DEVICE AND METHOD FOR AUTONOMOUSLY LOCATION OF A MOBILE VEHICLE ON A RAILWAY TRACK

Description

Previous Art

[0001] The present invention generally relates to location systems and in particular to a device and a method for autonomous location of a mobile vehicle in a railway network comprising at least one railway track using a cartographic database.

[0002] The localization of a railway vehicle moving on railway tracks of a railway network is generally carried out non-autonomously from sensors, called 'track sensors', deployed all along the railway tracks. A railway vehicle can be located using an odometry technique between two successive track sensors. The localization can be carried out in a railway reference frame by determining the section of the railway on which the railway vehicle is located. The returned localization can be qualified as integral if the risk that the railway vehicle is not on the determined section is lower than an acceptable limit. A localization technique combining track sensors and odometry is all the more integral as the distance separating two successive track sensors is low. However, such a known localization technique is expensive to deploy and involves high maintenance costs.

[0003] In other existing approaches, integrated and autonomous localization systems that can do without the deployment of sensors on the track have been proposed. The localization provided by such localization systems can be carried out in a terrestrial reference frame by means of a localization device fitted to the railway vehicle which receives and processes navigation signals emitted by geolocation satellites such as GPS (acronym for 'Global Positioning System' literally meaning 'Global Positioning System') and GNSS (acronym for 'Global Navigation Satellite System' literally meaning 'Global Navigation Satellite System') type satellites. The transition from a location in a terrestrial reference frame to a location in a railway reference frame can be carried out using a cartographic database of the railway network.Such a map database includes the geographical coordinates of various elements of the railway network such as railway tracks. The map database may further include a description of certain elements of the environment adjacent to the railway network that may interfere with the navigation signals received by the railway vehicle. The integrity of such autonomous location systems therefore depends on the integrity of the map database used.

[0004] An example of a rental system using a map database has been described for example in GB 2480102.

[0005] The construction of an integrated cartographic database requires topographic measurements to be carried out on the railway network and its adjacent environment before processing the resulting measurement point cloud to model elements of the railway network and certain objects in the adjacent environment. Such processing requires a classification of each measurement point according to whether it characterizes the railway network or its adjacent environment.

[0006] The classification of measurement points is generally carried out manually. However, such manual classification is complex to carry out and time-consuming. The processing of the classified points must then be carried out by devices whose resources in terms of storage memory and computing speed are often insufficient to efficiently handle the large volume of data (in the order of terabytes) corresponding to the classified points. This results in a reduction in the reliability and response time of the vehicle tracking device using such a database.

[0007] Modeling the railway tracks of a railway network is often done using computer-aided design (CAD) tools initially designed to model road infrastructures. A railway application requires sub-decameter accuracy when acquiring topographic data and a protection radius characterizing the error ellipsoid of a few meters. However, the requirement of a railway application in terms of integrity is very high (of the order of 10 -7< defects per hour). On the other hand, modeling the railway tracks is not sufficient to reconstruct the topology of the railway network which requires the identification of other elements of the railway network such as junctions. Design tools require increased expertise and know-how from the user. Several manipulations are necessary to extract information from acquisition measurements.

[0008] Furthermore, existing tools do not allow the topology of a railway network to be efficiently reconstructed from a geo-localized and classified measurement point cloud, taking into account the specificities of the railway application.

[0009] To model certain objects in the environment adjacent to the railway network, there are learning-based tools capable of extracting and modeling these objects from a geo-localized and classified measurement point cloud. Such tools allow recognizing objects in the adjacent environment. However, the functionalities of such tools are limited to extracting objects from the adjacent environment and do not allow analyzing their effects on the navigation signals received by a railway vehicle moving in their vicinity.

[0010] There is therefore a need for an improved device and method for locating vehicles in a railway network. General definition of the invention

[0011] The invention improves the situation by proposing a method and a system for locating in a railway reference system a mobile vehicle moving on railway tracks of a railway network according to independent claims 1 and 13.

[0012] Other embodiments are defined in the dependent claims.

[0013] The embodiments of the invention thus provide a method and a device allowing precise and integrated localization of a mobile vehicle moving on the railways of a railway network. Such localization can be guaranteed at each point of the railway network and at each moment. It is advantageously carried out with an optimal response time and reduced computational complexity thanks to taking into account the particularities of railway networks in terms of precision and integrity. Brief description of the drawings

[0014] Other characteristics and advantages of the invention will appear with the aid of the following description and the figures of the appended drawings in which: There Figure 1 represents a system for locating mobile vehicles moving in a railway network according to embodiments of the invention, The Figure 2 is a flowchart representing the steps in the process of generating the cartographic database, The Figure 3 is a flowchart representing the steps implemented to determine the topology of a railway network and the modeling of the geometry of railway tracks of the network, according to certain embodiments of the invention, The Figure 4 is a flowchart representing the steps implemented to determine geo-spatial vectors from a cloud of geo-located and classified measurement points, according to certain embodiments of the invention, The Figure 5is a flowchart representing the steps implemented to model the geometry of a segment of a railway track, according to certain embodiments of the invention, The Figure 6 is a flowchart representing the steps implemented to achieve an association between modeled railway track segments and objects of the adjacent environment, according to certain embodiments of the invention, The Figures 7 And 8 show two examples of association between a modeled railway track segment and an object of the adjacent environment, according to embodiments of the invention, and The Figure 9 is a localization system according to embodiments of the invention. Detailed description

[0015] There Figure 1 depicts a stand-alone location and navigation system 100 in which an integrated map database 102 may be used according to certain embodiments of the invention.

[0016] The autonomous location and navigation system 100 can be deployed in a railway vehicle moving on a railway track of a railway network in order to provide instantaneously and precisely the location of the railway vehicle as well as navigation data usable for navigation in autonomous mode. More generally, the autonomous location and navigation system 100 can be deployed on any type of mobile vehicle moving on a trajectory constrained by a dedicated track, such as a train.

[0017] The autonomous location and navigation system 100 may further comprise a location device 101 and a tracking device 103.

[0018] The location device 101 may comprise a navigation signal receiver configured to receive navigation signals transmitted by location systems such as GPS or GNSS type systems. The location device 101 may use such signals to determine location parameters associated with the moving vehicle. The location parameters may comprise position data representing the three-dimensional position of the vehicle in a global geographic reference frame linked to the Earth, and / or the orientation of the vehicle and / or the speed of movement of the vehicle. For each determined three-dimensional position, the location device 101 may associate an error ellipsoid and / or an alarm signal which may be triggered if a condition relating to the determined position relative to the ellipsoid is verified.In one embodiment, the alarm signal is triggered if the considered position of the vehicle lies within the error ellipsoid. The alarm signal may be stopped when the considered position of the vehicle passes outside the error ellipsoid after a predefined time delay.

[0019] The location device 101 may further comprise an inertial unit configured to provide location parameters that are less precise than those determined from navigation signals, without using any external information. The inertial unit may be used in the absence of available navigation signals in order to enable continuous location of the mobile vehicle. Such a situation may occur, for example, when a mobile vehicle is moving through a tunnel. The inertial unit may also be used in the event of degradation in the quality of navigation signals. Such degradation may originate from multi-path propagation caused by objects in the environment adjacent to the mobile vehicle. Objects in the adjacent environment that may induce such degradation include, for example, buildings, vegetation, road signs, etc.

[0020] In one embodiment of the invention, the location device 101 may additionally comprise a module for hybridizing the location parameters provided respectively by the navigation signal receiver and by the inertial unit.

[0021] The integrated map database 102 is configured to store a set of data describing the topology of the railway network (hereinafter referred to as "topology description data") comprising at least one railway track, in which a mobile vehicle can move. The data of the integrated map database 102 may further include data describing the format of such a network (hereinafter referred to as "network format description data"). The topology of the railway network and the network format constitute key elements of autonomous navigation of mobile vehicles. As used herein, the term "autonomous navigation" refers to navigation in a vehicle without an active driver.

[0022] The integrated cartographic database 102 may further include data describing certain objects in the environment adjacent to the railway tracks of the railway network and likely to disrupt navigation signals received by the mobile vehicle (hereinafter referred to as “environmental object description data”).

[0023] The topography description data may include identification data for some of the constituent elements of the railway network associated with the database, such as railway tracks, junctions and / or stopping points (also called 'terminuses'). The identification data associated with the constituent elements of the network may also include interconnection data representing the interconnections between the different constituent elements.

[0024] Each element of the railway network described by data from the database 102 can be associated with location data in the global geographic reference system linked to the Earth associated with the location device 101.

[0025] The network format corresponding to a railway network can represent a set of possible journeys on the railway network, each journey comprising a set of segments corresponding to a railway track in the network, for a given start point and end point. The network format can be determined from chaining characteristics in the network by identifying, for example, for each railway track connecting a start point and an end point the railway segments that constitute it and the junctions encountered by a mobile vehicle moving on the railway network. A junction is defined with respect to a direction of travel and can be seen as a zero-length segment that has one entrance and two exits.

[0026] The tracking device 103 may be configured to determine the location of the mobile vehicle in a railway reference frame by identifying the railway track segment on which the mobile vehicle is located. The tracking device may receive the location parameters provided by the location device 101 and then query the integrated map database 102 using an input query comprising at least some of these parameters to identify candidate railway track segments on which the mobile vehicle can be located. The tracking device 103 may use navigation rules in the railway network to determine the railway track segment on which the mobile vehicle is located. The tracking device 103 may further be configured to determine the abscissa of the mobile vehicle on the current railway track segment in the reference frame considered.

[0027] The embodiments of the invention advantageously allow semi-automatic generation of the integrated database 102.

[0028] There Figure 2illustrates the database generation method implemented to generate the integrated cartographic database 102, according to certain embodiments of the invention. In step 201, a geo-located and classified measurement point cloud describing in three-dimensional manner a railway network and its adjacent environment is received. Each measurement point can be geo-located in the sense that the location of this measurement point in a given measurement reference frame is known precisely. Such a measurement reference frame can be the terrestrial measurement reference frame. The coordinates of the measurement points can be provided in a GPS / GNSS trace type file. The geo-located measurement point cloud can be obtained from topographic surveys previously carried out using one or more topography techniques.Examples of surveying techniques include, but are not limited to, mobile mapping system surveying, also known as Mobile Mapping System (MMS), satellite surveying, and conventional surveying using a transportable surveying device. MMS systems have the advantage of allowing very high-quality measurements in terms of resolution (less than 10 centimeters) with increased geolocation quality through the use of a hybrid geolocation system comprising a GPS / GNSS location device associated with an inertial unit. MMS systems can use multiple optical cameras and laser remote sensing devices to build a three-dimensional model of the analyzed environment by implementing dedicated tools and vision algorithms.

[0029] Each geo-located measurement point can further be classified into at least two groups of geo-located measurement point clouds comprising a group of point clouds associated with the ground and the railway network, and a group of point clouds associated with the adjacent environment.

[0030] Such classification into two groups of measurement point clouds can be performed by applying one or more classification algorithms. Such classification algorithms can be executed on a computer system or on a cloud computing service. Such classification of geo-located measurement points offers savings in time, accuracy and computing resources compared to conventional classification approaches that are performed manually by operators.

[0031] In step 201, the coordinates of certain constituent elements of the railway network, in the same measurement reference frame as the geo-localized measurement point cloud, may further be received. Such constituent elements may include junctions and stopping points.

[0032] In step 201, a subsampling operation may be applied to the received geo-located and classified measurement point cloud. Such a subsampling operation makes it possible to reduce the size of the geo-located and classified measurement point cloud processed by the database generation method. The subsampling factor associated with such a subsampling operation may be the same for the three dimensions of the geo-located and classified measurement point cloud. Alternatively, a subsampling factor may be associated with each dimension of the measurement point cloud.

[0033] In step 202, the geo-localized measurement point cloud associated with the ground and the railway network can be separated from the geo-localized measurement point cloud associated with the environment adjacent to the railway network. Such a separation advantageously allows parallel processing of two geo-localized measurement point clouds.

[0034] In step 203, the topology of the railway network is determined from the geo-located and classified measurement point cloud associated with the ground and the rails. The step of determining the topology of the railway network may include an identification of the railway tracks constituting the railway network, junctions and stopping points. The railway tracks may be modeled by geo-spatial vectors. Each of the geo-spatial vectors may include the coordinates in a measurement reference frame of the centerline of the associated railway track.

[0035] Junctions may be identified by analyzing the intersections between the various identified railways as modeled by the geospatial vectors. Stopping points may be identified by detecting the ends of the identified railways. According to embodiments of the invention, the locations in the measurement reference system of the junctions and stopping points may be provided by step 201.

[0036] In step 204, the railways, represented by the geospatial vector associated with them, are received in order to model the geometry of the railways by standard or usual geometric shapes. The modeling of the geometry of a railway may comprise a subdivision of the railway into several railway segments. An analytical equation may be associated with each railway segment to describe its geometry. An association between each modeled railway segment and the elements of the railway network (other segments, junction, stopping points, etc.) to which it is connected may be determined, which defines the format of the railway network.

[0037] Step 205 is performed from the geo-localized measurement point cloud associated with the adjacent environment and the modeled railway track segments are received. The environment adjacent to the railway network may be a hundred meters wide centered on the railway tracks of the railway network. Such an environment may include stations, platforms, bridges, tunnels, etc. The measurement point cloud associated with the adjacent environment may be transformed into several objects of the adjacent environment. The objects of the adjacent environment may be modeled by standard or usual three-dimensional geometric shapes by determining, for each object of the modeled adjacent environment, a set of geometric parameters including the dimensions of the object and the distances separating the object from the modeled railway track segments.In step 205, an association is determined between each modeled railway track segment and objects of the modeled adjacent environment likely to disturb navigation signals received by a mobile vehicle moving on the railway track segment considered.

[0038] In step 206, the modeled railway segments, the identified adjacent environment objects, and the associations between the segments and the objects as provided by step 205 are used to characterize propagation risks affecting navigation signals received by a mobile vehicle traveling on the railway segments. The propagation risk characterization may be performed at multiple points on each modeled railway segment associated with one or more adjacent environment objects. The propagation risk characterization provides propagation risk parameters such as, for example and without limitation, the type of propagation risk and the dimensions of the mobile vehicle associated with the identified propagation risk.

[0039] According to one embodiment of the invention, elements of the map database may be described in one or more description files having a selected representation format such as XML format. Such elements of the database may include modeled railway segments, junctions, stopping points, objects of the adjacent environment and associations between the modeled segments and the objects of the adjacent environment.

[0040] There Figure 3 illustrates the steps implemented to generate modeled railway track segments from a geo-localized and classified measurement point cloud associated with the ground and the railway tracks according to an embodiment of the invention. Such steps correspond to steps 203 and 204 of the Figure 2The point cloud can be measured by a mobile surveying system associated with an inertial unit. The coordinates of the measurement points can be organized in geolocation files having the structure of a GPS / GNSS track. Each geolocation file can contain a set of data completely describing the coordinates of the measurement points associated with a railway. Coordinates of secondary railways can be described separately in other geolocation files.

[0041] In step 301, geo-located measurement points associated with the ground and the railways and positions of junctions and stopping points on the railways are received.

[0042] In step 302, the received geo-located measurement points are transformed into one or more geo-spatial vectors. Each geo-spatial vector may be associated with a railway track and may comprise the three-dimensional coordinates of a predefined number of points of the rails of the associated railway track. Alternatively, each geo-spatial vector may comprise the three-dimensional coordinates of a predefined number of points of the centerline of the associated railway track. The spacing between the points of the rails or the centerline may be constant. The geo-spatial vector may further comprise heading, slope, and superelevation measurements associated with each of the points of the rails or the centerline of the railway track. Step 302 may be implemented by applying a vision algorithm implemented on a local computer or on a cloud computing service.

[0043] In step 303, geospatial vectors associated with the different railways are subdivided to identify other elements of the railway network including junctions and stopping points. The identification of such elements can be carried out from their coordinates provided by step 301. The subdivision step 303 can further comprise the construction of a graph representing the topology of the railway network. The graph can comprise a set of edges, nodes connecting the edges and leaf nodes connected to a single edge. The edges of such a graph can represent segments of the railways associated with geospatial vectors and the nodes can represent junctions or stopping points (leaf nodes) of the railway network.At the subdivision step 303, the railway track segments connected to each of the nodes (junctions and stopping points) of the railway network, as well as the segments associated with each of the railway tracks can for example be determined.

[0044] In step 304, the geometry of the railway tracks of the railway network, represented by the geospatial vectors, may be modeled using analytical equations. The modeling step 304 may include a subdivision of each geospatial vector of the railway network into several segments in order to facilitate the modeling. A standard deviation measuring the error between the measured curve of a railway track and the curve resulting from the analytical equations may be determined. The modeling step 304 may further be configured so that the standard deviation associated with each of the segments of the railway network is less than a predefined modeling error threshold.

[0045] In step 305, the segmentation of the railway network is performed from the modeled railway track segments as well as other elements of the railway network. Step 305 may, for example, comprise a definition of the direction of switching in each of the junctions of the railway network and a direction of movement of mobile vehicles on each of the modeled railway track segments. Step 305 may further comprise a saving of the elements of the railway network as well as the format of the network in description files having a chosen description format (such as XML format for example).

[0046] There Figure 4is a flowchart representing the steps implemented to create a geospatial vector from a geolocated and classified measurement point cloud associated with the ground and the railways according to one embodiment of the invention. In such an embodiment, the geospatial vector represents the centerline of a railway designating a running path for a railway vehicle such as a train comprising two rail strands whose spacing is kept constant by an attachment to sleepers. According to other embodiments of the invention, the geospatial vector may represent one or more rail strands of the railway.

[0047] In step 401, one of the rail wires of each railway track is selected by analyzing the associated measurement points and their coordinates as provided by the geolocation file of the measurement point cloud. For a railway track comprising two rail wires, the right rail defined with respect to the direction of movement of the mobile vehicle can for example be selected. The remainder of the description will be made with reference to such an example of rail wire selection, by way of non-limiting example.

[0048] The measurement points associated with the selected rail wire are then subsampled in step 402. The subsampling factor can be of the order of ten points. Such an order of magnitude of the subsampling factor constitutes a compromise between measurement accuracy and computational complexity.

[0049] In step 403, the subsampled measurement points are grouped into subgroups of subsampled measurement points such that two successive subgroups of subsampled measurement points share at least one measurement point. Such an overlap between the subgroups of measurement points guarantees the continuity of the measurements that will be extracted.

[0050] In step 404, a moving average is applied to the subgroups of sub-sampled measurement points in order to calculate a three-dimensional average position for each of the subgroups. Such an average position can be defined in the same measurement reference system as that associated with the geo-localized measurement point cloud.

[0051] In step 405, a geospatial vector representing the right track of the railway is created by grouping the calculated three-dimensional average positions and respecting the order in which the positions are obtained.

[0052] In step 406, the geospatial vector thus obtained is filtered so as to eliminate the positions from measurement points likely to be impacted by noise from an automated classification.

[0053] In step 407, the measurement points associated with the left rail wire of the railway track are analyzed in a similar manner as for the right rail wire of the same railway track, to determine a second geospatial vector representing the left track.

[0054] At step 408, the geospatial vector representing the centerline of the railway track is determined from the geospatial vectors representing the associated left and right rail wires. The coordinates of the centerline of the railway track can be calculated by averaging the right and left rail vector points.

[0055] There Figure 5is a flowchart representing the modeling process implemented to model a geospatial vector representing a railway track using one or more analytical equations. The modeling process may include one or more iterations of a set of modeling steps and may receive a modeling error threshold defined per unit length. The analytical equations that can be used to model the geometry of the central axis of a railway track may correspond to a set of basic geometric shapes such as a straight line, a circular arc, and a clothoid. In practice, the shape of a railway track may be approximated to one of the basic geometric shapes even if an exact match is not obtained. A modeling error that may be of the mean quadrature type may be associated with this match.

[0056] In step 501 of the modeling method, a geospatial vector representing a railway track is received in order to be modeled with one of the identified basic geometric shapes. The basic geometric shape retained for modeling a geospatial vector is that associated with a minimal modeling error. Step 501 can further be configured to provide for each modeled geospatial vector parameters comprising the geometric shape retained, the parameters of the analytical equation making it possible to bring the geometry of the geospatial vector closer to the geometric shape retained and the modeling error associated with such a reconciliation.

[0057] In step 502 of the modeling method, the modeling error as provided by modeling step 501 is compared to a modeling threshold. Such a modeling threshold may be determined by multiplying the modeling threshold per unit length and the length of the railway represented by the geospatial vector. Iterations of the modeling method may be stopped if the modeling threshold is greater than the modeling error.

[0058] In step 503, the modeled geospatial vector having a modeling error greater than the modeling threshold is divided into two segments which may have the same length. Each of the two segments may be modeled in a similar manner as the complete geospatial vector using an iterative approach consisting of dividing each segment having a modeling error greater than the modeling error threshold into two sub-segments. The method for modeling a geospatial vector representing a railway provides several segments, each of which is modeled by an analytical equation meeting a modeling error threshold.

[0059] There Figure 6illustrates the steps implemented to identify and associate with the modeled segments objects adjacent to the railway network likely to disturb navigation signals received by a mobile vehicle moving on the modeled segments of railway track of the railway network, according to an embodiment of the invention.

[0060] In step 601, the cloud of geo-located and classified measurement points associated with the environment adjacent to the railway network is filtered according to predefined filtering criteria. Such a filtering operation makes it possible, for example, to eliminate measurement points corresponding to adjacent objects of low height relative to the height of the mobile vehicle receiving navigation signals. The risk that adjacent objects of low height disturb navigation signals can be neglected. Step 601 may further comprise a sub-sampling operation applied to the measurement points associated with adjacent objects whose height is of the same order of magnitude as the height of the mobile vehicle. The sub-sampling factor may be chosen so as to allow the identification of adjacent objects while reducing the complexity of the processing.

[0061] In step 602, the filtered and downsampled measurement points are transformed into adjacent objects, the adjacent objects being associated with object parameters such as the geographical location of the object, dimensions (e.g., length, width, and / or height), the distance separating the object from the nearest railway track, and / or the relative orientation of the object with respect to the trajectory. The resulting adjacent objects may further be synthesized into a standard geometric figure such as a parallelepiped, a cylinder, a pyramid, etc.

[0062] In step 603, the obtained adjacent objects are associated with segments of the railway tracks of the railway network. An object may be associated with a railway track segment if navigation signals received by a mobile vehicle moving on the railway track segment are likely to be disturbed by the object in question. An adjacent object may therefore be associated with more than one railway track segment. Railway track segments may not be associated with any adjacent object.

[0063] In step 704, the main physical phenomenon likely to disturb navigation signals with respect to each association between a railway segment and an adjacent object is determined. Such a physical phenomenon may be, for example, multipath propagation, non-line-of-sight propagation between the transmitter and receiver of the navigation signals, diffraction, etc. The physical phenomenon responsible for the disturbance of navigation signals may further be characterized by specifying the geometric parameters relating to the association between the modeled segment and the adjacent object and to the railway vehicle receiving the navigation signals. Such geometric parameters may include the distance separating the two elements of the association, the dimensions of the adjacent object and the dimensions of the railway vehicle.

[0064] There Figure 7shows an example of an association between a track segment and an adjacent object in which multipath propagation is the main physical phenomenon responsible for the disruption of navigation signals. The object is synthesized into a parallelepiped-like geometric shape, of height 'H' and distant from the central axis of the track sub-segment by a distance 'D'. For a moving vehicle of height 'M', the multipath propagation phenomenon can be quantified using a mathematical equation relating the parameters 'M' and 'D' to the elevation angle 'alpha' between the rail level and the direction in which the navigation signal is received. Such an equation can be written in the following form: H - M / D > tangente alpha

[0065] There figure 8shows a second example of an association in which a railway segment is associated with two adjacent objects. In such an association, the non-line-of-sight propagation between the transmitter and the receiver of navigation signals is the main physical phenomenon responsible for the disturbance of navigation signals. The two objects implemented in such an association are synthesized in a geometric shape of the parallelepiped type. The two adjacent objects are of heights H1 and H2 and are distant from the centerline of the railway segment by respective distances D1 and D2. For a moving vehicle of height M, the phenomenon of non-line-of-sight propagation between the transmitter and the receiver of navigation signals can manifest itself when the elevation angle as defined above satisfies the following relationships: H 1 - 4 M / D 1 > tg alpha ET H 2 - 4 M / D 2 > tg alpha ET tg alpha > H 2 - H 1 / D 1 + D 2

[0066] There Figure 9is a diagram representing an autonomous localization system 100 of a mobile vehicle moving on a railway track of a railway network implementing the localization method, according to one embodiment of the invention.

[0067] The autonomous location system 100 may comprise a location device 101 configured to: Determining location parameters associated with the mobile vehicle in a geographic reference frame, Determining one or more candidate railway track segments by querying a map database based on at least some location parameters, and Locating the mobile vehicle from the candidate railway track segments returned by the map database 102.

[0068] The location system 100 may further include a map database generation unit 91 for generating information or alerts on possible modifications to the database 102 (evolution of the environment).

[0069] The map database generation unit 91 may comprise: A receiving module 9100 configured to receive a cloud of measurement points classified and geo-located in a coordinate system, said cloud of measurement points being associated with said railway network and with said adjacent environment, A topology determination module 9102 configured to determine the topology of said railway network from geo-located and classified measurement points associated with said railway network and a plurality of additional elements of said railway network, A modeling module 9104 configured to model the geometry of the railway tracks of said railway network into a plurality of modeled railway track segments, An adjacent object determination module 9106 configured to determine from geo-located and classified measurement points associated with said adjacent environment one or more adjacent objects, each of said adjacent objects being associated with one or more modeled railway track segments,and An identification module 9108 configured to identify for each association between an adjacent object and one or more modeled railway track segments information representing the disturbances suffered by said navigation signals.

[0070] In one embodiment, the map database generation unit 91 may further comprise: A comparator 9111 capable of determining changes in the risk level 704 associated with the railway network and the adjacent environment. Such a comparator 9111 may in particular signal elements of the geo-localized point cloud associated with the adjacent environment in a plurality of adjacent objects likely to disrupt the navigation signals; An alert recording module 9112 capable of detecting the change in risk level (704) in the database; An alert analysis module 9113 capable of qualifying the alert by measurement redundancy 9112; An association module 9114 capable of associating the confirmed alerts 9113 with at least one of the modeled segments 102.

[0071] The embodiments of the invention thus make it possible to reconstruct the geometry and topology of a railway network from a geo-localized and classified measurement point cloud. They also make it possible to qualify the risks linked to the 3D nature of the environment adjacent to the railway network, by transforming the 3D measurements of a classified point cloud and its trace (GPS / GNSS) into 3D geo-spatial vectors for large-scale semi-automated production, without requiring a drawing or manual guidance step with CAD software.

[0072] Those skilled in the art will understand that the system or subsystems according to embodiments of the invention may be implemented in various ways by hardware, software, or a combination of hardware and software, including in the form of program code that may be distributed as a program product, in various forms. In particular, the program code may be distributed using computer-readable media, which may include computer-readable storage media and communication media. The methods described in this disclosure may be implemented, in particular, in the form of computer program instructions executable by one or more processors in a computer computing device. These computer program instructions may also be stored in a computer-readable medium.

[0073] Furthermore, the invention is not limited to the embodiments described above as a non-limiting example. It encompasses all variant embodiments which may be envisaged by those skilled in the art and which are covered by the scope of the appended claims.

Claims

1. A method for locating, in a railroad reference frame, a mobile vehicle traveling on railroad tracks of a railroad network, comprising the steps of: - determining location parameters associated with the position of the mobile vehicle in a geographical reference frame based on a plurality of navigation signals of a GNSS or GPS system received by said mobile vehicle, - determining one or more railroad track segments on which the mobile vehicle is likely to be located by querying a cartographic database based on at least one of the location parameters, - locating the mobile vehicle based on the railroad track segments provided by the cartographic database and at least one of the location parameters, said cartographic database comprising data representing a vectorial description of at least one railroad track of the railroad network and data representing a description of one or more objects of the environment adjacent to the railroad tracks of the railroad network likely to interfere with the navigation signals received by said mobile vehicle, the method comprising a step of generating said cartographic database comprising the steps of: - receiving (201) a measurement point cloud classified and geolocated in a coordinate system, said measurement point cloud being associated with said railroad network and with said adjacent environment, - determining (203) the topology of said railroad network based on geolocated and classified measurement points associated with said railroad network and a plurality of additional elements of said railroad network, - modelling (204) the geometry of the railroad tracks of said railroad network as a plurality of modelled railroad track segments, the step of modelling the geometry of the median axis of a railroad track represented by a geospatial vector comprises one or more iterations of the following steps: i. determining a current analytical model of the geometry of the median axis of the railroad track represented by a geospatial vector using at least one analytical equation, ii. measuring the standard deviation between the median axis of the railroad track and the associated curve returned by the analytical model, iii. dividing the geospatial vector into two elements if the measured standard deviation is greater than a predefined error threshold, steps i. to iii. being iterated for as long as a standard deviation between the analytical model and the geospatial vector is greater than the predefined error threshold. - determining (205), from geolocated and classified measurement points associated with said adjacent environment, one or more adjacent objects, each of said adjacent objects being associated with one or more modelled railroad track segments, - identifying (206), for each association between an adjacent object and one or more modelled railroad track segments, information representing the interference that said navigation signals are subjected to.

2. The method according to claim 1, characterised in that it further comprises a step of saving the elements comprising the modelled railroad track segments, the additional elements of the railroad network, and / or the adjacent objects in files having a given representation format.

3. The method according to claim 1, characterised in that said additional elements of the railroad network comprise junctions and stopping points.

4. The method according to one of the preceding claims, characterised in that the geolocated and classified measurement point cloud is received from at least one mobile topography system associated with a hybrid geolocation system comprising a location device associated with an inertial unit.

5. The method according to one of the preceding claims, characterised in that said location parameters comprise a position in three dimensions, a movement speed and / or an orientation parameter.

6. The method according to one of the preceding claims, characterised in that said step of determining the topology of the railroad network comprises the steps of: - transforming elements of said geolocated point clouds associated with said railroad network into a plurality of geospatial vectors, each of the geospatial vectors corresponding to a railroad track, - subdividing said railroad network by defining, for each geospatial vector, a plurality of elements comprising junctions and / or stopping points.

7. The method according to claim 6, characterised in that each railroad track comprises two lines of rails, and in that the step of vectorising said geolocated and classified measurement point cloud associated with the ground and with the railroad tracks into a plurality of geospatial vectors comprises the steps of: - identifying the points of said geolocated point cloud that are associated with a chosen rail line of the railroad track, - subsampling said identified points using a predefined subsampling factor, - grouping the subsampled points into a plurality of subsets of points, two successive subsets of points sharing at least one measurement point, - calculating the average position in three dimensions of each subset of points using a moving average, - creating a vector associated with the chosen rail line comprising the calculated average positions in three dimensions, - eliminating those components of the created vector which are likely to be affected by noise from an automated classification, which provides a geospatial vector, - determining the one or more vectors associated with the other rail line of the railroad track based on the geospatial vector determined for the selected rail, and - calculating the geospatial vector associated with the median axis of the railroad track.

8. The method according to claim 7, characterised in that each geospatial vector further comprises orientation, slope and cant measurements extracted from each subset of identified points.

9. The method according to claim 6, characterised in that the subdivision step uses a graph comprising a set of edges, and nodes connecting the edges, the edges representing railroad tracks associated with geospatial vectors and the nodes representing junctions or stopping points on the rail network.

10. The method according to one of the preceding claims, characterised in that it further comprises determining a network format corresponding to said railroad network by associating each modelled railroad track segment with one or more elements of the railroad network.

11. The method according to one of the preceding claims, characterised in that the step of defining adjacent objects and of associating between an adjacent object and one or more modelled segments comprises the steps of: - subsampling the measurement points associated with the adjacent environment by a predefined subsampling factor, - transforming the subsampled measurement points into one or more adjacent objects, - associating the adjacent objects with the modelled railroad track segments.

12. The method according to one of the preceding claims, characterised in that navigation signal propagation properties are assigned to each association between an adjacent object and one or more modelled segments as a function of a plurality of parameters comprising the height of the adjacent object and the distance separating the adjacent object from the median axis of the modelled segment.

13. A system for locating, in a railroad reference frame, a mobile vehicle traveling on railroad tracks of a railroad network, comprising a location device (101) configured for: - determining location parameters associated with the position of the mobile vehicle in a geographical reference frame based on a plurality of navigation signals of a GNSS or GPS system received by said mobile vehicle, - determining one or more railroad track segments on which the mobile vehicle is likely to be located by querying a cartographic database based on at least one of the location parameters, - locating the mobile vehicle based on the railroad track segments provided by the cartographic database and at least one of the location parameters, characterised in that said cartographic database comprising data representing a vectorial description of at least one railroad track of the railroad network and data representing a description of one or more objects of the environment adjacent to the railroad tracks of the railroad network likely to interfere with said navigation signals received by said mobile vehicle, and in that the location system (100) further includes a unit for generating cartographic databases (91) comprising: - a receiving module configured to receive a measurement point cloud classified and geolocated in a coordinate system, said measurement point cloud being associated with said railroad network and with said adjacent environment, - a topology-determining module configured to determine the topology of said railroad network based on geolocated and classified measurement points associated with said railroad network and a plurality of additional elements of said railroad network, - a modelling module configured to model the geometry of the railroad tracks of said railroad network as a plurality of modelled railroad track segments, where modelling the geometry of the median axis of a railroad track represented by a geospatial vector comprises: i. a determination of a current analytical model of the geometry of the median axis of the railroad track represented by a geospatial vector using at least one analytical equation, ii. performing a measurement the standard deviation between the median axis of the railroad track and the associated curve returned by the analytical model, iii. performing a division of the geospatial vector into two elements if the measured standard deviation is greater than a predefined error threshold, reiterating the steps of determining (i), measuring (ii) and dividing (iii) for as long as a standard deviation between the analytical model and the geospatial vector is greater than the predefined error threshold, - a module for determining adjacent objects, configured to determine, from geolocated and classified measurement points associated with said adjacent environment, one or more adjacent objects, each of said adjacent objects being associated with one or more modelled railroad track segments, - an identifying module configured to identify, for each association between an adjacent object and one or more modelled railroad track segments, information representing the interference that said navigation signals are subjected to.