EP3415400A1

EP3415400A1 - System and method for determining the position of a guided vehicle

Info

Publication number: EP3415400A1
Application number: EP17290075.5A
Authority: EP
Inventors: Pierre-Olivier GUISLAIN; André Marco; Christina Zaldivar Aparicio
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 2017-06-12
Filing date: 2017-06-12
Publication date: 2018-12-19

Abstract

The present invention concerns a method and a system for determining a position of a guided vehicle (1) within a railway network, said system comprising: - a camera (311, 312) installed on-board the guided vehicle (1) for acquiring an image of a portion (21, 22) of track; - a trackside panel (34, 35) installed along said portion (21, 22) of track and comprising an identifier (341, 351) and at least one fixed size feature; - an identifier database (33) storing for each identifier (341, 351) a set of data comprising at least the position of the trackside panel (34, 35); - a processing unit (32) configured for detecting if the image acquired by the camera (311, 312) comprises a trackside panel (34, 35), identifying the identifier (341, 351) of any detected trackside panel (34, 35), and retrieving in the identifier database (33) the set of data corresponding to said identifier (341, 351), the processing unit being further configured for calculating the position of the guided vehicle (1) from the position of the trackside panel (34, 35) and a calculation of the distance separating the camera (311, 312) from the trackside panel (34, 35).

Description

The present invention concerns a system and a method for safely and efficiently determining the position of a guided vehicle on a railway network.
The present invention is essentially related to the localization of a guided vehicle within a railway network. The expression "guided vehicle" refers to public transport means such as subways, trains or train subunits, etc., as well as load transporting means such as, for example, freight trains, for which safety is a very important factor and which are guided along a route or railway by at least one rail, in particular by two rails.
Identifying the precise position of a guided vehicle on a railway network is important for ensuring the safety of guided vehicle movements. Different systems are known for determining the position of a guided vehicle on a railway network. They are for instance the use of reference points with known position located on the track and read by the guided vehicle, the latter comprising usually sensors for measuring the speed and/or the displacement and/or the acceleration of the guided vehicle. From the knowledge of the position provided by the reference point and information provided by the sensors, it is then possible to determine the current position of the guided vehicle. Other systems involve Global Positioning Systems (GPS) cooperating with sensors (for measuring the speed and/or the displacement and/or the acceleration of the guided vehicle), said sensors enabling to determine the position of the guided vehicle in areas where a GPS signal cannot be received.
An objective of the present invention is to propose a new system and method for safely and efficiently determining the position of a guided vehicle within a railway network.
For achieving said objective, the present invention proposes notably a localization system and method as disclosed by the objects of independent claims. Other advantages of the invention are presented in the dependent claims.
The localization system according to the invention is configured for determining the position of a guided vehicle within a railway network. It comprises in particular on-board components, ground components, and it is configured for automatically determining the position of a guided vehicle, i.e. in a manner free of any user intervention. The localization system according to the invention comprises:

a camera system comprising at least one camera for acquiring an image of a portion of track of the railway network on which the guided vehicle is located, said camera being notably characterized by a magnification M. The camera according to the invention is an optical device comprising an optical system with at least one lens, said optical system making rays of light from the scene or object to be imaged converging onto an electronic sensor wherein the image of the scene or object is formed, the electronic sensor delivering then an output signal used for recording or processing the captured images. Said camera system is typically an on-board component of the guided vehicle;
a trackside panel configured for being installed along said portion of track, said trackside panel comprising an identifier and at least one feature with a fixed size, i.e. whose size is predefined and known, hereafter called "fixed size feature". The trackside panels are typically a ground components installed at different locations along tracks of the railway network. According to the present invention, the identifier is in particular configured for uniquely identifying the trackside panel which comprises it, i.e. the identifier of a trackside panel is different from an identifier of any other trackside panel of the railway network, in other words, each identifier is preferentially unique;
an identifier database storing for each identifier, and therefore for each trackside panel installed along a portion of track, a set of data comprising at least the position of the trackside panel comprising the identifier. The identifier database might be an on-board component or a ground component located at a remote location compared to the guided vehicle and communicating with at least one of the on-board component of the localization system;
a processing unit connected to the camera system and identifier database, and configured for detecting if the image acquired by the camera comprises a trackside panel, identifying the identifier of the trackside panel if any, and retrieving the set of data corresponding to said identifier in the identifier database, the processing unit being further configured for calculating the position of the guided vehicle from the position of the trackside panel extracted from the set of data and a calculation of the distance separating the camera of the camera system from the trackside panel comprising the identified identifier. The processing unit is typically an on-board component configured for communicating and exchanging data with the identifier database located either on-board or remotely compared to the guided vehicle. The processing unit may then communicate the position of the guided vehicle to a control system of the guided vehicle, and/or to a remote control center, and/or show said position on a map of the railway network.

For instance, the position of the trackside panel is defined as the distance separating a starting point on the railway network (for instance the place wherein the guided vehicle started moving) from the place where the trackside panel is located, said starting point serving as origin for the measurement of the distance (in meters) separating said starting point from the trackside panel. In this case, the position of the guided vehicle is also defined in function of said starting point. In particular, if the camera is a front facing camera, then the position of the guided vehicle is calculated by the processing unit by subtracting from the position of the trackside panel the distance separating the camera of the camera system from the trackside panel. And at the opposite, if the camera is facing backward, then the position of the guided vehicle is calculated by the processing unit by adding to the position of the trackside panel the distance separating the camera of the camera system from the trackside panel.
The present invention also concerns a localization method for determining the position of a guided vehicle within a railway network, said method being configured for automatically determining said position, and comprises the following steps:

acquiring an image of a portion of track of the railway network by means of a camera of a camera system installed on-board the guided vehicle;
detecting, by means of a processing unit, if the acquired image comprises a trackside panel installed along said portion of track, wherein the trackside panel comprises an identifier and has at least one feature having a fixed size;
identifying, by means of said processing unit, the identifier and retrieving a set of data corresponding to the identified identifier in an identifier database, wherein the identifier database comprises, for the identifier of each trackside panel installed along a track of the railway network, a set of data comprising at least the position of the trackside panel comprising the identifier;
calculating, by means of the processing unit, the position of the guided vehicle from the position of the trackside panel and a calculation of the distance separating the camera from the trackside panel comprising the identified identifier.

In particular, said distance separating the camera from the trackside panel is calculated according to the following equation, with M being the magnification of the camera acquiring the image of the trackside panel: $M = \frac{{Image}_{Size}}{{Panel}_{Size}} = \frac{{Image}_{Distance}}{{Panel}_{Distance}}$
wherein Image_size is the size, on an electronic sensor of the camera, of the image, formed on said electronic sensor, of the fixed size feature of the trackside panel and might be therefore measured by the processing unit in images acquired by the camera, Panel_size is the size of said fixed size feature on the trackside panel and is therefore a known parameter, Image_Distance is the distance along the principal axis of the optical system of the camera separating the image plane on the electronic sensor from the optical center of an imaging lens of said optical system and is therefore a known parameter, Panel_Distance is the distance along the principal axis of the optical system of the camera separating the trackside panel from the optical center of the imaging lens of said optical system and is the single unknown of Eq. 1. The processing unit according to the invention is preferentially configured for using Eq. 1 for automatically determining the unknown parameter Panel_Distance from a measure of the size of the image of the fixed size feature on the electronic sensor of the camera, and from the knowledge of the known parameters Image_Distance and Panel_size. The electronic sensor of the camera is located at the position of the image plane of the optical system of the camera so that images of scenes/objects acquired by the camera are formed on said electronic sensor.
Further aspects of the present invention will be better understood through the following drawings, wherein like numerals are used for like and corresponding parts:

Figure 1: schematic representation of a localization system according to the invention.
Figure 2: illustration of the optical geometry used for the imaging process of the trackside panel.
Figure 3: examples of results obtained when calculating the unknown parameter Panel_Distance.

Figure 1 illustrates a preferred embodiment of the localization system according to the invention. Said Figure 1 shows a guided vehicle 1 moving on a portion of track 2 of a railway network. The localization system according to the invention is configured for determining the position of the guided vehicle 1 with respect to the railway network. It comprises notably a camera system 31, a processing unit 32, an identifier database 33 and one or several trackside panels 34, 35 installed along tracks 2 of the railway network.
The camera system 31 is configured for being installed on-board the guided vehicle 1 and for acquiring an image of a portion of track 2 of the railway network. The camera system 31 according to the invention refers to any system capable of acquiring an image or a sequence of consecutive images of one or several scenes. It comprises notably at least one camera or preferentially a plurality of cameras 311, 312, wherein each camera comprises in particular an optical system 3110 for converging light of the scene to be acquired (in the present case, a portion of track 2) on an electronic sensor 3111 (see Fig. 2), which is for instance a CCD image sensor. The camera 311, 312 of the camera system 31 is preferentially configured for pointing to a scene comprising a portion of the track 2 on which the guided vehicle 1 is moving or has to move. For instance, said camera might be a front facing camera pointing forward along the track (corresponding to ref. 311 in Fig. 1) or a rear facing camera, pointing backward along said track 2 (corresponding to ref. 312 in Fig. 1). In particular, the camera system 31 comprises a front facing camera 311 that might be installed on-board the guided vehicle 1, for instance at the front end of the guided vehicle 1, so as to acquire images of the portion 21 of track located in front of the guided vehicle and/or a rear facing camera 312 that might be installed on-board the guided vehicle 1, preferentially at the rear end of the guided vehicle 1, so as to be able to acquire a portion 22 of track located backward the guided vehicle 1. The images acquired by each camera of the camera system are then sent to the processing unit 32.
Said processing unit 32 is configured for analyzing and processing the image acquired by each of the cameras 311, 312 of the camera system 31. It comprises typically one or several processors and optionally a memory. The images acquired by the camera system 31 are preferentially streamed to the processing unit over a communication link using wired or wireless known techniques. The processing unit 32 is configured for detecting if the image acquired by the camera system 31 comprises or not a trackside panel 34 (i.e. an image of the trackside panel 34 installed along the portion of track 2), and for detecting and identifying an identifier 341 of the latter.
Indeed, each trackside panel 34, 35 according to the invention is configured for being installed along a track 2 of the railway network wherein the guided vehicle is moving or has to move and comprises an identifier 341, 351. An identification of the identifier 341, 351 by the processing unit 32 provides to the latter access to a set of data comprised in the identifier database 33, wherein said set of data comprises at least the position within the railway network of the trackside panel 34, 35 comprising said identifier, and optionally a direction. The trackside panel 34, 35 is typically a board installed on a support, e.g. a pole 4, located along the track 2 so has to be visible for the camera 311, 312 of the camera system 31 from an upstream or downstream position of the guided vehicle 1. The trackside panel 34, 35 may have two faces, each face comprising said identifier or a different identifier, one face being visible for at least one camera of the camera system (preferentially for a front-camera) and the other face being visible for at least another camera of the camera system (preferentially a rear-camera). According to this embodiment, the identifier of one of the faces of the trackside panel might be identified when the guided vehicle is approaching the trackside panel, for instance in images acquired by a front-camera of the camera system, and the identifier of the other face of the trackside panel might be identified when the guided vehicle is moving away from the trackside panel, for instance in images acquired by a rear-camera of the camera system. The trackside panel 34, 35 may further comprise a specific feature like its shape and/or its color and/or a sign configured for improving its detection and recognition by the processing unit 32 in images acquired by the camera system 31. In particular, said sign might be part of the identifier, for instance by using a QR code comprising recognition patterns.
The identifier 341, 351 is typically a mark on the trackside panel 34, 35, for instance a two-dimensional drawing shown on the trackside panel notably in a location suitable for being seen/imaged by the camera of the camera system. Preferentially, the identifier might be encoded in the form of a matrix barcode (two dimensional barcode also called QR code) represented or shown on the trackside panel. The identifier 341 of each trackside panel 34 is in particular unique, so that the detection and identification of the identifier 341 of the trackside panel 34 by the processing unit 32 enables a determination by the processing unit 32 of a single set of data corresponding to the trackside panel 34 comprising the detected and identified identifier 341. The identifier 341 according to the invention is in particular a key or comprises a key to an entry into the identifier database, in that its identification and recognition by the processing unit 32 enables a retrieval by the latter of said data set comprising information related to the trackside panel 34 comprising the (identified and recognized) identifier 341.
The identifier database 33 might be installed on-board the guided vehicle 1 or remotely. In any case, the processing unit 32 and the identifier database 33 are able to communicate with each other in order to retrieve information associated to identifiers detected and identified by the processing unit 32. Said identifier database 33 comprises sets of data for each trackside panel 34, 35 installed along tracks 2 of the railway network, wherein each set of data comprises at least the position of the trackside panel 34, 35 with respect to the railway network, and optionally a direction in relation with said trackside panel 34, 35. The identifier database 33 comprises thus stored locations and optionally directions for each of the trackside panels installed along tracks of the railway network, wherein retrieval of the information related to a detected trackside panel is realized through the identification of its identifier and the finding in the identifier database of the information associated to said identifier. Optionally and advantageously, said identifier database might be updated in real time in order to add to a set of data corresponding to an identifier (and thus to a trackside panel) other information or data relevant for the portion of track equipped by the trackside panel comprising said identifier in order to automatically inform a guided vehicle or an operator of the guided vehicle about some issues in relation with said portion of track wherein the trackside panel is installed, like a warning regarding the presence of objects on the rails or a failure of a device, or the presence of another guided vehicle.
Preferentially, the processing unit 32 is configured for using stable and robust known computer vision techniques for real-time detection and identification of the identifier. Said techniques are for instance based on convolutional neural networks and combined or not to boosting techniques like AdaBoost. In particular, the processing unit might be configured for detecting a specific feature of the trackside panel, like its color and/or its shape and/or a sign shown on the trackside panel. Each time a trackside panel 34, 35 comprising an identifier 341, 351 is detected by the processing unit 32, then the latter proceeds to the identification of the identifier shown on the trackside panel 34, 35, said identification providing a key for an entry in the identifier database and thus for retrieving the set of data corresponding to the detected and identified identifier and that is stored in the identifier database 33.
For instance, if the identifier is a code, like a two dimensional code, shown on the trackside panel, then the processing unit is configured for detecting said code on the trackside panel, reading said code and determining a key corresponding to said code, wherein said key enables the retrieval of the set of data corresponding to the identifier in the identifier database. The same techniques as those used for reading conventional (bar- or two-dimensional) codes might be used. For instance, if the identifier is a drawing, then the identifier database may comprise, for each set of data associated to an identifier, a normative representation of said identifier. In this case, the processing unit might be configured for comparing the detected identifier to normative representations of identifiers stored in said identifier database in order to find a normative representation matching the detected identifier (i.e. a normative representation corresponding to the detected identifier or identical to the detected identifier, for instance with same geometrical features), said normative representation being the key giving access to the information stored in the identifier database and related to the trackside panel comprising the detected identifier. The reading or identification of the identifier by the processing unit enables the latter to find the right set of data corresponding to the trackside panel comprising said identifier. In other words, the identifier according to the invention comprises or is therefore the key to an entry in the identifier database. Once the processing unit retrieves the set of data of the identifier database that corresponds to the detected identifier, then it proceeds to the determination of the position of the guided vehicle from the information stored in said set of data, notably the stored position of the trackside panel comprising the detected identifier.
Figure 2 illustrates the optical geometry used for the imaging process of the trackside panel 34 and will be used to explain the determination of the position of the guided vehicle 1 within the railway network. It shows a trackside panel 34 comprising an identifier 341, a camera 311 comprising an optical system 3110 with a lens 3112 for which the assumption of a thin lens is preferentially valid and an electronic sensor 3111, wherein Panel_Distance is the distance along the principal axis A between the trackside panel 34 and the optical center 3113 of the lens 3112, Panel_size is the size of the fixed size feature of the trackside panel 34 (for instance its width W), Image_Distance is the distance along the principal axis A between the optical center 3113 of the lens 3112 and the electronic sensor 3111 located in the image plane, and Image_size is the size of the fixed size feature in the image formed on the electronic sensor 3111 (for instance the width W' of the trackside panel 34 measured on the image of said trackside panel 34 on the electronic sensor 3111).
As previously explained, the position of the trackside panel 34 is extracted by the processing unit 32 from the set of data stored in the identifier database 33 for the identifier 341 of said trackside panel 34. Said position is the position of the trackside panel with respect to the railway network. In order to determine the position of the guided vehicle with respect to the railway network, the processing unit 32 still has to determine the position of the guided vehicle 1 with respect to the position of the trackside panel.
For this purpose, the processing unit 32 is configured for calculating the distance separating the trackside panel 34 from the camera 311 installed on board the guided vehicle 1. Then, since the position of the camera 311 within the guided vehicle 1 is fixed and known (for instance, said position being 1 meter from the front end of the guided vehicle), it is then possible to determine the position of the front end and/or rear end of the guided vehicle compared to the trackside panel 34. In order to simplify the explanations, a thin lens 3112 approximation is taken for the optical system 3110 of the camera when calculating the distance separating the trackside panel 34 from the camera 311.
As previously explained, all trackside panels according to the invention comprise at least one feature that has a fixed size or dimension, i.e. the so-called fixed size feature. The size or dimension of said feature might be stored in the identifier database within the set of data corresponding to the identifier shown on the trackside panel 34. Preferentially, all trackside panels 34 comprise an identical fixed size feature, i.e. a fixed size feature which is the same for all trackside panels 34 of the railway network, wherein the dimension or size of said same fixed size feature is in particular stored in a memory of the processing unit 32, so that it is ready for use for the processing unit 32 when calculating the Panel_Distance from Eq. 1. Said fixed size feature might be for instance the width and/or height of the trackside panel 34, or the width and/or height of the identifier 341, or might be a drawing represented on the trackside panel and that has a fixed size. The processing unit 32 is in particular configured for detecting and recognizing said fixed size feature in the images acquired by the camera 311. The size or dimension of the fixed size feature is used by the processing unit 32 as a fixed parameter and will be referred to hereafter as Panel_size when calculating the Panel_Distance by the processing unit 32. Optionally, the magnification M of the optical system of the camera system 311 is also known, and thus might be stored in a memory of the processing unit 32 and/or in the identifier database 33.
Then, from Eq. 1, we have: $\begin{matrix} {Panlel}_{Distance} & = \frac{{Panel}_{Size}}{{Image}_{Size}} \cdot {Image}_{Distance} \\ = (\frac{{Panel}_{Size}}{{Pixel}_{Size} \cdot {Number}_{Pixel}}) \cdot {Image}_{Distance} \end{matrix}$
wherein the size of the image of the fixed size feature on an electronic sensor 3111 of the camera 311, i.e. Image_size, has been replaced by the number of pixels, "Number_Pixel", representing said fixed size feature times the size of a pixel, "Pixel_size", of the electronic sensor 3111. According to the present invention, the processing unit 32 is configured for calculating the Panel_Distance from Eq. 2, wherein

the Panel_size is a known parameter, for instance the width W of the trackside panel 34, a typical value for the Panel_size being 1 meter;
Image_Distance is a known parameter specific to the design of the optical system of the camera 311, a typical value for Image_Distance being 0.1 meter;
Pixel_size is also a known parameter: for instance, if considering a 4 megapixel CCD camera with 2000 pixels per line, the typical value for Pixel_size is then 10-5 meter;
Number_Pixel is determined by the processing unit 32 when detecting and recognizing the fixed size feature in the image acquired by the camera 311 by determining how many pixels are used for representing said fixed size feature on the electronic sensor 3111.

Finally, for a front camera measuring distances to trackside panels located in the front of the guided vehicle, the position P of the front end of the guided vehicle is given by: $P = P' - {Panel}_{Distance}$
wherein P' is the position of the trackside panel 34 within the railway network and the position P of the front end of the guided vehicle is for instance the length of railway network from the starting point of the guided vehicle until its current position, which is in this case assimilated to the position of the camera under the assumption that the optical center of the optical system is located at the front end of the guided vehicle. For a rear camera installed at the rear end of the guided vehicle, facing backward (like camera 312 in Fig. 1) and configured for measuring a distance to a trackside panel 35 located downward compared to the direction of travel of the guided vehicle (for instance an already passed trackside panel 35), the position P will be given by $P = P' + {Panel}_{Distance} + L$
wherein P' is the position of the trackside panel 35 within the railway network and the position P of the front end of the guided vehicle is for instance the length of railway network from the starting point of the guided vehicle until its current position, which is in this case assimilated to the position of the rear camera according to the assumption that the optical center of its optical system is located at the rear end of the guided vehicle plus the length L of the guided vehicle, i.e. the distance separating its front end from its rear end. Of course, the position of the guided vehicle and trackside panel might be defined differently, like for instance by geographic coordinate. In this case, the processing unit may use a map of the railway network and may receive information regarding the travelling direction of the guided vehicle compared to direction at which the camera acquiring images is pointing in order to determine the position of the guided vehicle on the railway network map compared to the position of the trackside panel.
For optical systems not satisfying a thin lens approximation, the same principles as explained above might be applied mutatis mutandis for calculating the distance separating the trackside panel from the camera, and therefore the guided vehicle.
Figure 3 presents then a table of results obtained when calculating the Panel_Distance from Eq. 2 using the typical values previously given for Panel_size, Image_Distance and Pixel_size, wherein the accuracy for the measurement of the Panel_Distance is given by: $Accuracy [m] = \frac{{Pixel}_{Size} \cdot {Panel}^{2}_{Distance}}{{Panel}_{Size} \cdot {Image}_{Distance}}$
wherein 1 pixel accuracy has been considered.
Optionally, the trackside panel may comprise two sides, each side comprising an identifier, one identifier on one side designed for being imaged by a camera pointing forward and on identifier on the other side designed for being imaged by a camera pointing backward with respect to a forward direction of movement of a guided vehicle. The trackside panel may also comprise other information that might be encoded, for instance the distance to the next and/or previous trackside panel, and/or the identifier of the next and/or previous trackside panel, in order for instance to improve the efficiency and safety of the system by providing redundant information.
Optionally, the camera system design presented in Fig. 1, with one camera pointing forward at the front end of the guided vehicle and one camera pointing backward at the rear end of the guided vehicle might also be used for determining the length of the guided vehicle by determining the position of the rear end and front end of the guided vehicle, the length of the guided vehicle being obtained by the processing unit by calculating the difference between said positions. Other designs of the camera system 31 may enable a determination of the length of the guided vehicle. For instance, the front end camera and the rear end camera may both point forward for imaging the same trackside panel and determining the distances between each of said cameras and said trackside panel. Again, the processing unit is then configured for calculating the difference between the obtained distances for determining the length of the guided vehicle. To summarize, if the length of the guided vehicle is an unknown parameter to be determined by the localization system according to the invention, then the processing unit is configured for calculating said length from the determination of the distance between the camera located at the front end of the guided vehicle and the camera located at the rear end of the guided vehicle, wherein either the two cameras are configured for pointing to a same direction and thus to a same trackside panel, or the two cameras are configured for pointing to opposite direction (for instance one camera pointing forward and one camera pointing backward) each to a different trackside panel, and the length of the guided vehicle is obtained through the difference of the positions of the cameras, i.e. from the Panel_Distances measured by the processing unit for each of the cameras and the knowledge of the position within the railway network of each of the imaged trackside panel.
In conclusion, the present invention proposes a new way for determining the position of a guided vehicle within a railway network, wherein trackside panels comprising identifiers are installed along tracks of the railway system and a camera system is configured for imaging said trackside panels, a processing unit processing the acquired images in order to measure the size of a fixed size feature of the trackside panel in the acquired image and identify the identifier, wherein the identification of the identifier provides access to a set of data comprising the position of the trackside panel and optionally a direction of travel corresponding to the identifier, and the determination of said size provides a measure of the distance separating the guided vehicle from the trackside panel, allowing thus the precise determination of the guided vehicle within the railway network.

Claims

Localization system for determining a position of a guided vehicle (1) within a railway network comprising:
- a camera system (31) comprising at least one camera (311, 312) installed on-board the guided vehicle (1) for acquiring an image of a portion (21, 22) of track of the railway network;

- a trackside panel (34, 35) configured for being installed along said portion (21, 22) of track, said trackside panel (34, 35) comprising an identifier (341, 351) and at least one feature with a fixed size, hereafter called "fixed size feature";

- an identifier database (33) storing for each identifier (341, 351) a set of data comprising at least the position of the trackside panel (34, 35) comprising said identifier (341, 351);

- a processing unit (32) configured for detecting if the image acquired by the camera (311, 312) comprises a trackside panel (34, 35), identifying the identifier (341, 351) of any detected trackside panel (34, 35), and retrieving in the identifier database (33) the set of data corresponding to said identifier (341, 351) in order to extract the position of the trackside panel (34, 35), the processing unit being further configured for calculating the position of the guided vehicle (1) from the position of the trackside panel (34, 35) extracted from said set of data stored in the identifier database (33) and a calculation of the distance (hereafter Panel_Distance) separating the camera (311, 312) of the camera system (31) from the trackside panel (34, 35) detected in the image acquired by the camera (311, 312) and that comprises the identified identifier (341, 351).
Localization system according to claim 1, wherein the processing unit (32) is configured for calculating the Panel_Distance separating the camera (311, 312) from the detected trackside panel (34, 35) using the following equation: ${Panel}_{Distance} = (\frac{{Panel}_{Size}}{{Image}_{Size}}) \cdot {Image}_{Distance}$
wherein Image_size is the size of the image of the fixed size feature of the detected trackside panel (34, 35) on an electronic sensor (3111) of the camera (311, 312), Panel_size is the size of said fixed size feature of the trackside panel (34, 35), Image_Distance is the distance separating the electronic sensor (3111) from the optical center (3113) of an imaging lens (3112) of said optical system (3110).
Localization system according to claim 2, wherein the Image_size is determined by the processing unit (32) from a signal output of the electronic sensor (3111), wherein the processing unit (32) is configured for automatically determining, from said signal output, the number of pixels representing said fixed size feature in the image acquired by the camera (311) and multiplying said number by the size of a pixel of the electronic sensor (3111) in order to determine the Image_size.
Localization system according to claim 2 or 3, wherein the Panel_size is stored within the set of data corresponding to the identifier of the trackside panel in the identifier database or in a memory of the processing unit (32).
Localization system according to one of the claims 1 to 4, wherein the camera is a front facing camera and the position of the trackside panel is defined as a distance along a way followed by the guided vehicle on the railway network separating a starting point on the railway network from the place where the trackside panel is installed, the processing unit (32) being then configured to calculate the position of the guided vehicle (1) by subtracting from the position of the trackside panel the calculated Panel_Distance for said front facing camera.
Localization system according to one of the claims 1 to 5, wherein a camera of the camera system (31) is facing backward and the position of the trackside panel is defined as a distance along a way followed by the guided vehicle on the railway network separating a starting point on the railway network from the place where the trackside panel is installed, the processing unit (32) being then configured for calculating the position of the guided vehicle (1) by adding to the position of the trackside panel the calculated Panel_Distance for said camera facing backward.
Localization system according to one of the claims 1 to 6, wherein the identifier (341, 351) is a matrix barcode.
Localization system according to one of the claims 1 to 7, wherein the camera system (31) comprises a first camera installed at the front end of the guided vehicle and a second camera installed at the rear end of the guided vehicle, wherein the processing unit is configured for calculating a length of the guided vehicle from the calculated position for the first camera and second camera.
Localization method for determining a position of a guided vehicle (1) within a railway network, the method comprising the steps:
- acquiring an image of a portion (21, 22) of track (2) of the railway network by means of a camera (311, 312) of a camera system (31) installed on-board the guided vehicle (1);

- detecting if the acquired image comprises an image of a trackside panel (34, 35) installed along said portion (21, 22) of track, wherein the trackside panel (34, 35) comprises an identifier (341, 351) and a feature having a fixed size;

- identifying the identifier (341, 351) and retrieving a set of data corresponding to the identified identifier in an identifier database (33), wherein the identifier database comprises, for the identifier of each trackside panel installed along a track of the railway network, a set of data comprising at least the position of the trackside panel (34, 35) comprising the identifier (341, 351);

- calculating the position of the guided vehicle (1) from the position of the trackside panel (34, 35) and a calculation of the distance (hereafter Panel_Distance) separating the camera from the trackside panel (34, 35) comprising the identified identifier.
Localization method according to claim 9, wherein calculating the Panel_Distance comprises using the equation: ${Panel}_{Distance} = (\frac{{Panel}_{Size}}{{Image}_{Size}}) \cdot {Image}_{Distance}$
wherein Image_size is the size of the image of the fixed size feature of the detected trackside panel (34, 35) on an electronic sensor (3111) of the camera (311, 312), Panel_size is the size of said fixed size feature of the trackside panel (34, 35), Image_Distance is the distance separating the electronic sensor (3111) from the optical center (3113) of an imaging lens (3112) of said optical system (3110).
Localization method according to claim 10, wherein the Image_size is determined from a signal output of the electronic sensor (3111) by multiplying a number of pixels by the size of a pixel of the electronic sensor (3111), wherein said number of pixels is the number of pixels representing said fixed size feature in the image acquired by the camera (311) and is determined from the signal output.
Localization method according to one of the claims 9 to 11, wherein the camera is a front facing camera and the position of the trackside panel is defined as a distance along a way followed by the guided vehicle on the railway network separating a starting point on the railway network from the place where the trackside panel is installed, a processing unit (32) being configured to calculate the position of the guided vehicle (1) by subtracting from the position of the trackside panel the calculated Panel_Distance for said front facing camera.
Localization method according to one of the claims 9 to 12, wherein a camera of the camera system (31) is facing backward and the position of the trackside panel is defined as a distance along a way followed by the guided vehicle on the railway network separating a starting point on the railway network from the place where the trackside panel is installed, the position of the guided vehicle (1) being calculated by adding to the position of the trackside panel the calculated Panel_Distance for said camera facing backward.
Localization method according to one of the claims 9 to 13, wherein the identifier (341, 351) is a matrix barcode.
Localization method according to one of the claims 9 to 14, wherein the camera system (31) comprises a first camera installed at the front end of the guided vehicle and a second camera installed at the rear end of the guided vehicle, wherein the processing unit is configured for calculating a length of the guided vehicle from the calculated position for the first camera and second camera.