METHOD FOR DETERMINING THE TRAJECTORY OF AN AUTONOMOUS VEHICLE

Description

Technical field of the invention

[0001] The present invention relates to a method for determining the trajectory of an autonomous vehicle. The invention also relates to a system, product, program, and corresponding recording medium. State of the art

[0002] Autonomous vehicles are vehicles capable of controlling their trajectory without the assistance of a driver. To achieve this, autonomous vehicles are equipped with various sensors connected to an electronic control unit. Using the information provided by these sensors, the electronic control unit calculates a "theoretical," or "optimal," trajectory. The electronic control unit then sends commands to a steering system to control the steering of the vehicle's front wheels so that the vehicle follows the theoretical trajectory. The driver no longer needs to steer but simply checks that the vehicle's trajectory is correct. However, the trajectory followed by an autonomous vehicle is often different from the one the driver would have followed if they had been controlling the vehicle themselves.This difference in behavior may surprise or worry the driver and passengers and / or it may generate discomfort by causing the vehicle to undergo unpredictable lateral accelerations.

[0003] Thanks to US publication 2015 / 0166069 A1, we know of a method for determining a vehicle's command, comprising a step of recording a driver's preferences, a step of identifying a predetermined scenario, and then applying a default command based on the preferences and the predetermined scenario. However, such an autonomous driving system does not allow for optimal comfort for the driver and passengers. Furthermore, the driver's confidence in the autonomous steering system, in other words, the acceptability of such an autonomous steering system, remains limited. JP 2009 / 227196 A and US 2017 / 0137033 A1 also describe vehicle control systems. Object of the invention

[0004] The aim of the invention is to provide a method for determining the trajectory of an autonomous vehicle that overcomes the aforementioned drawbacks and improves upon methods known in the prior art. In particular, the invention enables the implementation of a method for determining the trajectory of an autonomous vehicle that increases comfort for the driver and passengers. The method according to the invention aims to provide optimal confidence in the steering system of the autonomous vehicle.

[0005] The invention is defined by the independent claims.

[0006] The calculation of the first theoretical trajectory and / or the second theoretical trajectory and / or the customized trajectory may include calculating the steering angle of the vehicle's front wheels and / or calculating the vehicle's yaw. The measurement of the actual trajectory followed may include measuring the steering angle of the vehicle's front wheels and / or measuring the vehicle's yaw.

[0007] According to the invention, the calculation of the correction factor includes a comparison of the yaw of the vehicle following the first theoretical trajectory with the yaw of the vehicle measured during the second stage.

[0008] The calculation of the second theoretical trajectory may include calculating a yaw of the vehicle, and the fifth step may include multiplying the yaw of the vehicle during the second theoretical trajectory by the correction factor.

[0009] The determination process may include a sixth step of personalized vehicle trajectory control, the sixth step comprising: a first sub-step of calculating a first extreme trajectory of the vehicle and, possibly, a second extreme trajectory of the vehicle, then a second sub-step of comparing the personalized trajectory of the vehicle with the first extreme trajectory of the vehicle and, possibly, with the second extreme trajectory of the vehicle.

[0010] The determination process may include an eighth step of calculating a steering command for the vehicle's steering wheels so that the vehicle follows the customized trajectory.

[0011] The invention also relates to a steering system for an autonomous vehicle, the steering system comprising hardware and / or software means implementing a process as defined above, in particular hardware and / or software elements designed to implement a process as defined above.

[0012] The steering system may include: a yaw sensor, and / or a steering wheel steering sensor, and / or a vehicle speed sensor, and / or a camera, and / or a GPS sensor.

[0013] The invention also relates to a motor vehicle, comprising a steering system as defined above.

[0014] The invention also relates to a computer program product comprising program code instructions stored on a medium readable by an electronic control unit for implementing the steps of a process as defined above when said program is run on an electronic control unit. The invention further relates to a program product for an electronic control unit that can be downloaded from a communication network and / or stored on a data medium readable by an electronic control unit and / or executable by an electronic control unit, the program product comprising instructions which, when the program is executed by an electronic control unit, cause the unit to implement a process as defined above.

[0015] The invention also relates to a data storage medium, readable by an electronic control unit, on which is stored a program for an electronic control unit comprising program code instructions for implementing a process as defined above. The invention further relates to a storage medium readable by an electronic control unit comprising instructions which, when executed by an electronic control unit, cause it to implement a process as defined above. Brief description of the drawings

[0016] These objects, features and advantages of the present invention will be described in detail in the following description of a particular embodiment, given by way of non-limiting example, with reference to the accompanying figures, among which: There figure 1is a schematic view of a motor vehicle according to one embodiment of the invention. figure 2 is a flowchart representing the steps of the determination process according to one embodiment of the invention. figure 3 is a top-down view of a car race. The Figures 4 and 5 These are schematic top-down views of a vehicle and its trajectory on a road. figures 6 and 7 are graphs representing the yaw of a vehicle along a route. figure 8 is a top-down view of a car racetrack. figure 9 is a graph representing the lateral deviation of a vehicle along a path. Figure 10 is a graph representing the steering angle of a vehicle's steering wheels along a path. figure 11 is a synoptic diagram of a determination method according to an embodiment of the invention. Description of a method of implementation

[0017] There figure 1This schematically illustrates a motor vehicle 1 according to an embodiment of the invention. The vehicle 1 comprises four wheels 2, two of which, at the front of the vehicle, are steering wheels, i.e., steerable to steer the vehicle. Alternatively, the vehicle 1 could comprise a different number of wheels, for example, three or six wheels, and / or have more or fewer steering wheels. The vehicle 1 may be, in particular, a passenger car, a commercial vehicle, a truck, or a bus. The vehicle 1 conventionally comprises a steering system 3 including a steering column 4 and a steering wheel 5 fixed to the end of the steering column 4. The steering system further comprises an actuator 7 capable of interacting with the steering column 4 or directly with the steering wheels to steer the steering wheels of the vehicle 1.Thus, the vehicle can be used as an autonomous vehicle, that is, as a vehicle capable of navigating a road without driver intervention. The steering system 3 also includes an electronic control unit 8 equipped with a microprocessor 9 and a memory 10. The electronic control unit 8 is capable of issuing commands to the actuator 7 so that the latter orients the steering wheels of the vehicle 1.

[0018] Various sensors from the vehicle 1 are connected to the electronic control unit 8, including: a yaw sensor 11, a steering wheel angle sensor 12, a vehicle speed sensor 13, a camera 14, and a GPS sensor 15. These sensors can be integrated into the steering system 3 and connected, for example, via a wired or wireless connection to the electronic control unit 8. These sensors may also be used by other vehicle systems. They can be connected to the electronic control unit 8 via a CAN (Controlled Area Network) data bus.

[0019] The yaw sensor 11 is a sensor measuring the angular velocity of the vehicle around an axis perpendicular to the plane on which the vehicle rests. When the vehicle is on a horizontal surface, the yaw sensor 11 therefore measures the angular velocity of the vehicle around a vertical axis. Throughout this application, the term "yaw" refers to yaw rate, which is a physical value expressed in radians per second. The yaw sensor 11 can be connected to an electronic vehicle stability control unit, also known as an "ESP" unit.

[0020] The steering wheel angle sensor 12 is a sensor capable of measuring the angular position, or in other words, the orientation, of the steering wheels 12. Typically, this sensor can be a steering wheel angle sensor, meaning a sensor for the angular position of the steering wheel 5 or the steering column 4 attached to the steering wheel. This sensor can be connected to the steering wheel 5 or the steering column 4 provided that the steering wheel or steering column is itself coupled to the steering wheels, i.e., that their angular positions are correlated under all circumstances. If the orientation of the steering wheel 5 and / or the steering column 4 can be decoupled from the steering angle of the steering wheels, particularly when the vehicle is operating autonomously, a different sensor will be used to measure the angular position of the steering wheels.The steering wheel steering sensor 12 can also be connected to the vehicle's electronic trajectory control unit.

[0021] The speed sensor 13 can consist of a set of sensors integrated into each of the wheels 2 and measuring their rotational speed. The speed sensor 13 can also be connected to the vehicle's electronic trajectory control unit.

[0022] Camera 14 is capable of detecting the environment in front of the vehicle. Specifically, it can detect road markings such as the presence and position of a continuous or broken lane marking, generally white or yellow, used to delineate the edges of a roadway or lanes. Advantageously, camera 14 can also detect other indicators useful for the correct positioning of the vehicle on the roadway, such as arrows marked on the road, signs, obstacles, and other vehicles on the roadway.

[0023] The GPS 15 sensor (Global Positioning System) provides information to calculate the vehicle's GPS coordinates. The vehicle's position is accurate enough to pinpoint its location on the road.

[0024] Finally, the steering system 3 also includes a switch 16, conveniently accessible to the driver. Switch 16 can be in three distinct positions. The first position corresponds to the vehicle's first operating mode, M1, in which the driver manually controls the vehicle's trajectory. In this first operating mode, the driver controls the vehicle's direction by turning the steering wheel. A power steering mechanism may reduce the effort required to turn the wheels; however, steering is solely at the driver's discretion. Additionally, the vehicle's speed may or may not be controlled by cruise control or a speed limiter.

[0025] A second position of switch 16 corresponds to a second operating mode M2 ​​of the vehicle in which the trajectory of vehicle 1 is defined completely autonomously by the electronic control unit 8. In this second operating mode, vehicle 1 follows a theoretical trajectory.

[0026] A third position of the switch 16 corresponds to a third operating mode M3 of the vehicle in which the vehicle 1 follows a personalized trajectory defined according to an embodiment of the invention.

[0027] We will now detail the process for determining this personalized trajectory with reference to the figure 2illustrating a method comprising eight steps E1 to E8. The first three steps E1 to E3 are performed during a first phase P1 in which the vehicle's trajectory is manually controlled by the driver. The first phase P1 is an initialization phase, in other words, a parameterization or learning phase, of the determination method according to the invention. The four steps E4, E5, E6, and E8 are performed during a second phase P2 in which the vehicle's trajectory is controlled autonomously. The seventh step E7 can be performed at any time before the eighth step E8.

[0028] The first step, E1, is a calculation step for a first theoretical trajectory of the vehicle along a first route. This first route can be arbitrary. For example, the first route could be the route illustrated on the figure 3 or on the figure 8 The illustrated route on the figure 3It includes a starting line, indicated by "Starting point," and a finishing line, indicated by "Ending point," and is traversed in the direction shown by the "Driving direction" arrow. This first stage can be performed while the vehicle follows the initial route under manual control. During this stage, the vehicle is controlled manually by the driver. Therefore, the electronic control unit does not send a command to actuator 7 to direct the vehicle to follow the initial theoretical trajectory. The initial theoretical trajectory can be defined, in particular, using the vehicle's sensors 11, 12, 13, 14, and 15, and according to the shape of the road surface on which the vehicle is traveling. It is defined completely objectively, that is, entirely independently of the driver's driving style.For example, the first theoretical trajectory, which could also be called the "ideal" or "optimal" trajectory, might be one that positions the vehicle equidistant from the right and left edges of a roadway. (Referring to...) figure 4The first theoretical trajectory TT1 can also be a trajectory positioning the vehicle 1 equidistant from boundary lines LD marked on the right and left sides of the roadway. The first theoretical trajectory can also be a trajectory maintaining the vehicle at a given distance from a single boundary line. Regardless of the calculation method used to define the theoretical trajectory, the microprocessor 9 of the electronic control unit 8 executes an algorithm based solely on the information provided by the vehicle's sensors 11, 12, 13, 14, and 15. The first theoretical trajectory does not take into account the driver's behavior when operating the vehicle in the first operating mode M1, i.e., manual mode. This first theoretical trajectory is stored in the memory 10 of the electronic control unit 8.Alternatively, the first theoretical trajectory could also be calculated by a computer not on board the vehicle and / or be available in a database not on board the vehicle and then be recorded as digital data in the memory 10 of the electronic control unit 8. As a note, the theoretical yaw, i.e. the yaw of the vehicle following the first theoretical path at the same speed, can be calculated by knowing the first theoretical trajectory and the speed of the vehicle.

[0029] In a second step E2, the actual trajectory followed by the vehicle along the first route is measured. The actual trajectory followed by the vehicle, in other words, the real trajectory, is independent of the first theoretical trajectory. With reference to an example illustrated on the figure 5We observe that the actual trajectory TR is closer to the inside of the turn than the first theoretical trajectory TT1. This actual trajectory is also recorded in memory 10 of the electronic control unit 8.

[0030] According to an alternative embodiment of the method according to the invention, the first theoretical trajectory could be recorded while the vehicle follows the first path while being controlled autonomously according to the second operating mode M2. Then the actual trajectory would be measured during a second pass of the vehicle over the first path while being manually controlled.

[0031] In a third step E3, a correction factor F, or error rate, is calculated by comparing the initial theoretical trajectory, calculated in the first step E1, with the trajectory actually followed by the vehicle, recorded in the second step E2. In an example outside the scope of the invention, this factor could be, for instance, a fixed value, a value dependent on the steering angle of the vehicle's front wheels, a value dependent on the vehicle's speed, or a value dependent on the distance between the actual trajectory and the initial theoretical trajectory. According to the invention, the factor is a value dependent on the vehicle's yaw rate.

[0032] Specifically, the correction factor F can be calculated by comparing the yaw of the manually controlled vehicle with the theoretical yaw, that is, the yaw of the vehicle following the first theoretical path at the same speed. Alternatively, other methods can be considered for calculating the correction factor F.

[0033] In step E4, a second theoretical trajectory is calculated along a second path. The second path may be different from or identical to the first path. The second theoretical trajectory is calculated according to the same principle as the first theoretical trajectory. It can be calculated using one of the calculation methods described for calculating the first theoretical trajectory in step E1. If the first path is identical to the second path, then the first theoretical trajectory is logically identical to the second theoretical trajectory.

[0034] In step E5, a personalized vehicle trajectory is calculated. This calculation is based on the second theoretical trajectory calculated in step four and on a correction factor F also calculated in step four. This calculation may include, for example, multiplying the vehicle's yaw rate (if it were to follow the second theoretical trajectory) by the correction factor F.

[0035] The first and / or second theoretical trajectory, and / or the trajectory actually followed by the vehicle, and / or the customized trajectory can be recorded in various forms in memory 10. For example, these trajectories can be recorded as the steering angle of the vehicle's front wheels as a function of time and / or distance traveled. Alternatively, or in addition, these trajectories can also be recorded as the yaw rate of the vehicle as a function of elapsed time and / or distance traveled. Finally, as an alternative or in addition, the trajectories can also be recorded as a distance from a guideline marked on the road or as a GPS track comprising a set of GPS coordinates.

[0036] The graph of the figure 6illustrates in more detail a first example of the process for determining the vehicle's personalized trajectory. In this example, the first route is identical to the second route and corresponds to the route illustrated on the figure 3 Each trajectory is characterized by a yaw curve of the vehicle as a function of time: the yaw rate, expressed in radians per second, is displayed on the y-axis. The time, expressed in seconds, is displayed on the x-axis. The time required to complete the first trajectory is therefore approximately 75 seconds.

[0037] A first curve, C11, represents the vehicle's yaw rate as a function of time when the vehicle follows the initial theoretical trajectory. This first curve, C11, is obtained at the end of the first step, E1. A second curve, C12, represents the vehicle's yaw rate as a function of time when the vehicle is steered manually by the driver. This second curve, C12, is obtained at the end of the second step, E2. During this step, the driver followed a trajectory that was generally wider around the curves than for the initial theoretical trajectory, in order to reduce the vehicle's yaw rate. The correction factor, F, can be calculated, for example, by dividing the average of the second curve by the average of the first curve.Next, the second theoretical trajectory is calculated: since the second route in this example is identical to the first, the second theoretical trajectory is identical to the first theoretical trajectory and is therefore represented by the first curve, C11. Finally, the vehicle's personalized trajectory is calculated by multiplying the first curve, C11, by the correction factor, F. This yields a third curve, C13, representing the vehicle's personalized trajectory on the second route. It can be observed that the personalized trajectory results in a generally lower yaw than the yaw of the first theoretical trajectory, which corresponds well to the driver's driving style during the second step, E2—that is, a rather cautious driving style.

[0038] The graph of the figure 7illustrates a second example of the process for determining the vehicle's personalized trajectory. The first and second paths also correspond to the path of the figure 3 The determination method is identical to that presented previously with reference to the figure 6Therefore, the first curve C21, corresponding to the first theoretical trajectory, is identical to the first curve C11 seen previously. However, this time, while the vehicle was controlled manually, the driver followed a trajectory that was generally more to the inside of the curves than for the first theoretical trajectory, thus increasing the vehicle's yaw. The second curve C22, obtained at the end of the second step E2, therefore exhibits a generally higher yaw than the first theoretical trajectory. Applying the same calculation method as seen in the previous example, we obtain the third curve C23, representing a new customized trajectory for the vehicle on the second route.We observe that this new personalized trajectory produces an overall higher yaw than the yaw of the first theoretical trajectory, which corresponds well to the driver's driving style during the second stage E2, i.e. a rather sporty driving style.

[0039] A third example of the use of the determination method according to the invention is illustrated by the figures 8, 9 And 10 . There figure 8 This illustrates, from a top view, a course marked by a guideline LD on the ground. This course includes a slight right turn V1 followed by a left turn DT and is traversed in the direction of arrow F1. figure 9 is a graph representing the lateral deviation (expressed in meters) of the vehicle from the guideline LD as a function of the time taken to follow the path of the figure 8The first curve, C31, represents the lateral deviation obtained when the vehicle is controlled autonomously, following a theoretical trajectory along the guideline LD. It can be observed that the vehicle does not deviate by more than a few centimeters, or at most a few tens of centimeters, from the guideline LD. The second curve, C32, represents the lateral deviation obtained when the vehicle is controlled manually. It can be observed that the vehicle deviates by several tens of centimeters, or even a meter, on either side of the guideline LD. It should be noted that such a deviation from the guideline does not pose any risk to driver safety because the roadway is sufficiently wide. These deviations from the guideline LD reflect the driver's driving style.

[0040] There Figure 10is a graph representing the steering angle of the steering wheels as a function of the time required to follow the path of the figure 8A first curve, C41, represents the steering angle when the vehicle is controlled autonomously, following a theoretical trajectory. This curve includes several inflection points, P, corresponding to variations in the steering angle. These inflection points, P, can generate lateral accelerations and / or jolts felt by the driver and / or passengers. A second curve, C42, represents the steering angle when the vehicle is controlled manually. This second curve does not include any inflection points, or its inflection points are significantly less pronounced than those of the first curve.Compared to the trajectory obtained through autonomous vehicle control, as represented by curve C41, the trajectory obtained through manual vehicle control deviates further from the guideline LD but also produces less lateral acceleration and / or vibration. A third curve, C43, represents the steering angle when the vehicle follows a customized trajectory according to the invention. This third curve is almost identical to the second curve. It is therefore possible to reproduce a vehicle behavior, in other words, a driving style, very close to the behavior or driving style of a driver manually controlling the vehicle's trajectory.

[0041] In a sixth step, E6, the vehicle's personalized trajectory is checked. This check aims to verify that the road dimensions allow the personalized trajectory to be followed, or in other words, that the vehicle's trajectory can deviate from the theoretical trajectory. This sixth step includes a first substep, E61, which calculates a first extreme trajectory for the vehicle and a second extreme trajectory. The first extreme trajectory can correspond to the innermost path of a curve without the vehicle leaving the roadway or losing traction. The second extreme trajectory can correspond to the outermost path of a curve without the vehicle leaving the roadway or losing traction.These trajectories can be determined using vehicle sensors such as the yaw sensor 11, the steering wheel angle sensor 12, the speed sensor 13, the camera 14 and the GPS sensor 15, as well as possibly using an electronic vehicle trajectory control unit.

[0042] In a second substep, E62, the vehicle's customized trajectory is compared with the first and second extreme trajectories. If the customized trajectory falls between these two extreme trajectories, then a steering command can be issued. If the customized trajectory does not fall between these two extreme trajectories, then the theoretical trajectory is used to issue a steering command. Specifically, extreme yaw values ​​can be associated with the two extreme trajectories. A maximum extreme yaw value corresponds to the inside of a turn. A minimum extreme yaw value corresponds to the outside of a turn. It is then verified that the yaw of the customized trajectory is indeed contained between these two extreme values.

[0043] Alternatively, this sixth step E6 could be omitted, for example, if the road surfaces on which the vehicle is traveling do not present a risk of running off the track or loss of traction, or if the correction factor F is defined in such a way as to induce only minor changes to the theoretical trajectory. The check could also be performed against a single extreme trajectory, such as the inside or outside of a curve.

[0044] In step E7, a vehicle operating mode is selected from the second and third operating modes M2 and M3, or from the three operating modes M1, M2, and M3 described previously. The driver can activate switch 16 for this purpose. The second operating mode M2 ​​may only be accessible after an initialization phase P1. Step E7 may not be part of the personalized trajectory determination process itself, but rather of a broader operating process encompassing the personalized trajectory determination process, which would then only include steps E1 to E6 and E8. Step E7 may be performed before steps E1 to E6. Thus, if at the end of step E7 the driver selects the first or second operating mode, steps E1 to E6 are not necessarily performed.The choice of operating mode can be made during the second operating phase P2 to switch from the second operating mode M2 ​​to the third operating mode M3 or vice versa.

[0045] In the eighth step E8, a steering command for the vehicle's front wheels is calculated. The electronic control unit 8 sends a steering command to the actuator 7 based on the position of switch 16 and the result of the custom trajectory check performed in the sixth step E6. If the switch is in its first position, the vehicle is controlled manually. If switch 16 is in its second position, the vehicle is controlled autonomously and follows the theoretical trajectory. If the switch is in its third position and the custom trajectory check performed in the sixth step E6 allows it, then the vehicle follows the custom trajectory.

[0046] There figure 11In other words, this summarizes the previously described determination process. Moving from right to left in this diagram (i.e., from the end of the process to the beginning), a block B1 implements the eighth step, E8, which is to say, it generates a steering command for the front wheels based on a trajectory command received upstream. Upstream of block B1, a block B2 represents the sixth step, E6, which is the selection of a vehicle operating mode from among the three operating modes M1, M2, and M3 described previously. Three different trajectory commands can therefore reach block B2. The first trajectory command, represented by block B3, corresponds to a manual trajectory command, i.e., one resulting from the driver's action on the steering wheel. The second trajectory command, represented by block B4, corresponds to a theoretical trajectory command.The third trajectory command, symbolized by block B5, corresponds to a custom trajectory command. As explained previously, the custom trajectory command is obtained by combining the manual trajectory command with the theoretical trajectory command. This combination is performed by block B6, known as the "learning module." Block B6 implements, in particular, the third step E3, in which the correction factor F is calculated by comparing a manual trajectory command with a theoretical trajectory command. Then, block B7 implements the sixth step E6, in which the custom trajectory is checked.

[0047] Thanks to this invention, a method exists for adapting the trajectory of an autonomous vehicle based on the driver's habits and / or driving style. Personalized trajectory control combines a theoretical optimal trajectory with the driver's habits or driving style. The driver does not need to manually configure their desired driving style; simply using the vehicle in manual mode at least once is sufficient to define it. The vehicle's trajectory is thus more predictable for the driver, who is therefore more inclined to trust the autonomous vehicle's steering system.

[0048] The first phase, P1, known as the initialization phase, can be repeated as often as necessary to refine the correction factor or to adapt it to the driver's changing driving habits. The electronic control unit can store different correction factors associated with different drivers of the vehicle. Thus, after a driver is identified, their assigned correction factor can be used. The correction factor can also be manually selected from several previously stored options. Alternatively, it can be determined automatically based not only on the driver but also on the passengers in the vehicle.

Claims

1. Method for determining a path of a vehicle (1) capable of controlling its path autonomously, the method comprising a first phase (P1) that takes place while the path of the vehicle is being controlled manually by a driver, the first phase (P1) comprising: - a first step (E1) of computing a first theoretical path of the vehicle, - a second step (E2) of measuring a path actually followed by the vehicle, and - a third step (E3) of computing a correction factor (F), the third step (E3) comprising a comparison of the first theoretical path with the path actually followed, the computation of the correction factor (F) comprising comparing the yaw of the vehicle following the first theoretical path with the yaw of the vehicle measured in the second step (E2), and a second phase (P2) that takes place while the path of the vehicle is being controlled autonomously, the second phase (P2) comprising: - a fourth step (E4) of computing a second theoretical path of the vehicle, and - a fifth step (E5) of computing a personalized path of the vehicle, this computation being based on the second theoretical path and on said correction factor (F) .

2. Determining method according to the preceding claim, characterized in that the computation of the first theoretical path and / or of the second theoretical path and / or of the personalized path comprises computing a steering angle of the steered wheels of the vehicle and / or computing a yaw of the vehicle, and / or in that the measurement of the path actually followed comprises measuring a steering angle of the steered wheels of the vehicle and / or measuring a yaw of the vehicle.

3. Determining method according to one of the preceding claims, characterized in that the computation of the second theoretical path comprises computing a yaw of the vehicle, and in that the fifth step comprises multiplying the yaw of the vehicle on the second theoretical path by the correction factor (F).

4. Determining method according to one of the preceding claims, characterized in that it comprises a sixth step (E6) of checking the personalized path of the vehicle, the sixth step comprising: - a first sub-step (E61) of computing a first extreme path of the vehicle and, optionally, a second extreme path of the vehicle, then - a second sub-step (E62) of comparing the personalized path of the vehicle with the first extreme path of the vehicle and, optionally, with the second extreme path of the vehicle.

5. Determining method according to one of the preceding claims, characterized in that it comprises an eighth step (E8) of computing a steering command for the steered wheels of the vehicle so that the vehicle follows the personalized path.

6. Method for operating a steering system (3) for an autonomous vehicle (1), characterized in that it comprises a step (E7) of choosing an operating mode of the vehicle from among: - an operating mode (M2) in which the vehicle is controlled autonomously to follow a theoretical path, or - an operating mode (M3) in which the vehicle is controlled autonomously to follow a personalized path determined by the determining method according to one of Claims 1 to 5.

7. Steering system (3) for an autonomous vehicle (1), characterized in that it comprises hardware means (7, 8, 9, 10, 11, 12, 13, 14, 15, 16) and / or software means that implement a method according to one of the preceding claims, and especially hardware elements (7, 8, 9, 10, 11, 12, 13, 14, 15, 16) and / or software elements designed to implement a method according to one of the preceding claims.

8. Steering system (3) according to the preceding claim, characterized in that it comprises: - a yaw sensor (11), and / or - a sensor (12) of the steering angle of the steered wheels, and / or - a vehicle speed sensor (13), and / or - a camera (14), and / or - a GPS sensor (15).

9. Motor vehicle (1), characterized in that it comprises a steering system according to either of Claims 7 and 8.

10. Computer program product comprising program code instructions stored on a medium that is readable by an electronic control unit (8) with a view to implementing the steps of a method according to any one of Claims 1 to 6 when said program is run by an electronic control unit (8).

11. Data storage medium that is readable by an electronic control unit (8) and on which is stored a program for an electronic control unit (8) comprising program code instructions for implementing a method according one of Claims 1 to 6.