RU2772796C2 - Methods for positioning points or lines of interest on the railway, positioning and control of the machine for intervention on the railway - Google Patents

Abstract

FIELD: railway transport.

SUBSTANCE: invention relates to means of maintenance and monitoring of the railway track. Method for determining the location of points or lines (A, B, C, D) of interest on a railway track (22) by means of a railway track positioning system (12) comprising a linear camera (26) and an odometer (28) moving along the railway track (22 ) in the advancing direction (100), the odometer (28) repeatedly collects instantaneous position data of the line camera (26) relative to the railway track (22) in the advancing direction (100), and the line camera (26) directed at the railway track (22) repeatedly collects instantaneous linear optical data along the line (50) of the instantaneous measurement. Then, by processing at least instantaneous linear optical data and instantaneous positioning data, a bitmap image of the area of ​​the surface of the railway track (22) is built, and at least one spatial indexing marker (56) of a predetermined characteristic and its position relative to the support rail are identified on this bitmap image. (22A) railway track (22). Points or lines (A, B, C, D) of interest are identified on the raster image, and the coordinates of these points or lines of interest are determined in the two-dimensional system O, x, y of the location reference associated with the spatial indexing marker (56) and the reference line (22A).

EFFECT: increased accuracy and lead time of determining the locations of points or lines of interest on the railway track.

16 cl, 5 dwg

Description

FIELD OF TECHNOLOGY

[0001] The present invention relates to the use of a railway vehicle for surveying a railway track or for working on a railway track, in particular for its construction, maintenance, maintenance, repair, renovation or dismantling. It also refers to the early determination of the locations of points of interest on a railway track, if necessary defining the boundaries of areas of interest for subsequent intervention on the track by an intervention device carried and maneuvered by a vehicle moving along the indicated track. It also relates to converting such a location for later use, in particular for later inspection of the track by means of a survey device or for subsequent intervention on the track by means of an intervention device.

BACKGROUND OF THE INVENTION

[0002] US 4,986,189 describes an intervention machine for the maintenance or repair of a railroad track that includes intervention devices designed to operate while the machine is advancing along the railroad track in the direction of progress of the work. In order to anticipate the presence of obstacles and allow automatic control of the positioning of these devices, a measuring rod is located on the front of the machine. This measuring rod is located horizontally and perpendicular to the direction of the track, and also contains aligned sensors to enable detection of the transverse position of the rail. Another sensor, consisting of a camera, monitors obstacles in the way. The measuring rod is additionally equipped with an odometer. The sensor signals are transmitted to the fixture control circuit with a delay that depends on the odometer signal and a predetermined distance between the measuring rod and the fixtures. In this way, it is possible to position the measuring rod at a distance from fixtures without the risk of interference. But this mode of operation assumes that the predetermined distance between the measuring rod and the fixtures is known with high accuracy. Since the required positioning accuracy of the fixtures is centimeter, in order to create a common reference frame, it is necessary to provide an extremely rigid common chassis for supporting the intervention fixtures and the measuring rod. This also assumes that the perpendicularity of the measuring rod with respect to the path is accurate. In addition, odometric errors associated, for example, with the slip or angle inscription of the stylus wheel relative to the neutral line of the rail, are cumulative. Finally, this system does not handle curved paths.

[0003] In "A High-Performance Inspection and Maintenance System of Track using Continuous Scan" by Masato Ukai and Nobuhiko Nagahara of the 11th World Congress on Railway Research, 2016 (“ A High-Performance Inspection and Maintenance System of Track using Continuous Scan Image ” by Masato Ukai and Nobuhiko Nagahara, 11th ^WCRR 2016), describes a system for early analysis of a railway track for later intervention by a maintenance vehicle. This analytical system is mounted on a specialized railway vehicle that can travel along the track at a maximum speed of 45 km/h. The analysis system uses a line camera located on the vehicle across the track, connected to the odometer in such a way as to synchronize the odometer pulses with line pictures taken by the line camera. This system provides the ability to build a continuous raster (two-dimensional) image of the railway track. Bitmap analysis, which is not performed in real time, allows the detection of objects having a predetermined characteristic, and in particular obstacles, to determine areas of the path where tampering is possible and "forbidden" areas where automatic intervention is not possible. To correct odometer measurement errors, especially in curves, it is proposed to regularly re-adjust the odometer signal by recognizing predetermined markers located on the path whose absolute position is known. The analysis system makes it possible to create an intervention program that can be executed later by a maintenance vehicle carried by a maintenance vehicle traveling on a railway track. But the subsequent use of measurements by the maintenance vehicle requires the latter to have its own means of positioning relative to the path. In addition, potential defects associated with the perpendicularity between the line camera and the path are not taken into account.

DISCLOSURE OF THE INVENTION

[0004] In particular, the present invention aims to overcome the shortcomings of the prior art and to provide means for providing accurate early location of points or lines of interest on a railway track, optionally delimiting areas of interest for subsequent intervention by a device carried and controlled by an intervention machine. moving along the specified path.

[0005] For this purpose, according to a first aspect of the present invention, there is provided a method for determining the location of a railway track, performed by a system for determining the location of a railway track, comprising at least one linear camera directed to the railway track and one or more odometers, and the determining system the location of the railway track is made with the possibility of moving along the railway track in the direction of advancement, and the method for determining the location includes the following actions:

- repeated collection using the specified odometer (odometers) of data on the progress of the system for determining the location of the railway track on the railway track in the direction of advancement,

- repeated collection of instantaneous linear optical data along the line of instantaneous measurement with a line camera aimed at the railway track,

- by processing at least instantaneous linear optical data and, if necessary, advance data, building a bitmap image of the area of the surface of the railway track,

- by processing the constructed bitmap, identifying in the constructed bitmap at least one spatial indexing marker of a predetermined characteristic, and by processing at least the advance data, determining the curvilinear abscissa of the spatial indexing marker and the position of said spatial indexing marker relative to the reference line of the railway track, and

- identification of points or lines of interest in the constructed raster image and determination in the two-dimensional reference frame of the location associated with the spatial indexing marker and with the specified reference line, the coordinates of the specified points or lines of interest.

[0006] The line camera has the advantage that it is highly insensitive to parasitic movements, in particular to vibrations resulting from the movement of the positioning system on the railway track, which can be further enhanced when the positioning system is attached to the machine for intervention on the railway track or coupled to it. In addition, compared with matrix technologies, the line image pixel formation time is shorter. Thus, the delay and positioning inaccuracy that can be caused by long exposures and too long shaping times are reduced.

[0007] Due to the fact that a spatial index marker and a reference line are taken as a local reference system for determining the coordinates of points or lines of interest, a data set is constructed that can subsequently be used by the transformation system itself, equipped with the ability to identify a spatial index marker and a reference line .

[0008] The line optical data currently collected corresponds to the line of the bitmap. The construction of a raster image based on linear optical data is simple, since there is no problem of overlap between successive images, which would be typical for data collection through a matrix camera. The curves are also "naturally" flattened in the bitmap built from the line camera's instantaneous line data, providing a simplified yet relevant view for the operator.

[0009] According to an embodiment, the acquisition of instantaneous linear optical data is triggered by receiving advance data.

[0010] If the spatial resolution of the odometer is high, it is possible to run a linear optical data acquisition line for each odometer pulse or even for all N pulses, where N is any non-zero integer that allows for a constant spatial step for successive acquisitions of linear optical data lines .

[0011] According to an embodiment, instantaneous linear optical data and advance data are collected in a synchronized manner.

[0012] In particular, if the spatial resolution of the odometer is lower than the resolution required to build a bitmap, one can, for example, observe the time interval T between two successive pulses I-1 and I of the odometer, divide this time interval by a predetermined non-zero integer N, and start the collection lines linear optical data with a constant time interval T/N between pulse I and pulse I+1 of the odometer. The number of lines of measurement is counted up to the pulse I+1, and from thisa posterioriwithdraw spatial step for collecting linear optical data in the direction of advance between pulses I and I+1. In practice, variations in the rate of advance of the system for position determination at the scale of observation are small, and the assumption of a constant speed between two successive odometer pulses leads to negligible distortion of the bitmap. In general, and regardless of the supported algorithm, it is possible to run a line camera, for example, by means of an electronic card for generating pulse sequences or a special software block.

[0013] According to an embodiment, the advance data and the instantaneous linear optical data are timestamped, with the bitmap preferably aligned with the timestamps. Thus, it is possible to determine the spatial step, potentially variable, between two consecutive lines of linear optical data, based on the knowledge of the spatial resolution of the odometer and the time intervals observed between two odometer pulses and between two measurement lines of the linear camera.

[0014] In accordance with an embodiment, the points or lines of interest constitute the boundaries of the area of interest, preferably represent the vertices of a quadrangle constituting the area of interest. Preferably, the positioning system is configured to identify in the bitmap the intersections between the track sleepers or track sleeper neutral lines on the one hand and the railway track rails or rail neutral lines on the other hand, these intersections constituting at least some of the points or lines of interest.

[0015] The coordinates of the points of interest may be Cartesian coordinates in a reference system that takes as origin a spatial index marker, as an X-axis a reference line, and as a Y-axis an axis perpendicular to this reference line. In this assumption, the coordinates contain the distance relative to the reference line, measured perpendicular to this reference line, and the distance relative to the spatial indexing marker, measured parallel to this reference line.

[0016] The reference line may be locally determined by at least two point data or point and steering vector data, or any other equivalent means. According to an embodiment, the reference line is the neutral line of one of the rails of the railroad track, or a line constructed from the neutral lines of the rails of the railroad track, such as a median line between two rails of a railroad. In practice, it is possible to determine the neutral line of a rail, for example, by locating the edges of that rail in a bitmap built from instantaneous linear data. It is also possible to resort to an orientation matrix camera, as will be described below.

[0017] Preferably, the railroad track positioning system orientation device repeatedly determines the railroad track positioning system's angular orientation data relative to the reference line, wherein the coordinates of the points or lines of interest are determined depending on said orientation data. Knowing the angle between the linear camera and the reference line in the projection plane of the raster image makes it possible to correct the coordinates of the points of interest and the spatial indexing marker. Of course, if there is a non-right angle between the line camera and the reference line, the distance seen by the line camera between the point of interest and the rail is greater than the actual distance measured perpendicular to the reference line. Naturally, the distance between two neighboring pixels of the linear camera is known and is set by the calibration parameters.

[0018] In accordance with an embodiment, the positioning system orientation device comprises at least one first probe for detecting the orientation of the positioning system relative to the first rail of the railway track constituting the first orientation rail, and preferably includes a second probe for detecting the orientation of the positioning system relative to the second rail of the railroad track constituting the second orientation rail.

[0019] In accordance with another embodiment, the orientation device of the position determination system comprises at least one first orientation matrix camera located facing the first of the rails of the railway track constituting the first orientation rail, and the orientation device takes pictures using the first orientation matrix camera and processes said images in such a way as to detect on them the orientation of the first orientation rail relative to the target of the first orientation matrix camera, and preferably contains the second orientation matrix camera located facing the second of the rails of the railway track constituting the second orientation rail, and the orientation device takes pictures using the second orientation matrix camera and processes said images so as to detect the orientation of the second orientation rail relative to the target of the second orientation matrix camera. Snapshots of the orientation line camera(s) may also allow the reference line to be determined. Various options are available for zooming in and matching data from the line camera and orientation camera(s). According to the first method, it is possible to calibrate the railway track location system in such a way as to characterize, by means of calibration parameters, the relative position of the line camera and the orientation device consisting of the orientation camera(s). Based on this calibration data, it becomes possible to transform, on the bitmap built on the basis of the instantaneous linear optical data, the guide line of the rail detected in the image of the orientation camera. This first method can be implemented without detecting the neutral line of the rail, or in particular the reference line, on a raster image built on the basis of instantaneous linear optical data. On the contrary, it is possible to use only the angular orientation data provided by the orientation device, and the reference line is calculated only on the basis of the raster image data built on the basis of the instantaneous linear data.

[0020] If necessary, corrections can be made through inertial measurements taken at the midpoint of the rod, or through optical velocity measurement technologies (which involve obtaining closely spaced matrix images using an orientation camera). These approaches can also be used to select an appropriate image (with low specific rotational dynamics) to determine the average angle of the rod inscription relative to the rail.

[0021] If necessary, on the same rail or on two different rails, you can combine the camera and the probe. Thus, to estimate the orientation of the line camera, it is possible to interpolate between the orientation data transmitted by the two orientation matrix cameras. The presence of at least one sensor (probe or matrix camera) above each rail of the railway track also makes it possible to compensate for the absence of a rail on the track section, in particular when the positioning system moves along a track device, for example, a switch. It is also possible to select a specific orientation of a rail in curves, such as a rail located inside that curve.

[0022] Preferably, the orientation data is timestamped.

[0023] In accordance with an embodiment, the positioning system distinguishes between sleeper-containing railroad track surface areas and track surface areas between sleepers, with the orientation device transmitting orientation data only once for each of these areas between sleepers. Orientation variations between two sleepers are small, so it's best to limit the amount of data to be processed.

[0024] Preferably, the orientation data is used to determine the reference line.

[0025] To enable human intervention at identified points of interest or areas of interest, a bitmap image may be rendered on the viewing screen of the positioning system. Preferably, points or lines of interest are visually identified on the viewing screen.

[0026] It may be possible to validate and/or invalidate at least some of the points or lines of interest or area of interest, or to qualify the area of interest as a tamper zone or forbidden zone after an input is made on the input human-machine interface.

[0027] In accordance with another aspect of the invention, the present invention relates to a positioning method performed by a measuring node, containing a positioning system and a conversion system, and the location system includes a line camera and an odometer, the conversion system includes one or more matrix conversion cameras located at a distance from the line camera of the positioning system in the direction of advance and behind it, and the method includes: a positioning step performed by the positioning system and implementing the positioning method described earlier; then the conversion step, performed by the conversion system, which includes the following steps:

- collection by means of the transformation matrix camera(s) of a set of one or more transformation matrix images in the spatial frame of reference of the transformation system;

- depending on the progress data collected by the odometer, identifying in the specified set of one or more transformation matrix images of the spatial indexing marker and determining the coordinates of the specified spatial indexing marker and the characteristics of the reference line data in the spatial reference system of the transformation system, and

- calculation of the transformed coordinates of points or lines of interest in the spatial reference frame of the transformation system depending on the coordinates of the spatial indexing marker and the characteristics of the reference line data in the spatial reference frame of the transformation system, and the coordinates of the points or lines of interest in the positioning reference frame.

[0028] In accordance with another aspect of the invention, the present invention relates to a method for controlling an intervention machine advancing along a railway track in the direction of advance and containing an intervention device mounted on an intervention chassis, by means of a measuring assembly comprising a location system and a conversion system, wherein the position determination system comprises at least one line camera and an odometer, the transformation system comprises at least one transformation matrix camera located at a distance from the location determination system line camera in the direction of advance and behind it, the transformation matrix camera is integral with the intervention chassis intervention device, and the method includes: a positioning procedure such as described previously, and an intervention procedure including positioning the intervention device depending on the coordinates of points or and lines of interest in the transformation reference frame.

[0029] The intervention tool can be of any type, such as a tamping tool or a driving tool.

[0030] The method in accordance with this aspect of the present invention may include various embodiments that reproduce the characteristics of all embodiments described in relation to the first aspect of the present invention or part thereof.

[0031] In accordance with another aspect of the invention, the present invention relates to a railway vehicle containing a positioning system containing a line camera and an odometer, and the positioning system is capable of performing one of the methods described previously, in one or other of its alternatives.

[0032] In accordance with another aspect of the invention, the present invention relates to a method for determining the location of a railway track performed by a system for determining the location of a railway track advancing along a railway track in the direction of advance, and the method of determining the location includes the following steps:

- repeated collection of instantaneous linear optical data along the line of instantaneous measurement using the line camera of the railway track positioning system directed to the railway track,

- repeated collection by means of the orientation device of the railway track positioning system of the orientation data of the railway track positioning system relative to the reference line of the railway track,

- by processing at least instantaneous linear optical data, the construction of a possibly distorted raster image of the area of the surface of the railway track,

- identification of points or lines of interest in the specified probably distorted bitmap, and

- determining the corrected coordinates of points or lines of interest depending on the probably distorted coordinates of points or lines of interest in the frame of reference of the probably distorted bitmap and orientation data.

[0033] The line camera has the advantage of being highly insensitive to vibrations resulting from the positioning system advancing on the railway track, which can be further enhanced when the positioning system is attached to the railway intervention vehicle. or linked to it.

[0034] The instantaneous linear optical data currently collected corresponds to the line of the bitmap. As the positioning system advances, the instantaneous measurement line moves relative to the railroad track.

[0035] Knowing the angle between the linear camera and the reference line in the projection plane of the bitmap allows correction of the coordinates of the points of interest and the spatial indexing marker. Of course, if there is an oblique angle between the line camera and the reference line, the distance seen by the line camera between the point of interest and the rail is greater than the actual distance measured perpendicular to the reference line.

[0036] In accordance with an embodiment, it is provided that, in order to acquire orientation data, the first probe of the orientation device detects the orientation of the positioning system relative to the first rail of the railway track constituting the first orientation rail, and preferably the second probe of the orientation device detects the orientation of the positioning system relative to the second rail of the railway track constituting the second orientation rail.

[0037] In accordance with another embodiment, it is provided that, in order to collect orientation data, the first orientation matrix camera of the orientation device, located facing the first of the rails of the railway track constituting the first orientation rail, takes pictures processed by the orientation device to detect the orientation of the first rail on them orientation with respect to the target of the first orientation matrix camera, and preferably a second orientation matrix camera located facing the second of the rails of the railway track constituting the second orientation rail takes pictures and processes said pictures processed by the orientation device to detect the orientation of the second orientation rail with respect to the target second orientation matrix camera.

[0038] If necessary, on the same rail or on two different rails, you can combine the camera and the probe. Thus, to estimate the orientation of the line camera, it is possible to interpolate between the orientation data transmitted by the two orientation matrix cameras. The presence of at least one sensor (probe or matrix camera) above each rail of the railway track also makes it possible to compensate for the absence of a rail on the track section, in particular when the positioning system moves along a track device, for example, a switch. It is also possible to select a specific orientation of a rail in curves, such as a rail located inside that curve.

[0039] Preferably, the orientation data is timestamped.

[0040] In accordance with an embodiment, the positioning system detects areas of a railroad track surface containing sleepers and areas between sleepers of the track surface, with the orientation device providing orientation data only once for each of the areas between the sleepers. Orientation variations between two sleepers are small, so it's best to limit the amount of data to be processed.

[0041] Preferably, the reference line is the neutral line of one of the rails of the railroad track, or a line constructed from the neutral lines of the rails of the railroad track. Preferably, the orientation data is used to construct a reference line based on the neutral lines of the railway track rails.

[0042] According to a particularly advantageous embodiment, one or more rail track positioning system odometers repeatedly collect progress data of the rail track positioning system on the railroad track in the direction of advance.

[0043] According to the first implementation of this embodiment, the acquisition of instantaneous linear optical data is triggered by receiving advance data.

[0044] If the spatial resolution of the odometer is high, it is possible to run a linear optical data acquisition line for each odometer pulse, or even for all N pulses, where N is any non-zero integer that allows for a constant spatial step for successive acquisitions of linear optical data lines .

[0045] According to the first implementation of this embodiment, instantaneous linear optical data and advance data are collected in a synchronized manner.

[0046] In particular, if the spatial resolution of the odometer is lower than the resolution required to build a bitmap, one can observe the time interval T between two consecutive pulses I-1 and I of the odometer, divide this time interval by a predetermined non-zero integer N, and start the collection lines linear optical data with a constant time interval T/N between pulse I and pulse I+1 of the odometer. The number of measurement lines is counted up to the I+1 pulse, and from this the posterior spatial step is derived for collecting linear optical data in the direction of advance between the I and I+1 pulses. In practice, variations in the rate of advance of the system for position determination at the scale of observation are small, and the assumption of a constant speed between two successive odometer pulses leads to negligible distortion of the bitmap.

[0047] According to an embodiment, advance data and instantaneous linear optical data are timestamped, with the bitmap preferably aligned with the timestamps. Thus, it is possible to determine the spatial step, potentially variable, between two consecutive lines of linear optical data based on the knowledge of the spatial resolution of the odometer and the time intervals observed between two odometer pulses and between two measurement lines of a linear camera.

[0048] In accordance with a particularly advantageous embodiment, the method of determining the position is that:

- by processing a possibly distorted bitmap on said image, at least one spatial indexing marker of a predetermined characteristic is identified,

- by processing the advance data and the orientation data, the curvilinear abscissa of the spatial indexing marker and the position of this spatial indexing marker relative to the reference line of the railway track are determined, and

- corrected coordinates of points or lines of interest are determined in a local two-dimensional reference system associated with a spatial indexing marker and with a reference line.

[0049] Due to the fact that a spatial index marker and a reference line are taken as a local reference system for determining the coordinates of points or lines of interest, a data set is constructed that can subsequently be used by the transformation system itself, equipped with the ability to identify a spatial index marker and a reference line .

[0050] In accordance with an embodiment, the points or lines of interest constitute the boundaries of the area of interest, preferably represent the vertices of a quadrangle constituting the area of interest. Preferably, the positioning system is configured to identify in the bitmap the intersections between the track sleepers or neutral lines of the track sleepers on the one hand and the rails of the railway track or neutral lines of the rails on the other hand, these intersections constituting at least some of the points or lines of interest.

[0051] The coordinates of the points of interest may be Cartesian coordinates in a reference system taking a spatial index marker as the origin, a reference line as the X-axis, and an axis perpendicular to the reference line as the Y-axis. In this assumption, the coordinates contain the distance relative to the reference line, measured perpendicular to this reference line, and the distance relative to the spatial indexing marker, measured parallel to this reference line.

[0052] To enable human intervention at identified points of interest or areas of interest, a bitmap image may be rendered on the viewing screen of the position determination system. Preferably, the viewing screen provides visual identification of points or lines of interest.

[0053] It may be possible to validate and/or invalidate at least some of the points or lines of interest or area of interest, or qualify the area of interest as a tampering area or forbidden area after an input is made on the input human-machine interface.

[0054] In accordance with another aspect of the invention, the present invention relates to a railway vehicle comprising a positioning system containing a line camera, at least one matrix camera and, preferably, an odometer, and the positioning system is configured to perform one of the methods, previously described, in accordance with one or the other of its alternatives.

[0055] In accordance with another aspect of the invention, the present invention relates to a railway vehicle containing a positioning system containing a line camera and an odometer, as well as an orientation device, which may include, in particular, a probe or a matrix camera. This vehicle may be autonomous or may be hitched to an intervention vehicle supporting an intervention device or a path survey device.

[0056] In accordance with another aspect of the invention, the present invention relates to a railway vehicle equipped with a positioning system located in the first part of the specified vehicle, and a conversion system located in the second part of the specified vehicle, located at a distance from the first part in the direction advancing the vehicle and behind it, and the location system contains a linear camera and an odometer, and possibly also an orientation device, which may contain, in particular, a probe or a matrix camera, and the conversion system contains at least one conversion matrix camera.

[0057] In accordance with a preferred embodiment, the railway vehicle is a machine for constructing or maintaining a railway track, which further comprises at least one device for intervention on the railway track, located in the third part of the vehicle, located at a distance from the second part in direction of advance and behind it.

[0058] The present invention is also directed to providing means for determining the position of a set of one or more devices carried by a railway vehicle based on previously collected data relating in particular to the position of points or lines of interest in a positioning frame of reference. .

[0059] For this purpose, in accordance with another aspect of the invention, a method is provided for controlling a set of one or more devices mounted on an intervention railway vehicle advancing along a railway track in the direction of advance, performed by a conversion system comprising a conversion chassis mounted on said a railway vehicle for intervention, and one or more transformation array cameras attached to the transformation chassis, the method comprising the following steps:

- receiving data characterizing the curvilinear abscissa of the spatial indexing marker of a known characteristic and the position of the spatial indexing marker relative to the reference line of the railway track, as well as the coordinates of points or lines of interest in a two-dimensional reference frame for determining the position associated with the spatial indexing marker and with the specified reference line,

- collection by means of the transformation matrix camera(s) of a set of one or more transformation matrix images in the spatial reference frame of the transformation system fixed relative to the transformation chassis;

- collection with the help of an odometer of data on the progress of the conversion system relative to the railway track,

- identifying a spatial indexing marker in a specified set of one or more transformation matrix images depending on advance data and curvilinear abscissa data,

- defining the characterization of the data of the spatial indexing marker and the reference line in the spatial reference frame of the conversion chassis,

- calculation of the transformed coordinates of points or lines of interest in the spatial reference frame of the transformation system depending on the characteristics of the data of the spatial indexing marker and the reference line in the spatial reference frame of the transformation system, and the coordinates of the points or lines of interest in the positioning reference frame.

[0060] Preferably, the conversion chassis is fixed relative to the main chassis of the railway vehicle and, if necessary, can only be integral with this main chassis. Preferably, the rail vehicle is provided with a plurality of undercarriages that run on the rail track while supporting the main chassis.

[0061] The coordinates of the points of interest may be Cartesian coordinates in a reference system taking a spatial index marker as the origin, a reference line as the X-axis, and an axis perpendicular to the reference line as the Y-axis. In this assumption, the coordinates contain the distance relative to the reference line, measured perpendicular to this reference line, and the distance relative to the spatial indexing marker, measured parallel to this reference line.

[0062] Preferably, the method further includes an intervention procedure including positioning said set of one or more fixtures depending on the coordinates of points or lines of interest in the transform reference frame and position data of said set of one or more fixtures relative to the transform chassis.

[0063] Said set of one or more devices may be of any type, such as a tamping device or a driving device.

[0064] According to an embodiment, said set of one or more fixtures is movable relative to the conversion chassis, the intervention procedure comprising collecting position data of said set of one or more fixtures relative to the conversion chassis by means of a position measuring device.

[0065] In accordance with an embodiment particularly suitable for an intervention tool, the intervention procedure comprises a command to move said set of one or more devices relative to the conversion chassis.

[0066] In accordance with an embodiment, the points or lines of interest constitute the boundaries or characteristics of the area of interest, preferably represent the vertices or sides of a quadrangle constituting the area of interest. Preferably, receiving coordinates of points or lines of interest in a two-dimensional positioning reporting system associated with a spatial index marker and with a reference line, includes receiving data for qualifying the area of interest as a zone of possible interference or a forbidden zone, and positioning the specified set of one or more devices are performed in a unique way if the area of interest is a zone of possible interference.

[0067] According to an embodiment, the reference line is a neutral line of one of the rails of the railroad track, or a line constructed from the neutral lines of the rails of the railroad track.

[0068] According to an embodiment, determining the characterization of the spatial index marker data and the reference line includes repeatedly determining by the orientation device of the data conversion system the angular orientation of the track location system relative to the reference line.

[0069] In particular, it can be provided that the positioning system orientation device comprises at least one first probe for detecting the orientation of the positioning system relative to the first rail of the railway track constituting the first orientation rail, and preferably includes a second probe for detecting the orientation of the system positioning relative to the second rail of the railroad track constituting the second orientation rail.

[0070] In addition, it can be provided that the orientation device of the conversion system comprises at least one first orientation matrix chamber located facing the first of the rails of the railroad track constituting the first orientation rail. Preferably, the first orientation matrix camera consists of a first conversion camera from among the conversion camera(s). The orientation device takes pictures using the first orientation matrix camera and processes said pictures in such a way as to detect the orientation of the first orientation rail with respect to the target of the first orientation matrix camera.

[0071] Preferably, the conversion system orientation device comprises a second orientation matrix chamber located facing a second of the rails of the railroad track constituting the second orientation rail. Preferably, the second orientation matrix camera consists of a second conversion camera from among the conversion camera(s). The orientation device takes pictures with the help of the second orientation matrix camera and processes these pictures in such a way as to detect on them the orientation of the second orientation rail with respect to the target of the second orientation matrix camera.

[0072] If necessary, on the same rail or on two different rails, you can combine the camera and the probe. Thus, to estimate the orientation of the conversion chassis, it is possible to interpolate between the orientation data transmitted by the two orientation matrix cameras. The presence of at least one sensor (probe or matrix camera) above each rail of the railway track also makes it possible to compensate for the absence of a rail on a track section, in particular when the positioning system is advanced along a track device, for example, a switch. It is also possible to select a specific rail orientation in curves, such as a rail located inside a curve.

[0073] In accordance with another aspect of the invention, the present invention relates to a railroad intervention vehicle, comprising a set of one or more railroad intervention devices, as well as a conversion system, comprising a conversion chassis supported by the railroad intervention vehicle, and one or more conversion matrix cameras attached to the conversion chassis.

[0074] Various aspects of the present invention, as well as various embodiments, can naturally be combined.

BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Other characteristics and advantages of the present invention will become apparent upon reading the following description with reference to the accompanying drawings, in which:

- figure 1 shows a schematic side view of a railway intervention vehicle equipped with a location system and a conversion system for implementing a method in accordance with an embodiment of the invention;

- figure 2 shows a schematic top view of some elements of the railway intervention machine according to figure 1;

- figure 3 shows a schematic view of the system for determining the location of the vehicle according to figure 1;

- Fig. 4 is a schematic view of the vehicle conversion system of Fig. 1;

- figure 5 shows a schematic view of the area of the railway track directly in accordance with the positioning system of figure 3.

[0076] For the sake of clarity, throughout the drawings, the same or like elements are designated by the same reference numerals.

DETAILED DESCRIPTION OF THE INVENTION

[0077] Figures 1 and 2 illustrate a railway intervention vehicle 1 consisting of a first location railway vehicle 2 , here also having a traction function, and a second towed intervention railway vehicle 3 . The railway positioning vehicle 2 comprises a main chassis 4 supported by a plurality of bogies 5 spaced apart in the longitudinal direction of the railway vehicle 2, while the railway intervention vehicle 3 is a semi-trailer whose main chassis 6 is at one end is pivotally attached to the fixing element 7 of the railway vehicle 2 for positioning and is supported at the opposite end by the undercarriage 8 '. Mounted on the main chassis 6 of the intervention railway vehicle 3 is a transport trolley 9 which moves along the track by means of the trolleys 10 and supports a set of one or more intervention devices 11 . A set of one or more actuators (not shown) enables the transport trolley 9 to move relative to the main chassis 6 parallel to the longitudinal direction of the main chassis 6 '. The railway intervention vehicle 1 is equipped with a positioning system 12 located in the railway vehicle 2 positioning and a conversion system 16 located on the transport trolley 9 of the railway intervention vehicle 3 at a distance from the positioning system in the direction 100 of advancing the intervention machine 1 and behind it, and a set of one or more devices 11 for intervention on the railway track 22 is located on the transport trolley 9 of the railway intervention machine 3 at a distance from the conversion system 16 in the advance direction 100 and behind it.

[0078] The location system 12 includes a line camera 26 and an odometer 28 connected to a processing unit 30 '. The line camera 26 may optionally be comprised of several sensor units aligned along the same measurement line. The line camera 26 is mounted on the location chassis 32 mounted under the main chassis 4 of the railway vehicle 2 location. The odometer 28 is mounted on one of the undercarriages 11 . one.

[0079] As shown in FIG . 3, the location system 12 further comprises a human-machine interface 34 located in the control cabin 36 of the intervention machine 1 . This human-machine interface 34 comprises a screen 38 and an input human-machine interface 40 , which may be embedded in the screen if said screen is a touch screen, or may consist of, for example, a keyboard or a control stick.

[0080] The conversion system 16 shown schematically in FIG . 4 comprises a control unit 42 coupled to at least one conversion matrix chamber 44A and to a controller 46 for controlling said set of one or more fixtures 11 . The conversion matrix 44A is here supported by the chassis 45 of the transport truck 9 of the intervention railway vehicle 3 . If necessary, the control units 42 , 30 of the conversion system 16 and the location system 12 can be combined into one unit. The set of one or more tools 11 may be of any type, in particular a tamping tool, a driving tool or a bolting tool.

[0081] By design, the rails of the railroad track 22 locally define a reference plane, horizontal or inclined, in accordance with the inclination of the track. Since the railroad positioning vehicle 2 is traveling on the railroad track 22 , the chassis 4 of the positioning device is considered to be parallel to this reference plane, which is an acceptable approximation for the needs of locating the railroad track 22 and controlling said set of one or more fixtures 11 . The photocells of the linear camera 26 are directed along the direction perpendicular to the reference plane.

[0082] On the reference plane as shown in FIG . 5, the orientation of the measurement line 50 α determined by the line camera 26 is measured in the path reference plane at an angle δ with respect to the perpendicular 200 to the reference line 122A, which in this first embodiment is the neutral line one of the rails 22a, 22b of the railway track 22 taken as the supporting rail 22A. This orientation is considered unknown. It may vary depending on the radius of curvature of the railway track 22 , the position of the line camera 26 relative to the bogies of the positioning railway vehicle 2 , and depending on the yaw of the positioning railway vehicle 2 relative to the railway track 22 .

[0083] With each pulse of the odometer 28 determine the curvilinear abscissa of the positioning system 12 along the reference line 122A .

[0084] At this point, the line camera 26 captures instantaneous linear optical data constituting a measurement line that spans the entire width of the track 22 , optionally including the width of the sleepers. This input is repeated and successive lines of measurement enable the construction of a two-dimensional bitmap having a step that is a function of the distance traveled between the two measurements.

[0085] In accordance with the first embodiment, the odometer 28 emits a pulse each time it travels a known base distance in the advancing direction 100 of the rail vehicle 2 positioning on the rail track 22 , these successive pulses being used to trigger the line camera 26. Then the spatial the spacing between two consecutive lines of a bitmap is known and constant.

[0086] According to another embodiment, the line camera 26 is fired at time intervals determined by dividing the observed time interval between two successive odometer 28 pulses. Thus, the time T elapsed between pulses I-1 and I of the odometer is observed, this time T is divided by a predetermined non-zero integer N, and the period T/N is used as the time interval between two firings of the line camera in the time interval separating pulse I and pulse I+1 of the odometer. A posteriori , a trigger is observed between pulses I and I+1 of the odometer, and from this the space step between two triggers of the line camera between pulses I and I+1 is derived. If it is assumed that the rate of advance varies little on this scale, it can be assumed that between two successive impulses this step will be constant. Other assumptions can also be made, considering, for example, that between two pulses the change in speed (acceleration or deceleration) is constant, and, as a result, there will be a linearly varying spatial step between the two pulses I and I+1.

[0087] According to another embodiment, the line camera 26 is fired at regular intervals and the visual image data is timestamped. The odometer pulses 28 are also time-stamped in such a way that it is possible to determine by interpolation the path traversed between two successive lines of the raster image, which may vary.

[0088] According to an alternative embodiment of this last embodiment, the line camera 26 is fired at intervals that are not necessarily regular, so as to condense the measurements in the area of interest where increased accuracy is useful, and place them at a greater distance in areas without interest, which provides the possibility of optimizing the amount of data.

[0089] The bitmap resulting from successive inputs of the line camera 26 is distorted due to the angle δ between the perpendicular to the reference line 122A and the line camera 26. If a Cartesian reference system is considered in the reference plane, in which the X-axis is parallel to the reference line 122A , and the Y axis is perpendicular to the X axis, it is observed that two points P1 , P2 of the railway track 22, simultaneously measured by the linear camera 26 at a distance from each other, appear to have the same abscissa on the distorted raster image, and in the Cartesian reference system defined earlier, actually have a different abscissa, and this difference is proportional to the distance between the two points and the sine of the specified angle δ. In addition, two points P1 , P3 of the railroad track 22, which have the same ordinate in the Cartesian reference system defined earlier, appear to be distant from each other in the distorted bitmap, their apparent distance depending on the cosine of the specified angle δ.

[0090] It is possible to "correct" the distorted image by evaluating this angle δ.

[0091] A first estimate of the angle δ can be obtained by comparing the distance measured on a given line camera measurement line between two known points of interest with the known distance between those points. Thus, on a given measurement line, it is possible to estimate the distance measured between two points, one of which is located in the center of the support rail 22A and the other is located in the center of the other rail 22B of the railway track 22. The relationship between the distance E between the rails and the measured distance D in absolute the value is equal to the cosine of the angle δ:

[0092] But, in addition to the fact that this estimate of the angle δ does not provide the ability to determine its sign, the accuracy of the estimate for small values of the angle δ is low, and the derivative of the cosine function has a value close to zero. It also assumes that the actual distance between the rails is constant and known to the required accuracy.

[0093] The second estimation method can be implemented based on matrix images by marking on the distorted bitmap a predetermined object present on the path, the contour or certain dimensions of which are known, and by comparing the visible contour or dimensions measured on the distorted bitmap with this known contour or actual dimensions. But in practice, the objects that could serve for this comparison are not located at sufficiently close intervals on the railway track. In other words, using this method may result in angle δ estimates that are too far apart.

[0094] To estimate the angle δ, it is preferable to resort to the use of the device 52 orientation attached to the chassis 32 location.

[0095] In accordance with an embodiment, the orientation device 52 includes a probe attached to the location chassis.

[0096] In accordance with another embodiment, the orientation device 52 comprises at least one first orientation matrix camera 54A attached to the location chassis 32 and having a viewing direction perpendicular to the reference plane of the railway track 22 facing one of the two rails 22A . 22B, for example, support rail 22A . The orientation bitmap transmitted by the orientation matrix camera enables the direction of the reference line 122A to be identified by image processing and its orientation in the orientation bitmap, thus giving direct access to the orientation of the positioning chassis 32 at an angle δ to the line camera 26 .

[0097] In order to limit the processing power required to process the orientation image of the first orientation array camera 54A , only angle δ measurements can be performed when necessary, in particular when it is estimated or determined that the angle δ should have changed.

[0098] In practice, it turns out that over the space between the sleepers, the observed changes in the angle δ are small. Thus, it is advantageously possible to use an odometer28 to launch the first matrix camera54Aorientation each time along the reference rail22Awas passed a predetermined distance, this distance being preferably equal to the space between the sleepers. You can also use image processing operations that are also performed on a distorted bitmap transmitted by a line camera.26, and launch the image using the first matrix camera54A orientation each time a new space between the sleepers is found in the distorted bitmap.

[0099] It is also possible to select to start the first orientation array camera 54A when, due to other measured data, it is determined that there is a possibility that the angle δ has been changed. For example, an accelerometer located on the location chassis 32 can be used for this purpose. It is also possible to use changes in the distance D mentioned above, measured between the rails 22A ', 22B' by means of the linear camera 26 '. Naturally, it is also possible to combine trigger modes, for example by combining systematic triggering by means of the odometer 28 and additional triggering depending on measured or monitored data.

[00100] With known amplitude and sine of the angle δ, it is possible to "correct" the distorted bitmap obtained from the successive snapshots of the line camera 26 and the measurements of the distance traveled obtained by the odometer 28 .

[00101] You can also limit this correction to a few points of interest identified in the distorted image in a given space between sleepers. In addition, this is a preferred solution to limit the required processing power, since the distortion observed in the distorted image does not prevent the location of points of interest in the uncorrected distorted image by processing this image.

[00102] In practice, and provided that the measurement of the angle δ has been triggered in advance, a spatial indexing marker of a predetermined feature is located on the distorted bitmap, for example, the center 56 of the bolt, located near and on the predetermined side of the support rail 22A , or any other predefined path component, such as a fastener or arrowhead. In particular, this determination can be made by comparing the distorted bitmap with various predetermined shapes of the predetermined path component, distorted according to the angle δ, or by means of an artificial neural network or, in particular, a pre-trained artificial intelligence unit, in particular deep learning, for example, through pixel-by-pixel image segmentation technology (SegNet type), or through object detection technology (RFCN type). After identifying this spatial index marker 56 in the distorted image, its apparent distance relative to the reference line 122A is read and the actual distance between the spatial index marker 56 and the reference line 122A measured perpendicular to the reference line 122A as a function of the angle δ is calculated. This then, using the spatial indexing marker 56 and the reference line 122A (assumed to be rectilinear in terms of the space between the sleepers), gives a local two-dimensional positioning reference frame for the space between the sleepers under consideration. To get an idea, this reference system may have as reference the projection O of the spatial indexing marker onto the reference line 122A perpendicular to the reference line 122A, as the X axis the reference line 122A oriented in the advancing direction 100 of the vehicle 2 , and as the axis Y is perpendicular to the X-axis, passing through the origin O of the reference (and through the spatial indexing marker 56 ).

[00103] Next, focus on the area of interest bounded by the transverse edges 58 ', 60 of two consecutive sleepers 62 ', 64 and the inner edges 66 ', 68 of two rails 22A ', 22B (the potential curvature of which can be neglected at this scale). On the way, this area forms a quadrilateral, which can be defined by the intersections A , B , C , D between the edges of the sleepers 58 , 60 and the inner edges 66 , 68 of the two rails 22A , 22B . In the distorted bitmap, the quad image itself is distorted, but is nevertheless detectable through its vertices. Then, to obtain the coordinates of the vertices A , B , C , D in the positioning reference system, it is sufficient to determine the coordinates of these vertices on the distorted image and apply the necessary correction depending on the angle δ.

[00104] In practice, the contrast of the images is not always sufficient to enable the edges of the sleepers and the rail to be determined in the distorted bitmap image in the immediate vicinity of the intersection points A , B , C , D. According to an alternative, therefore, it is preferable to define the area of interest as a quadrilateral bounded by intersections A' , B' , C' , D' between the neutral lines 162 , 164 built for each of the sleepers 62 , 64 and the neutral lines 122A , 122B built for each of the rails 22A , 22B . Neutral lines are built through image processing throughout the space between the sleepers.

[00105] Also focus on identifying potential obstacles in the quadrangular area of interest < A , B , C , D > or <A',B',C',D'>. The presence or absence of such obstacles makes it possible to qualify this area of interest as a zone of possible interference (allowed zone) or a zone of possible exclusion (forbidden zone). Regardless of the detection of obstacles, the dimensions of the area of interest can be used to qualify that area of interest as a zone of possible interference (if the space between the sleepers is sufficient to allow subsequent intervention) or as a forbidden zone.

[00106] The above operations (calculating the angle δ, locating the spatial index marker 56 and calculating its distance to the reference line 122A, locating the quadrilateral consisting of points of interest < A , B , C , D > or <A' , B' , C' , D' > and calculation of their coordinates in the O, x, y frame of the location reference given by the spatial indexing marker 56 and the reference line 122A, the location of possible obstacles) is performed cyclically for each space between the sleepers and initiated either by the odometer 28, or by recognizing the edge of the sleeper in the distorted bitmap. In practice, each cycle and each space between the sleepers is assigned a serial number.

[00107] These processing operations are performed using shape recognition algorithms. The human-machine interface 34 allows them to be invalidated if necessary (if the automatic recognition is assumed to be reliable by default) or validated (at least in the learning mode as long as the confidence level of the shape recognition is insufficient). To this end, the operator can view a distorted image, which is displayed on the control screen 38 of the human-machine interface 34 . A quad < A , B , C , D > or <A' , B' , C' , D' > is materialized on the screen with an overlay, e.g. by means of a colored quad, with identified probable obstacles being flagged, if necessary, according to predefined visual conventions (arrow, outline, etc.). It may be possible for the operator, when in validation mode, to click on the area of interest to confirm its status. On the contrary, it is possible to provide the possibility that when in maintenance mode, the operator clicks the pointer in the areas of interest to cancel its status. Naturally, a large number of alternatives can be provided for the interaction between the operator and the positioning system 12 in accordance with the required ergonomics and tasks.

[00108] At the end of the location procedure described earlier, for each space between sleepers in the local frame <O, x, y> coordinates of points A , B , C , D or A' , B' , C' , D' are available of interest, defining the boundaries of the zone [A, B, C, D] or [A', B', C', D'] of interest, qualified as allowed or prohibited. It is also possible to have other data available, such as the line between the sleepers.

[00109] This data is transmitted to the control unit 42 of the conversion system 16 so that it can benefit from it at the moment when, after the intervention railway vehicle 3 continuously moves forward, the conversion system 16 is located at a height of a predetermined space predetermined by the system 12 location. Due to the curvature of the railroad track 22 , the position of the transport trolley 9 , compared to the predetermined space between the sleepers, when the matrix camera 44A of the conversion system 16 is located above the specified space between the sleepers, is different from the position previously obtained by the chassis 4 of the positioning system 12 , when the line camera 26 was located above the same space between the sleepers.

[00110] The transformation system 16 is directed to allowing the transformation, i.e. changing the Cartesian reference system, of the coordinates determined by the positioning system 12 in the positioning reference frame <O, x, y> associated with a given space between sleepers, to coordinates, which can be used at the level of the transformation system 16 , in particular to control said set of one or more fixtures 11 .

[00111] For this purpose, the matrix camera 44A of the conversion system 16 is located facing the support rail 22A and has a sufficient field width to capture the image of both the support rail 22A and the spatial index marker 56, it should be recalled that it was selected near the reference rail 22A.

[00112] First of all, the transformation system 16 must be configured to determine which location data is currently inherent in the space between the sleepers viewed by the matrix camera 44A of the transformation system 16 .

[00113] However, knowledge of the geometry of the intervention machine 1 may not be sufficient even for a rough estimate of the distance between the odometer 28 of the location system 12 and the matrix camera 44A of the conversion system 16 , since the transport cart 9 is assumed to be movable relative to the main chassis 6 of the railway vehicle 3 means of intervention. Thus, an additional measurement is required, which can be performed by means of a position sensor of the transport cart 9 in relation to the main chassis 6 or by means of an additional odometer 70 integrated with the undercarriage 10 of the transport cart 2 '.

[00114] The combined measurements of the position sensor of the transport cart 9 and the odometer 28 of the location system 12 or, alternatively, the measurements of the odometer 70 of the conversion system 16 provide the ability to determine with a reasonable degree of certainty which location data is currently characteristic of the space between the sleepers viewed by the matrix camera 44A conversion systems 16 .

[00115] The conversion matrix 44A is connected to a conversion reference frame fixed relative to the transport cart 9 . By processing the bitmap of the transformation matrix camera 44 , the transformation unit 42 identifies the reference rail 22A , constructs the neutral line 122A that constitutes the reference line, and on this bitmap determines its orientation, which gives direct access to the orientation angle γ of the reference system O, x, y determining the location relative to the transformation reference frame. The transformation unit 42 also identifies the spatial indexing marker 56 and projects the spatial indexing marker 56 onto the reference rail 22A perpendicular to the latter, which defines the origin O of the position reference system <O, x, y> and its coordinates in the transformation reference frame. Then, the transform unit 42 can convert to the transform reference frame the coordinates of the points A , B , C , D of interest that the positioning system 12 transmitted to it in the positioning reference frame <O, x, y>.

[00116] Based on this, to control the specified set of one or more fixtures 11 , the controller 46 can be transferred to the transformed coordinates of the area of interest and its qualification (as a zone of possible interference). The controller 46 then generates commands that enable said set of one or more devices 11 to intervene or not interfere with the area of interest A , B , C , D , the boundaries of which are thus defined. If desired, there may be one or more degrees of freedom of movement of said set of one or more fixtures 20 relative to the conversion chassis 45 supporting the conversion camera(s) 44A , 44B . The command may comprise rotating said set of one or more fixtures 20 about an axis perpendicular to the reference plane, or moving said set of one or more fixtures 20 transversely to optimize its positioning relative to the space between sleepers. The command may also include raising or lowering a specified set of one or more fixtures 20 in accordance with the qualification of the area of interest as a zone of possible interference or as a prohibited zone.

[00117] Naturally, the examples shown in the drawings and discussed above are for illustrative and non-limiting purposes only. It is expressly contemplated that various illustrated embodiments of the invention may be combined to provide other embodiments of the invention.

[00118] The reference line selected during the position determination step is not necessarily the line on which the angle δ is estimated. If necessary, the neutral line 122B of the rail 22B can be selected as a reference line and the angle δ with respect to the rail 22A can be measured.

[00119] Alternatively, the reference line constructed by the location system 12 is virtual in the sense that it is not associated with a particular rail 122A, 122B. For example, it may be the middle line 222 of the railroad track, built on the basis of the neutral lines 122A and 122B of the rails 22A , 22B of the railroad track 22 . Thus, features such as interruptions of one of the rails at the level of the track device are eliminated.

[00120] This median line can be constructed by processing a distorted bitmap, by performing a point-by-point search in each measurement line of the line camera 26 for the middle of a segment bounded by the centers of the two rails 22A , 22B , or, preferably, by first constructing the neutral lines 122A , 122B each rail 22A , 22B , and then a line located at an average distance from the neutral lines 122A , 122B .

[00121] In this assumption of a virtual reference line built and used by the location system 12 , the transformation system 16 must also be configured to rebuild this virtual reference line. For this purpose, the conversion system preferably includes a second matrix conversion chamber 44B positioned above and facing the second rail 22B .

[00122] The second conversion matrix camera 44B is attached to the conversion chassis 45 of the transport truck 9 such that the relative position of the two conversion matrix cameras 44A , 44B is known and calibrated. Although each conversion array camera 44A ', 44B ' has only a reduced field of view width to allow viewing only of the rail 22A ', 22B over which it is located, it is possible to determine the position of the midline 222 between the neutral lines 122A' , 122B by determining the position of the first rail's neutral line 122A . 22A on the bitmap of the first conversion camera 44A , by determining the position of the neutral line 122B of the second rail 22B on the bitmap of the second conversion camera 44B , and by calculating the center of the straight line segment between the two neutral lines 122A , 122B based on these measurements and the distance calibration data between the two cameras. 44A , 44B conversions.

[00123] In accordance with an alternative embodiment of the positioning system 12 , it comprises a second orientation matrix camera 54B attached to the chassis 4 of the positioning vehicle 4 and having a viewing direction perpendicular to the reference plane of the railway track 22 facing the other rail 22B. The second orientation matrix camera 54B can be used to give a second value of the angle δ measured relative to the second rail 22B . Various algorithms can be implemented to use this data. For example, it is possible to assign a confidence index for each bitmap transmitted by one or another matrix camera of orientation, for example, depending on the contrast of the image, the presence or absence of a rail in the image, or any other criterion, in order to keep for each space between sleepers that measurement of the angle δ, which has the best confidence index. It is also possible to combine the measurements taken to calculate the "average" angle δ. Since the distance between the two matrix cameras 54A, 54B is known and calibrated, these cameras can also be used to determine the virtual reference line 222 mentioned above. Finally, two independent measurement and calculation paths can be provided, one of which uses the first orientation matrix camera 54A and the first transformation matrix camera 44A with respect to the first reference line 122A , and the other uses the second matrix camera 54B and the second transformation matrix 44B relative to the first reference line 122 V. Optionally, it is also possible to provide two odometers at the level of the surveillance system, one for each rail 22 A, 22 B, each dedicated to one of the two measurements and path calculations.

[00124] In the location step, the identification of points of interest is not limited to the identification of vertices A , B , C , D or A' , B' , C' , D' of the likely intervention quadrilateral. Other types of points of interest can be identified, such as the coordinates of the centers of the screw heads to be screwed or replaced. The interference zone is not necessarily a quadrilateral, but can be any polygon. It is also possible to try to identify - instead of points A , B , C , D of interest or in addition to them - lines of interest, for example, straight lines constituting the neutral lines 162 , 164 of sleepers 62 , 64 , each of which is located in the corresponding center of the sleeper 62 . 64, or a straight line 163 constituting a line passing between the sleepers, which is the axis of symmetry for the previous straight lines 162 , 164 , located at an average distance between two sleepers 62 , 64. In the local reference system <O, x, y>, the coordinates of such straight lines can be, for example, the coordinates of two points belonging to it or the coordinates of a point of this line and an angle.

[00125] Although the foregoing description has focused on the space between the sleepers, the positioning system can also analyze sections of the track containing sleepers, in particular, to find points of interest on them that require intervention.

[00126] The processing of the odometer 28 , line camera 26 and orientation device 52 by the positioning system 12 is performed in real time or with a slight delay in such a way that it can be used by the conversion system 16 belonging to the same constantly advancing machine 10 for intervention. The distance between the location system 12 located in the front area 14 of the vehicle 2 location in the forward direction 100 , on the one hand, and the conversion system 16 , located at a distance and towards the rear of this vehicle, and on the other hand, mainly are used in particular to enable the operator to validate or invalidate the points A , B , C , D of interest or their qualifications.

[00127] Alternatively, it is possible to advance the intervention machine 10 in the direction to determine the location, and then return it in order to perform transformation and control of the specified set of one or more devices during the second pass. During this conversion, the vehicle may move in a direction opposite to the direction of advance at the time of positioning, or in the same direction.

[00128] Discussed railway intervention machine 10 may consist of one or more vehicles articulated together. Thus, if necessary, the location system 12 , the conversion system 16 and said set of one or more fixtures 11 can be on the same vehicle. In accordance with another embodiment, the location system 12 may be mounted on a carriage moving along a track and pivotally connected to said front by means of a roller block carrying the conversion system 16 and said set of one or more fixtures 11 '.

[00129] It is also possible to provide that the positioning system 12 is mounted on an autonomous positioning vehicle, independent of the intervention vehicle 10 , carrying said set of one or more devices 11 . In this latter assumption, it is advantageous to provide the vehicle with an absolute positioning unit location, such as a GPS unit, in order to be able to assign to each space between sleepers coordinates in the absolute positioning frame that are accurate enough to distinguish between two consecutive spaces between sleepers.

[00130] If necessary, said set of one or more attachments 20 can be fixed relative to the main chassis 6 of the intervention vehicle 3 . The position of said set of one or more fixtures 20 can then be approximately estimated based on the odometer 28 data of the positioning system 12 alone, in order to determine which positioning data is currently relevant for conversion.

[00131] Although the present description is particularly focused on using the transform procedure to control an intervention tool, the transform can also be advantageously used for an accurate track survey operation by a survey tool carried by a chassis remote from the location chassis.

[00132] The human-machine interface 34 may also be remote from the vehicle 10 , such as located at a remote control station.

Claims (27)

1. The method of determining the location of the railway track (22), performed by the system (12) for determining the location of the railway track, containing at least one linear camera (26) directed to the railway track (22), and one or more odometers (28), moreover, the system (12) for determining the location of the railway track is configured to move along the railway track (22) in the direction (100) of advancement, and the method for determining the location includes the following actions: - repeated collection with the help of the specified odometer (odometers) (28) of the data of the advancement of the system (12) for determining the location of the railway track on the railway track (22) in the direction (100) of advancement, - repeated collection using a linear camera (26) aimed at the railway track (22), instantaneous linear optical data along the line (50) of instantaneous measurement, - by processing at least instantaneous linear optical data and, if necessary, advance data, the construction of a bitmap image of the area of the surface of the railway track (22), - by processing the constructed raster image, identifying on the constructed raster image at least one spatial indexing marker (56) of a predetermined characteristic and, by processing at least advance data, determining the curvilinear abscissa of the spatial indexing marker (56) and the position of said spatial indexing marker (56 ) relative to the reference line (122A, 122B, 222) of the railway track (22), and differs in that: - points or lines (A, B, C, D, A', B', C', D', 162, 163, 164) of interest are identified on the constructed raster image, and in a two-dimensional reference system for determining the location associated with the spatial indexing marker (56) and with a reference line (122A, 122B, 222), determine the coordinates of the indicated points or lines of interest. 2. The method of claim. 1, characterized in that the collection of instantaneous linear optical data is triggered by receiving advance data. 3. The method of claim. 1, characterized in that the instantaneous linear optical data and advance data are collected in a synchronized manner. 4. The method according to any one of the preceding claims, characterized in that the advance data and the instantaneous linear optical data are timestamped, and the bitmap is preferably aligned depending on said timestamps. 5. The method according to any of the preceding claims, characterized in that the points or lines (A, B, C, D, A', B', C', D', 162, 163, 164) of interest constitute the boundaries of the area of interest, preferably , represent the vertices of the quadrilateral that constitutes the area of interest. 6. The method according to any of the preceding claims, characterized in that the reference line (122A, 122B, 222) is the neutral line of one of the rails (22A, 22B) of the railway track (22) or a line built on the basis of the neutral lines of the rails (22A , 22B) railway track (22). 7. The method of determining the position according to any one of the preceding claims, characterized in that the orientation device (52) of the system (12) for determining the location of the railway track repeatedly determines the angular orientation data of this system (12) for determining the location of the railway track relative to the reference line (122A, 122B, 222), wherein the coordinates of points or lines (A, B, C, D, A', B', C', D', 162, 163, 164) of interest are determined depending on said orientation data. 8. The method according to claim 7, characterized in that the positioning system orientation device (52) comprises at least one first probe for detecting the orientation of the positioning system (12) relative to the first rail (22A) of the railway track constituting the first rail (22A ) orientation, and preferably contains a second probe for detecting the orientation of the positioning system (12) relative to the second rail (22B) of the railway track constituting the second orientation rail (22B). 9. The method according to any one of paragraphs. 7-8, characterized in that the orientation device (52) of the positioning system comprises at least one first orientation matrix camera (54A) located facing the first of the rails of the railway track constituting the first orientation rail (22A), and the device (52 ) orientation takes pictures using the first orientation matrix camera and processes these pictures in such a way as to detect on them the orientation of the first orientation rail (22A) relative to the target of the first orientation matrix camera (54A), and preferably contains a second orientation matrix camera (54B) located facing the second of the rails of the railway track, constituting the second orientation rail (22B), and the orientation device (52) takes pictures using the second orientation matrix camera and processes these pictures in such a way as to detect the orientation of the second orientation rail (22B) relative to the target of the second matrix camera (54V) orientation. 10. The method according to any one of paragraphs. 7-9, characterized in that the location system (12) is configured to distinguish between areas of the surface of the railway track (22) containing sleepers (62, 64) and areas between sleepers of the track surface (22), wherein the orientation device is configured to transmission of orientation data only once for each of the zones between the sleepers. 11. The method according to any one of paragraphs. 7-10, characterized in that the orientation data is used to determine the reference line. 12. The method of determining the location according to any of the previous paragraphs, characterized in that it includes the reproduction of a bitmap image on the viewing screen (38) of the system (12) positioning. 13. The method of determining the location according to p. 12, characterized in that it includes visual identification on the viewing screen (38) of points or lines (A, B, C, D, A', B', C', D', 162, 163, 164) of interest. 14. The method of determining the location according to any one of paragraphs. 12-13 in combination with claim 5, characterized in that it includes validating and/or invalidating at least some of the points or lines of interest or area of interest, or qualifying the area of interest as a possible tampering area or forbidden area after performing an input on a person -machine interface (40) input. 15. Positioning method performed by means of a measuring unit containing a positioning system (12) and a conversion system (16), wherein the positioning system (12) contains a linear camera (26) and an odometer (28), the conversion system (16) contains one or more conversion matrix cameras (44A, 44B) located at a distance from the linear camera (26) of the positioning system (12) in the direction (100) of advancement and behind it, and the method includes: the stage of positioning performed by the positioning system (12) locations and implementation of the method according to any of the preceding paragraphs; and then the conversion step performed by the conversion system (16) and includes the following steps: collecting with the transformation matrix camera(s) (44A, 44B) a set of one or more transformation matrix images in the spatial reference of the transformation system; depending on the progress data collected by the odometer (28), identifying in the specified set of one or more matrix images of the transformation of the spatial index marker (56) and determining the coordinates of the specified spatial index marker (56) and the characteristics of the reference line data (122A, 122B, 222) in the spatial reference frame of the transformation system, calculation of the transformed coordinates of points or lines (A, B, C, D, A', B', C', D', 162, 163, 164) of interest in the spatial reference frame of the transformation system depending on the coordinates of the spatial indexing marker (56) and characteristics of the reference line data (122A, 122B, 222) in the spatial reference of the transformation system, and point or line coordinates (A, B, C, D, A', B', C', D', 162, 163, 164 ) of interest in the (0, x, y) position reference system. 16. Method for controlling an intervention machine (10) advancing along a railway track (22) in the advance direction (100) and containing an intervention device (20) mounted on an intervention chassis (48) by means of a measuring unit containing a system (12) location and a conversion system (16), wherein the location system (12) contains at least one linear camera (26) and an odometer (28), the conversion system (16) contains at least one conversion matrix camera (44A, 44B) , located at a distance from the linear camera (26) of the positioning system (12) in the direction (100) of advance and behind it, the matrix camera (44A, 44B) of the transformation is made in one piece with the chassis (48) of the intervention device (20) for intervention and the method includes: - the positioning procedure according to clause 15; - intervention procedure, including positioning of the device (20) for intervention depending on the coordinates of points or lines (A, B, C, D, A', B', C', D', 162, 163, 164) of interest in the frame of reference transformations.

Images