Method for tracking the trajectory of a motor vehicle

Abstract

The invention relates to a method for steering a motor vehicle (10) traveling in a lane of a road, comprising the steps of: - acquiring initial data relating to lane markings, - estimating a lane trajectory, - determining an initial steering command for the motor vehicle using a controller and based on said lane trajectory. According to the invention, the method also includes, when the motor vehicle is following a third vehicle (20), the steps of: - acquiring secondary data relating to the third vehicle, - evaluating a third vehicle trajectory (29), - determining a second steering command for the motor vehicle using said controller and based on the third vehicle trajectory, and - selecting one of the two steering commands, said steering actuator being controlled according to this command. Figure for the abstract: Fig. 4

Description

Title of the invention: Method for tracking the trajectory of a motor vehicle Technical field of the invention

[0001] The present invention relates generally to driving aids for motor vehicles.

[0002] The invention relates more specifically to a method for controlling a motor vehicle traveling on a traffic lane of a road, this method comprising steps of: - acquisition of initial data relating to the edge lines of said traffic lane, - estimation of a traffic lane trajectory taking into account said first data, - determination of a first control instruction for the motor vehicle by means of a controller and as a function of said traffic lane trajectory, and - control of the motor vehicle by a steering actuator.

[0003] The invention also relates to a motor vehicle suitable for implementing such a method. It applies more particularly to cars and other motorized vehicles traveling on roads. State of the art

[0004] In an effort to make motor vehicles safer, they are currently being equipped with driving assistance systems and even highly automated driving systems.

[0005] These are typically lane centering systems (better known by the English acronym LCA for “Lane Centering Assist”).

[0006] In order to function, such a system needs to know the position of the edges of the traffic lane taken by the vehicle. Currently, it is known to use a sensor, such as a camera, which incorporates image processing means in order to determine the position of the lane edge marking lines.

[0007] It is understood that the detection of the edges of the traffic lane must be extremely reliable to prevent the vehicle from leaving it.

[0008] Unfortunately, this detection does not always give complete satisfaction.

[0009] This is particularly the case when the camera is unable to detect these lane edges (for example because the marking lines are erased, or because the light is unfavourable, or because the marking lines are hidden by a truck in front of the vehicle) or when the markings are absent (for example example because the lane divides into two lanes or because two lanes merge into one lane).

[0010] Under these conditions, the current solution is to deactivate the LCA function. This solution is impractical for the driver, who must then take control of the vehicle and then manually reactivate the LCA function as soon as possible.

[0011] Another function known as target tracking (commonly called “Autosteer” function) is also known, which consists of locating the vehicle preceding the vehicle in question (hereinafter respectively called “target vehicle” and “ego vehicle”) and following it. This function typically allows a truck driver to follow another truck in complete safety.

[0012] Within the framework of this target tracking function, two methods are known. The first consists of locating the target vehicle and following it. This method has the disadvantage that when turning, the ego vehicle will tend to cut the turn, which can be dangerous. The second method consists of following the trajectory of the target vehicle.

[0013] In summary, the LCA and target tracking functions operate in separate and different ways, making them difficult to manage together. Presentation of the invention

[0014] The present invention proposes, on the contrary, to combine these two functions by using for this purpose a single controller, which receives as input different data but of the same nature depending on whether one or the other of the functions is selected.

[0015] More particularly, the invention proposes a method as defined in the introduction, in which, when the motor vehicle follows a third-party vehicle, steps are provided: - acquisition of second data relating to the third-party vehicle, - evaluation of a third-party vehicle trajectory taking into account said second data, - determination of a second control instruction for the motor vehicle by means of said controller and as a function of the trajectory of the third-party vehicle, and - selection of one of the first and second control instructions, said steering actuator being controlled according to the selected control instruction.

[0016] Thus, thanks to the invention, the controller uses either the traffic lane trajectory (which corresponds for example to the center line of the traffic lane), or the trajectory of the third-party vehicle in order to determine the vehicle's steering instruction. The development and fine-tuning of a single controller is therefore sufficient to ensure the implementation of the two LCA and target tracking functions since these two trajectories are distinct but of the same nature, so that the controller can process them in the same way. In addition, managing the transition from one function to another is simpler, and can be smoother for the driver.

[0017] Other advantageous and non-limiting characteristics of the method according to the invention, taken individually or in all technically possible combinations, are the following: - it is planned to determine a level of reliability of said first pilot instruction; - the second control instruction is selected only if the said reliability level is lower than a threshold; - said reliability level is calculated based on at least one of the following information: a quality of marking of the edge lines, a criterion of visibility of the road by sensors for acquiring the first data, a geometric characteristic of the traffic lane, a dynamic characteristic of the motor vehicle; - it is planned to measure and record successive positions of the third-party vehicle; - at the evaluation stage, the trajectory of the third-party vehicle is determined based on said successive positions; - it is planned to estimate at least one position of the third-party vehicle; - at the evaluation stage, the third-party vehicle trajectory is determined based on a difference between the estimated position and one of the successive measured positions; - the successive positions being measured at successive time steps, when measuring and recording a new position of the third-party vehicle at a current time step, at least part of said successive positions already recorded are corrected taking into account the movement of the motor vehicle between the current time step and the time step preceding the current time step; - the controller varies according to a lateral deviation between a trajectory of the motor vehicle and the trajectory of a third-party vehicle, measured at a distance in front of the motor vehicle which varies according to the speed of the motor vehicle; - when at a first time step, one of the first and second control instructions is selected and, at a second time step following the first time step, the other of the first and second control instructions is selected, the steering actuator is controlled according to the two control instructions for a determined duration after the second time step; - when at a first time step, the first control instruction is selected and, at a second time step following the first time step, the second control instruction is selected, the second control instruction is filtered for a duration at least equal to the time necessary for the motor vehicle to travel the distance separating, at the time of the second time step, the vehicle au- third-party vehicle vehicle.

[0018] The invention also provides a motor vehicle comprising: - first acquisition means adapted to acquire, when the motor vehicle is moving on a traffic lane and is preceded by a third-party vehicle, first data relating to the edge lines of said traffic lane, - second acquisition means adapted to acquire second data relating to the third-party vehicle, - a steering actuator adapted to control the motor vehicle, and - a computer programmed to implement a control method as mentioned above.

[0019] Of course, the various features, variants and embodiments of the invention may be combined with each other in various combinations to the extent that they are not incompatible or mutually exclusive. Detailed description of the invention

[0020] The description which follows with reference to the appended drawings, given as non-limiting examples, will make it clear what the invention consists of and how it can be implemented.

[0021] In the attached drawings:

[0022] [Fig.l] is a schematic view of a motor vehicle suitable for implementing a method according to the invention;

[0023] [Fig.2] is a diagram illustrating algorithmic blocks of a vehicle computer automobile of [Fig.l];

[0024] [Fig.3] is a graph illustrating an example of variation of a time constant used in the process of [Fig.l];

[0025] [Fig.4] is a diagram showing the motor vehicle of [Fig.l], seen from above, and a vehicle in front of it;

[0026] [Fig.5] is a diagram showing the motor vehicle of [Fig.l] with two successive moments.

[0027] In [Fig.l], a motor vehicle 10 is shown which is suitable for implementing the invention.

[0028] This is a car. Alternatively, it could be another type of vehicle (truck, motorcycle, etc.).

[0029] In this figure, the motor vehicle 10 is traveling on a traffic lane 31 of a road 30. It can be seen that the road 30 has two lateral marking lines 34, 35 (which delimit it) and a central marking line 33 delimiting the two traffic lanes 31, 32.

[0030] A traffic lane is herein defined as the part of a road on which a only one vehicle is allowed to circulate at a time. Such a traffic lane is usually demarcated between marking lines.

[0031] A road (or roadway) is defined as a set of traffic lanes. In the example considered here for illustrative purposes, this road 30 therefore comprises two traffic lanes 31 on which vehicles can travel in the same direction.

[0032] In [Fig.4], the motor vehicle 10 and a third-party vehicle preceding it on the same traffic lane are shown.

[0033] The motor vehicle 10 which is the subject of the present invention is therefore the one which follows the other. It will hereinafter be called the ego vehicle 10. The third-party vehicle will be called the target vehicle 20.

[0034] The target vehicle 20 could be of any type. It will only be considered that it is in motion and traveling on the same traffic lane as the ego vehicle 10, preceding it.

[0035] As shown in [Fig.l], the ego vehicle 10 conventionally comprises a passenger compartment in which there are in particular a seat for the driver 20 of the vehicle and a steering wheel 12.

[0036] This ego vehicle 10 conventionally comprises a powertrain, a braking system and a steering system for turning the vehicle. Conventionally, the steering system comprises an electronically controllable power steering actuator, the powertrain comprises an electronically controllable motor control actuator, and the braking system comprises an electronically controllable braking actuator.

[0037] The ego vehicle 10 further comprises an electronic and / or computer processing unit (hereinafter called calculator 11) comprising at least one microprocessor, at least one memory and input and output interfaces.

[0038] Thanks to its input interfaces, the computer 11 is adapted to receive different input data which come from sensors or third-party computers.

[0039] Among these sensors, there is for example provided a front camera making it possible to locate the edges of the traffic lane 31 taken, and a remote sensor (RADAR and / or LIDAR) making it possible to locate objects in the environment of the ego vehicle 10 (in particular the target vehicle 20).

[0040] Thanks to its output interfaces, the computer is adapted to control the power steering actuator, the engine control actuator, and the braking actuator.

[0041] Thanks to its memory, the computer stores a computer application, consisting of computer programs comprising instructions whose execution by the computer allows the implementation of a function for automatically maintaining the vehicle in its lane (hereinafter called the LCA function), and of a function target tracking (hereinafter referred to as AUS function).

[0042] To summarize, the LCA function will be favored to laterally control the ego-vehicle 10 in order to keep it in its lane. The AUS function will be used when the lane edge markings are not visible or cannot be used, so that the LCA function must be interrupted.

[0043] In [Fig.2], a part of the software architecture of the computer 11 has been shown diagrammatically, which makes it possible to choose which function to activate and how to implement it.

[0044] A first block B1 is designed, during a first step, to receive the data from the sensors.

[0045] These data could already be processed by processors embedded in the sensors. Typically, the camera could have a processor adapted to establish the mathematical equations of the trackside lines.

[0046] However, here, it will be considered that these are raw, unprocessed data. When it receives these data, the computer merges them, that is to say it processes them in a combined manner so as to deduce information on the track edge lines (typically their equations) and information on the objects present in the environment of the ego-vehicle 10 (typically the position of the target-vehicle 20).

[0047] The fusion operation may typically consist of comparing the data from the camera and the remote sensor in order to obtain new reliable and precise data.

[0048] A second block B2 receives this information and is designed to choose, during a second step, which function to activate from among the aforementioned LCA and AUS functions. Typically, it can favor the LCA function as long as the information on the lane edge lines is deemed reliable, and the AUS function otherwise (provided that a target vehicle 20 precedes the ego vehicle 10). This block B2 sends this information to a block B5 responsible for transmitting a steering instruction to the power steering actuator. This steering instruction is presented here in the form of an angle to be applied to the steering wheel or to the wheels.

[0049] A third block B3 is adapted to determine, during a third step, an equation of the trajectory 29 of the target vehicle 20 and to transmit this equation to the fourth block B4.

[0050] This fourth block B4 makes it possible to generate, during a fourth step, two control instructions for this actuator, respectively associated with the two functions LCA and AUS, then to transmit these instructions to block B5 so that it selects the one most suited to the situation.

[0051] We can then describe in more detail the method implemented by the computer 11 to automatically control the motor vehicle in its lane using the LCA and AUS functions.

[0052] This method will be implemented in a loop, at regular time steps (i.e. with a step constant sampling).

[0053] Initially, the computer 11 acquires the equations of the marking lines 33, 35 which border its taxiway. It can deduce from this the equation of the center line of its taxiway 31. This equation will define a “taxiway trajectory 39”.

[0054] In a well-known manner, the computer stores a controller K which makes it possible, taking this equation into account, to determine a first control instruction for the power steering actuator within the framework of the LCA function.

[0055] This controller K is a mathematical operator which receives as input different variables among which at least one is a function of the trajectory of taxiway 39 and which provides as output the piloting instruction.

[0056] In practice, this controller K receives as input a deviation between the trajectory 19 of the ego vehicle 10 and the trajectory of the traffic lane 39. This deviation will be better defined below.

[0057] To determine a second control instruction for the power steering actuator within the framework of the AUS function, the computer first acquires the trajectory of the target vehicle (hereinafter referred to as the target trajectory 29).

[0058] The idea will in fact be to use the same controller K for the LCA function and the AUS function, by submitting to it as input: - the difference between the trajectory 19 of the ego vehicle 10 and the trajectory of the traffic lane 39 to calculate the first piloting instruction, - the difference between the trajectory 19 of the ego vehicle 10 and the target trajectory 29 to calculate the second piloting instruction.

[0059] To determine the target trajectory 29, the block B2 receives as input data relating to the target vehicle 20, and in particular its position in a reference frame attached to the ego vehicle 10.

[0060] It will be noted that this position will be recorded in the form of two coordinates (x, y) expressed in a reference frame (Oego(t), Xego(t), Yego(t)) which is attached to the vehicle-ego 10 and whose position therefore varies over time (see [Fig.5]).

[0061] It will be considered here that the center Oego(t) is located at the level of the ego-vehicle 10, for example on its front face. The abscissa axis Xego(t) is oriented forward, in the longitudinal axis of the ego-vehicle 10. The ordinate axis Yego(t) is oriented laterally, towards the left, at right angles to the abscissa axis.

[0062] Each position of the target vehicle 20 is stored in the memory of the computer 11. It can be noted here that only the last N positions of the target vehicle 20 will remain recorded in this memory (with N a predetermined constant, preferably greater than 10). The oldest position of the target vehicle 20 will therefore be deleted each time a new position is recorded.

[0063] In practice, the part of the memory of the computer 11 associated with the storage of these positions (hereinafter called “register”) will have a fixed and predefined size.

[0064] This register has two fields, one for the axial coordinate and the other for the lateral coordinate of the position P of the target vehicle 20.

[0065] It also includes N records corresponding to the coordinates of the target vehicle 20 at the N previous time steps.

[0066] Since the position of the reference frame attached to the ego-vehicle 10 evolves over time, the idea is, at each time step where a new position recording is made, to correct the other positions stored in the memory. For this, a reference frame change matrix is ​​used.

[0067] To detail this operation, we can refer to [Fig.4] and introduce the following hypotheses.

[0068] The time step AT will correspond to the sampling step, that is to say the minimum time elapsing between two recordings of positions of the target vehicle 20.

[0069] The longitudinal speed of the vehicle-ego 10 will be noted vxEgo and will be expressed in m / s. This speed is measured by sensors at the wheels of the vehicle. This longitudinal speed represents a component of the total speed vego of the vehicle.

[0070] The angular velocity of the ego-vehicle 10 will be noted y and will be expressed in rad / s. This velocity is measured by an inertial sensor embedded in the ego-vehicle 10.

[0071] The drift angle of the ego-vehicle 10 will be noted [3 and will be expressed in radians.

[0072] These speeds and angles will be considered constant during the time step AT.

[0073] We will also consider the following hypothesis:

[0074] [Math.l] —— ~ usmi —2~ i — ——

[0075] The coordinates of the position P of the target vehicle 20 in the reference system associated with the vehicle-ego 10 at time t will be noted P yO_ega{t) LP

[0076] The coordinates of the vehicle-ego 10 at time t+1 (i.e. t+At) in the re ferential associated with the ego-vehicle 10 at time t will be noted y,O_eg <j(t) Egdt+\).

[0077] We can introduce the following matrix M of change of reference:

[0078] [Math.2] ' cos(yAT) - sintyAT) ' . sin(yAT) cos(yAT) .

[0079] Then we can write:

[0080] [Math.3] Xp XO_egc(t) P' rO_egc{t) ' ^EgoÇt+l) -yO_egdJ). • Ego{ ï+1). + M. Xp \}O_ega(t+l) P .

[0081] Or again:

[0082] [Math.4] yO_egc{t+1) A p ' xO_egdt) rO_ego(f) ' AP ' ' XEgdl+l).'P ' - yO_egdt) _ yO^egdt) LP -Egop+OJ

[0083] To simplify, we can therefore correct the coordinates of the positions of the points previously recorded in the register using the following equation:

[0084] [Math.5] xo_ego(t) _ a __y___* Xp ÀA 1 Vxego cos( / j) x sm A Tvx_eso— L- p

[0085] To summarize, at each new time step, the calculator proceeds as follows: - the coordinates of the oldest point P are deleted from its memory, which frees the first record of the memory register, - the coordinates of each record are moved into the previous record, which frees the last record from the memory register, - these coordinates are all updated in the aforementioned manner to take into account the movement of the reference frame attached to the ego-vehicle 10, - the new coordinates of point P are stored in the last recording.

[0086] It can be noted here that the computer can only carry out these operations if the movement of the target vehicle from the last recorded position is greater than a threshold, which will avoid unnecessary recordings if the target vehicle is stationary or traveling at low speed and will allow the measurements to be filtered in order to avoid generating inconsistencies.

[0087] Taking into account the positions stored in the register, block B4 of the calculator is able to calculate the coefficients of a function illustrating the trajectory of the target vehicle 20 in the last reference point considered attached to the ego vehicle 10.

[0088] Here, the function considered is a polynomial function. It is preferably of order 3. Thus, the equation of the trajectory of the target vehicle 20 is defined as follows:

[0089] [Math.6] y = cQ + cpc + c2^2 + CyX3

[0090]

[0091]

[0092]

[0093] The objective is therefore to determine the values ​​of the coefficients c0, cb c2, c3 which are such that the polynomial curve passes as close as possible to the N points recorded in the register. For this, one could use a least squares minimization method. However, the solution used here is different. We can first write, for the N points Pi with coordinates (xi5 y;) correspond

[0094] corresponding to the last N positions of the target vehicle 20 in the reference frame attached to the ego vehicle 10 at the current time: [Math.7] x^ x| xj xf vO Y1 y 2 [Ay A y An Ay j '^o' q C2 .C 3.

[0095] We can also introduce the variable V. which corresponds to the lateral coordinate

[0096]

[0097]

[0098] measured from point Pi (unlike the variable y; which corresponds to the estimated lateral coordinate). This estimated coordinate is calculated at each time step and is recorded in a third field of the register, based on the previous lateral coordinates. We can also introduce the coefficients Q (1=0,..,3) of the trajectory. Then, this coefficient matrix can be determined by the following equation: [Math. 8] <T C2 .C3. = arg [Math. 8]

[0099]

[0100] To simplify, we can therefore write: [Math.9] '0 / C2 S 3, with [Math.9] And [Math.9]

[0101] In other words, the equation Math.7 gives us the polynomial function of order 3 to estimate the values ​​of the variable y;, and the calculation of the equation Math.8 allows us to find the values ​​of the coefficients C so that the points of coordinates (xi, are as close as possible to the points with coordinates (x;, y;).

[0102] At this stage, block B4 of the computer will therefore be able to calculate the second control instruction for the power steering actuator, the one which would allow the ego vehicle 10 to follow the target trajectory 29, using the controller K.

[0103] Conventionally, the trajectory tracking of the target vehicle 20 is never exact and tracking delays and / or instabilities are often observed. Moreover, during the initialization of the AUS function, the ego-vehicle 10 is often offset from this trajectory and / or oriented in a different direction, so that it cannot follow it Exactly.

[0104] The use of the aforementioned controller K makes it possible to reduce the tracking delay as much as possible. Such a controller is for example described in document FR3082162.

[0105] This controller K uses for this a state vector which includes a state variable formed by the lateral deviation between the ego-vehicle 10 (or its trajectory 19) and the target trajectory 29.

[0106] To estimate this lateral deviation, several types of strategy can be used (see [Fig.4]).

[0107] The first strategy consists of piloting the ego-vehicle 10 by taking into account, as lateral deviation, the lateral distance ey between the ego-vehicle 10 and the target trajectory 29, at the level of the center of gravity of the ego-vehicle 10. This first strategy makes it possible to avoid cutting corners when the two ego-vehicles 10 and target 20 follow each other in a corner. It has the disadvantage of generating instabilities in the trajectory tracking, which are at least unpleasant for the passengers of the ego-vehicle 10, or even dangerous.

[0108] The second strategy consists of piloting the ego-vehicle 10 by taking into account, as lateral deviation, the lateral distance yk between the trajectory 19 of the ego-vehicle 10 and the target trajectory 29, at a predefined aiming distance in front of the ego-vehicle 10. This strategy reduces instabilities but can generate lane departures when cornering.

[0109] The third strategy, which is preferentially adopted here, consists of using a variable aiming distance.

[0110] In practice, when the longitudinal distance between the ego vehicle 10 and the target vehicle 20 is greater than a predetermined distance threshold Sd, the aiming distance 1a varies as a function at least of the speed of the ego vehicle 10.

[0111] The higher this speed, the higher the risk of corner cutting, and therefore the smaller the aiming distance will be.

[0112] It may be noted here that as a variant or in addition, this aiming distance could also vary as a function of the curvature of the target trajectory 29 and / or the lateral distance yk.

[0113] In the case where the longitudinal distance between the ego vehicle 10 and the target vehicle 20 is less than the predetermined distance threshold Sd, the aiming distance 1a will be chosen equal to this longitudinal distance.

[0114] In any case, at this stage, the calculator was therefore able to calculate: - the first driving instruction within the framework of the LCA function (allowing the ego-vehicle 10 to remain centered in its lane), and - the second driving instruction within the framework of the AUS function (allowing the ego vehicle 10 to follow the target vehicle 20).

[0115] In practice, the LCA function remains activated as soon as conditions allow, the AUS function taking over from the LCA function when conditions are no longer favorable.

[0116] Block B2 is more specifically responsible for determining when to use the AUS function rather than the LCA function.

[0117] In this case, the AUS function can only be activated if there is a target vehicle preceding the ego vehicle 10 on the same traffic lane.

[0118] In this case, the computer relies on the reliability level of the first control instruction and that of the second control instruction to determine which instruction to use.

[0119] In practice, these reliability levels are calculated based on the following information: - a quality of marking of the marking lines, - a criterion of visibility of the road 30 by the camera, - a geometric characteristic of the traffic lane 31, such as its width, - a dynamic characteristic of the motor vehicle 10 such as its speed, - a criterion of visibility of the target vehicle 20 by the remote sensor.

[0120] Typically, if the track edge markings are no longer visible and the target vehicle 20 is clearly visible, the computer can select the second control instruction (within the framework of the AUS function).

[0121] The transition between one and the other of the functions can be done directly or indirectly. Before detailing the possible transitions, we can describe in more detail the states that the LCA and AUS functions can take.

[0122] These functions can take: - an activated LCA state in which the first developed piloting instruction is selected to pilot the ego-vehicle 10, - an activated AUS state in which the second developed piloting instruction is selected to pilot the ego-vehicle 10, - a deactivated state in which none of these instructions are used to control the ego-vehicle 10, - an error state in which none of these instructions are used to control the ego-vehicle 10 since the acquired data are not sufficient to implement the AUS and LCA functions (for example because the camera or the remote sensor is faulty), and - an inhibited state in which the piloting instructions are momentarily not used to pilot the ego-vehicle 10 (for example because the environment is deemed too dangerous for this).

[0123] In practice, the calculator can directly pass: - from the activated LCA state to one of the four states, or vice versa, - from the activated AUS state to one of the four states, - from any state (except error state) to AUS enabled state, - from error state to disabled state, or vice versa.

[0124] It is understood that when switching from the LCA function to the AUS function, or vice versa, the control instruction of the power steering actuator may suddenly vary. It is therefore preferentially provided to filter this variation to smooth the control instruction received by the power steering actuator. This filtering operation is advantageously carried out for a determined duration.

[0125] A first filtering consists of applying a low-pass filter, for example of the first order, to the input signals of the controller K.

[0126] Here, this filter is thus applied to the following inputs: - the measured steering angle signal, - the measured lateral distance signal ey or yk, - the target vehicle trajectory curvature signal 20, - the signal of the derivative of the curvature, and - possibly, signals relating to the track (width, etc.).

[0127] Here we use a single low-pass filter (with its time constant which can vary) to switch from the LCA function to the AUS function and vice versa.

[0128] A second filtering consists of ensuring a switch from one of the first and second control instructions to the other of these control instructions, so that the instruction applied by the power steering actuator varies continuously (and not in stages).

[0129] Typically, by noting Co the steering instruction to be sent to the power steering actuator, Ci the first steering instruction (calculated according to the LCA function) and C2 the second steering instruction (calculated according to the AUS function), this second filtering could consist of carrying out the following calculation:

[0130] [Math. 10] Co = « .Cj + ( 1 - « ).C?

[0131] The term a would then be a variable which varies continuously between 0 and 1 during the duration of the filtering, for example in an affine manner, at a speed which would be either constant or variable (for example depending on the speed of the ego-vehicle 10).

[0132] However, the solution employed here is a little different.

[0133] It consists of filtering the input signals by means of the aforementioned low-pass filter, in applying a suitable time constant to them.

[0134] This time constant is preferably adaptive, that is to say that it can vary according to parameters relating to the ego-vehicle 10 or to the target-vehicle 20 or to the environment.

[0135] Here, this time constant varies according to the speed of the ego-vehicle 10. The filtering is thus carried out for a duration which adapts to the situation.

[0136] In practice, when switching from the LCA function to the AUS function, this duration is chosen to be at least equal to the time required for the ego vehicle 10 to reach the position that the target vehicle 20 had at the time of this switch from one function to the other. In other words, the filtering remains active at least as long as the distance between the ego vehicle 10 and the point P corresponding to the position of the target vehicle 20 at the time of activation of the AUS function is strictly greater than zero. Here, the filtering is active for this duration, then it gradually decreases afterwards.

[0137] On the other hand, when switching from the AUS function to the LCA function, this duration is chosen to be shorter. Here, the filtering is intended to be activated when switching to the LCA function and then to immediately begin to decrease thereafter, in a progressive manner.

[0138] As an example, [Fig.3] shows the variation of this time constant.

[0139] In this example, the LCA function is initially activated (before time to), then the The AUS function is activated in turn in place of the LCA function at time t0. Filtering then takes place during the durations ôtl and ôt2. During the duration ôt3, the AUS function remains activated without filtering. The LCA function is then reactivated at time tb so that filtering takes place during the duration ôt4. During the duration ôt5, the LCA function remains activated without filtering.

[0140] At time t0 and for duration ôtl, the time constant takes a non-zero value. Its value is determined as a function of the speed of the vehicle-ego 10. In this case, the higher the speed, the lower this time constant.

[0141] Then, as soon as the ego-vehicle 10 reaches the initial position of the target-vehicle 20, the value of the time constant is continuously decreased until it becomes zero (during the duration ôt2). Here, this decrease is planned to follow a predetermined slope ramp.

[0142] During the duration ôt3, the AUS function remains activated without any filtering.

[0143] As soon as the LCA function can be reactivated, filtering is restored.

[0144] Thus, during the duration ôt4, the time constant takes an initially non-zero value. This value is preferably determined as a function of the speed of the ego-vehicle 10. Then this value of the time constant is continuously decreased until it becomes zero. Here again, this decrease is intended to follow a predetermined slope ramp.

[0145] Finally, during the duration ôt5, the LCA function remains activated without any filtering.

[0146] The present invention is in no way limited to the embodiment described and shown, but those skilled in the art will be able to provide any variant in accordance with the invention.

Claims

Claims

1. Method for controlling a motor vehicle (10) traveling on a traffic lane (31) of a road (30), comprising steps of: - acquiring first data relating to edge lines (33, 35) of said traffic lane (31), - estimating a traffic lane trajectory (39) taking into account said first data, - determining a first control instruction for the motor vehicle (10) by means of a controller (K) and as a function of said traffic lane trajectory (39), and - steering the motor vehicle (10) by a steering actuator, the steering method also comprising, when the motor vehicle (10) follows a third-party vehicle (20), steps of: - acquisition of second data relating to the third-party vehicle (20), - evaluation of a third-party vehicle trajectory (29) taking into account said second data, - determination of a second control instruction for the motor vehicle (10) by means of said controller (K) and as a function of the trajectory of the third-party vehicle (29), and - selection of one of the first and second steering instructions, said steering actuator being controlled according to the selected steering instruction, in which it is provided to measure and record successive positions (Pi) of the third-party vehicle (20) in which, in the evaluation step, the third-party vehicle trajectory (29) is determined as a function of said successive positions (Pi), and in which, the successive positions (Pi) being measured at successive time steps, during the measurement and recording of a new position (Pn) of the third-party vehicle (20) at a current time step, at least part of said successive positions (Pi) already recorded are corrected taking into account the movement of the motor vehicle (10) between the current time step and the time step preceding the current time step.

2. A driving method according to claim 1, in which it is provided to estimate at least one position of the third-party vehicle (20) and in which, in the evaluation step, the trajectory of the third-party vehicle (29) is determined as a function of a deviation between the estimated position and one of the positions successive (Pi) measured.

3. A driving method according to claim 1 or 2, wherein the controller (K) is a function of a lateral deviation between a trajectory (19) of the motor vehicle (10) and the trajectory of a third vehicle (29), measured at a distance (la) in front of the motor vehicle (10) which varies as a function of the speed of the motor vehicle (10).

4. Control method according to one of claims 1 to 3, in which, when at a first time step, one of the first and second control instructions is selected and, at a second time step following the first time step, the other of the first and second control instructions is selected, the steering actuator is controlled according to the two control instructions for a duration (ôt2, ôt4) determined after the second time step.

5. A driving method according to one of claims 1 to 4, wherein, when at a first time step, the first driving instruction is selected and, at a second time step following the first time step, the second driving instruction is selected, the second driving instruction is filtered for a duration (ôtl, ôt2) which is at least equal to the time necessary for the motor vehicle (10) to travel the distance separating, at the time of the second time step, the motor vehicle (10) from the third-party vehicle (20).

6. Control method according to one of claims 1 to 5, in which it is provided to determine a reliability level of said first control instruction and in which the second control instruction is selected only if said reliability level is lower than a threshold.

7. A driving method according to claim 6, wherein said reliability level is calculated as a function of at least one of the following information: - a quality of marking of the edge lines (33, 35), - a criterion of visibility of the road (30) by sensors for acquiring the first data, - a geometric characteristic of the traffic lane (31), - a dynamic characteristic of the motor vehicle (10).

8. Motor vehicle (10) comprising: - first acquisition means adapted to acquire, when the motor vehicle (10) is moving on a traffic lane (31) and is preceded by a third-party vehicle (20), first data relating to the edge lines (33, 35) of said traffic lane (31), - second acquisition means adapted to acquire second data relating to the third-party vehicle (20), and - a steering actuator adapted to control the motor vehicle (10), characterized in that it further comprises a computer (11) programmed to implement a control method in accordance with one of claims 1 to 7.

Images