WO2024042278A1

WO2024042278A1 - Methods and systems for controlling lane-change authorisations in wet weather

Info

Publication number: WO2024042278A1
Application number: PCT/FR2023/050993
Authority: WO
Inventors: Hamza El Hanbali; Yassine Et-Thaqfy; Zoubida LAHLOU
Original assignee: Stellantis Auto Sas
Priority date: 2022-08-26
Filing date: 2023-06-29
Publication date: 2024-02-29
Also published as: FR3139092A1

Abstract

The invention relates to a system (2) for controlling lane-change authorisations of a vehicle (1) from a first lane (11) to a second lane (12), comprising - a computer (4) configured to: determine a first demarcation line (6) separating the first and second lanes and a second demarcation line (7) demarcating the second lane together with the first line, based on an image covering the second lane; calculate a variation in the width of the second lane, based on the first and second lines; determine a centre line (8) of the second lane, based on the image; and, when the variation is greater than a threshold value, calculate a distance between the first line and the centre line; the system being configured so as not to authorise a lane change into the second lane when the distance is outside a confidence interval.

Description

Title of the invention: Methods and systems for controlling authorizations to change traffic lanes in rainy weather The present invention claims priority from French application 2208544 filed on 08.26.2022, the content of which (text, drawings and claims) is hereby incorporated by reference.

[0001] The present invention relates to methods and systems for controlling authorizations to change the lane of a motor vehicle, particularly in rainy weather.

[0002] The relative position of the motor vehicle in relation to the traffic lane demarcation lines is essential for several driving assistance functions, including in particular the semi-automatic lane change assistance function (better known as English name “Semi-Automatic Lane Change” or SALC). This function allows, on command from the driver and under his supervision, to carry out a lane change maneuver from the current lane of the motor vehicle towards the center of a target lane which is adjacent to it. To do this, this function is based, among other things, on the detection of the ground marking lines delimiting the adjacent destination traffic lane and its width. The authorization of a lane change is therefore dependent on reliable detection of these demarcation lines.

[0003] For this, existing methods and systems for controlling lane change authorizations are configured to detect the delimitation lines of the adjacent destination traffic lane and verify that its width is greater than a predefined threshold value to authorize a change of lane. This width is calculated from the polynomials representative of the detected boundary lines. The availability of the semi-automatic lane change assist function requires precise detection of the boundary lines of the target adjacent lane and a suitable width of the latter.

[0004] However, in rainy weather or, more generally, in the presence of water on the roadway, the recent passage of a motor vehicle in the adjacent destination traffic lane causes shiny traces of wheels to appear which may be confused with a delimitation line (cases of false positives). It results an inaccurate identification of this traffic lane and the lateral position of the motor vehicle in relation to it and, consequently, a possible inappropriate availability of the semi-automatic lane change assistance function or an interruption mid- path of this function if it is already running. Such an error is detrimental to the SALC function, as well as it may harm the safety of road users.

[0005] An object of the present invention is to remedy the aforementioned drawbacks.

Another object of the present invention is to improve the control of the availability of the semi-automatic lane change assistance function.

[0007] For this purpose, it is proposed, firstly, a method for controlling authorizations to change the lane of a motor vehicle from a first lane to a second adjacent destination lane, this method comprising the following steps:

- acquisition of at least one image covering the second channel;

- determination, from said at least one image on the basis of a first set of predefined criteria, of a first delimitation line separating the first lane and the second lane and of a second delimitation line delimiting with the first boundary line the second lane;

- calculation, from the first delimitation line and the second delimitation line, of a width of the second lane;

- calculation of a variation over time of said width;

- measurement of a lateral separation distance between the motor vehicle and an obstacle separated from the motor vehicle by the first delimitation line;

- determination, from said at least one image on the basis of a second set of predefined criteria different from the first set and/or from said lateral separation distance, of a center line of the second channel;

- calculation, when said variation is greater than a predefined threshold value, of a distance between the first delimitation line and the center line;

- non-authorization, when the distance between the first line of delimitation and the center line is outside a predefined confidence interval, of a lane change to the second lane of traffic.

[0008] Various additional characteristics can be provided, alone or in combination:

- the method presented above further comprises a step of detecting the presence of rain, the distance between the first delimitation line and the center line being calculated when the presence of rain is further detected ;

- the first set comprises a contrast criterion for detection of the first delimitation line and the second delimitation line in said at least one image;

- the calculated variation is a first derivative with respect to time of said width.

[0009] Secondly, a system is proposed for controlling authorizations to change the lane of a motor vehicle from a first lane to a second adjacent destination lane, this system comprising:

- an image sensor for acquiring at least one image covering the second channel;

- a calculator configured to determine, from said at least one image on the basis of a first set of predefined criteria, a first delimitation line separating the first lane and the second lane and a second delimitation line delimiting with the first boundary line the second lane; calculate, from the first boundary line and the second boundary line, a width of the second lane; calculate a variation over time of said width;

- a distance sensor capable of measuring a lateral separation distance between the motor vehicle and an obstacle separated from the motor vehicle by the first delimitation line; the calculator being, furthermore, configured to determine, from said at least one image on the basis of a second set of predefined criteria different from the first set and/or from said lateral separation distance, a center line from the second way; calculating, when said variation is greater than a predefined threshold value, a distance between the first boundary line and the center line; the system being configured not to authorize, when the distance between the first delimitation line and the center line is outside a predefined confidence interval, a change of lane towards the second traffic lane.

[0010] Various additional characteristics can be provided, alone or in combination:

- the system presented above further comprises a rain sensor capable of detecting the presence of rain, the distance between the first delimitation line and the center line being calculated when the presence is further detected rain ;

[0011] Thirdly, a motor vehicle is proposed comprising the system presented above.

[0012] Other characteristics and advantages of the invention will appear more clearly and concretely on reading the following description of embodiments, which is made with reference to the appended drawings in which:

[0013] Figure [Fig. 1] schematically illustrates a motor vehicle provided with a system for controlling lane change authorizations according to various embodiments;

[0014] Figure [Fig.2] schematically illustrates the steps of a method for controlling lane change authorizations according to various embodiments.

Referring to Figure 1, a system 2 is displayed for controlling authorizations to change the lane of a motor vehicle 1 from a first lane 11 to a second adjacent lane 12 of destination .

The first traffic lane 11 is the current traffic lane of the motor vehicle, that is to say on which the motor vehicle 1 is currently traveling. The second track 12 of traffic is the neighboring (or adjacent) traffic lane to which the driver wishes to change lanes. The second traffic lane 12 can be immediately to the left (as illustrated in Figure 1) or to the right of the first traffic lane 11 (in particular, for an overtaking or folding maneuver of the automobile vehicle 1 according to the road traffic rules).

The system 2 comprises at least one image sensor 3 (in particular, a camera) for the acquisition (step 21 in FIG. 2) of at least one image covering the second target traffic lane 12 (or destination). The image sensor 3 at least partially covers the environment at the front of the automobile vehicle 1. One or more cameras used in frontal vision can be arranged to capture images of the second traffic lane 12 in front of the automobile vehicle 1 as it moves. Multi-sensor fusion algorithms for combining images from several image sensors 3 can be considered.

[0018] The system 2 further comprises a calculator 4 configured to carry out image processing to determine (step 22 in FIG. 2), from the images acquired by the image sensor 3, - a first line 6 of boundary separating the first traffic lane 11 and the second traffic lane 12; and a second delimitation line 7 delimiting with the first demarcation line 6 the second traffic lane 12.

The first delimitation line 6 is a separation line between the first traffic lane 11 and the second adjacent destination traffic lane 12. The second boundary line 7 is the next boundary line after the first boundary line 6 to the left (as illustrated in Figure 1) or to the right of the automobile vehicle 1. The first delimitation line 6 and the second delimitation line 7 are two successive delimitation lines detected on the acquired images covering the second traffic lane 12.

The second demarcation line 7 can be a real ground marking line (continuous or discontinuous, single or double, or even separation or roadside) delimiting longitudinally, with the first demarcation line 6, the second traffic lane 12 adjacent target or a linear trace in the water of the wheels of a vehicle traveling on the second traffic lane 12.

The determination of the first and second delimitation lines 6, 7, from an image acquired by the image sensor 3, is based on a first set of predefined criteria. Starting from the hypothesis that the demarcation lines must present, with respect to the roadway, a contrast greater than a predefined threshold contrast, said first set comprises, in one embodiment, a contrast criterion for detection of the first and the second delimitation line 6, 7 in an image acquired by the image sensor 3. Alternatively or in combination, the first set of criteria may include color, dimensions, regularity or, more generally, any other distinctive criterion of the delimitation lines. Algorithms known from the state of the art (based, for example, on a Hough transform, a generalized Hough transform or equivalent) can be used for this purpose. Each of the first and second delimitation lines 6, 7 is determined in a reference point linked to the automobile vehicle 1 by polynomial interpolation, in particular a third degree polynomial interpolation, of points representative of these lines detected in the acquired images . The reference frame linked to the automobile vehicle 1 is a Cartesian coordinate system whose abscissa axis corresponds to the direction of the longitudinal movement of the automobile vehicle 1 on the road 10 and the ordinate axis corresponding to the direction of the lateral movement of the automobile vehicle 1. The constant coefficient (that is to say, the coefficient of the monomial of degree 0) of the polynomial of a boundary line represents the lateral position of this boundary line with respect to the automobile vehicle 1 (or, also, the ordered to the origin of the corresponding boundary line in the reference linked to the automobile vehicle 1).

From the first delimitation line 6 and the second delimitation line 7, the calculator 4 calculates (step 23 in Figure 2) a width of the second traffic lane 12. In one embodiment, a width of the second traffic lane 12 is given by the average of the constant coefficient of a first polynomial representing the first delimitation line 6 and the coefficient constant of a second polynomial representing the second delimitation line 7.

The calculator 4 is, furthermore, configured to calculate (step 24 in Figure 2) a variation over time (or, equivalently, along the road 10 with the movement of the automobile vehicle 1) of the width of the second adjacent destination traffic lane 12. This variation is, in one embodiment, the first derivative with respect to time of the width of the second traffic lane 12 (the difference between the values of this width calculated at two successive instants divided by the time step between these two instants ). A first temporal (or, equivalently, spatial) derivative quantifies the variation in the width of the second traffic lane 12 per unit of time (or per unit of distance traveled by the automobile vehicle 1).

[0025] Furthermore, the system 2 comprises a distance sensor 5 capable of measuring (step 25 in FIG. 2) a lateral separation distance between the automobile vehicle 1 and an obstacle separated from the automobile vehicle 1 by the first line 6 of delimitation. This distance sensor 5 is, in one embodiment, a contactless distance measuring sensor such as an ultrasonic sensor, an optical distance measuring sensor, an infrared distance measuring sensor or, more generally, a radar . Said obstacle is, for example, a user of route 10, a vertical border of route 10, a road bench, a sidewalk, a central or lateral reservation of route 10 or any other fixed or mobile obstacle separated from the automobile vehicle 1 by the first delimitation line 6 and making it possible to deduce directly or in combination with information extracted from the images acquired by the image sensor 3 a location of the center of the second adjacent traffic lane 12 of destination.

The calculator 4 is, in fact, able to determine (step 26 in Figure 2), from an image acquired by the image sensor 3 and/or from the lateral separation distance measured by the distance sensor 5, a center (or middle) line 8 of the second traffic lane 12.

[0027] In one embodiment, assuming that the width of the second traffic lane 12 is equal to a predefined value (for example, associated with the type of road 10 on which the vehicle is traveling automobile vehicle 1 and/or the position of the latter), the computer 4 determines, from the lateral separation distance, a center line 8 of the second traffic lane 12. Alternatively or in combination, the center line 8 is determined by the calculator 4 from the images acquired on the basis of a second set of predefined criteria different from the first set. This second set can include any distinctive criterion of an object present on the images acquired by the image sensor 3 making it possible to deduce directly or indirectly the center line 8 by measuring a relative or absolute lateral separation distance between this object and the vehicle 1 automobile. This second set includes, for example, a criterion of contrast, color, position, dimension, and/or orientation relating to an element of the road infrastructure such as a border of the road 10 or signage road on the ground or vertical, or a criterion relating to one or more road users such as the number, the direction of movement, the trajectory, or its lateral distance in relation to the vehicle 1 automobile.

Advantageously, the determination of a center line 8 of the second traffic lane 12 is based on a fusion of data extracted from the images acquired and/or from the distance sensor 5. The second center line 7 is, in one embodiment, represented by a polynomial of degree 3. Advantageously, the delimitation lines 6, 7 and the center line 8 are determined according to two different or decorrelated methods (sets of different criteria or search for different elements in the acquired images).

[0029] In one embodiment, when the variation in the width of the second traffic lane 12 is greater than a predefined threshold value, the calculator 4 calculates (step 27 in Figure 2) a distance between the first line 6 of delimitation and center line 8. This distance is assumed to be the half-width of the second adjacent destination traffic lane 12. A value of this distance is given by the difference between the constant coefficient of the polynomial representing the center line 8 and the constant coefficient of the polynomial representing the first delimitation line 6.

[0030] In one embodiment, the system 2 monitors the variation in the width of the second traffic lane 12 and triggers the measurement step 25 and the step 26 of determining the center line 8 as soon as this variation is greater than the predefined threshold value. A significant variation in the width of the second adjacent destination traffic lane 12 is considered by the system 2 as resulting from inaccurate detection of the second delimitation line 7, particularly in rainy weather where this second delimitation line 7 can be confused with a trace of the wheels of a vehicle traveling on the second lane of traffic. In rainy weather, the determination of the first boundary line 6 and the center line 8 is considered by the system 2 to be more reliable than that of the second boundary line 7.

[0031] In one embodiment, system 2 further comprises a rain sensor capable of detecting the presence of rain. The half-width of the second traffic lane 12 (distance between the first delimitation line 6 and the center line 8) is calculated when the presence of rain is also detected.

System 2 is configured not to authorize (step 28 in Figure 2), when the calculated half-width is outside a predefined confidence interval, a change of lane towards the second traffic lane 12. The confidence interval is, for example, 1.2m to 2m, given that the width of a traffic lane is most often between 2.4m and 4m.

[0033] System 2 considers, in fact, that a significant variation in the width of the second traffic lane 12 in combination with a half-width of the latter outside the confidence interval results from inaccurate detection of the second delimitation line 7. Consequently, the system 2 prevents or refuses a lane change (in other words, deactivates or makes unavailable the semi-automatic lane change assistance function) towards this second lane 12 of which the second demarcation line 7 is detected imprecisely. In this case, if the driver orders a lane change to the second traffic lane 12 by operating the indicator or the like, the system 2 makes the semi-automatic lane change assistance function unavailable.

[0034] Advantageously, the embodiments described above make it possible to improve the security and behavior of the function semi-automatic lane change assistance, particularly in rainy weather.

Claims

[Claim 1] Method for checking authorizations to change the lane of a motor vehicle (1) from a first lane (11) of traffic to a second adjacent destination traffic lane (12), this method comprising the following steps:

- acquisition (21) of at least one image covering the second channel (12);

- determination (22), from said at least one image on the basis of a first set of predefined criteria, of a first demarcation line (6) separating the first channel (11) and the second channel (12) and a second delimitation line (7) delimiting with the first delimitation line (6) the second lane (12);

- calculation (23), from the first delimitation line (6) and the second delimitation line (7), of a width of the second lane (12);

- calculation (24) of a variation over time of said width;

- measurement (25) of a lateral separation distance between the motor vehicle (1) and an obstacle separated from the motor vehicle (1) by the first boundary line (6);

- determination (26), from said at least one image on the basis of a second set of predefined criteria different from the first set and/or from said lateral separation distance, of a center line (8) from the second track (12);

- calculation (27), when said variation is greater than a predefined threshold value, of a distance between the first delimitation line (6) and the center line (8);

- non-authorization (28), when the distance between the first delimitation line (6) and the center line (8) is outside a predefined confidence interval, of a change of lane towards the second lane ( 12) circulation.

[Claim 2] Method according to the preceding claim, characterized in that it further comprises a step of detecting the presence of rain, the distance between the first delimitation line (6) and the center line (8). being calculated when the presence of rain is also detected.

[Claim 3] Method according to claim 1 or 2, characterized in that the first set comprises a contrast criterion for detection of the first delimitation line (6) and the second delimitation line (7) in said at least one picture.

[Claim 4] Method according to any one of the preceding claims, characterized in that the calculated variation is a first derivative with respect to time of said width.

[Claim 5] System (2) for controlling authorizations to change the lane of a motor vehicle (1) from a first lane (11) of traffic to a second adjacent destination traffic lane (12), this system comprising:

- an image sensor (3) for acquiring at least one image covering the second channel (12); - a calculator (4) configured to determine, from said at least one image on the basis of a first set of predefined criteria, a first delimitation line (6) separating the first channel (11) and the second channel ( 12) and a second delimitation line (7) delimiting with the first delimitation line (6) the second lane (12); calculating, from the first boundary line (6) and the second boundary line (7), a width of the second lane (12); calculate a variation over time of said width;

- a distance sensor (5) capable of measuring a lateral separation distance between the automobile vehicle (1) and an obstacle separated from the automobile vehicle (1) by the first boundary line (6); the calculator (4) being, in addition, configured to determine, from said at least one image on the basis of a second set of predefined criteria different from the first set and/or from said lateral separation distance, a center line (8) of the second track (12); calculating, when said variation is greater than a predefined threshold value, a distance between the first boundary line (6) and the center line (8); the system (2) being configured not to authorize, when the distance between the first delimitation line (6) and the center line (8) is outside a predefined confidence interval, a change of lane towards the second traffic lane (12).

[Claim 6] System according to the preceding claim, characterized in that it further comprises a rain sensor capable of detecting the presence of rain, the distance between the first delimitation line (6) and the line (8) center being calculated when the presence of rain is also detected.

[Claim 7] System according to the preceding claim, characterized in that the first set comprises a contrast criterion for detection of the first delimitation line (6) and the second delimitation line (7) in said at least one image.

[Claim 8] System according to the preceding claim, characterized in that the calculated variation is a first derivative with respect to time of said width.

[Claim 9] Motor vehicle (1) comprising the system (2) of any one of claims 5 to 8.