximum operator.

[0100] Alternatively, the reference speed V REF is an average of the speeds of the same axle. According to another variant, the reference speed corresponds to the average of the speeds of the wheels of an axle, each of which may be weighted by the torque allocated to it. In the specific case where the rear axle of the vehicle illustrated in Figure 11 is considered, the reference speed V REF can be expressed, for example, as follows:

[0101] V REF — (V RL + V RR ) / 2 - R RL * a + R RR * a) / 2

[0102] a =

[0103] R R = (V RL +V RR )·b2 “ 2*(V RL -IZ RR )

[0104] R F = √(R R 2 +L2 2 -L1 2 )

[0105] R RL = R R - b2 / 2

[0106] R RR = R R + b2 / 2

[0107] R FL = R F - b1 / 2 = √(R R 2 + L2 2 - L1 2 ) - b1 / 2

[0108] R FR = R F + b1 / 2 = √(R R 2 + L2 2 - L1 2 ) + b1 / 2

[0109] with R R the turning radius of the rear part of the vehicle, R F the turning radius of the front part of the vehicle, a the angular velocity around the center of rotation C, b the track of the front part of the vehicle, b2 the track of the rear part of the vehicle, the distance between the pivot and the axle center of the front part of the vehicle, and L2 the distance between the pivot and the axle center of the rear part of the vehicle,

[0110] Then, during step S120, a turning radius R of the vehicle is determined, based on representative wheel speed data and at least one dimension of the vehicle. This step S120 is implemented, for example, by the MOD_TURN sub-module of the MOD_OBS observation module described with reference to Figure 3.

[0111] In a particular implementation illustrated by Figure 9, assuming that the vehicle 1000-1 has four wheels and is articulated at its center, and that the inner wheels (e.g., the front-left and rear-left wheels) have the same speed v ieft> that l es The outer wheels (e.g., the front right and rear right wheels) have the same speed v right and that the center of the front axle and the rear axle also have the same speed, then:

[0112] v left = V FL = V RL = Ri xa = (R3- xa

[0113] Vright = V FR = V RR = R2x a = (R3+ b / 2) xa

[0114] and R3= (V left - V right ) xb / 2

[0115] with a the angular velocity around the center of rotation C, b the wheelbase of the vehicle, and R3 the turning radius of the vehicle 1000.

[0116] In another specific implementation illustrated by Figure 10, assuming that vehicle 1000-2 has three wheels, then:

[0117] V FL = R1x a

[0118] V FR = R2x a = (R1+ b) xa

[0119] V R = R3x a = √((R1+ b / 2) 2 + a 2 ) xa

[0120] a = V FR~ V FL b VFL xb

[0121] and Ri = V FR -V FL

[0122] with a the angular velocity around the center of rotation C, b the wheelbase of the vehicle, a the track width of the vehicle, V R the speed of the rear wheel and R1 the radius of rotation of the front-left wheel of the vehicle 1000-2.

[0123] In another embodiment illustrated by Figure 11 – and corresponding to a generalization of the embodiment illustrated with reference to Figure 9 – assuming that the vehicle 1000-3 has four wheels and is articulated, then:

[0124] V FL = R FL xa = (R F - b1 / 2) xa

[0125] V FR = R FR xa = (R F + b1 / 2) xa

[0126] V RL = R RL xa = (R R - b2 / 2) xa

[0127] V RR = R RR xa = (R R + b2 / 2) xa

[0128] R F = √(R R 2 + L2 2 - L1 2 )

[0129] RR = -b1V FL V RL +b2V FL 2 ±√(b1 2 V RL 4 -2b1b2V FL V RL 3 +b2 2 V FL 2 V RL 2 -4L1 2 V FL 2 V RL 2 +4L1 2 V RL 4 +4L2 2 V FL 2 V RL 2 -4L2 2 V RL 4 ) 2(V FL 2 -V RL 2 )

[0130] with a the angular velocity around the center of rotation C, b the track of the front part of the vehicle, b2 the track of the rear part of the vehicle, the distance between the pivot and the axle center of the front part of the vehicle, L2 the distance between the pivot and the axle center of the rear part of the vehicle, R R the turning radius of the rear part of the vehicle and RF the turning radius of the front part of the vehicle.

[0131] Returning to Figure 6, the overall torque control process further includes a step S130 for estimating the effective slip k t of each of the vehicle's wheels, based on a representative data point of wheel motor rotation speeds, the vehicle's turning radius, and a representative data point of a reference speed V REF wheels. This S130 step is implemented for example by the MOD_SR sub-module of the MOD_OBS observation module described with reference to figure 3.

[0132] In a particular implementation mode, the effective slippage k t The slip rate of each wheel of the vehicle is expressed as a rate called the "slip rate". Generally, a slip rate k t of a wheel 100; of radius y having an estimated linear speed V {and being driven by a wheel motor 10 j having a rotational speed a)i is expressed as follows:

[0133] k( = with absQ the absolute value operator.

[0134] Considering again the implementation method illustrated in Figure 9, and assuming that a driving torque is applied, that the front left wheel is the slowest (and therefore does not slip or slips very little), then the reference speed V REF corresponds to the speed V FL of this front left wheel.

[0135] Furthermore, V FL = o) FL XR FL and k FL = = 0

[0136] The slip rate k RL of the left rear wheel is expressed as follows:

[0137] k RL = (ω RL xr - V RL ) / abs(V RL ) = (ω RL xr - V FL ) / abs(V FL ) RL abs(VRL ) abs(V FL )

[0138] The slip rate k FR of the right front wheel is expressed as follows: R3+| ω FR x r-(V FL x - g) k FR = (ω FR xr - V FR ) / abs(V FR )

[0139] FR abs(V FR ) R3+| abs(V FL x - "3-2

[0140] And the slip rate k RR of the right rear wheel is expressed as follows: R3+| ω RR x r-(V FL x - 5)

[0141] k RR = (ω RR xr - V RR ) / abs(V RR ) RR abs(V RR ) R3+| abs(V FL x - |) “3-

[0142] Thus, this value k tThe representative value of an effective wheel slip is determined based on data specific to the wheel motor, such as its rotational speed, and based on at least one dimension of the vehicle, such as the radius of at least one wheel, a track width, and / or a wheelbase. In other words, this value k t representative of an effective wheel slip is not determined from data representative of a vehicle orientation generated by a steering angle sensor of the vehicle's steering wheel or generated by a steering angle sensor installed at the pivot joint of an articulated vehicle.

[0143] The overall torque control process further includes a step S140 for estimating the vehicle speed, based on the steering radius and a reference speed. This step S140 is implemented, for example, by the MOD_VS sub-module of the MOD_OBS observation module described with reference to Figure 3.

[0144] Considering again the implementation method illustrated in Figure 9, and still assuming that a driving torque is applied, that the front left wheel is the slowest, and that consequently the reference speed V REF corresponds to the speed V FL of this left front wheel, the speed V VEC The vehicle's power is expressed as follows: V VEC = (V FL x RS ~2

[0145] Then, during step S150, a global torque setpoint TG is determined, based on the estimated speed V VECof the vehicle and a speed setting V SET This S150 step is implemented, for example, by the CTRL_SD controller for determining an overall torque setpoint, as described with reference to Figure 3. This CTRL_SD controller of determination is for example a PID regulator ("Proportional, Integral, Derivative") with an anticipatory action ("feedforward" according to Anglo-Saxon terminology).

[0146] The overall torque control process further includes a step S160 for determining a torque limit TL for each of the vehicle's wheel motors FL TL FR TL RL TL RL depending on the effective slip of the wheel driven by the wheel motor, and depending on the optimal slip k SETof this wheel 100. The S160 step includes the steps S1600, S1610, S1620 and S1630 described below with reference to Figure 7, and is for example implemented by the CTRL_SR controller for determining a limit torque described with reference to Figure 3.

[0147] In a particular implementation mode, these limit torque values ​​TL FL TL FR TL RL TL RL are further determined according to at least one criterion among a slip / skip angle, a vehicle load, attitude or trim, yaw rate (this is an angular rate), the temperature of the wheel motors 10j and the electrical consumption of the wheel motors 10j.

[0148] Then, during an S170 step, a torque value T FL , T FR , T RL , T RR is allocated to each of the wheel motors, according to the TL torque limits FL TL FR TL RL TL RRand the overall torque setpoint TG. This step is implemented for example by the MOD_TQ torque allocation module described with reference to Figure 3.

[0149] As a first illustrative example, if the total torque (TG) is 40 Nm, and the vehicle has four wheel motors, each with a torque limit of 20 Nm, then the total torque value is equally distributed among the four wheel motors. In other words, each wheel motor will receive a command to apply a torque of 10 Nm.

[0150] According to a second illustrative example, if the overall torque TG is 40Nm, and the vehicle includes four wheel motors 10 i=FLFR , RL, RR and that the limiting pairs are as follows: TL FL = 2, TL FR = 20, TL RL = 3, TL RR = 20, so the wheel motor 10j =FL will be allocated a couple T FL with 2Nm corresponding to its limiting torque, the wheel motor 10 i=RLwill be allocated a couple T RL of 3Nm corresponding to its torque limit, and the remaining 35 Nm to be allocated will be equally distributed between the last two motors 10 j=FR,RR In other words, the two 10j engines =FR RR will be allocated a couple T FR , T RR of 17.5 Nm.

[0151] Finally, the overall torque control process includes an S180 step in which the torques allocated during the S170 step are transmitted to the respective wheel motors, and applied by them.

[0152] It is important to reiterate at this point that the number of wheels (100) equipping vehicle 1000 does not constitute a limiting factor of the invention. The preceding developments regarding torque allocation can, in fact, be easily adapted by a person skilled in the art if the vehicle is equipped with a number of wheels other than four.

[0153] Figure 7 represents, in the form of a flowchart, a particular method of implementing step S160 of determining a limit torque of at least one wheel motor of a vehicle 1000.

[0154] As illustrated in Figure 7, the control process includes a first step S1600 during which at least one value k SET representative of an optimal slip of at least one wheel 1001 of the vehicle 1000 driven by at least one wheel motor 10j. In a particular implementation mode, this value k SETThe representative value of optimal slip corresponds to the optimal slip rate of a wheel. In a particular implementation, several optimal slip rates are obtained, each associated with one of the vehicle's wheels. In another particular implementation, the controller is dedicated to controlling the torque of a single wheel motor, and a single optimal slip rate associated with the wheel whose wheel motor is controlled by this controller is then obtained during this S1600 step.

[0155] The optimum slip rate is determined, for example, based on the type of surface, and / or the type of wheel or tire.

[0156] In a particular implementation mode, this optimum slip rate is determined by a machine learning model.

[0157] In specific implementation modes, this machine learning model is implemented in the form of neural networks (convolution, perceptron, autoencoder, recurrent, etc.). According to one particular implementation, the neural networks considered are recurrent neural networks of the "long short-term memory" (LSTM) type.

[0158] Furthermore, it is important to note that there are no limitations on the type of training technique used to obtain this machine learning model. Any technique implementing a learning algorithm (or "machine learning" in English) and providing, as output, a value representing the optimal slip of a wheel, given input data, can be considered within the context of the invention (for example, support vector machine, regression). logistics, etc). In other words, the learning model is independent of the training method considered to train this model.

[0159] In addition, any training criterion known to a person skilled in the art can be considered during the training phase of this machine learning model, such as the least squares method or cross-entropy minimization.

[0160] The control process further includes a step S1610 for obtaining the k values t representative of an effective slip of the 100 wheels; determined during the S130 step of estimating the effective slip / of each of the wheels described with reference to Figure 6.

[0161] The control process further includes a step S1620 during which the representative values ​​of an optimal slip k SET and an effective slippage k tare compared. This S1620 comparison step is implemented for each of the vehicle's wheels.

[0162] Finally, an S1630 step is implemented during which so-called "limit torque" values ​​are determined for each of the vehicle's wheel motors.

[0163] Figure 8 represents, in flowchart form, a particular method of implementing a global method of controlling the torque of a wheel motor, for example executed by the torque controller of Figure 4. Unlike the method described with reference to Figure 7, this global method of torque control is implemented by a controller dedicated to controlling the torque of a single wheel motor, and for example embedded within this wheel.

[0164] As illustrated in Figure 8, this overall torque control process includes a first step, S200, for obtaining inertial data. This S200 data acquisition step is implemented, for example, by the MOD_AL inertial data acquisition sub-module described with reference to Figure 4. As discussed in more detail below, this inertial data is determined, for example, by an inertial measurement unit (IMU) installed within the drive system, such as IMU 15 described with reference to Figure 12. This inertial data corresponds to linear accelerations and / or angular velocities of rotation. As an example, we will subsequently consider the case where this inertial data corresponds to linear accelerations, referred to as "measured linear accelerations."

[0165] The overall control process further includes a step S210 for obtaining the rotational speed ω j, with i = FL, FR, RL or RR of a wheel motor of the vehicle. This step is implemented for example by the MOD_VR submodule described with reference to figure 4.

[0166] Then, during step S220, the torque controller 200 calculates an estimated linear speed, based on the rotational speed of the wheel motor obtained during step S210, and the radius R t of the wheel driven by this wheel motor. More precisely, this estimated linear speed is expressed as V t = ω j x R i .

[0167] The method further includes a step S230 of determining that the estimated linear acceleration over a predetermined period (e.g., 20 s) is greater than a first threshold value, while the measured linear acceleration is less than a second threshold value (e.g., 5 m / s²). 2) during this same period. These steps S220 and S230 are, for example, implemented by the MOD_DET module described with reference to Figure 4.

[0168] If this is the case (e.g., if the estimated linear acceleration during the predetermined period is greater than the first threshold value, while the measured linear acceleration is less than the second threshold value during the same period), a step S240 is implemented in which the torque of the wheel driven by the aforementioned wheel motor is controlled. More precisely, a torque T i The application to be carried out by the wheel motor is determined.

[0169] As an example, consider a four-wheeled vehicle traveling in a straight line, with all four wheels aligned. Assume that during its journey, one of the wheels rolls onto a patch of black ice and begins to "slip." In other words, the angular velocity ω of this wheel increases significantly, without a corresponding increase in the vehicle's linear speed. Since the vehicle maintains a relatively constant speed, the linear acceleration detected by the inertial measurement unit (IMU) in a horizontal plane and in the direction of travel is low or even zero. In this particular implementation, the control device 200 detects this difference between the linear acceleration determined from the inertial data and the speed estimated by measuring the wheel motor's rotation, and adjusts the torque accordingly.

[0170] Figure 13 shows modules embedded in a torque controller, according to a particular embodiment of the invention. This torque controller is configured to control the torque of a single wheel motor, and for example embedded within that wheel.

[0171] As illustrated in Figure 13, this torque controller includes:

[0172] - a MOD_OBS_SL slope observation module connected to a TCH tachometer, and to an IMU inertial measurement unit;

[0173] - a MOD_OBS_AT altitude observation module connected to the MOD_OBS_SL slope observation module and to the IMU inertial measurement unit;

[0174] - a MOD_PRED module for predicting estimated rotation speeds, connected to the MOD_OBS_SL module for observing a slope, the MOD_OBS_AT module for observing an altitude and a MOD_OBS_TR module for determining a traction force.

[0175] - and a MOD_DEC decision module connected to the MOD_PRED prediction module, as well as to the MOD_OBS_SL slope observation module.

[0176] In a particular implementation mode, the MOD_OBS_SL slope observation module takes as inputs wheel rotation speeds tOj, as well as a raw acceleration a RAW measured by an inertial measurement unit (IMU). This MOD_OBS_SL slope observation module is configured to generate, from these inputs, a so-called "refined" acceleration (resp. a refined rotational velocity (tOj)), that is, an acceleration devoid of the projection term of the support reaction opposing gravity. This refined acceleration is determined, for example, by applying a Kalman filter to the raw acceleration a RAW measured by the IMU inertial measurement unit. This MOD_OBS_SL slope observation module is also configured to determine longitudinal accelerations. Land transversal a T .

[0177] The MOD_OBS_AT altitude observation module takes these longitudinal accelerations as input. L and transversal a T , as well as wheel accelerations R0T measured by the IMU inertial measurement unit. It is configured to estimate the roll R and pitch P of the part fixed to the wheel chassis, for example by applying a complementary filtering algorithm or Kalman filtering to these input data.

[0178] The MOD_PRED prediction module, meanwhile, takes longitudinal accelerations as input. L and transversal a TThe estimated roll (R) and pitch (P), a traction force determined by the MOD_OBS_TR module, and vehicle parameters such as mass, wheel radius, etc., are used. This MOD_PRED module is configured to predict different wheel accelerations (or different rotational speeds). OYEST wheel) corresponding to different adhesion conditions, using a set of data maintained over a horizon of several time steps.

[0179] In specific implementation modes, this MOD_PRED prediction module includes a machine learning model, for example implemented in the form of neural networks (convolution, perceptron, autoencoder, recurrent, etc.).

[0180] Furthermore, it is important to note that there are no limitations on the type of training technique used to obtain this machine learning model. Any technique implementing a learning algorithm (or "machine learning" in English terminology) and providing, as output, an estimated acceleration (or rotational speed) of the wheel as a function of grip conditions can be considered within the context of the invention (for example, support vector machines, logistic regression, etc.). In other words, the learning model is independent of the training method used to train it.

[0181] In addition, any training criterion known to a person skilled in the art can be considered during the training phase of this machine learning model, such as the least squares method or cross-entropy minimization.

[0182] Finally, the MOD_DEC decision module takes these different accelerations (respectively, rotational speeds tOj) as input. £ST ) of the predicted / estimated wheel, as well as the "refined" acceleration (resp. the refined rotational speed (tOj)), to estimate the slip / slip of this wheel, and thus apply a torque control algorithm, for example similar to that described with reference to Figure 5.

Claims

Demands

1. A method for controlling the torque of a wheel motor (10) of a vehicle (1000), the method being implemented by a torque controller (200) mounted within the vehicle and comprising: - a determination (S130, S230) of an effective slip of at least one wheel (100) of the vehicle (1000) driven by said wheel motor, implemented without obtaining, by the torque controller, a data representative of a vehicle orientation; and,; - a control (S170, S240) of the torque of said wheel motor, as a function of the determination of an effective slip of said wheel.

2. A method of torque control according to claim 1, the vehicle (1000) comprising a plurality of wheels (100) equipped with a wheel motor (10), the controller being configured to control the torque of each of the wheel motors.

3. A method for torque control according to claim 2, comprising a determination (S130) of a value ( / q) representative of the effective slip of each of the wheels as a function of a steering radius (R) of the vehicle, and a determination (S120) of said steering radius (R) as a function of representative wheel speed data (100) of the plurality and as a function of at least one dimension of said vehicle.

4. A torque control method according to claim 3, wherein at least one dimension of said vehicle comprises a radius value of a wheel of the plurality, at least one track value and / or at least one wheelbase value of the vehicle (1000).

5. A method for torque control according to claim 2, comprising a determination (S130) of a value ( / q) representative of the effective slip of each of the wheels as a function of a steering radius (R) of the vehicle, and obtaining said steering radius (R) from an association table linking a steering radius to at least one dimension of said vehicle.

6. A method for torque control according to any one of claims 3 to 5, the value ( / q) representing the effective slip being further determined as a function of the rotational speeds ω of the wheel motors and a reference speed (VRE) F ) wheels of plurality.

7. A method for torque control according to any one of claims 3 to 6, further comprising a determination (S160), for each of the wheel motors (10) of the vehicle (1000), of a limiting torque as a function of the value (λj) representing the effective slip (λj) of said wheel (100), and as a function of a value representing an optimal slip k SET ) of said wheel (100).

8. A method of torque control according to claim 7, further comprising an allocation (S170) of torque to each wheel motor of the vehicle, as a function of an overall torque setpoint and said limit torques.

9. A method for controlling torque according to claim 8, the overall torque setpoint being determined as a function of a setpoint (V SET ) of vehicle speed and a speed (V VEC estimated ) of the vehicle (1000), the method further comprising an estimation (S140) of the speed (V VEC) of the vehicle as a function of the turning radius (R) and the reference speed (V REF ) wheels of plurality.

10. A method of torque control according to claim 1, the vehicle (1000) comprising a plurality of wheels (100) equipped with a wheel motor (10), and the controller is configured to control the torque of a single wheel motor of one wheel (100), without obtaining data representative of the rotational speeds of the other wheel motors of the vehicle.

11. A torque control method according to claim 10, wherein the determination (S130, S230) of effective slip comprises a determination that the rotational speed of the controlled wheel motor increases significantly during a predetermined period, while the linear acceleration of said vehicle is less than a threshold value during said period.

12. A control method according to claim 11, the linear acceleration of said vehicle being determined by an inertial measurement unit equipping said vehicle.

13. Computer program (PROG) comprising instructions for implementing a torque control method according to any one of claims 1 to 12, when said program is executed by a processor.

14. Torque controller (200) of at least one wheel motor (10) of a vehicle (1000), the controller being configured to implement a torque control method according to any one of claims 1 to 12.

15. Vehicle (1000) comprising a plurality of wheel motors (10), and in which is mounted a torque controller (200) according to claim 14.

Images