FR3085082A1 - ESTIMATION OF THE GEOGRAPHICAL POSITION OF A ROAD VEHICLE FOR PARTICIPATORY PRODUCTION OF ROAD DATABASES - Google Patents

Abstract

The present invention relates to systems (100) and methods (500) for processing participatory road data and a set of computer programs making it possible to determine the geographical position of road vehicles for the participative production of road databases. The general principle of the invention is based on the determination of the geographical position of road vehicles for the participative production of road databases. Participatory road data are obtained by participatory production of road vehicles circulating on a road network. These data are gathered in a so-called "cloud" computer system. In the invention, only raw data is collected and then used to calculate the geographic position of road vehicles, according to a homogeneous calculation method.

Description

The present invention relates to the determination of the geographical position of road vehicles. More specifically, it relates to systems and methods for processing participatory road data and a set of computer programs making it possible to evaluate the geographical position of road vehicles, this position being subsequently used for the implementation of geographical data. of road data by participatory production.

The road scenes are images of the road and its surrounding environment, acquired by a road vehicle in circulation. Nowadays, road scene image databases are used more and more for the implementation of driver assistance systems.

Beyond the simple visual exploitation, an offline analysis of these images makes it possible to detect, recognize and locate road objects associated with the roadway. For example, we can find out the positioning of fixed objects on the shoulder which represent a potential danger for a vehicle leaving the road, and which can be useful for the vehicle to locate on the road. The road objects of interest are primarily objects of vertical orientation having a "bole" part such as trees, posts, bridge piers, panels and delineators and objects with horizontal "bole", such as slides.

It is known that the estimate of the position of road objects is evaluated in a relative manner with respect to the geographical position of the road vehicle carrying the camera having captured the images. However, it is also known that the absolute location of vehicles (for example by GPS) is intrinsically flawed, making it particularly imprecise. In addition, it is known that the method of calculating the geographical position of road vehicles is not homogeneous according to the models and brands of absolute location sensors which are carried in road vehicles. Thus, several road vehicles capturing images of road scenes to detect and recognize the same road object can produce inconsistent geographic positioning results.

Consequently, it is necessary to be able to determine the position of the road vehicles in a precise and homogeneous manner, in order to facilitate the detection, the recognition, and the localization of road objects in the image bases of road scenes.

The present invention therefore aims to solve the aforementioned drawbacks. For this, in a first aspect, the invention proposes a method for processing participatory road data.

In a second aspect, the invention proposes a computer program with a program code making it possible to implement the method of the first aspect.

In a third aspect, the invention provides a non-transient storage medium on which a computer program according to the second aspect of the invention is stored.

Finally, in a fourth aspect, the invention proposes a system for processing participatory road data.

Thus, the invention relates to a process for processing participatory road data comprising the following steps:

• during the acquisition, at a given moment, of at least one image representative of a road scene from a remote sensing sensor on board in each of a plurality of road vehicles, a measurement step, during which measurement, by a satellite positioning receiver on board the road vehicle, of a plurality of distances, called raw satellite positioning data, between the satellite positioning receiver and a plurality of satellites in view of the satellite positioning receiver at the time given, • a transmission step, during which the image and the raw satellite positioning data are transmitted by a transmitter on board the road vehicle to at least one participative data processing server, • a step of wait, during which one waits for a predetermined period of time to elapse before performing an ac step quisition, during which we acquire, at the level of the participative data processing server, additional positioning data by satellites which are associated with the given moment, • a first calculation step, during which we calculate, at the level of the processing server participatory data, the coordinates of the geographic position of the road vehicle at the given time, from raw satellite positioning data and additional satellite positioning data.

In one example, the additional satellite positioning data comprise data taken, alone or in combination, from: orbital data, space-time data, raw satellite positioning data which relate to at least one ground station of known position and arranged at a predetermined distance from the road vehicle, and measurements relating to the conditions of atmospheric propagation of the satellite signals.

According to a first embodiment:

• in the acquisition step, we acquire proprioceptive data from the road vehicle, and • in the first calculation step, we calculate the geographic position of the road vehicle at the given time from, in addition, proprioceptive data from the road vehicle .

According to a second embodiment, the method further comprises, after the transmission step, an integrity verification step, during which a mechanism for verifying the integrity of the raw positioning data by satellites is applied.

According to a third embodiment, the method further comprises, after the calculation step, an image processing step, during which at least one road object is identified in the image representative of the road scene, and determines the coordinates of the geographical position of the identified road object, from the coordinates of the geographical position of the road vehicle at the given time and at least the image representative of the road scene at the given time.

In a variant of the third embodiment, the method further comprises a second calculation step, during which one calculates, for each identified road object, an average of the coordinates of the geographical position of the identified road object, from coordinates of the geographical position of the identified road object associated with all or part of the plurality of road vehicles.

The invention also covers a computer program with program code for executing the process steps according to the first aspect of the invention when the computer program is loaded into the computer or executed in the computer.

The invention further covers a non-transient storage medium on which a computer program according to the second aspect of the invention is stored.

Finally, the invention covers a participatory road data processing system comprising:

• a plurality of road vehicles each comprising a remote sensing sensor for acquiring, at a given time, at least one image representative of a road scene, a satellite positioning receiver for measuring a plurality of distances, called raw positioning data by satellites, between the satellite positioning receiver and a plurality of satellites in view of the given satellite positioning receiver and a transmitter for transmitting the image and the raw satellite positioning data to at least one data processing server participative, • at least one participative data processing server comprising a clock for delivering a time from the system, a receiver for receiving, as a function of the clock, the raw images and data for positioning by satellites as well as complementary data of satellite positioning, for each of the plural ity of road vehicles, and which are associated at the given time, and • in which, each of the plurality of road vehicles and the participatory data processing server further comprises at least one processor coupled to a storage medium according to the third aspect of the invention.

Other characteristics and advantages of the invention will be better understood on reading the description which follows and with reference to the appended drawings, given by way of illustration and in no way limiting.

- Figure 1 shows the participatory road data processing system according to the invention.

- Figure 2 shows a road vehicle according to the invention.

FIG. 3 represents an image illustrating a road scene acquired by a remote sensing sensor on board the road vehicle of FIG. 2.

- Figure 4 shows a participatory data processing server according to the invention.

FIG. 5 represents a process implemented by the processors of FIGS. 2 and 4.

For reasons of clarity, the elements represented are not necessarily represented on the same scale, with respect to each other, unless otherwise stated.

The general principle of the invention is based on the determination of the geographical position of road vehicles circulating on a road network, for the participative production of road databases. Participatory road data is then obtained by participatory production of road vehicles. These data are gathered in a cloud computing system (“cloud computing system”, in English). In the invention, only raw data is collected and then used to calculate the geographic position of road vehicles, according to a homogeneous calculation method.

FIG. 1 illustrates a system 100 for processing participatory road data according to the invention. Participatory road data (“road data crowdsourcing” in English) means a set of data acquired by participatory production. These data are associated with one or more road scenes which are collected by a large number of road vehicles, a priori, anonymously, so as to feed one or more road databases (also known as the "road model> >). These databases can then be used in the context of road asset management, road safety studies or even driver assistance systems. In the invention, the term “road vehicle” means any vehicle equipped with an engine (generally an internal combustion or electric motor) intended to move it on the road and capable of transporting people or loads (for example, a car or a motorcycle ).

The system 100 comprises a plurality of road vehicles 110, at least one server 120 for processing participative data and at least one server 130 for complementary satellite positioning data. The road vehicles 110, the server 120 for processing participative data and the server 130 for additional satellite positioning data are functionally coupled together via a wireless communication network 200 of known type.

FIG. 2 illustrates a road vehicle 110 according to the invention. In the example of FIG. 2, the road vehicle 110 comprises a remote sensing sensor 111 of known type, a receiver 112 for positioning by satellites of known type, a transmitter 113 of known type and a processor 114 of known type. In the invention, the transmitter 113 is adapted to operate with the wireless communication network 200.

In FIG. 2, the remote sensing sensor 111 is configured to acquire at least one image representative of a road scene when the road vehicle 110 is traveling on a road network. In one example, the remote sensing sensor 111 is a camera. In another example, the remote sensing sensor 111 is a scanning remote sensing sensor such as a lidar, a radar, a SAR (“Synthetic Aperture Radar”, in English) or a sonar.

FIG. 3 illustrates an image 400 representing a road scene acquired by the remote sensing sensor 111 on board the road vehicle 110 (not shown) traveling on a portion of a road network 410 in the traffic direction 420.

In the example of FIG. 3, the image 400 comprises a road object 430 associated with the portion of the road network 410. In a particular implementation, the road scene comprises a plurality of road objects 430. The term “object” is understood to mean road 430, any object associated with a portion of a road network, either permanently or temporarily. This can be, for example, vertical (e.g. a traffic sign) and horizontal (e.g. a road mark, retarder, roundabout) traffic sign. It can also be a question of pavement irregularities (eg a pothole, an area of ice) or an event associated with the portion of the road network (eg an accident). Finally, it can also be a road vehicle.

Returning to FIG. 2, the satellite positioning receiver 112 is configured to measure a plurality of distances, called raw satellite positioning data, between the satellite positioning receiver 112 and a plurality of satellites 300, in view of the receiver 112 satellite positioning. In one example, the satellite positioning receiver 112 receives signals from a GNSS satellite positioning system, such as the American GPS system, the Russian GLONASS system and / or the European GALILEO system.

Raw satellite positioning data includes raw data that can be obtained from known methods such as phase measurement or code measurement, and indicators of the quality of these measurements.

Still in FIG. 2, the transmitter 113 is configured to send data to at least one server 120 for processing participatory data.

FIG. 4 illustrates a server 120 for processing participatory data according to the invention. In the example of FIG. 3, the server 120 for processing participative data comprises a clock 121 of known type, a receiver 122 of known type and a processor 123 of known type. In the invention, the receiver 122 is adapted to operate with the wireless communication network 200. The server 120 is also connected and configured to receive additional positioning data by satellite from at least one external server such as the server 130 additional satellite positioning data.

In Figure 4, the clock is configured to deliver a time from system 100.

In addition, the receiver 122 is configured to receive data from the road vehicle 110.

With reference to FIGS. 1 to 4, the processor 114 of each road vehicle 110 and the processor 123 are configured to implement a method 500 for processing participatory road data, as illustrated in FIG. 5.

In practice, the processor 114 is firstly configured for, during the acquisition, at a given moment, of at least one image representative of a road scene from the remote sensing sensor 111, to implement a step of measure 510 during which the raw satellite positioning data is measured by the satellite positioning receiver 112.

In addition, the processor 114 is configured to implement a transmission step 520, during which the image and the raw satellite positioning data are transmitted by the transmitter 113 to at least one server 120 for processing data. participatory data.

Thereafter, the processor 123 is configured to implement a waiting step 530 during which it is expected that a predetermined period of time by the clock 121 has elapsed before performing an acquisition step 540, during which we acquire, at the level of the server 120 for processing participatory data, complementary positioning data by satellites and which are associated with the given moment. In the example of FIG. 1, the additional satellite positioning data are obtained from the server 130 of additional satellite positioning data.

In a first example, the additional satellite positioning data includes orbital data such as ephemeris data and clock correction data.

In a second example, the additional satellite positioning data comprise raw measurements of at least one ground station of fixed position, the location of which is known and which is arranged at a predetermined distance from the road vehicle 110. In one implementation particular, the ground station is part of the RGP network ("permanent GNSS network") of IGN ("National Institute of Geographic and Forest Information >>; http://rgp.ign.fr).

In a third example, the additional satellite positioning data comprises spatiotemporal data. By spatiotemporal data is meant the data used in augmentation systems which have been developed to improve performance, in particular the accuracy of GNSS satellite positioning systems. Different augmentation systems exist. For example, there are systems which use differential complementary data to GNSS signals, obtained either from satellites of a constellation other than the GNSS constellation (SBAS augmentation system for GPS) or from a GNSS reference station at ground by a radio data link operating in VHP (GBAS augmentation system). There are also systems that allow the calculation of unknowns or entire ambiguities related to phase measurements, by using the differential increase through a reference station (RTK technique) or by obtaining information on the position of the satellites and their clock error, newer and better (PPP technique).

In a fourth example, the additional satellite positioning data includes measurements relating to the conditions of atmospheric propagation of satellite signals, for example, in the ionosphere and the troposphere.

In the invention, the waiting step 530 is aimed at making it possible to recover, with more or less precision, the additional satellite positioning data. According to the GNSS satellite position system, it is not possible to obtain, in real time, additional satellite positioning data, with sufficient precision. However, with a little waiting, it is possible to obtain the desired level of precision. For example, additional very precise satellite positioning data, of the order of 2.5 cm, can be retrieved online via Internet servers such as that of the IGS (“International GNSS Service”, in English; http : //www.igs.org/), and this, after twelve to eighteen days following the measurements of the receiver 112 for positioning by satellites. In the realm of road scene image databases, such delays do not affect the general creation process.

In addition, the processor 123 is configured to implement a calculation step 550 during which the coordinates of the geographic position of the road vehicle 110 at the given time are calculated at the time of the participatory data processing server 120, from the raw satellite positioning data, and additional satellite positioning data. In this step, it is possible to use conventional methods of trilateration and / or multilateration localization or estimation techniques of the Kalman filtering or particle filtering type.

It should be noted that in the invention, the participative data processing server 120 determines the coordinates of the geographical position of each road vehicle 110, by applying, in a homogeneous manner, a participatory data processing to the raw positioning data by satellites and to additional satellite positioning data. The problem of disparity in the methods for calculating the geographic coordinates of road vehicles 110 is therefore resolved, which is due to the use of different methods according to the manufacturers of road vehicles 110. In addition, the accuracy of the positions calculated is improved by taking advantage of additional correction data.

According to an implementation of the process 500 for processing participatory road data, the following operations are implemented:

• in the acquisition step, proprioceptive data of the road vehicle 110 is acquired, and • in the first calculation step, the geographic position of the road vehicle at the given time is calculated from, in addition, proprioceptive data of the vehicle road 110.

In this implementation, the calculation of the accuracy, the availability and the integrity of the geographical position of the road vehicle 110 is improved by using additional information provided by the instruments on board the road vehicle 110 which provide information on the movement of the vehicle. road 110. Then, by proprioceptive data is meant the data acquired from proprioceptive sensors of the road vehicle 110 such as the gyroscope or the odometer. This data can be used with known methods of data fusion such as hybridization.

According to another implementation of the method 500 for processing participatory road data, after the transmission step, an integrity verification step is carried out, during which an integrity verification mechanism for the raw positioning data is applied. by satellites (or RAIM for “Receiver Autonomous Integrity Monitoring”).

According to yet another implementation of the method 500 for processing participatory road data, after the calculation step, an image processing step is implemented, during which at least one road object 430 is identified in the image representative of the road scene, and the coordinates of the geographical position of the identified road object 430 are determined, from the coordinates of the geographical position of the road vehicle 110 at the given time and at least the image representative of the scene roadside at the given time. In this case, it is possible to use conventional methods of detection and recognition of road objects 430 which are based, for example, on the comparison of visual characteristics of an observation with references previously learned by a computer vision system.

In an example of this implementation, the processor 123 implements a second calculation step, during which one calculates, for each road object 430 identified, an average of the coordinates of the geographical position of the road object 430 identified, at from the coordinates of the geographic position of the identified road object 430 associated with all or part of the plurality of road vehicles. With this implementation, we take advantage of participatory production to refine the precision of the position of the geographic coordinates of an identified road object 430.

In a particular embodiment of the invention, the different steps of the method 500 for processing participatory road data are determined by instructions from computer programs. Consequently, the invention also relates to a program with a computer program code fixed on a non-transient storage medium, this program code being capable of executing the steps of the method 500 for processing participatory road data, when the computer program is loaded into the computer or executed in the computer.

The present invention has been described and illustrated in the present detailed description and in the figures. However, the present invention is not limited to the embodiments presented. Thus, other variants and embodiments can be deduced and implemented by the person skilled in the art on reading this description and the appended figures.

Claims (9)

1. Method (500) for processing participatory road data comprising the following steps: • during the acquisition, at a given time, of at least one image (400) representative of a road scene from a remote sensing sensor (111) on board in each of a plurality of road vehicles (110) , a measurement step (510), during which a plurality of distances, called raw satellite positioning data, between the satellite receiver and a satellite positioning receiver (112) on board the road vehicle are measured positioning by satellites and a plurality of satellites (300,) in view of the satellite positioning receiver at the given time, • an emission step (520), during which it is emitted, by a transmitter (113) on board the road vehicle , the image and the raw satellite positioning data to at least one server (120) for processing participative data, • a waiting step (530), during which a period of time is waited for p has elapsed before carrying out an acquisition step (540), during which one acquires, at the level of the participative data processing server, additional satellite positioning data which are associated with the given moment, • a first step of calculation (550), during which the coordinates of the geographic position of the road vehicle at the given time are calculated at the level of the participative data processing server, from the raw satellite positioning data and additional satellite positioning data . 2. Method according to claim 1, in which the additional satellite positioning data comprise data taken, alone or in combination, from: orbital data, space-time data, raw satellite positioning data which relate to at least a ground station of known position and disposed at a predetermined distance from the road vehicle, and measurements relating to the conditions of atmospheric propagation of the satellite signals. 3. Method according to any one of claims 1 to 2, in which: • in the acquisition step, we acquire proprioceptive data from the road vehicle, and • in the first calculation step, we calculate the geographic position of the road vehicle at the given time from, in addition, proprioceptive data from the road vehicle . 4. Method according to any one of claims 1 to 3, further comprising, after the transmission step, an integrity verification step, during which a mechanism for verifying the integrity of the raw data of satellite positioning. 5. Method according to any one of claims 1 to 4, further comprising, after the calculation step, an image processing step, during which at least one road object (430) is identified in the image. representative of the road scene, and the coordinates of the geographical position of the identified road object are determined, from the coordinates of the geographical position of the road vehicle at the given time and at least the image representative of the road scene at given time. 6. The method of claim 5, further comprising a second calculation step, during which one calculates, for each identified road object, an average of the coordinates of the geographic position of the identified road object, from the coordinates of the geographic position of the identified road object associated with all or part of the plurality of road vehicles. 7. A computer program with a program code for executing the process steps of a process according to any one of claims 1 to 6 when the computer program is loaded into the computer or executed in the computer. 8. A non-transient storage medium on which a computer program according to claim 7 is stored. 9. Participatory road data processing system (100) comprising: • a plurality of road vehicles (110) each comprising a remote sensing sensor (111) for acquiring, at a given moment, at least one image (400) representative of a road scene, a satellite positioning receiver (112) for measuring a plurality of distances, called raw satellite positioning data, between the satellite positioning receiver and a plurality of satellites in view of the satellite positioning receiver at the given time and a transmitter (113) for transmitting the image and the raw satellite positioning data to at least one participative data processing server, • at least one participative data processing server (120) comprising a clock for delivering system time, a receiver (122) for receiving, in clock function, images and raw satellite positioning data as well as additional positioning data by satellite, for each of the plurality of road vehicles, and which are associated at the given time, and in which, each of the plurality of road vehicles and the participatory data processing server further comprises at least one processor coupled to a storage medium according to claim 8.