Method for adapting the longitudinal speed of a remotely operated land vehicle to a curved trajectory

Abstract

Method for adapting the lateral acceleration of a remotely operated land vehicle to a curved trajectory. One aspect of the invention relates to a computer-implemented method (2) for adapting the lateral acceleration of a land vehicle (V) to a curved trajectory (C) on a surface, the vehicle being remotely controlled. Method (2) comprises a set of iterations for which, in each iteration, optimal longitudinal speed and optimal steering angle setpoints are obtained by solving a constrained finite-horizon optimal control problem. Figure to be published with the abstract: Figure 4.

Description

Title of the invention: method for adapting the longitudinal speed of a remotely operated land vehicle to a curved trajectory TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of teleoperation and robotics.

[0002] The present invention relates to a method for adapting a lateral acceleration of a land vehicle to a curved trajectory and a system for adapting a lateral acceleration of a remotely controlled land vehicle. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0003] In the field of robotics and land, air and / or maritime drones, teleoperation is the primary function for controlling vehicle mobility. Teleoperation consists of remotely controlling devices such as robots or drones / uninhabited vehicles. Teleoperation allows, for example, access to land areas inaccessible to human operators, places with high temperatures, high altitude areas, or areas to be explored, particularly in an underwater or space environment, for example. Thus, teleoperation allows an operator to carry out tasks remotely in potentially hostile environments.

[0004] Remotely operated devices generally comprise on-board computers which are controlled by control systems (at a control-command station) comprising a remote computer and a device such as a joystick, a controller, a tablet or a smartphone. In particular, the computers and the control systems are linked via wireless communication links of the radio link type according to protocols such as Bluetooth® or Wi-Fi® for example.

[0005] In particular, there are several types of remotely operated ground vehicles, including unmanned ground vehicles (UGVs) used in places where a human presence is impossible or unwise and enabling tasks such as reconnaissance or surveillance to be carried out, particularly in the military field for example. In addition, there are also remotely operated ground vehicles enabling soils to be probed using dedicated sensors and samples to be taken.

[0006] One of the main disadvantages of the teleoperation of land vehicles is the difficulty in maintaining the lateral stability of the vehicles when the trajectories are curved. Indeed, since the vehicles are controlled by teleoperators who are not in the vehicle, they do not physically feel the dynamics of these vehicles, i.e. their speeds and accelerations, and therefore cannot precisely adapt the speed and acceleration commands in real time. A Uncontrolled lateral acceleration of the vehicle when cornering can cause vehicles to skid or even roll over laterally.

[0007] Several solutions implemented to overcome this drawback are described below.

[0008] A first solution is presented in patent US11305767B2, entitled "Controlling movement of a vehicle", which describes a method consisting of modifying the maximum speed controllable by a remote operator according to a real-time estimate of the loss of grip of the tires on the road. The estimate is based on two variables: a first variable corresponding to the force applied by the steering actuator on the wheels on the one hand and a second variable corresponding to the normal force applied by the road on the wheels on the other hand. The first variable is a measurement by a sensor included in the vehicle's steering system.The second variable is the result of an estimation from several parameters and measurements such as: vehicle speed, vehicle acceleration or deceleration, vehicle weight, vehicle weight distribution, wheel load, wheel speed, tire pressure and / or vehicle orientation. The disadvantage of this solution is the need to have a multitude of different parameters and measurements mentioned above available to estimate wheel grip, which requires a plurality of sensors and complicates the calculations.

[0009] Another solution is presented in patent application US20230060236A1, entitled "Notification System and notification method", in which a method is described which uses longitudinal and lateral acceleration measurements of a remotely operated vehicle. When the lateral acceleration of the vehicle exceeds a first threshold, a notification is sent to the remote operator who is piloting the vehicle, the notification being a vibration emitted on the steering wheel of his cockpit. The operator then adjusts his controls via an accelerator pedal and the steering wheel to modify the speed and the steering angle. When, even after the operator's intervention to adjust the lateral acceleration, it continues to increase and exceeds a second threshold, an automatic limitation of the vehicle's speed is carried out.Disadvantages of this solution are the need to have real-time lateral acceleration measurement available and the necessary intervention of the teleoperator so that the latter can adjust his controls.

[0010] There are also solutions for managing lateral stability in a turn for non-remotely operated vehicles such as forklifts. Safety systems called "active stability gain systems" help to reduce the risk of lateral rollover. These systems significantly increase the level of safety for the operator on board the forklift.

[0011] Thus, there is a need to adapt the longitudinal and lateral controls of a remotely operated ground vehicle and to manage the lateral stability of the vehicle in a bend, from a solution requiring the fewest possible data sensors on the vehicle, and without real-time intervention from the teleoperator to adjust the vehicle controls. Summary of the invention

[0012] The invention provides a solution to the problems mentioned above, by making it possible to automatically adapt the lateral acceleration of a remotely operated land vehicle in a curved trajectory on the ground, taking into account at each given period different limitations linked in particular to the ground, to the properties of the vehicle and to the speed of the vehicle at the previous period. The invention comprises a set of iterations for which, in each iteration, optimal longitudinal speed and optimal steering angle setpoints are preferably obtained by solving an optimal control problem with a finite horizon under constraint. Thus, the invention makes it possible to automatically filter the longitudinal and lateral setpoints to avoid the vehicle overturning during a turn.

[0013] One aspect of the invention relates to a computer-implemented method of adapting a lateral acceleration of a land vehicle to a curved trajectory on ground, the land vehicle being remotely controlled, the method comprising, for each given iteration of a set of iterations of the method: • Determine, from a longitudinal speed setpoint intended for the vehicle and a steering angle setpoint of at least one wheel of the vehicle, an optimal longitudinal speed setpoint intended for the vehicle and an optimal steering angle setpoint intended for the wheel of the vehicle, the determination comprising: • Determining an optimal longitudinal acceleration prediction of the vehicle and an optimal steering angle prediction of the vehicle wheel by solving an optimization problem, the optimization problem being a function of a longitudinal speed prediction of the vehicle, the longitudinal speed setpoint of the vehicle, a steering angle prediction of the vehicle wheel, the steering angle setpoint of the wheel, a longitudinal acceleration prediction of the vehicle and a steering speed prediction of the vehicle wheel, the optimization problem satisfying: • a first constraint depending on a longitudinal speed value of the vehicle obtained in the previous iteration, • a second constraint depending on a steering angle value of the vehicle wheel obtained in the previous iteration, and • a third constraint on a prediction of a lateral acceleration of the vehicle depending on the properties of the ground, • Determination of the optimal longitudinal speed setpoint from an optimal longitudinal speed setpoint determined at the previous iteration, a sampling time Te, and the prediction of optimal longitudinal acceleration of the vehicle, • Determination of the optimal steering angle setpoint from an optimal steering angle setpoint determined at the previous iteration, the sampling time Te of the given iteration, and the prediction of the optimal steering speed of the vehicle wheel.

[0014] The term "remotely controlled land vehicle" means a land vehicle without a human operator inside the vehicle who controls the speed of the vehicle, and who receives commands for speed, acceleration, steering from a pilot located outside said vehicle. The remote controlled land vehicle can also optionally receive one or more additional commands, for example a yaw type command.

[0015] The invention advantageously makes it possible to adapt the longitudinal speed of the vehicle and the steering angle of the wheels in an optimal manner when a teleoperator commands a turning movement, so as to limit the lateral acceleration and to avoid the overturning of the teleoperated vehicle. In particular, when the steering control of the wheels is modified by a teleoperator in order to negotiate the curved trajectory, the invention makes it possible to automatically modify the longitudinal speed setpoint by reducing it in an optimal manner (and in particular by respecting the braking capacities of the vehicle) so as not to exceed the authorized lateral and longitudinal accelerations.

[0016] When the land vehicle resumes a non-curved trajectory, in particular a straight trajectory, the invention optionally makes it possible to automatically re-increase the longitudinal speed commanded to the vehicle up to the value requested by the teleoperator.

[0017] In addition to the characteristics which have just been mentioned in the preceding paragraph, the method according to one aspect of the invention may have one or more complementary characteristics among the following, considered individually or according to all technically possible combinations.

[0018] According to one embodiment, the optimization problem further satisfies a constraint on the prediction of the longitudinal acceleration of the vehicle and / or a constraint on the prediction of the steering speed of the wheel of the vehicle.

[0019] According to one embodiment, the first constraint is a constraint linking the vehicle longitudinal velocity prediction and vehicle longitudinal acceleration prediction, and the second constraint is a constraint linking wheel steering angle prediction and wheel steering speed prediction.

[0020] According to one embodiment, the method according to the invention comprises, before determining the optimal longitudinal speed setpoint and the optimal steering angle setpoint: • Receipt of a preliminary longitudinal speed instruction intended for the vehicle and a preliminary steering angle instruction intended for at least one wheel of the vehicle, • Determination : • An intermediate longitudinal speed instruction corresponding to the preliminary longitudinal speed instruction received or to a maximum longitudinal speed instruction of the vehicle, • An intermediate steering angle instruction corresponding to the preliminary steering angle instruction received or to a maximum steering angle instruction of the wheel, The longitudinal speed setpoint corresponding to the intermediate longitudinal speed setpoint and the steering angle setpoint corresponding to the intermediate steering angle setpoint.

[0021] Advantageously, when a teleoperator issues a preliminary longitudinal speed instruction or a preliminary steering angle instruction, these can be modified so as not to exceed respectively a maximum longitudinal speed or a maximum steering angle imposed by the capabilities of the vehicle for example.

[0022] According to one embodiment, the method according to the invention comprises at each iteration: • Control of the vehicle from the optimal longitudinal speed setpoint determined at said iteration, and from the optimal steering angle setpoint determined at said iteration, so as to adapt the lateral acceleration of the vehicle to the curved trajectory on the ground.

[0023] According to one embodiment, the method according to the invention comprises, after controlling the vehicle: • Receiving a measurement of the vehicle's longitudinal speed and a measurement of the vehicle's wheel steering angle.

[0024] According to one embodiment, the longitudinal speed value obtained at the previous iteration corresponds to a measurement of longitudinal speed of the vehicle received at the previous iteration and the steering angle value obtained at the previous iteration corresponds to a measurement of steering angle of the wheel of the vehicle received at the previous iteration.

[0025] Advantageously, this characteristic makes it possible to take into account the real dynamics of the vehicle, when solving the optimization problem and to obtain optimal longitudinal speed and optimal steering angle instructions, which makes it possible to have feedback on the real state of the vehicle.

[0026] According to one embodiment, the longitudinal speed value obtained at the previous iteration corresponds to an optimal longitudinal speed setpoint of the vehicle obtained at the previous iteration and the steering angle value obtained at the previous iteration corresponds to an optimal steering angle setpoint of the wheel of the vehicle obtained at the previous iteration.

[0027] Advantageously, in this embodiment, no measurement from the vehicle is necessary for calculating the optimal longitudinal speed and optimal steering angle setpoints. Thus, it is not necessary to obtain in this embodiment measurements from the sensors (odometers and / or inertial unit for longitudinal speed, angular encoders of the steering for the steering angle, inertial unit for heading / yaw speed).

[0028] According to one embodiment, the given iteration is denoted given iteration K, K being a non-zero natural integer, and the optimization problem is solved by minimizing a criterion, denoted Crit, over a finite prediction horizon denoted [K+1; K+N] of which each element K+i corresponds to a given iteration, with i between 1 and N, N being a non-zero natural integer, the criterion Crit to be minimized corresponding to the following quadratic sum: 100291 cry, = - *42 + -5j2 + 2 + w2 'with: * v%term 'a intermediate longitudinal speed setpoint; • a longitudinal velocity prediction at iteration K+i; • 5™term the intermediate steering angle setpoint; • a steering angle prediction at iteration K+i; • a prediction of longitudinal acceleration at iteration K+i; • a steering speed prediction at iteration K+i; • ^v, ^5, 7 ü weighting coefficients..

[0030] According to one embodiment, the time relating to each iteration is noted Te, and for each iteration K+i, the plurality of constraints is noted: * vli ~ d'T^a^, with rio equal to the longitudinal velocity value obtained in the previous iteration (Kl); • ô, — i + Te*ÿ> with ô0 equal to the steering angle value obtained in the previous iteration (Kl); * kl - with corresponding to a longitudinal acceleration maximum vehicle capacity; • |©1<Ô , corresponding to a maximum yaw acceleration of the h ^max vehicle; * \f{alat with ai“a prediction of lateral acceleration of the vehicle at " r lat ' the iteration K+i, f a function of the vehicle lateral acceleration prediction at iteration K+i and a maximum vehicle lateral acceleration, ' depending on the soil properties..

[0031] Another aspect of the invention relates to a system for adapting a lateral acceleration of a remotely controlled land vehicle to a curved trajectory on the ground, the system comprising a control station, located outside the land vehicle and a control unit, on board the land vehicle, the system being configured to implement the adaptation method according to one aspect of the invention.

[0032] Another aspect of the invention relates to a computer program comprising instructions which, when the program is executed by a computer, cause the latter to implement the method according to one aspect of the invention.

[0033] The invention and its various applications will be better understood upon reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0034] The figures are presented for information purposes only and in no way limit the invention. • [Fig.l] is a schematic representation of a land vehicle V following a curved trajectory C. • [Fig.2] represents a system for adapting lateral acceleration of a remotely controlled land vehicle according to an exemplary embodiment of the invention. • [Fig.3] is a block diagram illustrating the sequence of steps of a method for adapting lateral acceleration of a remotely controlled land vehicle in a curved trajectory according to a first embodiment. • [Fig.4] is a block diagram illustrating in detail a step of determining an optimal longitudinal acceleration setpoint and an optimal steering angle setpoint according to the first embodiment. • [Fig.5] is a block diagram illustrating the sequence of steps of a method for adapting lateral acceleration of a remotely controlled land vehicle in a curved trajectory according to a second embodiment. • [Fig.6] is a block diagram illustrating in detail a step in determining an optimal longitudinal acceleration setpoint and a optimal steering angle setting according to the second embodiment. DETAILED DESCRIPTION

[0035] The present invention is placed in the context of the remote operation of land vehicles and more particularly in curved trajectories.

[0036] By "land vehicle" is meant any motorized platform intended to travel on ground, the platform comprising at least one front wheel and one rear wheel, the front wheel and the rear wheel being aligned parallel to a longitudinal axis of the vehicle, propulsion actuators and steering motors of at least one wheel, brakes and control means intended to receive a command intended for the wheels, the motors and the brakes. The control means will be described in more detail below.

[0037] In particular, when a vehicle enters a bend, it is necessary to adapt its longitudinal speed and the steering angle of its wheels in order to avoid any risk of instability, of deflection due to centrifugal force or of wear of the tires for example due to excessive lateral acceleration. However, an operator or driver who controls the vehicle remotely does not physically feel the lateral and longitudinal accelerations of the vehicle in a bend and cannot adapt its speed (in particular the longitudinal speed and the steering angle of the wheels) optimally to negotiate the bend.

[0038] Thus, the invention aims to automatically and optimally adjust the speed and steering angle of the wheels of a remote-controlled vehicle in a bend, so as to adapt the lateral acceleration of the vehicle to the curved trajectory and to avoid any risk of instability of the vehicle.

[0039] The invention finds an application for any vehicle of the unmanned surface vehicle type, commonly called UGV (from the English: "unmanned ground vehicle"). More generally, the invention can be applied to any remotely operated rolling platform, in particular in the field of terrestrial robotics.

[0040] In order to describe the invention, a kinematic model of the vehicle is introduced below.

[0041] [Fig.l] represents a vehicle V having a curved trajectory C and its various kinematic parameters. In particular, the curved trajectory C will also be called turn C in the following.

[0042] The vehicle V is represented in a Galilean Cartesian frame of reference with axes: X, Y, Z and origin O. In particular, the plane (O, X, Y) represents the ground on which the vehicle V rolls.

[0043] In this example, the vehicle V has four wheels, including two front wheels and two rear wheels.

[0044] As previously described, the vehicle V comprises propulsion actuators and steering actuators, also called steering actuators, not shown.

[0045] The vehicle comprises a front axle Eav and a rear axle E^.

[0046] [Fig.l] represents kinematic parameters of the vehicle V and in particular: a longitudinal speed vi of the vehicle V, a steering angle ô of the front wheels 101, a yaw angle (or heading) of the vehicle 9, a yaw speed (or heading) 9 of the vehicle, a longitudinal acceleration ai and a lateral acceleration aim of the vehicle and a wheelbase L of the vehicle V. Alternatively, the invention can also make it possible to calculate a steering angle of the rear wheels 102 instead of the steering angle ô of the front wheels.

[0047] The wheelbase L is the distance between the front axle Eav and the rear axle Ear of the vehicle.

[0048] The longitudinal speed D corresponds to a speed of the vehicle whose direction is tangent to its trajectory, which in this case corresponds to bend C.

[0049] The yaw angle θ corresponds to an angle of rotation of the vehicle around the Z axis, relative to the X axis. In particular, in this case, that is to say when the vehicle is engaged in the bend C, the yaw angle θ corresponds to an angle between the direction of the longitudinal speed D and the X axis.

[0050] The yaw rate 9 corresponds to the rotation speed of the vehicle around the Z axis, which corresponds to the axis of revolution of the turn C. In particular, the steering actuators make it possible to control the yaw rate 9.

[0051] The steering angle δ of the front wheels corresponds to an angle formed by the projection of a longitudinal axis Al of the vehicle V and an axis I representing the line of intersection of the plane of one of the front wheels corresponding to the central plane of the tire, perpendicular to the axis of rotation of the wheel, and the plane (O, X, Y) representing the ground. In particular, the wheel steering actuators make it possible to control the steering angle δ.

[0052] The longitudinal acceleration ai corresponds to the acceleration of the vehicle whose direction is tangent to the curved trajectory C of the vehicle V.

[0053] The lateral acceleration al«t corresponds to the acceleration of the vehicle whose direction is perpendicular to the curved trajectory C of the vehicle V and in particular to the longitudinal speed

[0054] As described previously, control of lateral acceleration is essential to keep the vehicle V in its lane when negotiating a bend.

[0055] The lateral dynamics, i.e. more particularly the lateral acceleration and the yaw rate / steering angle of at least one wheel of a land vehicle is commonly modeled from a so-called bicycle model. The bicycle model is based on the following assumptions: the vehicle is symmetrical and the Roll (around the X axis) and pitch (around the Y axis) movements are neglected. The model is characterized by a representation of the four wheels of the vehicle by an equivalent front wheel on the front axle and an equivalent rear wheel on the rear axle. In the present example, the lateral dynamics of the vehicle are modeled by the bicycle model.

[0056] In particular, at low speed, the kinematic bicycle model is advantageously used: in this model, the speed of each wheel of the vehicle is assumed to have the same direction as said wheel. By low speed is meant, for example, a speed less than or equal to 50 kilometers per hour and preferably a speed less than or equal to 30 kilometers per hour.

[0057] The bicycle model is applied to the rear axle E.,r of the vehicle V.

[0058] We denote x(t) and y(t) the positions of the rear axle E^ of the vehicle V respectively along the X axis and along the Y axis. In order to simplify the positions x(t) and y(t) are denoted x, y.

[0059] In particular, the heading angle (or yaw) qXO of the vehicle depends on time but is noted to simplify the notations.

[0060] We note t the time variable.

[0061] The kinematic equations of the Bicycle model in the (O, X, Y) plane are as follows:

[0062] X — V^cns^q)), x being the time derivative of x.

[0063] y — Y being the time derivative of y.

[0064] ¢) = $ being the time derivative of (P and corresponding to the yaw rate (or heading) introduced previously.

[0065] Thus, the lateral acceleration ai«t depends on the longitudinal speed vi of the vehicle and the steering angle 5. Therefore, remote control of the vehicle with optimal longitudinal speed vi and steering angle ô values ​​makes it possible to adapt the lateral acceleration aiat to the negotiated bend in order to reduce any risk of instability of the vehicle for example.

[0066] One aspect of the invention thus relates to a method of adaptation, implemented by computer, of a lateral acceleration of a land vehicle to a curved trajectory on the ground, the land vehicle being controlled remotely.

[0067] In particular, the lateral acceleration of the land vehicle is adapted by determining optimal setpoint values ​​vi and ô to control the vehicle and thus negotiate the curved trajectory.

[0068] In the following, the terminologies "remotely controlled" and "remote-controlled" are used interchangeably.

[0069] In particular, the remotely operated land vehicle corresponds to the vehicle described above. and also preferably includes a plurality of sensors such as: a longitudinal speed sensor, a wheel steering angle sensor, in particular the front wheels, optionally a yaw rate (or heading) sensor, and a GPS position sensor.

[0070] The longitudinal speed sensor may comprise one or more odometers and / or an inertial unit.

[0071] The yaw rate sensor may comprise an inertial unit which corresponds to the inertial unit included in the longitudinal speed sensor.

[0072] The wheel steering angle sensor may comprise one or more angular encoders.

[0073] Further, the ground vehicle may include a camera acquiring a video stream.

[0074] Another aspect of the invention relates to an acceleration adaptation system lateral of a remotely controlled land vehicle in a curved trajectory, making it possible to implement the method according to the invention.

[0075] [Fig.2] represents the lateral acceleration adaptation system 1 according to an example of the invention.

[0076] The adaptation system 1 comprises a control station 11, external to the vehicle, and a control unit 12 intended to be installed in a land vehicle to be controlled remotely.

[0077] In particular, the control station 11 and the control unit 12 each have an antenna 111, 121 respectively, the antenna 111 and the antenna 121 being linked together by a wireless link, for example a radio link or a satellite link.

[0078] The control station 11 may comprise a computer (not shown) comprising a processor.

[0079] The control station 11 comprises a communication interface 112 with a human operator for example, also called pilot or teleoperator in the following, who can provide instructions Ce intended for the control unit 12 and therefore for the vehicle to be controlled remotely (not shown). In particular, the instructions Ce correspond to values ​​of longitudinal speeds and steering angles of wheels of the vehicle to be controlled remotely.

[0080] The control unit 12 comprises for example a computer 122 comprising a processor.

[0081] The computer 122 is configured to receive the instructions Ce sent by the control station 11 to the vehicle in which the control unit is installed.

[0082] Furthermore, the computer 122 is configured to modify the instructions Ce using the method according to the example set out below in order to obtain output instructions Cs directly transmitted to the vehicle.

[0083] Furthermore, the computer 122 of the control unit 12 is configured to receive

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090]

[0091]

[0092]

[0093]

[0094]

[0095] Capt data from the plurality of sensors included in the vehicle in which the control unit 12 is embedded. Thus, the adaptation system 1 according to the invention makes it possible to implement the method according to the invention comprising a plurality of steps described below. In particular, a plurality of iterations of the method according to the invention can be executed along the trajectory, to allow the vehicle to follow for example a curved trajectory. Each iteration includes a plurality of steps to adapt the lateral acceleration of the vehicle at each iteration. The iteration set includes for example M iterations, where M is a non-zero natural integer. The iterations of the set of iterations are carried out sequentially and are spaced apart in time by a time Te corresponding for example to a sampling time of the computer 122 of the on-board control unit 12 in the vehicle. Thus, the steps of the method according to the invention are repeated every period of duration Te, for example every period of duration Te while the vehicle is in the curved trajectory C. Consider a given iteration K in progress, among the set of iterations, K being a natural integer between 1 and M. For the given iteration K, a block diagram of the steps of the method called 2 according to a first embodiment is shown in [Fig.3]. The first embodiment of method 2 is called open loop mode. At iteration K, the method optionally comprises a step 21 of reception, by the control unit 12 on board the land vehicle, of a preliminary longitudinal speed instruction and a preliminary steering angle instruction of at least one wheel of the vehicle. For example, if the land vehicle has a front wheel and a rear wheel in the same plane, the preliminary steering angle setpoint gref is k intended for the front wheel and / or the rear wheel. For example, if the land vehicle has two front wheels and two rear wheels, the preliminary steering angle setting here is for the two front wheels. Alternatively or in combination, the preliminary steering angle setting is for both rear wheels. In particular, a preliminary steering angle setpoint g^f of at least one wheel of the vehicle is equivalent to a preliminary yaw rate setpoint ç>^ of the vehicle. Indeed, according to the kinematic formulas described previously, the speed yaw angle (p) of a vehicle and the steering angle 5 of the vehicle's wheels are related by the following formula: $ = ^tan(ô)-

[0096] Preliminary longitudinal speed and steering angle instructions are for example given by the pilot at the control station 11 which transmits them to the control unit 12 on board the vehicle, and more particularly to the computer 122 of the control unit.

[0097] After the step 21 of receiving preliminary instructions, the method 2 comprises an optional step 22 of determining an intermediate longitudinal speed instruction Vl^erm and an intermediate steering angle instruction Ql£Ierm.

[0098] In particular, the determination step 22 is carried out by comparing the preliminary longitudinal speed setpoint with a maximum longitudinal speed vmax and by comparing the preliminary steering angle setpoint with a maximum steering angle Ômax.

[0099] The maximum longitudinal speed vmax is a value imposed for example by the capabilities of the propulsion actuators of the remotely controlled vehicle. It can only be configured by the pilot in the direction of reduction.

[0100] The steering angle ômax is a value imposed for example by the capabilities of the steering (or steering) actuators of the remotely controlled vehicle. It can only be configured by the driver in the direction of reduction.

[0101] When a first condition Cl is fulfilled, that is to say when the preliminary longitudinal speed setpoint is greater than or equal to vmax, then the intermediate longitudinal speed is equal to v»iax.

[0102] When the first condition Cl is not fulfilled, that is to say when the preliminary longitudinal speed setpoint is strictly less than vmax, then the intermediate longitudinal speed setpoint is equal to v1^.

[0103] When a second condition C2 is fulfilled, that is to say when the preliminary steering angle setpoint xre / is greater than or equal to Smax, then the intermediate steering angle setpoint ^7 is equal to ÔfWlx.

[0104] When the second condition C2 is not fulfilled, that is to say when the steering angle setpoint ^^is less than ômax, then the intermediate steering angle setpoint g"'ferm is equal to .

[0105] Advantageously, the determination step 22 makes it possible to modify and limit the value preliminary longitudinal speed and steering angle instructions xref sent by the pilot, if these exceed predefined maximum thresholds, in order to avoid any damage due to actuator runaway for example.

[0106] Method 2 comprises a step 23 of determining optimal longitudinal speed and optimal steering angle instructions to adapt the lateral acceleration aiot of the vehicle to the negotiated curved trajectory C.

[0107] The optimal longitudinal speed setpoint is determined from a longitudinal speed setpoint preferably corresponding to the intermediate longitudinal speed setpoint V^crm.

[0108] The optimal steering angle setpoint is determined from a steering angle setpoint preferably corresponding to the intermediate steering angle setpoint Ql”tenn.

[0109] The determination step 23 comprises sub-steps represented in a block diagram in [Fig.4].

[0110] The determination step 23 comprises a sub-step 231 of determining a prediction of an optimal longitudinal acceleration of the vehicle and a prediction of an optimal steering speed ofpt of the vehicle.

[0111] The prediction of an optimal longitudinal acceleration tz^ of the vehicle and the prediction of an optimal steering speed ofpt of the vehicle are determined by solving a constrained optimization problem.

[0112] In particular, the constrained optimization problem is a constrained optimal control problem with a finite prediction horizon.

[0113] In particular, in this embodiment, the optimization problem is a function of at least one prediction of longitudinal speed of the vehicle, of the longitudinal speed setpoint, of at least one prediction of longitudinal acceleration of the vehicle, of at least one prediction of steering speed of at least one wheel of the vehicle, and of at least one prediction of steering angle of the wheel.

[0114] Preferably, the optimization problem is a function of: a sequence of longitudinal acceleration predictions (a / 0i <i<N, une séquence de prédictions de vitesse de braquage une séquence de prédictions de vitesse longitudinale (V / 0i<i<Net a sequence of steering angle predictions (ô;)i <i<N la consigne intermédiaire d'angle de braquage Q‘£term et la consigne intermédiaire de vitesse longitudinale

[0115] By optimal control problem with a finite prediction horizon we mean a problem for which we seek optimal prediction sequences which make it possible to minimize a criterion over a finite and discrete interval of instants (horizon).

[0116] In particular, the finite prediction horizon comprises N discrete instants, corresponding respectively to N iterations following the given iteration K in progress.

[0117] Thus, at the given iteration K in progress, the finite horizon corresponds to an interval of instants corresponding respectively to the iterations going from K+l to K+N, K+l cor corresponding to the iteration directly following the current given iteration K.

[0118] In particular, N is a non-zero natural integer. The choice of N falls within the know-how of the person skilled in the art in the field of automation and signal processing. N is, for example, greater than or equal to 3 and less than or equal to 10. In a possible example, N is equal to 5.

[0119] In the sequence of longitudinal acceleration predictions (a'0i£l£\, àl' corresponds to a longitudinal acceleration prediction at iteration K+i.

[0120] In the steering speed prediction sequence (ü'j)i <i<N, correspond à une prédiction de vitesse de braquage à l'itération K+i.

[0121] Thus, in the sequence of longitudinal velocity predictions (v»)i <i<N, correspond à une prédiction d'accélération longitudinale à l'itération K+i.

[0122] In the steering angle prediction sequence (Ô0i <i<N, correspond à une prédiction d'angle de braquage à l'itération K+i.

[0123] In particular, when solving the optimal control problem with a finite horizon, the criterion, noted Crit in the following, is minimized while respecting a plurality of constraints.

[0124] In the first embodiment of method 2, the minimized criterion Crit is a quadratic sum weighted over the finite prediction horizon N, the formula of which is as follows: 101251 c rit - +%(5k“""-8,) 2 +r^'

[0126] In particular, Qv, Qg ra rw are weightings which make it possible to adjust the filtering dynamics of the longitudinal speed and steering angle setpoints and are not configurable by the pilot. By filtering the setpoints is meant in particular the modification of the intermediate longitudinal speed setpoints and intermediate steering angle setpoints so as to obtain the optimal longitudinal speed and optimal steering angle setpoints thanks to the determination step 23.

[0127] The plurality of constraints to be respected when minimizing the criterion Crit is described below, for each iteration K+i, with i between 1 and N.

[0128] A first constraint links a longitudinal velocity prediction and a longitudinal acceleration prediction for the same iteration. The first constraint has the expression: Te* .

[0129] The first constraint also verifies: ■ t „ •

[0130] Thus, the first constraint depends on a prediction of initial velocity vzo.

[0131] In this first embodiment of method 2, v / n — Æ ' being an A lu / (ki) optimal longitudinal speed setpoint determined according to steps 21 to 23 during the previous iteration, i.e. at iteration K-1 preceding the given iteration K in course.

[0132]

[0133]

[0134]

[0135]

[0136]

[0137]

[0138]

[0139]

[0140]

[0141]

[0142]

[0143]

[0144] If the given iteration K is such that K is equal to 1, then v / o is initialized to a value predetermined by the pilot, for example a zero longitudinal speed value. So, when i is equal to 1, v.> = + T»* a>v / 1 {{ki) £ ti A second constraint links a steering angle prediction and a steering speed prediction for the same iteration and has the expression: 5, = + Te*(Oj. The second constraint also verifies: g — g + J" * ■' Thus, the second constraint depends in particular on a prediction of initial steering angle ô0. In this first embodiment of method 2, x — being a optimal steering angle setpoint determined according to steps 21 to 23 at the previous iteration, i.e. at iteration Kl preceding the current given iteration K. So, when i is equal to 1, we have Si - Ôq + If the current given iteration K is such that K is equal to 1, then ô0 is initialized to a value predetermined by the driver, for example a zero steering angle value. The first and second constraints cited above are equality constraints, reformulating on the prediction horizon N kinematic equations relating to the vehicle, in discrete time. In particular, rewriting the criterion Crit by applying the first and second constraints, Crit has the formula: Crit = + qJô!mcH'^ +raan2 +rwWi2 Thanks to the first and second conditions, the criterion Crit therefore only depends on the sequence of longitudinal acceleration predictions, the sequence of steering speed predictions and ô0, v^term and ginterm k A third constraint has the formula: j < ü'i'ux corresponding to a maximum longitudinal acceleration imposed for example by the capacities of the propulsion actuators. The maximum longitudinal acceleration @Tax is only configurable by the operator in the direction of decrease. In particular, the maximum longitudinal acceleration ctfu'x can be positive, which physically corresponds to an acceleration of the vehicle, or negative, which physically corresponds to a deceleration of the vehicle. A fourth constraint has the formula: corresponding to a maximum steering speed, imposed by the capabilities of the steering or steering actuators. It can only be configured by the driver in the direction of decrease.

[0145] A fifth constraint has the formula: ;)| < k / H • f(aiat being a iat * nonlinear function of lateral acceleration of formula vi? such that Z / i defined previously. Thus, = f(vu, ô,) . Preferably, f(aialJ) = aiatJ.

[0146] In particular, corresponds to a maximum lateral acceleration that the vehicle may have, said maximum lateral acceleration being configurable according to the properties of the ground on which the vehicle negotiates a curved trajectory C.

[0147] For example, soil properties correspond to soil states. More particularly, soil properties may correspond to a dry soil state, a wet soil state, or soil comprising grass, for example.

[0148] The maximum lateral acceleration Cl^x is determined and validated following test campaigns for example. By default it is configured at a lowest value, corresponding to the most unfavorable ground condition for the grip of the tires of the vehicle wheels.

[0149] The third, fourth and fifth constraints cited above are inequality constraints, reformulating on the prediction horizon N conditions to be respected by the vehicle in discrete time.

[0150] The criterion Crit is minimized for example with an algorithm by successive quadratic optimization, noted SQP for “Sequential Quadratic Programming” in English.

[0151] Thus, the resolution of the optimization problem makes it possible to obtain a sequence of prediction of longitudinal accelerations (a / 0i <i<N et une séquence de prédictions d'accélérations de lacet (9;)i<i<Nqui minimisent le critère Crit.

[0152] The prediction of the optimal longitudinal acceleration is chosen in the acceleration prediction sequence (a^)i<; <Net est de préférence égale au premier élément de ladite séquence qui est associé à l'itération K+l.

[0153] The element a / i corresponds to the prediction of the longitudinal acceleration of the vehicle at iteration K+L

[0154] The prediction of the optimal steering speed is chosen in the steering speed prediction sequence (^Oi^^and is preferably equal to the first element of said sequence which is associated with the iteration K+L. The element 6,'[ corresponds to a prediction of the steering angle at the iteration K+L

[0155] Advantageously, only the first term an of the sequence of longitudinal acceleration predictions (a«)i <i<Nest choisi tel que a^1 = a^ car, à l’itération K+l suivant l'itération donnée K en cours, et donc à la période d'échantillonnage suivante, les étapes 21 à 23 sont répétées et l'horizon de prédiction fini est décalé, ce qui correspond au principe d'horizon fuyant.

[0156] Indeed, the sequence of longitudinal acceleration predictions (aft)2 <i<N tel que i est greater than or equal to 2 is not useful because at the next iteration a new longitudinal acceleration prediction sequence (az0i <i<N minimisant le critère Crit est déterminée.

[0157] Thus, the acceleration prediction an corresponding to the iteration K+l following the current given iteration K is taken into account. It is therefore the principle of “Predictive Control” or “Receding Horizon Control” which is used to choose aT-

[0158] A similar reasoning applies to the choice of the optimal steering speed.

[0159] Step 23 comprises a sub-step 232 of determining the optimal longitudinal speed setpoint.

[0160] In particular, the optimal longitudinal speed setpoint is determined from the prediction of the optimal longitudinal acceleration a"pt, a sampling time Te, and the optimal longitudinal speed setpoint obtained at the iteration preceding the current given iteration K, i.e. the KL iteration

[0161] In particular, the sampling time Te corresponds to the duration of an iteration or in other words, to the time relative to each iteration which separates each iteration from a following iteration.

[0162] Preferably, yf / * = y^f\. + T^a°ipî-

[0163] Step 23 further comprises a sub-step 233 of determining the optimal steering angle setpoint.

[0164] In particular, the optimal steering angle setpoint ^pt is obtained from the prediction of the optimal steering speed of the sampling time Te, and from the optimal steering angle setpoint ^'Aob held at the iteration KL

[0165] In particular, the sampling time Te of each iteration corresponds to the duration of an iteration or in other words, to the sampling time Te separating each iteration from a following iteration.

[0166] Preferably,

[0167] Equivalently, and as described previously, the determination of the optimal steering angle setpoint ^pt is equivalent to a determination of an optimal yaw rate setpoint

[0168] Referring again to [Fig.3], the method 2 further comprises a step 24 of controlling the vehicle from the optimal longitudinal speed and optimal steering angle instructions so as to adapt the lateral acceleration aiat of the vehicle to the curved trajectory.

[0169] In particular, the control is carried out by transmitting the optimal longitudinal speed instruction to the vehicle and more particularly to the propulsion engines of the vehicle, and by sending the optimal steering angle instruction to the vehicle and more particularly to the vehicle's steering motors.

[0170] Thus, steps 21 to 24 advantageously make it possible to control the vehicle in the curved trajectory by automatically adapting the preliminary instructions issued by the operator in step 21 to conditions relating to the vehicle and the ground, without exceeding the maximum speeds and accelerations authorized for the vehicle.

[0171] Optionally, method 2 may comprise, following the control step 24, a step 25 of receiving a measurement of a longitudinal speed of the vehicle and a measurement of a steering angle of at least one wheel of the vehicle. vehicle.

[0172] The at least one wheel of the vehicle for which the steering angle is measured is the wheel to which the optimal steering angle instruction is issued.

[0173] In particular, the measurement of the longitudinal speed is transmitted by the odometers and / or the inertial unit of the vehicle to the control unit on board the vehicle. The measurement of the longitudinal speed thus corresponds to the actual longitudinal speed of the vehicle at iteration K and after the control step 24.

[0174] In particular, the measurement of the steering angle is transmitted by the angular encoders of the vehicle steering and is emitted to the control unit 12 on board the vehicle. The measurement of the steering angle corresponds to the actual steering angle of the vehicle at iteration K and after the control step 24.

[0175] Equivalently, the measurement of the steering angle can be obtained by measuring the yaw rate of the vehicle. It is recalled that the yaw rate <? d'un véhicule et l'angle de braquage ô des roues du véhicule sont liés par la formule suivante : $ = Jtan(ô)-

[0176] Thus, the first embodiment of the method described in steps 21 to 25 corresponding to the open-loop mode makes it possible to automatically adapt the lateral acceleration of the vehicle to the curved trajectory in an optimal manner while using the least possible data from the vehicle: the optimal instructions are in particular determined without the measurements from the sensors.

[0177] For the given iteration K in progress, a block diagram of the steps of method 2 according to a second embodiment is shown in [Fig.5].

[0178] The second embodiment of method 2 is called closed-loop mode.

[0179] In particular, the second embodiment of method 2 comprises the same

[0180]

[0181]

[0182] steps of reception 21, determination 22, control 24 and reception 25 than those described above in connection with the first mode. In particular, method 2 according to the second embodiment comprises a step 23' of determining optimal longitudinal speed and optimal steering angle instructions for adapting the lateral acceleration of the vehicle to the negotiated curved trajectory, the determination step 23' comprising differences with the determination step 23 of the first embodiment of method 2. A first difference between determination step 23 of the first embodiment and determination step 23' of the second embodiment is the criterion to be minimized when solving the optimal control problem. Indeed, in the determination step 23', the criterion to be minimized includes additional terms corresponding to filters modeling response times of the propulsion and steering angle actuators. These filters are described below.

[0183] A second difference between the determination step 23 and the determination step 23' is the use of the longitudinal speed and steering angle measurements obtained at the previous iteration, and not the optimal longitudinal speed and optimal steering angle setpoints obtained at the previous iteration, to determine the optimal longitudinal speed and optimal steering angle setpoints at the current given iteration K.

[0184] Thus, in the second embodiment of method 2, method 2 thus comprises the step 23' of determining optimal longitudinal speed and optimal steering angle instructions to adapt the lateral acceleration a lat of the vehicle to the negotiated curved trajectory C.

[0185] The optimal longitudinal speed setpoint is determined from a longitudinal speed setpoint preferably corresponding to the intermediate longitudinal speed setpoint

[0186] The optimal steering angle setpoint is determined from a steering angle setpoint preferably corresponding to the intermediate steering angle setpoint

[0187] The determination step 23' comprises sub-steps represented in a block diagram in [Fig.6].

[0188] The determination step 23' comprises a sub-step 231' of determining a prediction of the optimal longitudinal acceleration "? / J / of the vehicle and a prediction of the optimal steering speed ofpt of the vehicle by solving a constrained optimization problem.

[0189] As described in determination step 23, the constrained optimization problem is a constrained finite prediction horizon optimal control problem, the finite prediction horizon corresponding to that of determination step 23.

[0190] Similar to the first embodiment of method 2, the optimization problem is a function of: the sequence of longitudinal acceleration predictions (a,-0i <i<N, la séquence de prédictions de vitesses de braquage (w0i<i^N, la séquence de prédictions de vitesse longitudinale (v«)i<i<Net la séquence de prédictions d’angle de braquage la consigne intermédiaire d’angle de braquage Ql£term et la consigne intermédiaire de vitesse longitudinale v\^erm.

[0191] Furthermore, in this second embodiment, the optimization problem is also a function of a filter modeling the response time of the propulsion actuators and a function of a filter modeling the response time of the steering actuators. Furthermore, in this second embodiment, the optimization criterion is identical to that of the first mode but the equality constraints, describing the dynamics, are modified. In summary, in this second embodiment, a longitudinal acceleration prediction sequence makes it possible to produce a sequence of instructions for the propulsion actuators. Then, a prediction of the longitudinal speeds is obtained by modeling the dynamics of the propulsion actuators by a first-order low-pass filter in discrete time. An identical approach is also used for the steering angles. Finally, the response times to be used are characterized during tests on the vehicle.The plurality of constraints to be respected when minimizing the criterion Crit is described below, for each iteration K+i, with i between 1 and N.

[0192] The first constraint has the expression j _ L- ) $ 'U j * yc , with = vfq + Te* a}i. Thus, corresponds to the prediction of the longitudinal speed of the first embodiment.

[0193] In this second embodiment of method 2, being a longitudinal speed measurement received at step 25 of the previous iteration, i.e. at iteration Kl preceding the current given iteration K.

[0194] If the current given iteration K is such that K is equal to 1, then v«> is initialized to a value predetermined by the pilot, for example a zero longitudinal speed value.

[0195] Thus, when i is equal to 1, Vq = V^^+Te* Hq.

[0196] The second constraint has the expression: g _ 1 _ Zs j * § j + Te* ü\- Thus, 5^ corresponds to the prediction of the steering angle and a steering speed prediction of the first embodiment.

[0197] In this second embodiment of method 2, 50 = being a steering angle measurement received at reception step 25 of the previous iteration, i.e. at iteration K1 preceding the current given iteration K.

[0198] Thus, when i is equal to 1, we have Cq =

[0199] If the current given iteration K is such that K is equal to 1, then Ôq is initialized to a value predetermined by the pilot, for example a zero steering angle value.

[0200] In the second embodiment of method 2, the optimization problem also satisfies the third, fourth and fifth constraints described in the first embodiment of method 2.

[0201] The optimization criterion is minimized using the same methods as those described in the first embodiment of method 2.

[0202] Thus, the resolution of the optimization problem makes it possible to determine the sequence of prediction of longitudinal accelerations (az')i <i<n et la séquence de prédictions d’accélérations lacet dpji^xminimisant le critère d’optimisation.

[0203] The prediction of the optimal longitudinal acceleration a^pt is preferably equal to the first element aH.

[0204] The prediction of the optimal longitudinal acceleration (papt) is preferably equal to the first element lP s.

[0205] Step 23' further comprises a determination sub-step 232' identical to determination sub-step 232 and a determination sub-step 233' identical to determination sub-step 233.

[0206] Advantageously, the second embodiment of method 2 makes it possible to take into account the real dynamics of the vehicle thanks to the measurements of longitudinal speeds and steering angles used to resolve the optimization problem.

[0207] Thus, the computer 122 included in the on-board unit 12 makes it possible to implement the steps of method 2 according to the invention. More particularly, the steps of method 2 are coded in the Matlab or Simulink software for example, and are compiled into executable code in the computer 122 of the on-board unit 12 in the form of a computer code in C language for example.

[0208] When the land vehicle leaves the bend, and therefore no longer has a curved trajectory, the intermediate longitudinal speed and steering angle setpoints are sent by the control unit to the vehicle and respectively to the propulsion and steering actuators. Thus, when the vehicle has a straight trajectory for example, the intermediate setpoints are not modified because the lateral acceleration is zero.

Claims

1. Claims Method (2) for adapting, implemented by computer, a lateral acceleration of a land vehicle (V) to a curved trajectory (C) on a ground, the land vehicle being remotely controlled, the method (2) comprising, for each given iteration of a set of iterations of the method: Determining (23, 23'), from a longitudinal speed setpoint intended for the vehicle and a steering angle setpoint of at least one wheel of the vehicle, an optimal longitudinal speed setpoint intended for the vehicle and an optimal steering angle setpoint intended for the wheel of the vehicle, the determination (23, 23') comprising: • Determining (231, 231') an optimal longitudinal acceleration prediction of the vehicle and an optimal steering angle prediction of the wheel of the vehicle by solving an optimization problem, the optimization problem being a function of a longitudinal speed prediction of the vehicle, the longitudinal speed setpoint of the vehicle, a steering angle prediction of the wheel of the vehicle, the steering angle setpoint of the wheel, a longitudinal acceleration prediction of the vehicle and a steering speed prediction of the wheel of the vehicle, the optimization problem satisfying: a first constraint depending on a longitudinal speed value of the vehicle obtained at the previous iteration, a second constraint depending on a steering angle value of the vehicle wheel obtained at the previous iteration, and a third constraint on a prediction of a lateral acceleration of the vehicle depending on the properties of the ground, Determination (232) of the optimal longitudinal speed setpoint from an optimal longitudinal speed setpoint determined at the previous iteration, a sampling time Te, and the prediction of optimal longitudinal acceleration of the vehicle;

2.

3.

4. Determination (233) of the optimal steering angle setpoint from an optimal steering angle setpoint determined at the previous iteration, the sampling time Te of the given iteration, and the prediction of the optimal steering speed of the vehicle wheel. Method (2) according to the preceding claim in which the problem optimization further satisfies a constraint on the prediction of the longitudinal acceleration of the vehicle and / or a constraint on the prediction of the steering speed of the vehicle wheel. Method (2) according to one of the preceding claims in which: The first constraint is a constraint linking the vehicle longitudinal velocity prediction and the vehicle longitudinal acceleration prediction, and The second constraint is a constraint linking the wheel steering angle prediction and the wheel steering speed prediction. Method (2) according to one of the preceding claims comprising, before the determination (23, 23') of the optimal longitudinal speed setpoint and the optimal steering angle setpoint: Reception (21) of a preliminary longitudinal speed instruction intended for the vehicle and of a preliminary steering angle instruction intended for at least one wheel of the vehicle, Determination (22): From an intermediate longitudinal speed instruction corresponding to the preliminary longitudinal speed instruction received or to a maximum longitudinal speed instruction of the vehicle, From an intermediate steering angle setpoint corresponding to the preliminary steering angle setpoint received or to a maximum steering angle setpoint of the wheel, The longitudinal speed setpoint corresponding to the intermediate longitudinal speed setpoint and the steering angle setpoint cor- responding to the intermediate steering angle setting.

5. Method (2) according to one of the preceding claims comprising at each iteration: - Control (24) of the vehicle from the optimal longitudinal speed setpoint determined at said iteration, and from the optimal steering angle setpoint determined at said iteration, so as to adapt the lateral acceleration of the vehicle to the curved trajectory (C) on the ground.

6. Method (2) according to any one of claims 1 to 5 in which: - The longitudinal speed value obtained at the previous iteration corresponds to a longitudinal speed measurement of the vehicle received at the previous iteration, - The steering angle value obtained at the previous iteration corresponds to a steering angle measurement of the vehicle wheel received at the previous iteration.

7. Method (2) according to one of claims 1 to 5 in which: - The longitudinal speed value obtained at the previous iteration corresponds to an optimal longitudinal speed setpoint of the vehicle obtained at the previous iteration, - The steering angle value obtained at the previous iteration corresponds to an optimal steering angle setpoint of the vehicle wheel obtained at the previous iteration.

8. Method (2) according to the preceding claim in which the given iteration is denoted given iteration K, K being a non-zero natural integer, and the optimization problem is solved by minimizing a criterion, denoted Cnt, on a finite prediction horizon denoted [K+1; K+N] of which each element K+i corresponds to a given iteration, with i between 1 and N, N being a non-zero natural integer, the criterion Crit to be minimized corresponding to the following quadratic sum: C* = + q^'e'm-ô^ with: - ymterm |a intermediate longitudinal speed setpoint; - a longitudinal speed prediction at iteration K+i; - gnterm |a intermediate steering angle setpoint; - a steering angle prediction at iteration K+i; - an a longitudinal acceleration prediction at iteration K+i; - a steering speed prediction at iteration K+i; - 4$, ra, r"' weighting coefficients.

9. System (1) for adapting a lateral acceleration of a remotely controlled land vehicle (V) to a curved trajectory (C) on the ground, the system comprising a control station (11), located outside the land vehicle (V) and a control unit (12), on board the land vehicle (V), the system (1) being configured to implement the adaptation method (2) according to any one of the preceding claims.

10. A computer program product comprising instructions which, when the program is executed by a computer, cause the latter to implement the adaptation method according to any one of claims 1 to 8.

Images