FR3082044A1 - METHOD AND DEVICE FOR DETECTING THE TRACK ON WHICH A VEHICLE IS MOVING, ACCORDING TO THE DETERMINED DETERMINATIONS - Google Patents

Abstract

A detection method is used in a vehicle (V) traveling on a road comprising traffic lanes (VC1-VC2) defined by two boundaries, and analyzing the environment around it to determine first information representative of boundaries. This process includes: - a first step where the first information is analyzed to determine mathematical functions representative of the delimitations and confidence indices that these mathematical functions represent delimitations, - a second step where we determine from the functions mathematics retained because of their confidence indices of the widths separating neighboring delimitations, - a third step where we consider that there is a traffic lane present when two neighboring delimitations defining it have a width comprised in a chosen interval, and - a fourth stage where it is determined on which lane of traffic the vehicle is traveling (V).

Description

PATHWAY DETECTION METHOD AND DEVICE

TRAFFIC ON WHICH A VEHICLE IS MOVING, IN OPERATION

SPECIFIC DELIMITATIONS

The invention relates to land vehicles, possibly with at least partially automated (or autonomous) driving and possibly of the automobile type, and more precisely the detection of the traffic lanes on which such vehicles circulate.

In what follows, it is considered that a vehicle is driven at least partially automated (or autonomous) when it can be driven automatically (partial or total (without intervention of its driver)) during an automated driving phase, or manually (and therefore with the driver intervening on the steering wheel and / or pedals) during a manual driving phase.

Certain vehicles defined above, generally of the automobile type, include a satellite positioning device (or GNSS (“Global Navigation Satellite System”) - for example of the GPS (“Global Positioning System”), Galileo or Glonass type), loaded to determine their current position, and a navigation aid device, storing known environmental information and constituting road maps, possibly in high definition. Note that this navigation aid device may possibly include this satellite positioning device.

Some of these road maps notably define the number of traffic lanes on certain roads associated with very precise geographic positions. When the current GNSS position of the vehicle is sufficiently precise, the navigation aid device can determine the lane on which this vehicle is circulating in a road map. This knowledge of the lane on which this vehicle is traveling can in particular be used by a device for assisting in driving the vehicle (possibly of the ADAS (“Advanced Driver Assistance System”) type) to control its transverse and longitudinal positions on the route taken. This knowledge can also make it possible to materialize on a screen of the vehicle, for example that of the central handset, to inform the driver, the position of the vehicle relative to the traffic lanes of the road taken.

However, as the person skilled in the art knows, the road maps stored in a vehicle are very rarely up to date, and therefore the definition of a portion of stored road may not correspond to reality following temporary works or to a final modification. Consequently, these road maps can provide erroneous information, and thus make possible automated driving unsafe.

Furthermore, the precision of the determined position can vary significantly depending on the constellation of satellites used by the satellite positioning device on board the vehicle and according to the environment in which the vehicle is moving (open air or tunnel, open or closed parking, city or country). However, these variations in accuracy or availability of position information can temporarily cause errors in determining the presence or absence of lane (s) of traffic, and thus make possible automated driving unsafe.

It has certainly been proposed in patent document EP 1035509 to determine the boundaries of the lane of the road on which a vehicle is traveling by analysis of environmental information acquired by sensors on board this vehicle and representative of the environment at least in front of the latter. This environmental information generally comes from digital images acquired by at least one camera installed in a front part of the vehicle. Such a solution only makes it possible to determine the only traffic lane on which the vehicle is located, and again with a reliability which depends on the quality of the camera and on the quality of the analysis of the environmental information carried out by the circuits. analysis on board the vehicle.

The invention therefore aims in particular to improve the situation.

It proposes in particular for this purpose a detection method intended to be implemented in a vehicle traveling on a road comprising traffic lanes each defined by two delimitations, and analyzing the environment around at least part of it in order to determine initial information representative of boundaries.

This detection process is characterized by the fact that it includes:

a first step in which the first determined information is analyzed in order to determine mathematical functions representative of the delimitations and confidence indices representative of the confidence that these determined mathematical functions represent delimitations, and only the mathematical functions associated with a confidence index higher than a chosen threshold,

- a second step in which widths separating neighboring delimitations are determined from the mathematical functions used in at least one chosen location,

- a third stage in which it is considered that there is a traffic lane present when two neighboring delimitations which define it have a width comprised within a chosen interval, and

- a fourth stage in which it is determined on which lane of traffic present the vehicle is traveling.

Thanks to the invention, it is now possible to permanently determine the traffic lane on which the vehicle actually travels, whether or not it is equipped with a navigation aid device storing road maps.

The detection method according to the invention may include other characteristics which can be taken separately or in combination, and in particular:

- in its fourth step, it is possible to determine among the delimitations of the traffic lanes present those which are situated to the right of the vehicle and their first number, and those which are situated to the left of the vehicle and their second number, and one can determine the lane present traffic on which the vehicle travels as a function of these first and second determined numbers;

- in its first step we can determine mathematical functions of the polynomial type of degree three;

- in its first step, each confidence index associated with a mathematical function representing a delimitation can be determined as a function of the number of first pieces of information defining points detected in the environment and used to determine this mathematical function.

The invention also provides a computer program product comprising a set of instructions which, when executed by processing means, is capable of implementing the detection method described above to determine on which channel of present traffic is circulating a vehicle.

The invention also provides a detection device intended to equip a vehicle traveling on a road comprising traffic lanes each defined by two delimitations, and analyzing the environment around at least part of it in order to determine first representative information. of boundaries.

This detection device is characterized by the fact that it includes a processor:

- analyzing the first determined information in order to determine mathematical functions representative of the delimitations and confidence indices representative of the confidence that these determined mathematical functions represent delimitations, and retaining only the mathematical functions associated with a confidence index greater than a threshold chosen then

- determining from the mathematical functions retained widths separating neighboring delimitations in at least one chosen location, then

- considering that there is a traffic lane present when two neighboring delimitations which define it have a width included in a chosen interval and have confidence indices higher than a chosen threshold, then

- determining on which traffic lane the vehicle is traveling.

The invention also provides a vehicle, possibly of the automobile type, capable of traveling on a road comprising traffic lanes each defined by two delimitations, analyzing the environment around at least part of it in order to determine first representative information. of delimitations, and comprising a detection device of the type of that presented above.

For example, this vehicle may also include an assistance device controlling its driving in an at least partially automated manner as a function of the traffic lane present on which it is located and determined by its detection device.

Other characteristics and advantages of the invention will appear on examining the detailed description below, and the attached drawing, in which:

- Figure 1 schematically and functionally illustrates a vehicle located on one of the two lanes of traffic on a road and equipped with a sensor, analysis circuits and an exemplary embodiment of a detection device according to the invention, and

FIG. 2 schematically illustrates an example of an algorithm implementing a detection method according to the invention.

The object of the invention is in particular to propose a detection method, and an associated DD detection device, intended to allow the detection of the traffic lane VCk on which a vehicle V is located.

In what follows, it is considered, by way of nonlimiting example, that vehicle V is of the automobile type. This is for example a car. However, the invention is not limited to this type of vehicle. It concerns in fact any type of land vehicle which can circulate on land traffic lanes each defined by two delimitations.

There is schematically and functionally shown in FIG. 1 a vehicle V traveling on one of the two traffic lanes VCk (k = 1 or 2) of a road R, here framed by safety guardrails RS. It should be noted that the first taxiway VC1 is here separated from the right RS safety railing by an emergency stop band BAU.

This vehicle V comprises at least one sensor CP, analysis circuits CA and an exemplary embodiment of a detection device DD according to the invention.

This CP sensor is responsible for acquiring information in the environment which is located around at least part of the vehicle V.

It will be noted that the number of CP sensors here is equal to one (1), but it can take any value greater than or equal to one (1), since this makes it possible to acquire information in the environment which is located around at least part of vehicle V.

For example, the sensor CP can comprise at least one camera installed in a front part of the vehicle (for example on the windshield or on the interior rear-view mirror), and responsible for acquiring digital images at least in front of the vehicle V and on at least part of its two lateral sides.

As a variant or in addition, the vehicle V can also include at least one ultrasonic sensor, or at least one radar or lidar, or even at least one camera installed in a rear part. These other sensors can be installed at the front or at the rear or on at least one lateral side of the vehicle V.

The analysis circuits CA are arranged so as to analyze the information of the environment coming from each sensor CP in order to determine first information which is representative of delimitations of lanes of traffic VCk. In the example illustrated without limitation in FIG. 1, the analysis circuits CA are part of a CAL computer on board the vehicle V. But this is not compulsory. Indeed, the CA analysis circuits could include their own computer. Consequently, the CA analysis circuits can be produced in the form of a combination of electric or electronic (or "hardware") circuits or components and software (or computer or even "software") modules.

The first information can, for example, define points (having a longitudinal coordinate and a transverse coordinate) in a chosen reference frame attached to the vehicle V, and respectively representative of delimiting portions detected in a digital image acquired by the sensor CP. It will be noted that this first information can also define the heading of the vehicle V, and / or the estimate of the curvature of a supposed delimitation as well as possibly the estimate of the derivative of this curvature, and / or the width of a supposed delimitation, for example.

As mentioned above, the invention proposes in particular a detection method intended to make it possible to detect the traffic lane VCk on which the vehicle V is located at the time considered. It will be noted that the knowledge of the lane on which the vehicle V is traveling can possibly be used by a device for assisting in driving the DAC of the vehicle (possibly of the ADAS (Advanced Driver Assistance System) type), for example for controlling the transverse and longitudinal positions of this vehicle V on the route R taken.

This detection method can be at least partially implemented by the detection device DD which for this purpose comprises at least one processor PR, for example of digital signal (or DSP ("Digital Signal Processor")), possibly associated with the minus a memory.

In the example illustrated without limitation in FIG. 1, the detection device DD is part of the computer CAL which also includes the analysis circuits CA. But it is not compulsory. Indeed, the DD detection device could include its own computer. Consequently, the DD detection device can be produced in the form of a combination of electrical or electronic circuits or components and software modules.

As illustrated in FIG. 2, the detection method according to the invention comprises four steps.

In a first step 10 on (the processor PR) begins by analyzing the first information which has been determined by the analysis circuits CA, in order to determine mathematical functions yj (x) which are representative of the delimitations dj and of the confidence indices vq which are representative of the confidence that these determined mathematical functions yj (x) represent delimitations.

It should be noted that a traffic lane delimitation can take the form of a ground marking (such as a line) or a guardrail (possibly safety), for example.

It will be understood that a confidence index vq represents the confidence that one has in the fact that a mathematical function yj (x) effectively represents a delimitation dj of a traffic lane VCk.

In the example illustrated without limitation in FIG. 1, the processor PR detects a first delimitation di (associated with a first mathematical function yi (x)) just to the right of the vehicle V, a second delimitation d2 (associated with a second mathematical function y2 (x)) just to the left of vehicle V, a third delimitation ds (associated with a third mathematical function ya (x)) a little further to the left of vehicle V (i.e. to the left of the second delimitation d2), and a fourth delimitation RS (associated with a fourth mathematical function y4 (x)) to the right of the first delimitation di).

For example, in the first step 10 of the method one (the processor PR) can determine mathematical functions of the polynomial type of degree three yj (x) = cO + c1x + c2 x ² + c3x ³ , where x represents the distance, following the axis of movement of the vehicle V, separating the origin of the chosen reference frame of a point xi from a delimitation dj represented by the mathematical function yj (x). But other types of mathematical function can be used to represent the delimitations dj, and in particular a clothoid.

Also for example, in the first step 10 of the method one (the processor PR) can determine each confidence index vq associated with a mathematical function yj (x) representing a delimitation dj as a function of the number of first pieces of information defining points xi detected in the environment by the CA analysis circuits.

Each confidence index vq can depend for example on the visibility of the measurement sensor, on the quality of the approximation of the delimitation by a mathematical function (perfect correspondence over the entire measurement range or partial correspondence, for example), of the type of the delimitation detected (marking on the ground or guardrail), of the type of sensor having detected the delimitation (confidence index different depending on whether the delimitation is detected by a marking on the ground or by a camera-type sensor or by a laser). The processor PR then validates the use of the mathematical form of each delimitation dj for the calculation of the deviation of the points xi detected with respect to this delimitation dj.

For example, the processor PR can determine an average value of the deviations of each of the points xi detected with respect to the delimitation dj concerned, then can use this average value to determine a confidence index vq associated with the presence of a channel (channel from vehicle V, lane to the left of vehicle V and lane to the right of vehicle V).

Each confidence index vcj is used to validate or not the use of the associated mathematical function yj (x). If a confidence index vq is less than a chosen threshold, we do not validate the use of the associated mathematical function yj (x), and therefore we do not retain the latter (yj (x)) ·

In a second step 20 of the method on (the processor PR) determines from the mathematical functions y (x) retained (which he has just determined) widths Is separating neighboring delimitations dj and dp (with j '+ j) in at least one location chosen.

It will be noted that the width Is can be determined at a chosen location corresponding to a point xi of a delimitation dj or in several chosen locations included over a chosen distance and corresponding respectively to several points xi of a delimitation dj. In this last alternative, the width Is is for example equal to the average of the widths determined at each of the points xi (or places) chosen. Each chosen location corresponds to a point xi reported in the chosen reference frame in which the mathematical functions yj (x) are constructed. Each width at a point xi is equal to yr (xi) - yj (xi), when yj (xi) and yj (xi) represent two neighboring delimitations dj and dj ’detected.

For example, the points xi can be placed at a distance of between 1 meter and 70 meters from the origin of the reference frame attached to the vehicle V.

In a third step 30 of the method one (the processor PR) considers that there is a traffic lane present when two neighboring delimitations dj and dj ’which define it have a width Is included in a chosen interval.

This chosen interval corresponds to usual (or standard) widths of the traffic lanes of the roads. For example, it can be bounded by the smallest and largest of the lane widths commonly encountered. Thus, when the width Is between two neighboring delimitations dj and dp belongs to the chosen interval, it is certain that they effectively delimit a traffic lane on which a vehicle is normally authorized to circulate, and not a stop strip emergency BAU or a bicycle lane or a parking area.

For example, the bounds for this chosen interval can be 2.8 m and 4.5 m.

In the example illustrated without limitation in FIG. 1, the processor PR considers that there is a first traffic lane VC1 present, framed by the first di and second d2 adjacent delimitations, and a second traffic lane VC2 present, framed by the second d2 and third of neighboring delimitations. The BAU emergency stop strip, between the first delimitation di and the RS safety railing on the right, is considered to be too small to constitute a traffic lane.

In a fourth step 40 of the method one (the processor PR) determines on which traffic lane the vehicle V is traveling.

It will be understood that this traffic lane VCk on which the vehicle V is located is that which is bordered by the two adjacent dj boundaries which are located respectively just to its right and just to its left. In the example illustrated without limitation in FIG. 1, the processor PR considers that the vehicle V is located on the first traffic lane VC1 which is bordered by the first di and second d2 neighboring delimitations.

Thus, it is now possible to permanently determine the traffic lane VCk on which the vehicle V actually travels, whether or not the latter is equipped with a DAN navigation aid device storing road maps. In other words, the DD detection device makes it possible to partially replace a DAN navigation aid device when the current GNSS positions of vehicle V are imprecise or temporarily unavailable and / or when the road map is not up to date or does not correspond locally to reality. It will be noted that the detection device DD can also operate in parallel with a navigation aid device DAN, in order to check that the relative positioning of the vehicle V with respect to the traffic lanes VCk of the road R that it ( DAN) determines is in agreement with the result delivered by the DD detection device.

In the example illustrated without limitation in FIG. 1, the navigation aid device DAN is at least partially installed in the computer CAL which also includes the analysis circuits CA and the detection device DD. But he (DAN) could understand his own calculator. Consequently, a DAN navigation aid device can be produced in the form of software modules, or of a combination of electrical or electronic circuits or components and software modules. When the DAN navigation aid includes its own satellite positioning device (GNSS), it also includes a GNSS receiver, for example installed on the roof of vehicle V.

For example, in the fourth step 40 of the method one (the processor PR) can determine among the delimitations dj of the traffic lanes present those which are located to the right of vehicle V and their first number n1, and those which are located to the left of vehicle V and their second number n2. Then, one (the processor PR) can determine the traffic lane present on which the vehicle V circulates (here VC1) as a function of these first n1 and second n2 determined numbers.

In the example illustrated without limitation in FIG. 1, the first number n1 is equal to one, while the second number n2 is equal to two. Here, there is therefore no taxiway located to the right of vehicle V, but there is a taxiway located to the left of vehicle V. Vehicle V therefore travels on taxiway VC1 which is located further to the right of road R (which is normal when the latter (R) is suitable for left-hand drive vehicles).

Note that in another situation, the first number n1 could be equal to two, while the second number n2 could be equal to one. In this case, there is no taxiway located to the left of vehicle V, but there is a taxiway located to the right of vehicle V. Vehicle V therefore travels on the taxiway which is located further to the left of the road (this is normal when the road is suitable for right-hand drive vehicles or may mean that vehicle V is overtaking when the road is suitable for left-hand drive vehicles).

In yet another situation occurring in a three-lane road, the first n1 and second n2 numbers can be two. In this case, there is a taxiway located to the left of vehicle V and a taxiway located to the right of vehicle V. Vehicle V therefore travels on the central taxiway (which may mean that it is making a Over).

It will also be noted, as illustrated without limitation in FIG. 1, that the vehicle V can also optionally include a DAC assistance device responsible for controlling its driving in an at least partially automated manner as a function of the traffic lane VCk present on which it is located and determined by the DD detection device. In this case, the vehicle V is said to be “with automated (or autonomous) driving”.

It is recalled that here the term “vehicle with automated driving” means a vehicle which can be driven in an automated manner (partial or total (without intervention of its driver)) during an automated driving phase, under control (partial or total ) an assistance device, and possibly manually (and therefore under the sole control of its driver) during a manual driving phase. Such a DAC assistance device can constitute, for example, what a person skilled in the art calls an ADAS device (Advanced Driver Assistance System).

For example, this DAC assistance device can be responsible for determining commands for organs involved in the movements of vehicle V (such as for example the power steering, the engine, the gearbox, the clutch or the braking system. ). These commands are intended to make the vehicle V follow a trajectory determined as a function of the traffic lane VCk present on which it is located (determined by the detection device DD) and of dynamic objects (such as other vehicles or pedestrians) detected in the environment of vehicle V by sensors CP of the latter (V).

In the example illustrated without limitation in FIG. 1, the assistance device DAC is installed in the computer CAL which also includes the analysis circuits CA and the detection device DD. But it (DAC) could include its own calculator. Therefore, an assistive device

DAC can be implemented as software modules, or a combination of electrical or electronic circuits or components and software modules.

It will also be noted that the invention also provides a computer program product comprising a set of instructions which, when executed by processing means of electronic circuits (or hardware) type, such as for example at least part of the analysis circuits CA, is suitable for implementing the detection method described above to determine on which traffic lane present the vehicle V is traveling, for example to allow the lane change of the vehicle V on an adjacent lane from from the moment when the latter was validated (and more precisely considered as present) by the detection method described above.

It will also be noted that in FIG. 1 the detection device DD is very schematically illustrated with only its processor PR. This DD detection device can take the form of a box comprising printed circuits, or of several printed circuits connected by wired or non-wired connections. By printed circuit is meant any type of device capable of performing at least one electrical or electronic operation. As mentioned above, this detection device DD comprises at least one processor PR, for example of digital signal (or DSP (Digital Signal Processor)), a random access memory for storing instructions for the implementation by this processor of the method of detection as described above, and a mass memory for storing data intended to be kept after the implementation of the detection method. The digital signal processor PR receives at least the first information determined by the analysis circuits CA, to analyze it and use it in calculations, possibly after having shaped it and / or demodulated and / or amplified it, in a known manner in itself. The DD detection device can also include an input interface for receiving at least the first determined information, and an output interface for transmitting the results of its analyzes and calculations.

One or more steps of the detection process can be carried out by different components. Thus, the detection method can be implemented by a plurality of processors, random access memory, mass memory, input interface, output interface and / or digital signal processor. In these situations, the DD detection device can be decentralized, within a local network (several processors linked together for example) or a wide area network.

Claims (9)

1. Detection method for a vehicle (V) traveling on a road comprising traffic lanes (VCk) each defined by two delimitations, and analyzing the environment around at least part of it in order to determine first representative information of delimitations, characterized in that it comprises i) a first step (10) in which said first determined information is analyzed in order to determine mathematical functions representative of said delimitations and confidence indices representative of the confidence that these determined mathematical functions represent delimitations, and only the mathematical functions associated with a confidence index higher than a chosen threshold are retained, ii) a second step (20) in which widths separating neighboring delimitations are determined from at least the selected mathematical functions a chosen location, iii) a three ith step (30) in which it is considered that there is a traffic lane present when two neighboring delimitations which define it have a width comprised within a chosen interval, and iv) a fourth step (40) in which it is determined on which present traffic lane circulates said vehicle (V). 2. Method according to claim 1, characterized in that in said fourth step (10) one determines among the delimitations of the traffic lanes present those which are located to the right of said vehicle (V) and their first number, and those which are located to the left of said vehicle (V) and their second number, and the current lane on which said vehicle (V) travels is determined as a function of said first and second determined numbers. 3. Method according to claim 1 or 2, characterized in that in said first step (10) mathematical functions of the polynomial type of degree three are determined. 4. Method according to one of claims 1 to 3, characterized in that in said first step (10) each confidence index is determined associated with a mathematical function representing a delimitation as a function of the number of first pieces of information defining points detected in said environment and used to determine this mathematical function. 5. Computer program product comprising a set of instructions which, when executed by processing means, is capable of implementing the detection method according to one of the preceding claims to determine on which traffic lane present a vehicle is traveling (V). 6. Detection device (DD) for a vehicle (V) traveling on a road comprising traffic lanes (VCk) each defined by two delimitations, and analyzing the environment around at least part of it in order to determine first information representative of boundaries, characterized in that it comprises a processor (PR) i) analyzing said first determined information in order to determine mathematical functions representative of said boundaries and confidence indices representative of the confidence that these determined mathematical functions represent delimitations, and retaining only the mathematical functions associated with a confidence index higher than a chosen threshold, then ii) determining from the mathematical functions retained widths separating neighboring delimitations in at least one chosen location, then iii) considering that there is a traffic lane present when two neighboring delimitations which define it have a width comprised within a chosen interval, then iv) determining on which traffic lane present circulates said vehicle (V). 7. Vehicle (V) capable of traveling on a road comprising traffic lanes (VCk) each defined by two delimitations, and analyzing the environment around at least part of it in order to determine first information representative of delimitations, characterized in that it comprises a detection device (DD) according to claim 6. 8. Vehicle according to claim 7, characterized in that it comprises an assistance device (DAC) controlling its driving in an at least partially automated manner as a function of the traffic lane (VCk) present on which it is located and determined by said detection device (DD). 9. Vehicle according to claim 7 or 8, characterized in that it is of the automobile type.