WO2019220030A1

WO2019220030A1 - Method and device for assisting the automated driving of a vehicle by determining the safety of a proposed trajectory

Info

Publication number: WO2019220030A1
Application number: PCT/FR2019/050995
Authority: WO
Inventors: Rachid ATTIA; Vincent Abadie; Celine Etcheverry; Jean Michel PASCAL; Loic Chambrin; Audrey MANDARD; Nicolas Becker
Original assignee: Psa Automobiles Sa
Priority date: 2018-05-18
Filing date: 2019-04-29
Publication date: 2019-11-21
Also published as: FR3081142A1

Abstract

A method assists the automated driving of a first vehicle (V1) driving on a road (R), storing known information defining roads, and analysing the environment in front of it in order to determine first information representative of this road (R) and its environment. This method comprises a step that involves: -determining trajectories followed by second vehicles (V2) from first determined information, and estimated road markings of the road (R) from first determined information, corresponding known road markings and determined followed trajectories, then determining a reference trajectory for the first vehicle (V1) depending on these estimated road markings, -determining a driving space of the first vehicle (V1) depending on first determined information, and -considering that the reference trajectory can be followed when it is compatible with this driving space.

Description

METHOD AND DEVICE FOR ASSISTING THE AUTOMATED DRIVING OF A VEHICLE BY DETERMINING THE SAFETY OF A PROPOSED TRACK.

The present invention claims the priority of the French application 1854183 filed on May 18, 2013 whose content (text, drawings and claims) is hereby incorporated by reference.

The invention relates to vehicles, possibly of automobile type, which can circulate on roads and can be driven at least partially automated (or autonomous), and more specifically assistance to the automated driving of such vehicles.

In the following it is considered that a vehicle is driving 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 intervention of its driver on the steering wheel and / or the pedals) during a manual driving phase.

Certain vehicles, generally of automobile type, comprise several (at least two) sensors capable of acquiring, in the same time intervals, environmental information that is used by onboard analysis circuits to reconstitute their environment. Thanks to this environment information, it is in particular possible to determine markings on the ground, such as lines defining traffic lanes and / or passing lanes and / or emergency stop strips, static objects. (or fixed) or dynamic (or mobile, eg vehicles or pedestrians), and attributes that describe these objects.

These reconstructions and the attributes of associated objects are used by at least one assistance device embedded in the vehicle, for example of the type ADAS ("Advanced Driver Assistance System"), and responsible for assisting (totally or partially) driving of this vehicle. More specifically, the invention relates to methods and assistive devices that are responsible for determine the trajectory that a vehicle must follow according to the environment determined before it.

In order for such assistance devices to be able to carry out this trajectory determination to be followed, their vehicle must comprise a satellite positioning device (or GNSS (Global Navigation Satellite System) - for example of the GPS ("Global Positioning") type. System "), Galileo or Glonass) determining the current position of the vehicle, a memory storing known environment information defining roads (or road maps), and analysis circuits analyzing the environment in front of the vehicle to determine first information representative of this road and its environment. Among these first determined information are ground markings which are used by the assistance device of a vehicle to determine the trajectory to be followed by the latter (and in particular the lateral deviation, the heading and the curvature).

The data representative of these markings on the ground are generally acquired by at least one frontal camera (frequently installed at the top of the windshield).

The determination of trajectory to follow (or guidance) according to the data acquired by the front camera has at least two disadvantages. Indeed, the camera is sensitive to visibility conditions (including brightness and meteorology). In addition, the road markings of a road are not always available or exploitable, for example because they are in poor condition, or at least partially masked by water, mud, snow, ice, or static or mobile objects (such as other vehicles, especially in the case of a slowdown or traffic jam), or temporarily placed (for example in a construction area). In general, the detection performance of the current cameras can not guarantee the determination of the trajectory in a safe way for the guidance of the vehicle.

It is also possible to determine the trajectory of a vehicle according to high definition road map (or "HD Map") and precise vehicle positions determined by an advanced satellite positioning device (for example to differential measures). However, this solution also has several disadvantages. Indeed, it is quite expensive, and high-definition road maps that it uses currently cover very few road areas. In addition, the accuracy of the determined position can vary significantly according to the constellation of satellites used by a satellite positioning device on board the vehicle and according to the environment in which the vehicle operates (open air or tunnel, open or closed parking, city or countryside). However, these variations in accuracy or availability of position information can cause erroneous trajectories because the current position of the vehicle is used to locate it relative to the environmental information provided by the road maps (possibly associated with at least one embedded or remote database (but accessible by wave)). Moreover, GNSS positioning solutions coupled with a cartography can not withstand high operational safety requirements, particularly because of the accuracy of positioning and mapping, the updating of the cartography, the vulnerability to cyberattacks (such as spoofing), and therefore can not guarantee only safe guidance for the vehicle.

The object of the invention is in particular to improve the situation, and in particular to guarantee the determination of a trajectory for guiding the vehicle with high operating safety requirements.

It proposes for this purpose an assistance method intended to assist the automated driving of a first vehicle traveling on a road, determining its current position, storing known environmental information defining roads, and analyzing the environment in front of him and on his side to determine first information representative of this road and its environment.

This assistance method is characterized in that it comprises a step in which:

trajectories followed by second vehicles traveling on this road are determined from first determined information, and estimated ground markings of this road from first information. determined representative of ground markings of this road, corresponding ground markings defined by known stored environment information and of these determined tracked trajectories, then a reference trajectory for the first vehicle is determined based on these estimated ground markings ,

a space is determined (preferably in parallel) in which the first vehicle can circulate as a function of the first determined information, and

it is considered that this determined reference trajectory can be followed by the first vehicle when it is compatible with this determined circulation space.

This separation of the determinations into two independent parts (or branches) advantageously makes it possible in the end to provide a safe reference trajectory by exploiting all the elements of the environment, even though each part offers a level of safety that is not maximum due to the constrained performance of the environmental data sensors of the first vehicle.

The assistance method according to the invention may include other features that can be taken separately or in combination, and in particular:

in a first embodiment, in its step it is possible to determine the trajectories followed by the second vehicles traveling on the road from the first determined information, it is possible to verify whether first determined information representative of road markings of the road are compatible. with corresponding ground markings defined by stored known environment information, and if so, whether these determined first information representative of road markings of the road are compatible with the determined tracked trajectories, and if so it can be considered that the estimated ground markings are represented by these first determined information, then a reference trajectory for the first vehicle can be determined according to the estimated ground markings;

in a second embodiment, in its step, it is possible to determine the trajectories followed by the second vehicles traveling on the road from the first determined information, it is possible to verify whether first determined information representative of road markings of the road are compatible with corresponding ground markings defined by known environment information. stored and determined trajectories followed, then in the affirmative it can be considered that the estimated ground markings are represented by these first determined information, then a reference trajectory for the first vehicle can be determined according to the estimated ground markings;

in a third embodiment, in its step it is possible to determine the trajectories followed by the second vehicles traveling on the road from the first determined information, it is possible to determine estimated road markings from the first representative determined information. of road markings of the road and corresponding ground markings defined by stored known environmental information, and then it can be checked whether the estimated ground markings are compatible with the determined tracked trajectories, and if so can be determined. reference trajectory for the first vehicle based on the estimated ground markings;

in its step it is also possible to determine a value representative of a confidence level associated with the reference trajectory as a function of the determination of the estimated ground markings, and it can be considered that the determined reference trajectory can be followed by the first vehicle when it is compatible with the determined circulation space and that determined value is greater than a predefined threshold;

in its step, the circulation space can be determined as a function of first determined information representative of empty zones of the environment and / or zones of the environment containing static objects.

The invention also proposes a computer program product comprising a set of instructions which, when executed by processing means, is suitable for implementing an assistance method of the type presented above for assist the automated driving of a first vehicle traveling on a road.

The invention also proposes an assistance device intended to equip a first vehicle traveling on a road and comprising:

a satellite positioning device determining a current position of the first vehicle,

a memory storing known environment information defining routes, and

- Analysis circuits analyzing the environment in front of the first vehicle and on the sides of the latter to determine first information representative of the road and its environment.

This assistance device is characterized in that it comprises at least one processor:

determining trajectories followed by second vehicles traveling on the road based on first determined information, and estimated ground markings of said road from first determined information representative of ground markings of said road, corresponding defined ground markings by known stored environment information and these determined tracked trajectories, then determining a reference trajectory for the first vehicle based on these estimated ground markings,

determining a space in which the first vehicle can travel according to first determined information, and

Whereas this determined reference trajectory may be followed by the first vehicle when it is compatible with this determined circulation space.

The invention also proposes a vehicle, possibly of automotive type, suitable for driving on a road, and comprising:

a satellite positioning device determining a current position of the vehicle,

a memory storing known environment information defining routes, and

- analysis circuits analyzing the environment in front of the vehicle and on the sides of the vehicle to determine initial information representative of this road and its environment.

This vehicle is characterized in that it further comprises an assistance device of the type of that presented above.

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

FIG. 1 schematically and functionally illustrates a road comprising two traffic lanes on which a first vehicle equipped with a driving assistance device according to the invention and three second vehicles, and

- Figure 2 schematically and functionally illustrates the route of Figure 1 without the second vehicles but with the dashed materialization of a determined traffic space and dashed-points of a determined reference trajectory for the first vehicle.

The object of the invention is in particular to propose an assistance method, and an associated DA1 assistance device, intended to assist, by meeting high operational safety requirements, the automated (or autonomous) driving of a first V1 vehicle traveling on a road R.

In the following, we consider, by way of non-limiting example, that the first vehicle V1 is automotive type. This is for example a car. But the invention is not limited to this type of vehicle. It concerns indeed any type of vehicle that can travel on land traffic routes.

FIG. 1 schematically and functionally shows a road R comprising first VC1 and second VC2 traffic lanes having identical traffic directions. Note that the road may have only one VC1 traffic lane, or two VCk traffic lanes (k = 1 or 2) with opposite traffic directions, or have more than two lanes of traffic.

In the example shown non-limitatively, first V1 and second V2 vehicles travel on the first VC1 traffic lane (the first V1 following the second V2), and two second vehicles V2 flow on the second VC2 circulation lane.

The first vehicle V1 is notably equipped with analysis circuits CAN, an MP satellite positioning device, an MS memory, and an exemplary embodiment of a DA1 assistance device according to the invention.

The satellite positioning device MP is arranged to determine the position (in progress) of the first vehicle V1 at known times. This determination is made through messages that are transmitted by waves by a constellation of satellites (GNSS), for example GPS, Galileo or Glonass.

For example, and as shown in non-limiting manner in FIG. 1, the satellite positioning device MP may be part of an on-board navigation aid device DA2 (permanently or temporarily) in the first vehicle V1.

Note that in the example shown in non-limiting manner in Figure 1, the navigation aid device DA2 is part of a CAL calculator. But this is not obligatory. Indeed, the DA2 navigation aid device could include its own computer. Consequently, the navigation aid device DA2 can be implemented in the form of software (or computer or software) modules, or a combination of circuits or electrical or electronic components (or "hardware") and software modules.

The memory MS is responsible for storing known environment information that defines routes. For example, this environment information is part of road maps (possibly high definition), possibly downloaded via an MCN communication module embedded in the first vehicle V1.

For example, and as illustrated without limitation in FIG. 1, the memory MS may be part of the navigation aid device DA2. But this is not obligatory. Indeed, it could be independent of this DA2 navigation aid device.

The CAN analysis circuits are arranged to analyze the environment at least in front of the first vehicle V1 and on the sides of the latter (V1) in order to determine first information which is representative of the road R (on which this last (V1)) and its environment. It will be understood that this first information is associated respectively at positions derived from the current position of the first vehicle V1 (provided by the MP satellite positioning device or by inertial navigation means). Preferably, these positions respectively associated with the first information are positions relative to the first vehicle V1.

This first information defines at least ground markings of the road R (and in particular the boundaries of the traffic lanes VCk). But, they can also define, for example and without limitation, the curvature of the portion of the road R, the number of traffic lanes VCk of the road R (here temporarily equal to two (k = 1 or 2)), a Road R entrance and / or exit, R road signs, R road tunnel, R road traffic lights, RS Highway R safety guardrails, or traffic lanes. BAU emergency stop on Route R.

For example, some of the first information (including traffic signs and ground markings) can be detected by shape recognition, and their positions can be defined with respect to a reference attached to the first vehicle V1. Ground markings such as delimitations may, for example, be modeled by the CAN analysis circuits by mathematical polynomials, with as reference a point of the first vehicle V1.

The environment which is located at least in front of the first vehicle V1, and which is analyzed by the CAN analysis circuits, is defined (or represented) by data acquired by sensors CPj which are fixedly attached to the first vehicle V1 in suitable places for this purpose. For the invention to be implemented, the sensors CPj must comprise at least one front camera acquiring digital images in front of the first vehicle V1. This front camera is (or is part of) first sensors CP1 (j = 1) and is, for example, secured to the windshield PB, preferably in its upper part (or upper), or to the inner central rearview mirror of the first vehicle V1.

It will be noted that the first vehicle V1 may comprise other sensors CPj, as in the example illustrated without limitation in FIG. Thus, it may, for example, comprise second sensors CP2 (j = 2) installed in its shield (or bumper) before and responsible for acquiring environmental data in an area in front of him and on both sides. For example, these second sensors CP2 may comprise ultrasonic sensors, or even radars or lidars.

In general, the first vehicle V1 may comprise sensors CPj of at least one type, and for example of two, three or four different types. Therefore, each of them can provide digital images or cartographies or data on the presence and attributes of objects of various natures, static (or fixed) or dynamic (or mobile), and present in its area acquisition.

In the example illustrated non-limitatively in the single figure, the CAN analysis circuits are part of the CAL calculator which also comprises the navigation aid device DA2. But this is not obligatory. Indeed, the CAN analysis circuits could be part of another calculator than the CAL calculator, or could include their own calculator. Consequently, the CAN analysis circuits can be realized in the form of software modules, or a combination of electrical or electronic circuits or components and software modules.

As mentioned above, the invention notably proposes an assistance method intended to assist, in a safe manner, the automated (or autonomous) driving of the first vehicle V1 when traveling on the road R.

This assistance method can be at least partially implemented by the assistance device DA1 which comprises for this purpose at least one PR processor, for example digital signal processor (or DSP ("Digital Signal Processor")).

In the example shown non-limitatively in the single figure, the assistance device DA1 is part of the CAL calculator which also comprises the navigation aid device DA2 and the CAN analysis circuits. But this is not obligatory. Indeed, it could be part of another calculator than the CAL calculator, or could include its own calculator. Therefore, the assistance device DA1 can be realized in the form of software modules, or a combination of circuits or electrical or electronic components and software modules. The assistance method according to the invention comprises a step in which one (the processor PR) performs several determinations. More precisely, one (the processor PR) determines, on the one hand, trajectories which are followed by second vehicles V2 traveling on the road R on the basis of first information determined by the CAN analysis circuits, and, on the other hand, on the basis of the estimated ground markings of this route R from first information determined by the CAN analysis circuits and representative of the ground markings of this road R, corresponding ground markings which are defined by known environmental information and stored in the MS memory and these determined tracked trajectories, and then (the PR processor) determines a reference trajectory TR for the first vehicle V1 based on these estimated ground markings. A first part (or branch) of determinations is thus produced.

It will be understood that the trajectories followed by the second vehicles V2 are those on the road R in the environment close to the first vehicle V1 (here the three second vehicles V2 of Figure 1). Moreover each of these trajectories is constructed from a history of successive relative positions of a second vehicle V2 with respect to the first vehicle V1, determined by the CAN analysis circuits.

As illustrated without limitation in FIG. 2, the reference trajectory TR may, for example, be defined by two lines parallel to one another, spaced from each other by at least the width of the first vehicle V1, and here materialized by dashes-dots. Alternatively, the reference trajectory TR could be defined by a single line extending forwardly the center of the bumper (or shield) before the first vehicle V1.

In addition, one (the processor PR) also determines, optionally in parallel determinations of the first part (or branch) defined above, an EC space in which the first vehicle V1 can be moved in safety (ie without risk of collision with the surrounding static objects and the second vehicles V2, also on a safe surface without risk (no grass, no shoulder, in particular), according to first determined information. An example of an EC space corresponding to the situation illustrated in FIG. 1 is shown in dotted lines in FIG. thus realizes a second part (or branch) of determinations.

For example, in the process step one (the processor PR) can determine the circulation space EC as a function of first information determined by the CAN analysis circuits and which are representative of the empty zones of the environment and / or areas of this environment that contain static objects. The circulation space EC is the free space where the first vehicle V1 can evolve without danger. It is therefore delimited by the road infrastructure (slides, walls and the like, which are static objects). Furthermore, the presence of a static object (such as a tire debris) on which the vehicle can not roll makes this circulation space EC partially occupied and therefore not passable by the first vehicle V1. Also, some sensors can detect the limits of asphalt (surface detection) and it can thus limit the circulation space EC to parts of the road that are tarred (it is indeed dangerous to ride on grass or on a steep shoulder).

Then (the processor PR) considers that this reference trajectory TR determined can be followed by the first vehicle V1 when it is compatible with this circulation space EC just determined.

An example of reference trajectory TR of the first vehicle V1 compatible with the circulation space EC (because it is included in the circulation space EC) is illustrated in FIG. 2.

Thanks to this separation of the determinations into two independent parts (or branches), it is advantageous to provide a reference trajectory TR which is safe by exploiting all the elements of the environment, even though each part offers a level of safety which is not maximal because of the constrained performance of the sensors CPj of the first vehicle V1 which it uses the environment data.

It should be noted that the determination of the estimated ground markings can be done in at least three ways.

In a first way, one (the processor PR) can begin by determining the trajectories followed by the second vehicles V2 traveling on the road R from first information determined by the CAN analysis circuits. Then, one (the PR processor) can check whether first determined information representative of road markings of the road R are compatible with corresponding ground markings which are defined by known environment information and stored in the memory MS. Then, in the affirmative (and therefore in case of compatibility), one (the processor PR) can check whether these first determined information representative of road markings of the road R are compatible with the trajectories followed by the second vehicles V2 which have previously determined. Then, in the affirmative (and therefore in case of compatibility), one (the processor PR) can consider that the estimated ground markings are represented by these first determined information. Then, one (the processor PR) can determine a reference trajectory TR for the first vehicle V1 as a function of these estimated ground markings.

It will be noted that in the event of incompatibility between the first determined information representative of road markings of the road R and the corresponding and known ground markings, the (PR processor) does not determine a reference trajectory TR. Instead, one (the processor PR) can determine a safety path, included in the circulation space EC and intended to stop the first vehicle V1 as quickly as possible while ensuring the operating safety requirements. Thus, in case of incompatibility a safe path is determined and braking started, without leaving the circulation space EC, the objective being to avoid any risk of collision. Likewise, in the event of incompatibility between the first determined information representative of road markings of the road R and the trajectories followed by the second vehicles V2 and previously determined, the (PR processor) does not determine a reference trajectory TR. Instead, one (the processor PR) can determine a safety path, included in the circulation space EC and intended to stop the first vehicle V1 as quickly as possible while ensuring the operating safety requirements. It should be noted, however, that the detailed decision-making mechanisms are more complex, in particular because of the confidence level of the inputs, the number of inputs, the reliability of the raw information, and are based on design rules that take into account operating safety requirements. In a second way, one (the processor PR) can begin by determining the trajectories followed by the second vehicles V2 traveling on the road R from first information determined by the CAN analysis circuits. Then, one (the processor PR) can check whether first determined information representative of road markings of the road R are simultaneously compatible with corresponding ground markings which are defined by known environment information and stored in the memory MS and with the trajectories followed by the second vehicles V2 which have been previously determined. Then, in the affirmative (and therefore in case of double compatibility), one (the processor PR) can consider that the estimated ground markings are represented by these first determined information. Then, one (the processor PR) can determine a reference trajectory TR for the first vehicle V1 as a function of these estimated ground markings.

It will be noted that in the presence of at least one incompatibility, the (PR processor) does not determine a reference trajectory TR. Instead, one (the processor PR) can determine a safety path, included in the circulation space EC and intended to stop the first vehicle V1 as quickly as possible while ensuring the operating safety requirements.

In a third way, one (the processor PR) can begin by determining the trajectories followed by the second vehicles V2 traveling on the road R from first information determined by the CAN analysis circuits. Then, the PR processor can determine estimated road markings of the road R from first information determined by the CAN analysis circuits and representative of the road markings R of this road and corresponding ground markings defined by known environment information stored by the storage means MS. Here, in a way, a merging of the determined ground markings and the corresponding known and stored ground markings are carried out here. Then, one (the processor PR) can check whether these estimated ground markings are compatible with the trajectories followed by the second vehicles V2 which have been previously determined. Then, in the affirmative (and therefore in case of compatibility), one (the PR processor) can determine a trajectory of reference TR for the first vehicle V1 based on these estimated ground markings.

It will be noted that in the event of incompatibility between the estimated ground markings and the trajectories followed by the second vehicles V2 and previously determined, the (PR processor) does not determine a reference trajectory TR. Instead, one (the processor PR) can determine a safety path, included in the circulation space EC and intended to stop the first vehicle V1 as quickly as possible while ensuring the operating safety requirements.

It will also be noted that in the process step, the processor

PR) can also determine a value vc that is representative of a confidence level associated with the determined reference trajectory TR, based on the determination of the estimated ground markings. In this case, it is possible (the processor PR) to consider that the determined reference trajectory TR can be followed by the first vehicle V1 when it is compatible with the determined circulation space EC and that, at the same time, the determined value vc is greater than a predefined threshold. By way of example, if the value vc can vary between 1 and 5, the predefined threshold can be chosen equal to three.

This confidence value vc can, for example, be determined: - for ground markings, based on the confidence levels provided by the cameras,

for the trajectory of the second vehicles V2, the confidence level is calculated from, on the one hand, the confidence levels of the detection of these second vehicles V2, and, on the other hand, as a function of the number of vehicles having compatible trajectories (if five vehicles have compatible trajectories, the resulting overall trajectory is more reliable than if only two vehicles had compatible trajectories).

The circulation space EC is measured by different sensors and the overall level of confidence is calculated according to the confidence levels of the primary information given by these sensors.

If the value vc determined is less than or equal to the predefined threshold, the (PR processor) can determine a safety trajectory, intended to stop the first vehicle V1 as quickly as possible while guaranteeing the requirements. operating safety.

Preferably, when the processor PR is forced to determine a safety path, the assistance device DA1 generates an alert for at least passengers of the first vehicle V1 (including its possible driver). The alert of the passengers can be done by means of a text message displayed on at least one screen of the first vehicle V1, for example that of the dashboard or that of the central unit installed in or on the dashboard, and / or a sound message broadcast by at least one speaker of the first vehicle V1. For example, the alert may signal an impossibility of automated guidance of the first vehicle V1, and a triggering of a procedure of safe shutdown of the latter (V1).

It will also be noted that the invention also proposes a computer program product comprising a set of instructions which, when executed by processing means of the electronic circuit (or hardware) type, such as for example the PR processor, is adapted to implement the assistance method described above to assist the automated driving of the first vehicle V1 when traveling on the road R.

It will also be noted that in FIGS. 1 and 2, the assistance device DA1 is very schematically illustrated with only its processor PR. This assistance device DA1 can take the form of a housing comprising printed circuits, or of several printed circuits connected by wired or non-wired connections. The term printed circuit means any type of device capable of performing at least one electrical or electronic operation. As mentioned above, this assistance device DA1 may, for example, comprise at least one digital signal processor (or DSP (Digital Signal Processor)) PR, a RAM for storing instructions for implementation by this processor PR of the assistance method as described above, and a mass memory for storing data to be retained after the implementation of the assistance method. The digital signal processor PR receives at least the first information determined by the CAN analysis circuits and known environment information defining the route R and stored in the memory MS (as well as possibly the current positions determined by the device positioning by MP satellites) for analysis and use in calculations, possibly after formatting and / or demodulated and / or amplified, in a manner known per se. The assistance device DA1 may also comprise an input interface for receiving the first determined information and known environment information defining the route R (as well as possibly the current positions), and an output interface for the transmission of the results of its analyzes and calculations.

One or more substeps of the step of the assisting method may be performed by different components. Thus, the assistance 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 assistance device DA1 can be decentralized, within a local network (several processors connected together for example) or a wide area network.

Claims

A method of assisting the automated driving of a first vehicle (V1) traveling on a road (R), determining its current position, storing known environment information defining roads, and analyzing the environment in front of it and on its sides to determine first information representative of said road (R) and its environment, characterized in that it comprises a step in which a) paths are determined followed by second vehicles (V2) flowing on said road (R) from first determined information, and estimated ground markings of said road (R) from first determined information representative of ground markings of said road (R), corresponding ground markings defined by information known stored environment and said determined trajectories followed, then determining a reference trajectory for said first v based on said estimated ground markings, b) a space is determined in which said first vehicle (V1) can be driven as a function of first determined information, and c) said determined reference trajectory can be followed by said first vehicle (V1) when it is compatible with said determined circulation space.

2. Method according to claim 1, characterized in that in said step said trajectories followed by the second vehicles (V2) traveling on said road (R) are determined from first determined information, it is checked whether first determined information representative of ground markings of said road (R) are compatible with corresponding ground markings defined by known stored environment information, and if so, whether said determined first information representative of ground markings of said road (R) are compatible with said determined tracked trajectories, then in the affirmative it is considered that said estimated ground markings are represented by these first determined information, then a reference trajectory for said first vehicle (V1) is determined according to said estimated ground markings .

3. Method according to claim 1, characterized in that in said step is determined said trajectories followed by the second vehicles (V2) traveling on said road (R) from first determined information, it is checked whether first determined information representative of ground markings of said road (R) are compatible with corresponding ground markings defined by known stored environment information and said determined tracked trajectories, then if so, said estimated ground markings are assumed to be represented by these first determined information, then a reference path for said first vehicle (V1) is determined based on said estimated ground markings.

4. Method according to claim 1, characterized in that in said step is determined said trajectories followed by the second vehicles (V2) circulating on said road (R) from determined first information, determined ground markings of said estimated route (R) from determined first information representative of ground markings of said road (R) and corresponding ground markings defined by stored known environmental information, and then verifies whether said estimated ground markings are compatible with said determined trajectories followed, and in the affirmative said reference trajectory for said first vehicle (V1) is determined as a function of said estimated ground markings.

5. Method according to one of claims 1 to 4, characterized in that in said step is further determined a value representative of a confidence level associated with said reference trajectory based on the determination of said estimated ground markings, and it is considered that said determined reference trajectory can be followed by said first vehicle (V1) when it is compatible with said determined circulation space and said determined value is greater than a predefined threshold.

6. Method according to one of claims 1 to 5, characterized in that in said step said circulation space is determined based on first determined information representative of empty areas of said environment and / or areas of said environment containing static objects.

7. A computer program product comprising a set of instructions which, when executed by processing means, is adapted to implement the assistance method according to one of the preceding claims for assisting the automated driving of a first vehicle (V1) traveling on a road (R).

8. Assist device (DA1) for a first vehicle (V1) traveling on a road (R) and comprising i) a satellite positioning device (MP) determining a current position of said first vehicle (V1), ii) a memory (MS) storing known environment information defining routes, and iii) analysis circuits (CAN) analyzing the environment in front of said first vehicle (V1) and on the sides of the latter (V1) in order to determining first information representative of said road (R) and its environment, characterized in that it comprises at least one processor (PR) a) determining trajectories followed by second vehicles (V2) traveling on said road (R) from first determined information, and estimated ground markings of said road (R) from first determined information representative of ground markings of said road (R), corresponding ground markings defined by known stored environment information and said determined tracked trajectories, then determining a reference trajectory for said first vehicle (V1) as a function of said estimated ground markings, b) determining a space in which said first vehicle (V1) can circulate ) based on first determined information, and c) considering that said determined reference trajectory can be followed by said first vehicle (V1) when it is compatible with said determined circulation space.

9. Vehicle (V1) capable of traveling on a road (R) and comprising i) a satellite positioning device (MP) determining a current position of said vehicle (V1), ii) a memory (MS) storing information of known environment defining routes, and iii) analysis circuits (CAN) analyzing the environment in front of said vehicle (V1) and on the sides of the latter (V1) in order to determine first information representative of said road (R ) and its environment and the current positions associated with this first information, characterized in that it comprises in in addition to an assistance device (DA1) according to claim 8.

10. Vehicle according to claim 9, characterized in that it is automotive type.