FR3072182A1 - CALCULATOR, SYSTEM AND METHOD FOR GEOLOCATION OF A VEHICLE - Google Patents

Abstract

The subject of the present invention is a computer (140) for a motor vehicle (10), said computer (140) being configured to: receive a sequence of images generated by a video camera (130) of the vehicle (10), receive the position of the vehicle (10), determined by a satellite geolocation module (110) on board said vehicle (10), transmitting to a management server (20), via a communication network (30), the determined position of the vehicle (10), receiving from said management server (20) a list of objects located in the environment of the vehicle (10) and characterized by a descriptor, detecting, in said sequence of images, at least one object of said list of objects from its descriptor, determining the distance between the vehicle (10) and each detected object, and obtaining a refined position of the determined position of the vehicle (10) from said determined position and from the determined distance between the vehicle (10) and each detected object you.

Description

The present invention relates to the field of geolocation and relates more particularly to a method of geolocation of a vehicle, a geolocation system for a vehicle as well as a vehicle comprising such a geolocation system. The invention aims to make it possible to precisely determine the position of a vehicle from its geolocation in order in particular to exploit this position to assist the driver in driving the vehicle.

In known manner, a vehicle today embeds a plurality of systems among which a geolocation module which in particular makes it possible to guide the driver when he is associated with a card. Such a module determines the position of the vehicle from signals emitted by a constellation of satellites, for example of the GPS type. Along the way, these signals can be disturbed in various ways, notably by atmospheric elements or by terrestrial elements which deflect or attenuate them, which can lead to inaccuracy in the determined position which can reach several meters.

In order to remedy this drawback, it is known to use a differential geolocation system, in particular of the GPS type. In practice, such a system comprises a plurality of GPS receivers, called "reference", positioned on fixed land elements, for example signal lights or poles arranged along the roads. When the vehicle is moving near reference receivers, the latter transmit correction signals to it which it uses to correct the position determined by its on-board GPS module in order to obtain improved accuracy which can be of the order of a few centimeters. . However, this solution requires the use of a large number of reference receivers, which is complex to implement and expensive. In addition, this solution requires the availability of a means of communication of the cellular network or radio waves type. However, not all territories are covered and so-called "white" areas remain which are not covered by the network or radio waves, which makes this solution in practice not ideal. At the same time, the need for precise cartography is increasing with the trend to offer increasingly autonomous vehicles to provide not only assistance to the driver but also assistance in driving anywhere.

The invention aims to resolve these drawbacks by improving the accuracy of the location of a vehicle, in particular without using reference receivers communicating with the vehicle.

To this end, the invention firstly relates to a computer for a motor vehicle, said computer being configured for:

• receive a sequence of images generated by a video camera of the vehicle, • receive the position of the vehicle, determined by a geolocation module by satellite on board said vehicle, • transmit to a management server, via a communication network, the determined position of the vehicle, • receive from said management server a list of objects located in the environment of the vehicle, each object being characterized by a descriptor, • detect, in said sequence of images, at least one object from said list of objects from its descriptor, • determine the distance between the vehicle and each object detected, and • obtain a refined (that is to say corrected) position of the determined position of the vehicle from said determined position and from the distance determined between the vehicle and each object detected.

The descriptor is constructed from an image representation (pixels) of the object or part of the object and allows the object to be sufficiently characterized to distinguish it from others and thus guarantee the uniqueness of detection . The descriptor can include the type of object, its shape, color, geographic location or any other relevant characteristic.

The invention notably makes it possible to achieve almost centimeter-level precision, with an error of less than ten centimeters, which makes it possible to know the position of the vehicle well, in particular in order to drive it precisely in a road environment. The object or objects being characterized via their distance by their relative position relative to the vehicle, a differential type positioning is obtained, largely canceling out the effects of the hazards and errors introduced in particular by atmospheric elements on satellite signals and other errors (clock , propagation ...) which may affect the accuracy of the satellite location alone. The refined position of the vehicle can in particular be used to extend the perception of the sensors on board the vehicle. More precisely, the precise location of the vehicle associated with an on-board navigation module makes it possible to anticipate events present on the route such as an accident or debris, although they are not perceptible by the vehicle's sensors at the same time .

Preferably, the computer is configured to determine the distance between the vehicle and each object detected from the images of the sequence of images.

Alternatively or in addition, the computer can be configured to receive the distance between the vehicle and each detected object from a distance measurement module, preferably on board the vehicle.

For example, this distance measurement module can comprise a lidar or radar type unit configured to emit waves on the detected objects and receive in return waves reflected on said objects in order to measure the distance separating each object from the vehicle.

In one embodiment, the computer is configured to calculate the refined position of the vehicle in particular to avoid any subsequent exchange with the management server to refine the current position of the vehicle.

In another embodiment, the computer is configured to send to the management server the determined distance between the vehicle and each object detected so that said management server determines the refined position of the vehicle and sends this information back to the computer, always via said communication network. This makes it possible in particular to concentrate the computation load and resources on the management server rather than on the computer in order in particular to make it less powerful, less complex and / or less expensive.

The invention also relates to a motor vehicle comprising:

• a geolocation module able to determine the position of the vehicle from signals received from a plurality of satellites, • a communication module able to communicate with a management server via a communication network, in particular for sending the position of the vehicle determined by the geolocation module and to receive a list of objects determined by the management server from the determined position of the vehicle, said objects being located in the environment of the vehicle and being each characterized by a descriptor, • at least a video camera capable of generating a sequence of images at least partially illustrating the environment of the vehicle, and • a computer as presented above.

The invention also relates to a management server for determining the position of a motor vehicle, said management server being able to communicate with said vehicle via a communication network and comprising a memory area in which is stored a set of objects, each object being characterized by a descriptor, said management server being configured to:

• receive a determined position of the vehicle sent by said vehicle, • extract from its memory area, from a position received from a vehicle, a list of objects located in a predetermined area around said vehicle, said list comprising for each object a descriptor to characterize it, and • send said list to the vehicle.

In one embodiment, the management server is configured to:

• receive the distance, determined by the vehicle, between the vehicle and each object detected by said vehicle, • calculate the refined position of the vehicle from the position received from the vehicle and the determined distance from each detected object, and • send said position refined to the vehicle via the communication network.

The invention also relates to a system for determining the position of a motor vehicle, said system comprising a motor vehicle, as presented above, and a management server, as presented above.

The invention also relates to a method for determining the position of a motor vehicle, as previously presented, using a management server, as previously presented, said method comprising the steps of:

• generation, by the vehicle video camera, of a sequence of images representing at least in part the environment of the vehicle, • determination, by the geolocation module, of the (geographic) position of the vehicle, • sending, by the vehicle, to the management server, of the determined position of the vehicle, • reception, by the management server, of the determined position of the vehicle sent by said vehicle, • extraction, by the management server, of its memory area, in depending on the position of the vehicle received, of a list of objects located in the environment of said vehicle, said list comprising for each object a descriptor making it possible to characterize it, • sending of said list to the vehicle, • reception, by the vehicle , from said list, • detection, in said generated image sequence, of at least one object from the list of objects from its descriptor, • determination of the distance between the vehicle and each object detected, and • obtaining a refined position of the determined position of the vehicle from said determined position and the determined distance between the vehicle and each object detected.

Preferably, several objects are detected in the image sequence in order to improve the accuracy of the correction.

Advantageously, the computer detects all the objects listed in the list of objects.

In one embodiment, the distance between the vehicle and each object detected in the image sequence is determined from said image sequence.

In one embodiment, the vehicle further comprises a distance measurement module, the distance between the vehicle and each object detected in the image sequence is measured by said distance measurement module.

According to one aspect of the invention, the distance measurement module comprises a lidar or radar type unit configured to emit waves on the detected objects and receive in return waves reflected on said objects in order to measure the distance separating each object from the vehicle.

In one embodiment, the determined position of the vehicle is refined by the computer on the basis of the position of the vehicle and the distance determined between the vehicle and each object detected.

In another embodiment, the method further comprises the steps of:

• sending to the server for managing the distance determined between the vehicle and each object detected, • reception by the server for managing the distance, determined by the vehicle, between the vehicle and each object detected by said vehicle, • calculation, by the management server, of the refined position of the vehicle from the position received from the vehicle and of the determined distance from each detected object, and • sending, by the management server, of said refined position to the vehicle via the communication network, • reception by the vehicle of the refined position sent by the management server.

Other characteristics and advantages of the invention will become apparent from the following description given with reference to the appended figures given by way of nonlimiting examples and in which identical references are given to similar objects.

- Figure 1 schematically illustrates an embodiment of the system according to the invention.

- Figure 2 schematically illustrates an example of detection of objects arranged along a road.

- Figure 3 illustrates a first embodiment of the method according to the invention.

- Figure 4 illustrates a second embodiment of the method according to the invention.

There is shown schematically in Figure 1 an embodiment of the system according to the invention. The system makes it possible to precisely determine the position of said vehicle.

To this end, the system 1 comprises a vehicle 10 and a management server 20 capable of communicating with one another via a communication network 30.

The vehicle 10 comprises a geolocation module 110, a communication module 120, a video camera 130 and a computer 140.

The geolocation module 110 is configured to determine the position of the vehicle 10 from signals received from a plurality of satellites 2.

This geolocation module 110 is for example of the GPS, GALILEO, GLONASS, BEIDOU, etc. type.

The communication module 120 is configured to communicate with the management server 20 via the communication network 30.

In particular, the communication module 120 is configured to send to the management server 20 the position of the vehicle 10 determined by the geolocation module 110.

The communication module 120 is also configured to receive a list of objects, determined by the management server 20 from the determined position of the vehicle 10, said objects being located in the environment of the vehicle 10 and being each characterized by a descriptor as will be explained below.

With reference to FIG. 2, the video camera 130 is configured to generate a sequence of images illustrating at least in part the environment of the vehicle 10 and in particular at least in part an area Z around the vehicle 10, determined by the server of management 20. It will be noted that, in another embodiment and without limitation, the vehicle 10 could comprise more than one video camera 130 mounted at various locations on the vehicle 10 (front, side, rear, etc.).

The computer 140 is configured to perform a plurality of tasks.

First, the computer 140 is configured to receive a sequence of images generated by the video camera 130 and representing at least in part the environment of the vehicle 10.

The computer 140 is also configured to receive the position of the vehicle 10, determined by the geolocation module 110, and to transmit said position of the vehicle 10 to the management server 20, via the communication network 30.

The computer 140 is also configured to receive from the management server 20 a list of objects OBJ1, OBJ2 (FIG. 2) located in the environment of the vehicle 10 and each characterized by a descriptor, for example their type, their shape, their color. , their geographic position, etc. In one embodiment, the descriptor of each object OBJ1, OBJ2 includes its geographic position in order to allow the computer 140 to refine the position of the vehicle 10 as will be explained below.

The computer 140 is also configured to detect, in a sequence of images generated by said video camera 130, at least one object OBJ1, OBJ2 from the list of objects determined by the management server 30 from its descriptor.

Advantageously, all the objects in the list are detected, but it may happen that some of the objects are not detected insofar as the vehicle 10 is moving and the objects are scrolling or the image processing algorithm does not allow the detection of certain objects OBJ1, OBJ2. Preferably, the list containing at least two objects OBJ1, OBJ2, at least two objects OBJ1, OBJ2 are detected on the images in order to carry out a precise calculation of the position of the vehicle 10 as will be explained below.

With reference to FIG. 2, the computer 140 is configured to determine the distance d1, d2 between the vehicle 10 and each object OBJ1, OBJ2, of a list of objects OBJ1, OBJ2 received from the management server 20, detected in a sequence of images captured by the video camera 130. By the terms “distance d1, d2 between the vehicle 10 and each object OBJ1, OBJ2 >>, is meant the distance between a reference point of the vehicle 10, which may be the camera video 130 or any other reference point (for example the right front wheel), and the object OBJ1, OBJ2 in question, the refinement of the position of the vehicle 10 being carried out with respect to this reference point (which can be corrected by calculation for use in a vehicle driving assistance system 10, if necessary for example).

In one embodiment, the computer 140 is configured to determine the distance d1, d2 between the vehicle 10 and each object OBJ1, OBJ2 detected from the images of the sequence of images. Such a determination of the distance of an object OBJ1, OBJ2 from a sequence of images being known per se, it will not be further detailed here.

Alternatively or in addition, the computer 140 can be configured to receive the distance d1, d2 between the vehicle 10 and each object OBJ1, OBJ2 detected from a distance measurement module (not shown), preferably on board the vehicle 10 For example, this distance measurement module can include a lidar or radar type unit configured to emit waves on the detected objects OBJ1, OBJ2 and receive in return waves reflected on said objects OBJ1, OBJ2 in order to measure the distance d1 , d2 separating each object OBJ1, OBJ2 from the vehicle 10.

The computer 140 is also configured to obtain a refined position (that is to say corrected) of the determined position of the vehicle 10 from the position of the vehicle 10 and of the distance d1, d2 determined from each object OBJ1, OBJ2 detected. This refined position can be calculated for example by triangulation using the distances d1, d2 between the video camera 130 and the objects OBJ1, OBJ2 as well as their geographic positions, for example received from the management server 20 in the list of objects OBJ1, OBJ 2.

In one embodiment (described below with reference to FIG. 3), the computer 140 is configured to calculate the refined position of the vehicle 10 itself.

In another embodiment, the computer 140 is configured to send to the management server 20 the distances d1, d2 determined between the vehicle and the objects OBJ1, OBJ2 detected so that said management server 20 determines itself the refined position of the vehicle 10, then sends this information back to the computer 140, still via the communication network 30.

By way of example of calculation making it possible to obtain the refined position, one can use a circle centered on each object OBJ1, OBJ2 and whose radius corresponds to the distance d1, d2 determined between the vehicle 10 and said object OBJ1, OBJ2. Thus, the computer 140 can determine the intersection of at least two circles each centered on the object OBJ1, OBJ2 which, by being correlated with the geolocation of the vehicle 10 (for example by choosing that of the two intersections of the two circles which corresponds most at the determined position of the vehicle 10), makes it possible to deduce therefrom the refined position of the vehicle 10. It will be noted that, without using a correlation with the determined current position of the vehicle 10, it is possible to use particulate filtering holding account of the previous positions of the vehicle 10 in order to identify the refined position of the vehicle 10 among the two intersections of the two circles. By particle filtering and in a manner known per se, is meant the operation consisting in finding the possible solutions to the minimization of a function by taking different initialization hypotheses. In this case, the previous positions of previously detected objects can be used to make a prediction of their position at a next instant, these predictions then being used to refine the position of the vehicle 10.

Advantageously, the computer 140 can also be configured to minimize the error on the distances between the vehicle 10 and several objects when the number of these objects is greater than two.

The management server 20 is able to communicate with the vehicle 10 via the communication network 30 and includes a memory area (not shown) in which is stored a set of objects OBJ1, OBJ2 linked to a geographic map (called "cartographic base" >>) including the roads on which vehicle 10 can travel.

Each object OBJ1, OBJ2 can be characterized by a descriptor and can be associated with an area of the geographical map, in particular with an area Z in which or near which the vehicle 10 is located and which is therefore likely to be at least partly filmed by video camera 130.

The descriptor is constructed from a reference image representation (in pixels) of the object or part of the object which allows the object to be sufficiently characterized to distinguish it from others and thus guarantee uniqueness. of detection. The descriptor may include the type of object, its shape, color, geographic location or any other relevant characteristic of the object. Also, when the computer 140 analyzes the images, it detects an object from the list when it finds the characteristics of the descriptor sent by the management server 20 in the images.

The size of the geographic zone Z is adapted so that said geographic zone Z includes objects OBJ1, OBJ2 located near the vehicle 10, in the field of vision of the video camera 130, for example in front of the vehicle 10 and / or on the side of the vehicle 10 when the vehicle 10 is advancing and the video camera 130 is mounted on the front windshield of the vehicle 10.

In particular, the management server 20 is configured to receive a determined position from the vehicle 10 sent by said vehicle, to extract from its memory area, from a position received from the vehicle 10, a list of objects OBJ1, OBJ2 located in a predetermined area around said vehicle 10, said list comprising for each object OBJ1, OBJ2 its descriptor, and for sending said list to vehicle 10.

In one embodiment (described below with reference to FIG. 4), the management server 20 is configured to receive the distance d1, d2, determined by the computer 140, between the vehicle 10 and each object OBJ1, OBJ2 detected by said vehicle 10, to calculate the refined position of vehicle 10 from the position received from vehicle 10 and the determined distance d1, d2 of each object OBJ1, OBJ2 detected, and to send said refined position to vehicle 10 via the network communication 30.

In this case, as described previously for the computer 140 and by way of example, the management server 20 can calculate the refined position of the vehicle 10 using, in addition to the position determined by the geolocation module 110, the intersection of two circles centered on two objects OBJ1, OBJ2 and each radius of which is equal to the distance d1, d2 between the vehicle and the object in question OBJ1, OBJ2.

The invention will now be described in its implementation, in particular with reference to Figures 2 to 4.

According to a first embodiment illustrated in FIG. 3, first of all, in a step EO, the video camera 130 generates a sequence of images representing the environment of the vehicle 10, in particular at least part of an area Z in which the vehicle 10 travels, and transmits it to the computer 140.

In parallel, in a step E1, the geolocation module 110 determines the geographic position of the vehicle 10 from signals received from a plurality of satellites 2 and transmits it to the computer 140.

The computer 140 then sends, in a step E2, via the communication module 120 and the communication network 30, the position of the vehicle 10 thus determined to the management server 20.

On reception (step E3), the management server 20 determines, in a step E4, on the one hand, the geographic zone Z of predetermined size close to the current position of the vehicle 10, for example within a radius of 100 meters or in front the vehicle 10, in its cartographic base and, on the other hand, a list of objects OBJ1, OBJ2 located in said zone Z from the set of objects stored in its memory zone. The list includes for each object a descriptor indicating for example its type, its shape, its color, its geographical position (for example its geographical coordinates).

In the nonlimiting example of FIG. 2, the management server 20 detects two objects: a traffic sign OBJ1 and a three-color traffic light OBJ2. It will be noted that, more generally, any type of object could be used, such as for example a signaling arrow painted on route 3, a lamppost, etc. Similarly, it should be noted that the list of objects could include more or less than two objects.

The geographic zone Z determined by the management server 20 covers at least in part the field of the video camera 130 so that some of the objects OBJ1, OBJ2 detectable on the images generated by the video camera 130 correspond to objects OBJ1, OBJ2 located in said geographic area Z.

The management server 20 then sends, in a step E5, via the communication network 30, the list of objects OBJ1, OBJ2 determined.

On reception (step E6) by the computer 140, via the communication module 120, of said list, the computer 140 detects, in a step E7, in the sequence of images generated in parallel by the video camera 130 and which represents at at least in part the geographical zone Z determined by the management server 20 in step E4, one or more of the objects OBJ1, OBJ2 of the list received using the descriptor of each object OBJ1, OBJ2 to allow their recognition.

It will be noted that data relating to the position of the objects OBJ1, OBJ2 could be used in addition in order to limit the search area and improve the accuracy of the location by avoiding bad matches. For example, this can avoid looking for a sign on the road if the sign is positioned 3 meters high.

The computer 140 then determines, in a step E8, the distance d1, d2 between the vehicle 10 and each object OBJ1, OBJ2 detected in the sequence of images. This distance d1, d2 can be evaluated using a mathematical method on the images of the sequence of images or else be measured by a distance measurement module (not shown but as mentioned above) on board the vehicle 10 if applicable.

Finally, in a step E9A, the computer 140 corrects the determined position of the vehicle 10 using the distances d1, d2 determined for each object OBJ1, OBJ2 detected, for example, as described above, using the intersection of the circles centered on each object OBJ1, OBJ2 closest to the position determined by the geolocation module 110. In this case, a minimum of two objects OBJ1, OBJ2 makes it possible to refine the position of the vehicle 10 by using the intersection of the two corresponding circles. If a single object (OBJ1 or OBJ2) is detected by the computer 140, the possible refined positions of the vehicle 10 are on an arc of the corresponding circle and a prediction based on the previous positions of the object OBJ1, OBJ2 can then allow d refining the position of the vehicle 10. If none of the objects OBJ1, OBJ2 is detected, the position of the vehicle 10 corresponds to the position determined by the geolocation module 110 without more precision, but it is possible to refine this position using one or more predictions made from one or more objects OBJ1, OBJ2 previously detected (in particular by following their trajectory).

According to a second embodiment illustrated in FIG. 4, first of all, in a step E0, the video camera 130 generates a sequence of images representing the environment of the vehicle 10, in particular at least part of an area Z in which the vehicle 10 travels, and transmits it to the computer 140.

In parallel, in a step E1, the geolocation module 110 determines the geographic position of the vehicle 10 from signals received from a plurality of satellites 2 and transmits it to the computer 140.

The computer 140 then sends, in a step E2, via the communication module 120 and the communication network 30, the position of the vehicle 10 thus determined to the management server 20.

On reception (step E3), the management server 20 determines, in a step E4, on the one hand, the geographic zone Z of predetermined size close to the current position of the vehicle 10, for example within a radius of 100 meters or in front the vehicle 10, in its cartographic base and, on the other hand, a list of objects OBJ1, OBJ2 located in said zone Z from the set of objects stored in its memory zone. The list includes for each object a descriptor as mentioned above.

In the nonlimiting example of FIG. 2, the management server 20 detects two objects: a traffic sign OBJ1 and a three-color traffic light OBJ2. It will be noted that, more generally, any type of object could be used, such as for example a signaling arrow painted on route 3, a lamppost, etc.

The geographic zone Z determined by the management server 20 covers at least in part the field of the video camera 130 so that some of the objects OBJ1, OBJ2 detectable on the images generated by the video camera 130 correspond to objects OBJ1, OBJ2 located in said geographic area Z.

The management server 20 then sends, in a step E5, via the communication network 30, the list of objects OBJ1, OBJ2 determined.

On reception (step E6) by the computer 140, via the communication module 120, of said list, the computer 140 detects, in a step E7, in the sequence of images generated in parallel by the video camera 130 and which represents at at least in part the geographical zone Z determined by the management server 20 in step E4, one or more of the objects OBJ1, OBJ2 of the list received using the descriptor of the object OBJ1, OBJ2 to allow their recognition.

It will be noted that data relating to the position of the objects OBJ1, OBJ2 could be used in addition in order to limit the search area and improve the accuracy of the location by avoiding bad matches. For example, this can avoid looking for a sign on the road if the sign is positioned 3 meters high.

The computer 140 then determines, in a step E8, the distance d1, d2 between the vehicle 10 and each object OBJ1, OBJ2 detected in the sequence of images. This distance d1, d2 can be evaluated using a mathematical method on the images of the sequence of images or else be measured by a distance measurement module (not shown but as mentioned above) on board the vehicle 10 if applicable.

Then, the computer 140 sends to the management server 20, in a step E9B, the distance d1, d2 determined for each object OBJ1, OBJ2 detected, and the management server 20 corrects, in a step E9C, the determined position of the vehicle 10 from the determined position and distance d1, d2 of each object OBJ1, OBJ2 detected, for example as described previously using the intersection of the circles centered on each object OBJ1, OBJ2 closest to the GPS position of the vehicle 10 .

As before, a minimum of two objects OBJ1, OBJ2 makes it possible to refine the position of the vehicle 10 by using the intersection of the two corresponding circles. If a single object (OBJ1 or OBJ2) is detected by the computer 140, the possible refined positions of the vehicle 10 are on an arc of the corresponding circle and a prediction based on the previous positions of the object OBJ1, OBJ2 can then allow d refining the position of the vehicle 10. If none of the objects OBJ1, OBJ2 is detected, the position of the vehicle 10 corresponds to the position determined by the geolocation module 110 without more precision, but it is possible to refine this position using one or more predictions made from one or more objects OBJ1, OBJ2 previously detected (in particular by following their trajectory).

The method according to the invention makes it possible to determine the position of the vehicle in a precise and reliable manner, this position then being able to be used by the vehicle, in particular by a driving assistance system on board said vehicle, to control it precisely or to offer improved services to its users.

Claims (11)

1. Computer (140) for a motor vehicle (10), said computer (140) being characterized in that it is configured for: • receive a sequence of images generated by a video camera (130) of the vehicle (10), • receive the position of the vehicle (10), determined by a geolocation module (110) by satellite on board said vehicle (10), • transmit to a management server (20), via a communication network (30), the determined position of the vehicle (10), • receive from said management server (20) a list of objects (OBJ1, OBJ2) located in the environment of the vehicle (10) and characterized by a descriptor, • detecting, in said sequence of images, at least one object (OBJ1, OBJ2) from said list of objects (OBJ1, OBJ2) from its descriptor, • determine the distance (d1, d2) between the vehicle (10) and each object (OBJ1, OBJ2) detected, and • obtain a refined position of the determined position of the vehicle (10) from said determined position and the distance determined between the vehicle (10) and each object (OBJ1, OBJ2) d tected. 2. computer (140) according to claim 1, said computer (140) being configured to determine the distance (d1, d2) between the vehicle (10) and each object (OBJ1, OBJ2) detected from the images of the sequence d images. 3. Computer (140) according to one of claims 1 and 2, said computer (140) being configured to calculate the refined position of the vehicle (10). 4. Motor vehicle (10), said vehicle (10) being characterized in that it comprises: • a geolocation module (110) capable of determining the position of the vehicle (10) from signals received from a plurality of satellites (2), • a communication module (120), capable of communicating with a management server (20) via a communication network (30), in particular for sending the position of the vehicle (10) determined by the geolocation module (110) and for receiving a list of objects (OBJ1, OBJ2) determined by the management server (20) from the determined position of the vehicle (10), said objects (OBJ1, OBJ2) being located in the environment of the vehicle (10) and being each characterized by a descriptor, • at least one video camera (130) able to generate a sequence of images at least partly illustrating the environment of the vehicle, and • a computer (140) according to one of the preceding claims. 5. Management server (20) for determining the position of a motor vehicle (10), said management server (20) being able to communicate with said vehicle (10) via a communication network (30) and comprising a memory area in which a set of objects (OBJ1, OBJ2) is stored, each object (OBJ1, OBJ2) being characterized by a descriptor, said management server (20) being characterized in that it is configured for: • receive a determined position from the vehicle (10) sent by said vehicle (10), • extract from its memory area, from a position received from a vehicle (10), a list of objects (OBJ1, OBJ2) located in a predetermined area around said vehicle (10), said list comprising for each object (OBJ1, OBJ2) a descriptor making it possible to characterize it, and • send said list to the vehicle (10). 6. Management server (20) according to the preceding claim, said management server (20) being configured for: • receive the distance (d1, d2), determined by the vehicle (10), between the vehicle (10) and each object (OBJ1, OBJ2) detected by said vehicle (10), • calculate the refined position of the vehicle (10) from the position received from the vehicle (10) and the determined distance (d1, d2) of each object (OBJ1, OBJ2) detected, and • send said refined position to the vehicle (10) via the communication network (30) . 7. System (1) for determining the position of a motor vehicle (10), said system comprising a motor vehicle (10) according to claim 4, and a management server (20), according to one of claims 5 and 6. 8. Method for determining the position of a motor vehicle (10) according to claim 4, using a management server (20), according to one of claims 5 and 6, said method being characterized in what it includes the steps of: • generation (E0), by the video camera (130) of the vehicle (10), of a sequence of images representing at least in part the environment of the vehicle (10), • determination (E1), by the module of geolocation (110), of the position of the vehicle (10), • sending, by the vehicle (10), to the management server (20), of the determined position of the vehicle (10), • reception, by the management server (20), of the determined position of the vehicle (10) sent by said vehicle (10), • extraction, by the management server (20), of its memory area, as a function of the position of the vehicle (10) received, a list of objects (OBJ1, OBJ2) located in the environment of said vehicle (10), said list comprising for each object (OBJ1, OBJ2) a descriptor making it possible to characterize it, • sending of said list to the vehicle (10 ). • reception, by the vehicle (1), of said list, • detection, in said generated image sequence, of at least one object (OBJ1, OBJ2) of the list of objects (OBJ1, OBJ2) from its descriptor, • determination of the distance between the vehicle (10) and each object (OBJ1, OBJ2) detected, and • obtaining a refined position of the determined position of the vehicle (10) from said determined position and the distance determined between the vehicle (10) and each object (OBJ1, OBJ2) detected. 9. Method according to the preceding claim, wherein the determination of the distance (d1, d2) between the vehicle (10) and each object (OBJ1, OBJ2) detected in the sequence of images is carried out from said sequence of images. 10. Method according to one of claims 8 and 9, wherein the vehicle (10) further comprising a distance measurement module, the distance (d1, d2) between the vehicle (10) and each object (OBJ1, OBJ2 ) detected in the image sequence is measured by said distance measurement module. 1/2 2/2 Fig. 3 Fig. 4 FRENCH REPUBLIC PROPERTY INDUSTRIAL PRELIMINARY SEARCH REPORT based on the latest claims filed before the start of the search National registration number FA 849244 FR 1851855 DOCUMENTS CONSIDERED AS RELEVANT Relevant claim (s) Classification attributed to the invention by ΙΊΝΡΙ Category Citation of the document with indication, if necessary, of the relevant parts EP 3,078,937 A1 (HITACHI AUTOMOTIVE SYSTEMS LTD [JP]) October 12, 2016 (2016-10-12) * paragraph [0017] - paragraph [0021] * * paragraphs [0026], [0028], [0032] - [0034] * * paragraphs [0039] - [0043], [ 0051], [0063], [0064] - paragraph [0043] * * paragraph [0069] - paragraph [0077] * * paragraph [0087] - paragraph [0092] * * paragraphs [0096] - [0109], [0112 ] - [0114] * * paragraphs [0131] - [Θ133], [0136], [0138], [Θ139] * * figures 1,2,4,5,7-11,14 * 1-9 G01S19 / 40 G01S19 / 42 G01S19 / 14 B60W50 / 00 EP 3 279 611 Al (CLARION C0 LTD [JP]) February 7 2Θ18 (2018-02-07) * paragraphs [0023] - [0025], [0036], [0041] * * paragraph [0061] - paragraph [0065 ] * * paragraphs [0075] - [0089], [0103] - [0106], [0149], [0150] * * paragraph [0160] - paragraph [0171] * * figures 1-5,10 * 1-9 TECHNICAL AREAS RESEARCHED (IPC) G01C G01S G05D FR 2 997 512 Al (BOSCH GMBH ROBERT [DE]) May 2, 2014 (2014-05-02) * page 1, line 12 - line 32 * * page 2, line 19 - line 35 * * page 6, line 13 - page 7, line 2 * * page 8, line 3 - line 35 * * pages 1,2; figures 1-3 * 1,3,4,7 8,10