• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    To locate points or lines of interest (A, B, C, D, A ', B', C ', D', 162, 163, 164) of a railway track (22), using a railway tracking system (12) progressing on the railroad track (22), linear linear optical data (22) is acquired repetitively with linear linear camera (26) along an instantaneous measuring line (50), and, with an orientation device (52) of the railway tracking system (12), the orientation data of the railway tracking system (12) is acquired with respect to a reference line (22A) of the railway track (22). By processing at least one of the instantaneous optical linear data, a potentially distorted matrix image of an area of the surface of the railway track (22) is constructed, and points or lines of interest (A, B, C, D) are identified. , A ', B', C ', D', 162, 163, 164) in the potentially distorted matrix image, before determining rectified coordinates of the points or lines of interest (A, B, C, D, A ', B', C ', D', 162, 163, 164), as a function of potentially distorted coordinates of the points or lines of interest (A, B, C, D, A ', B', C ', D ', 162, 163, 164) in a potentially distorted matrix image repository and orientation data.
    公开号:FR3077553A1
    申请号:FR1850961
    申请日:2018-02-06
    公开日:2019-08-09
    发明作者:Milan Stupar；Youcef Chinoune
    申请人:Matisa Materiel Industriel SA；
    IPC主号:
    专利说明:

    METHOD FOR LOCATING POINTS OR LINES OF INTEREST ON A RAILWAY
    TECHNICAL FIELD OF THE INVENTION The invention relates to the use of a railway vehicle for monitoring a railway track or for working on a railway track, in particular for its construction, its upkeep, its maintenance, its repair , its renewal or its dismantling. It also relates to the early identification of points of interest on a railroad track, delimiting if necessary zones of interest for the subsequent intervention of an intervention tool on the track, carried and operated by a machine running on the way. It also relates to the transposition of such tracking for its subsequent use, in particular for a subsequent auscultation of the track using an auscultation device or for subsequent intervention on the track using a intervention tool.
    STATE OF THE PRIOR ART In the document US4986189 is described an intervention machine for the maintenance or the repair of a railway track, which includes intervention tools intended to work while the machine advances while rolling on the railway in a direction of progress of works. To anticipate the presence of obstacles and allow automated control of tool positioning, a measuring beam is placed at the front of the machine. This measurement beam is arranged horizontally and perpendicular to the direction of the track and has aligned sensors making it possible to detect the transverse positioning of the rail. Another sensor, consisting of a camera, monitors obstacles on the track. The measuring beam is also equipped with an odometer. The sensor signals are transmitted to a tool control circuit with a delay depending on the signal from the odometer and the predetermined distance between the measuring beam and the tools. It is thus possible to place the measuring beam at a distance from the tools, without the risk of interference with them. However, this operating mode presupposes that the predetermined distance between the measuring beam and the tools is known with precision. Insofar as the precision sought for the positioning of the tools is centimetric, it is necessary to provide an extremely rigid common frame to support the intervention tools and the measurement beam, in order to create a common reference frame. It also presupposes that the perpendicularity of the measurement beam with respect to the track is precise. In addition, odometric errors linked, for example, to slippage or the angular marking of the probe wheel in relation to the neutral line of the rail are cumulative. Finally, this system does not manage curved tracks.
    In the document "A High-Performance Inspection and Maintenance System ofTrack using Continuons Scan Image" by Masato Ukai and Nobuhiko Nagahara, llth WCRR 2016, is described a system for early analysis of a railway for a subsequent intervention using a maintenance vehicle. This analysis system is mounted on a dedicated rail vehicle, which can travel on the track at a maximum speed of 45 km / h. The analysis system uses a linear camera arranged on the vehicle across the track, coupled to an odometer, so as to synchronize pulses from the odometer with linear shots by the linear camera. The system allows the construction of a continuous (two-dimensional) raster image of the railway. The analysis of the matrix image, which is not carried out in real time, makes it possible to detect objects having a predetermined signature, and in particular obstacles, to determine areas of the route where intervention is possible and areas " prohibited ”or an automated intervention is not possible. To correct errors in the measurement of the odometer, especially in curves, it is proposed to regularly readjust the signal from the odometer by recognizing predetermined beacons arranged on the track and whose absolute position is known. The analysis system makes it possible to generate an intervention program which can be subsequently executed using a maintenance machine tool carried by a maintenance vehicle traveling on the railway. However, the subsequent use of the measurements by the maintenance vehicle requires that the latter have its own means of positioning relative to the track. In addition, any errors in the perpendicularity between the linear camera and the track are not taken into account.
    PRESENTATION OF THE INVENTION The invention aims in particular to remedy the drawbacks of the state of the art and to propose means allowing precise early identification of points or lines of interest on a railway track, delimiting where appropriate areas of interest for the subsequent intervention of a tool carried and driven by an intervention machine rolling on the track.
    To do this is proposed, according to a first aspect of the invention, a method of locating a railroad track, executed by a rail tracking system comprising at least one linear camera targeting the railroad track and one or more odometers , the railway tracking system progressing on the railway in a direction of progression, the tracking process comprising the following actions:
    linear repetition data is acquired with the odometer (s) from the railway tracking system on the railway in the direction of travel, linear optical data are repeatedly acquired with the linear camera aiming at the railway instantaneous along an instantaneous measurement line, by processing at least instantaneous linear optical data, and where appropriate progress data, a matrix image of an area of the surface of the railway line is constructed, by processing of the constructed matrix image, in the constructed matrix image, at least one spatial indexing tag of predetermined signature is identified, and by processing at least the progress data, a curvilinear abscissa of the spatial indexing tag is determined and positioning of the spatial indexing tag relative to a reference line of the railway, and we identify coordinates points or lines of interest in the constructed matrix image, and we determine, in a two-dimensional reference frame of reference linked to the spatial indexing beacon and the reference line, coordinates of the points or lines of interest.
    The linear camera offers the advantage of being very insensitive to parasitic movement, in particular the vibrations undergone by the tracking system progressing on the railway, which can be further accentuated if the tracking system is fixed or coupled to an intervention machine on the railway. In addition, the formation time of a pixel in the linear image is shorter compared to matrix technologies. This reduces latency and the uncertainty of positioning that would be introduced by prolonged exposure and too long a training time.
    By taking a spatial indexing tag and a reference line as local repository to define the coordinates of the points or lines of interest, a data set is constructed which can be used subsequently by a transposition system itself equipped to identify the spatial indexing tag and the reference line.
    The linear optical data acquired at a given time corresponds to a line of the matrix image. The construction of the matrix image from linear optical data is simple because there is no problem of overlap between successive images which would be inherent in an acquisition by a matrix camera. The curves are also "naturally" straightened in the raster image constructed from instantaneous linear data from the linear camera, which allows a simplified representation and nevertheless relevant to the operator.
    According to one embodiment, the acquisition of instantaneous linear optical data is triggered by the reception of progress data.
    If the spatial resolution of the odometer is high, it is possible to trigger a line for acquiring linear optical data at each pulse of the odometer, or even all the N pulses, N being any non-zero integer, which allows a constant spatial step for successive acquisitions of linear optical data lines.
    According to one embodiment, the instantaneous linear optical data and the progress data are acquired in a synchronized manner.
    In particular, if the spatial resolution of the odometer is less than the resolution sought for the construction of the matrix image, one can for example observe the time interval T between two successive pulses 1-1 and I of Pedometer, divide this time interval by a predetermined non-zero integer N, and trigger lines of acquisition of linear optical data at constant time interval T / N between pulse I and pulse 1 + 1 of Pedometer. We count the number of measurement lines up to the 1 + 1 pulse, and we deduct a posteriori the spatial acquisition step of the linear optical data in the direction of progression between the I and 1 + 1 pulses. In practice, the progression speed variations of the tracking system are small on the observation scale, and the hypothesis of a constant speed between two successive pedometer pulses leads to negligible deformation of the matrix image. More generally, and whatever the algorithm chosen, the linear camera can be triggered for example using an electronic pulse train generation card or a dedicated software block.
    According to one embodiment, the progress data and the instantaneous linear optical data are time stamped, the matrix image preferably being constructed as a function of the time stamps. It is thus possible to determine a spatial step, possibly variable, between two successive lines of linear optical data, from knowledge of the spatial resolution of the pedometer and of the time intervals observed between two pedometer pulses and between two measurement lines of the linear camera.
    According to one embodiment, the points or lines of interest constitute the boundaries of an area of interest, preferably the vertices of a quadrilateral constituting an area of interest. Preferably, the tracking system is able to identify, in the matrix image, intersections between railway sleepers or neutral lines of railway sleepers on the one hand and the rails of the railway track or neutral lines of rail d on the other hand, these intersections constituting at least some of the points or lines of interest.
    The coordinates of the points of interest can be Cartesian coordinates in a reference frame originating from the spatial indexing tag, as the abscissa axis the reference line and as the ordinate axis an axis perpendicular to the reference line. In this case, the coordinates include a distance from the reference line measured perpendicular to the reference line and a distance from the spatial indexing tag measured parallel to the reference line.
    The reference line can be determined locally by the data of at least two points, or by the data of a point and a directing vector, or by any other equivalent means. According to one embodiment, the reference line is a neutral line of one of the rails of the railway, or a line constructed from the neutral lines of the rails of the railway, for example a median line between the two rails of the railway. In practice, we can determine the neutral line of a rail for example by locating the edges of the rail on the matrix image constructed from instantaneous linear data. It is also possible to use a matrix orientation camera, as will be explained below.
    Preferably, it is repeatedly determined, with a device for orienting the rail tracking system, angular orientation data of the rail tracking system relative to the reference line, the coordinates of the points or lines of interest being determined based on orientation data. Knowing the angle between the linear camera and the reference line, in a plane of projection of the raster image, makes it possible to straighten the coordinates of the points of interest and of the spatial indexing beacon. Indeed, in the presence of a non-right angle between the linear camera and the reference line, the distance seen by the linear camera between a point of interest and the rail is greater than the real distance measured perpendicular to the reference line . Naturally, the distance between two adjacent pixels of the linear camera is known, and given by calibration parameters.
    According to one embodiment, the orientation device of the tracking system comprises at least a first probe for detecting an orientation of the tracking system relative to a first rail of the railroad track, constituting a first orientation rail , and preferably comprises a second probe for detecting an orientation of the tracking system with respect to a second rail of the railroad track, constituting a second orientation rail.
    According to another embodiment, the orientation device of the tracking system comprises at least a first matrix orientation camera disposed opposite a first of the rails of the railway, constituting a first orientation rail , the orientation device carrying out with the first matrix orientation camera shots and processing the shots so as to detect therein an orientation of the first orientation rail relative to a test pattern of the first matrix camera of orientation and preferably comprises a second matrix orientation camera disposed opposite a second of the rails of the railroad track, constituting a second orientation rail, the orientation device performing with the second matrix orientation camera of the holds and processing the shots so as to detect an orientation of the second orientation rail with respect to a target of the second raster orientation camera. The shots of the linear orientation camera (s) can also be used to determine the reference line. Different options are open to reconcile and reconcile the data from the line camera and the orientation camera (s). According to a first method, it is possible to perform a calibration of the rail tracking system, so as to characterize by calibration parameters the relative positioning of the linear camera and of the orientation device constituted by the orientation camera (s). From this calibration data, it becomes possible to transpose to the matrix image constructed from 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 more generally the reference line, in the matrix image constructed from instantaneous linear optical data. Conversely, it is possible to use only the angular orientation data supplied by the orientation device, the reference line being calculated from the only data of the matrix image constructed from instantaneous linear data.
    If necessary, corrections can be made using inertial measurements made at a midpoint of the beam or using optical speed measurement techniques (which involve taking close-up matrix images with the orientation camera). These approaches can also be used to select a suitable image (with low inherent rotation dynamics) to determine an average angle of inscription of the beam relative to the rail.
    We can optionally mix a camera and a probe, on the same rail or on two different rails. It is thus possible to interpolate between the orientation data delivered by the two matrix orientation cameras to estimate the orientation of the linear camera. The presence of at least one sensor (feeler or matrix camera) above each rail of the railway track also makes it possible to compensate for the absence of a rail on a section of the track, in particular when the tracking system progresses along a switchgear, for example a switch. You can also choose a specific orientation rail in the curves, for example the rail located inside the curve.
    Preferably, the orientation data is time-stamped.
    According to one embodiment, the tracking system differentiates areas of the surface of the railroad track comprising crosspieces and inter-transverse areas of the surface of the track, the orientation device only delivering orientation data '' once for each of the cross-beam zones. As the orientation variations are small between two crosspieces, it is advantageous to limit the mass of data to be processed.
    Preferably, the orientation data is used to determine the reference line.
    To allow human intervention on the points of interest or the areas of interest identified, it is possible to provide for the reproduction of the matrix image on a display screen of the tracking system. Preferably, visual identification is provided on the display screen of the points or lines of interest.
    One can provide for validation and / or invalidation of at least some of the points or lines of interest or of the area of interest or a qualification of the area of interest as possible intervention or prohibited area , following an input on a human-machine input interface.
    According to another aspect of the invention, it relates to a positioning method executed by a measurement assembly comprising a tracking system and a transposition system, the tracking system comprising a linear camera and an odometer , the transposition system comprising one or more matrix transposition cameras positioned at a distance and behind the linear camera of the tracking system in a direction of progression, the method comprising: a tracking phase executed by the tracking system and putting in implements the tracking process described above; then a transposition phase executed by the transposition system and comprising the following actions:
    we acquire with the matrix transposition camera (s) a set of one or more matrix transposition images in a spatial reference system of the transposition system;
    as a function of the progress data acquired by the odometer, the spatial indexing tag is identified in the set of one or more matrix transposition images and the coordinates of the spatial indexing tag and of the data are determined. characteristics of the reference line in the spatial reference system of the transposition system, and the transposed coordinates of the points or lines of interest in the spatial reference system of the transposition system are calculated, as a function of the coordinates of the spatial indexing tag and of the characteristic data of the reference line in the spatial reference system of the transposition system, and of the coordinates of the points or lines of interest in the reference reference system.
    According to another aspect of the invention, it relates to a method of driving an intervention machine progressing on a railway in a direction of progression and comprising an intervention tool mounted on a chassis intervention, using a measuring system comprising a tracking system and a transposition system, the tracking system comprising at least one linear camera and an odometer, the transposition system comprising at least one matrix camera of transposition positioned at a distance and behind the linear camera of the tracking system in the direction of progression, the matrix transposition camera being secured to the intervention frame of the intervention tool, the method comprising: a positioning procedure such as described above and an intervention procedure comprising the positioning of the intervention tool according to the coordinates points or lines of interest in the transposition repository.
    The intervention tool can be of any type, for example a tamping or tamping tool.
    The method according to this aspect of the invention may include various embodiments incorporating the characteristics of all or part of the embodiments set out in connection with the first aspect of the invention.
    According to another aspect of the invention, it relates to a railway vehicle comprising a tracking system comprising a linear camera and an odometer, the tracking system being capable of performing one of the methods described above, in one or other of their variants.
    According to another aspect of the invention, it relates to a method of locating a railroad track, executed by a rail locating system progressing on the railroad in a direction of progression, the locating method with the following actions:
    we acquire repetitively, with a linear camera of the railway tracking system targeting the railway, instantaneous linear optical data along an instantaneous measurement line, we acquire repetitively, with a device for orienting the system of railway tracking, orientation data of the railway tracking system with respect to a reference line of the railway, by processing at least instantaneous linear optical data, a potentially distorted matrix image of an area of the surface of the surface is constructed the railway, we identify points or lines of interest in the potentially distorted matrix image, and we determine rectified coordinates of the points or lines of interest, as a function of potentially distorted coordinates of the points or lines of interest in a repository of the potentially distorted matrix image and of the data of rientation.
    The linear camera offers the advantage of being very insensitive to the vibrations undergone by the tracking system progressing on the railway, which can be further accentuated if the tracking system is fixed or coupled to an intervention machine. on the railroad tracks.
    The instantaneous linear optical data acquired at a given instant correspond to a line of the matrix image. The instantaneous measurement line moves, relative to the railroad track, as the tracking system progresses.
    Knowledge of the angle between the linear camera and the reference line, in a projection plane of the matrix image, makes it possible to straighten the coordinates of the points of interest and of the spatial indexing beacon. Indeed, in the presence of a non-right angle between the linear camera and the reference line, the distance seen by the linear camera between a point of interest and the rail is greater than the real distance measured perpendicular to the reference line .
    According to one embodiment, it is expected that to acquire the orientation data, a first probe of the orientation device detects an orientation of the tracking system relative to a first rail of the railroad track, constituting a first rail orientation, and preferably a second feeler of the orientation device detects an orientation of the tracking system relative to a second rail of the railroad track, constituting a second orientation rail.
    According to another embodiment, it is expected that to acquire the orientation data, a first matrix orientation camera of the orientation device, disposed opposite a first of the rails of the railway constituting a first orientation rail, takes shots processed by the orientation device to detect an orientation of the first orientation rail relative to a test pattern of the first orientation matrix camera, and preferably a second matrix camera orientation arranged opposite a second of the rails of the railroad track, constituting a second orientation rail, takes shots and processes the shots, processed by the orientation device to detect an orientation of the second orientation rail relative to a target of the second matrix orientation camera.
    We can optionally mix a camera and a probe, on the same rail or on two different rails. It is thus possible to interpolate between the orientation data delivered by the two matrix orientation cameras to estimate the orientation of the linear camera. The presence of at least one sensor (feeler or matrix camera) above each rail of the railway track also makes it possible to compensate for the absence of a rail on a section of the track, in particular when the tracking system progresses along a switchgear, for example a switch. You can also choose a specific orientation rail in the curves, for example the rail located inside the curve.
    Preferably, the orientation data is time-stamped.
    According to one embodiment, the tracking system detects areas of the surface of the railway comprising sleepers and inter-sleeper areas of the surface of the track, the orientation device does not deliver the data orientation only once for each of the cross-beam areas. As the orientation variations are small between two crosspieces, it is advantageous to limit the mass of data to be processed.
    Preferably, the reference line is a neutral line of one of the rails of the railway, or a line constructed from the neutral lines of the rails of the railway. Preferably, the orientation data is used to construct the reference line from the neutral lines of the rails of the railway.
    According to a particularly advantageous embodiment, one acquires repeatedly, with one or more odometers of the rail tracking system, progress data of the rail tracking system on the railway in the direction of travel.
    According to a first implementation of this embodiment, the acquisition of instantaneous linear optical data is triggered by the reception of progress data.
    If the spatial resolution of the odometer is high, it is possible to trigger a line for acquiring linear optical data at each pulse of the odometer, or even all of the N pulses, N being any non-zero integer, which allows a constant spatial step for successive acquisitions of linear optical data lines.
    According to a second implementation of this embodiment, the instantaneous linear optical data and the progress data are acquired in a synchronized manner.
    In particular, if the spatial resolution of the odometer is less than the resolution sought for the construction of the matrix image, one can observe the time interval T between two successive pulses 1-1 and I of the odometer, divide this time interval by a predetermined non-zero integer N, and trigger lines of acquisition of linear optical data at constant time interval T / N between pulse I and pulse 1 + 1 of odometer. We count the number of measurement lines up to the 1 + 1 pulse, and we deduct a posteriori the spatial acquisition step of the linear optical data in the direction of progression between the I and 1 + 1 pulses. In practice, the progression speed variations of the tracking system are small on the observation scale, and the hypothesis of a constant speed between two successive pedometer pulses leads to negligible deformation of the matrix image.
    According to one embodiment, the progress data and the instantaneous linear optical data are time stamped, the matrix image preferably being constructed as a function of time stamps. It is thus possible to determine a spatial step, possibly variable, between two successive lines of linear optical data, from knowledge of the spatial resolution of the pedometer and of the time intervals observed between two pedometer pulses and between two measurement lines of the linear camera.
    According to a particularly advantageous embodiment, the tracking method is such that:
    by processing the potentially distorted matrix image we identify, in the potentially distorted matrix image, at least one predetermined signature spatial indexing tag, by processing the progress data and the orientation data, a curvilinear abscissa is determined of the spatial indexing beacon and a positioning of the spatial indexing beacon relative to the reference line of the railway, and the straightened coordinates of the points or lines of interest are determined in a local two-dimensional reference frame of reference linked to the spatial indexing tag and the reference line.
    By taking a spatial indexing tag and the reference line as a local reference frame to define the coordinates of the points or lines of interest, a data set is constructed which can be used subsequently by a transposition system itself equipped to identify the spatial indexing tag and the reference line.
    According to one embodiment, the points or lines of interest constitute the boundaries of an area of interest, preferably the vertices of a quadrilateral constituting an area of interest. Preferably, the tracking system is able to identify, in the matrix image, intersections between railway sleepers or neutral lines of railway sleepers on the one hand and the rails of the railway track or neutral lines of rail d on the other hand, these intersections constituting at least some of the points or lines of interest.
    The coordinates of the points of interest can be Cartesian coordinates in a reference frame originating from the spatial indexing tag, as the abscissa axis the reference line and as the ordinate axis an axis perpendicular to the reference line. In this case, the coordinates include a distance from the reference line measured perpendicular to the reference line and a distance from the spatial indexing tag measured parallel to the reference line.
    To allow human intervention on the points of interest or the areas of interest identified, it is possible to provide for the reproduction of the matrix image on a display screen of the tracking system. Preferably, visual identification is provided on the display screen of the points or lines of interest.
    One can provide for validation and / or invalidation of at least some of the points or lines of interest or the area of interest or a qualification of the area of interest as possible intervention or prohibited area , following an input on a human-machine input interface.
    According to another aspect of the invention, it relates to a railway vehicle comprising a tracking system comprising a linear camera, at least one matrix camera and, preferably, an odometer, the tracking system being suitable to execute one of the methods previously described, in one or other of their variants.
    According to another aspect of the invention, it relates to a railway vehicle comprising a tracking system comprising a linear camera and an odometer, as well as an orientation device, which may in particular include a probe or a matrix camera. This vehicle can be autonomous or can be coupled to an intervention vehicle supporting an intervention tool or a track monitoring center.
    According to another aspect of the invention, it relates to a railway vehicle, equipped with a tracking system positioned in a first part of the vehicle, and a transposition system positioned in a second part of the vehicle located at a distance and behind the first part in a direction of progression of the vehicle, the tracking system comprising a linear camera and an odometer, as well as possibly an orientation device, which may in particular include a feeler or a camera matrix, the transposition system comprising at least one matrix transposition camera.
    According to a preferred embodiment, the railway vehicle is a railway construction or maintenance machine, which further comprises at least one intervention tool on the railway, disposed in a third part of the vehicle located at distance and back from the second part in the direction of progression.
    The invention also aims to provide means for determining the positioning of a set of one or more tools carried by a railway intervention vehicle, from previously acquired data, in particular concerning the positioning of points or lines of interest in a benchmark.
    To do this, there is proposed, according to another aspect of the invention, a method for controlling a set of one or more tools mounted on a railway intervention vehicle progressing on a railway track in a direction of progression. , executed by a transposition system comprising a transposition chassis mounted on the railway intervention vehicle and one or more matrix transposition cameras fixed to the transposition chassis, the method comprising the following actions:
    reception of data characterizing a curvilinear abscissa of a spatial indexing beacon of known signature, and a positioning of the spatial indexing beacon relative to a reference line of the railway, and of coordinates of points or lines of interest in a two-dimensional reference frame of reference linked to the spatial indexing tag and to the reference line, acquisition with the matrix transposition camera or cameras of a set of one or more matrix transposition images in a spatial reference frame of the system transposition, fixed relative to the transposition frame;
    acquisition with an odometer of progress data of the transposition system with respect to the railway, identification of the spatial indexing tag in the set of one or more matrix transposition images as a function of the progress data and the data of curvilinear abscissa, determination of data characteristic of the spatial indexing tag and the reference line in the spatial reference frame of the transposition frame, calculation of transposed coordinates of the points or lines of interest in the spatial reference frame of the transposition system, as a function of the characteristic data of the spatial indexing beacon and of the reference line in the spatial reference system of the transposition system, and of the coordinates of the points or lines of interest in the location reference system.
    The transposition frame is preferably fixed relative to a main frame of the rail vehicle, and can if necessary form only one with this main frame. The railway vehicle is preferably provided with several rolling trains, which run on the railway track supporting the main chassis.
    The coordinates of the points of interest can be Cartesian coordinates in a reference frame originating from the spatial indexing tag, as the abscissa axis the reference line and as the ordinate axis an axis perpendicular to the reference line. In this case, the coordinates include a distance from the reference line measured perpendicular to the reference line and a distance from the spatial indexing tag measured parallel to the reference line.
    Preferably, the method further comprises an intervention procedure comprising the positioning of the assembly of one or more tools as a function of the coordinates of the points or lines of interest in the transposition reference frame and of positioning data. of one or more tools relative to the transposition frame.
    The set of one or more tools can be of any type, for example a tamping, tamping or bolting tool.
    According to one embodiment, the assembly of one or more tools is movable relative to the transposition frame, the intervention procedure comprising an acquisition of the positioning data of the assembly of one or more tools by relation to the transposition frame by a position measuring device.
    According to an embodiment particularly suitable for an intervention tool, the intervention procedure comprises a command to move the assembly of one or more tools relative to the transposition frame.
    According to one embodiment, the points or lines of interest constitute limits or characteristics of an area of interest, preferably vertices or sides of a quadrilateral constituting the area of interest. Preferably, the reception of coordinates of points or lines of interest in a two-dimensional reference frame of reference linked to the spatial indexing beacon and to the reference line, comprises the reception of data qualifying the area of interest as an area possible intervention or prohibited area, the positioning of the set of one or more tools being carried out only if the area of interest is a possible intervention area.
    According to one embodiment, the reference line is a neutral line of one of the rails of the railway, or a line constructed from the neutral lines of the rails of the railway.
    According to one embodiment, the determination of the characteristic data of the spatial indexing beacon and of the reference line comprises the determination, repeatedly, with an orientation device of the transposition system, of angular orientation of the railway tracking system with respect to the reference line.
    One can in particular provide that the orientation device of the tracking system comprises at least a first probe for detecting an orientation of the tracking system with respect to a first rail of the railway track, constituting a first orientation rail, and preferably comprises a second probe for detecting an orientation of the tracking system with respect to a second rail of the railway track, constituting a second orientation rail.
    One can also provide that the orientation device of the transposition system comprises at least a first matrix orientation camera disposed opposite a first of the rails of the railroad track, constituting a first orientation rail. The first orientation matrix camera is preferably constituted by a first transposition camera from among the transposition camera or cameras. The orientation device performs with the first matrix orientation camera shots and processing the shots so as to detect an orientation of the first orientation rail relative to a test pattern of the first matrix orientation camera .
    The orientation device of the transposition system preferably comprises a second matrix orientation camera arranged facing a second of the rails of the railway track, constituting a second orientation rail. The second orientation matrix camera is preferably constituted by a second transposition camera from among the transposition camera or cameras. The orientation device performs with the second matrix orientation camera shots and processing the shots so as to detect an orientation of the second orientation rail relative to a test pattern of the second matrix orientation camera .
    If necessary, it is possible to mix a camera and a probe, on the same rail or on two different rails. It is thus possible to interpolate between the orientation data delivered by the two matrix orientation cameras to estimate the orientation of the transposition frame. The presence of at least one sensor (feeler or matrix camera) above each rail of the railway track also makes it possible to compensate for the absence of a rail on a section of the track, in particular when the transposition system progresses along a switchgear, for example a switch. You can also choose a specific orientation rail in the curves, for example the rail located inside the curve.
    According to another aspect of the invention, it relates to a railway intervention vehicle comprising a set of one or more intervention tools on a railway track, as well as a transposition system comprising a transposition chassis supported by the railway intervention vehicle and one or more matrix transposition cameras fixed to the transposition chassis.
    The various aspects of the invention can naturally be combined, as well as the various embodiments.
    BRIEF DESCRIPTION OF THE FIGURES Other characteristics and advantages of the invention will emerge on reading the description which follows, with reference to the appended figures, which illustrate:
    Figure 1, a schematic side view of a railway intervention machine equipped with a tracking system and a transposition system for the implementation of a method according to an embodiment of the invention;
    Figure 2, a schematic top view of certain elements of the railway intervention machine of Figure 1;
    Figure 3, a schematic view of the tracking system of the vehicle of Figure 1;
    Figure 4, a schematic view of the transposition system of the vehicle of Figure 1;
    FIG. 5, a schematic view of a railroad area directly above the tracking system of FIG. 3.
    For the sake of clarity, identical or similar elements are identified by identical reference signs in all of the figures.
    DETAILED DESCRIPTION OF EMBODIMENTS In FIGS. 1 and 2 is illustrated a railway intervention machine 1 composed of a first tracking railway vehicle 2, here also having a traction function and of a second railway vehicle d 'towed intervention 3. The locating railway vehicle 2 is comprises a main chassis 4 supported by several undercarriages 5 spaced from one another in a longitudinal direction of the railway vehicle 2 while the railway intervention vehicle 3 is a semi trailer, the main chassis 6 of which is articulated at one end to an attachment 7 of the tracking railway vehicle 2 and supported at the opposite end by a running gear 8. On the main chassis 6 of the intervention railway vehicle 3 is mounted a shuttle 9 which runs on the track via undercarriages 10, and supports a set of one or more their intervention tools 11. A set of one or more actuators (not shown) makes it possible to move the shuttle 9 relative to the main chassis 6 parallel to a longitudinal direction of the main chassis 6. The railway intervention machine 1 is equipped a locating system 12 positioned in the locating railway vehicle 2 and a transposition system 16 positioned on the shuttle 9 of the railway intervention vehicle 3, at a distance and behind the locating system, in a direction of progression 100 of the intervention machine 1, the assembly of one or more intervention tools 11 on the railway 22 being disposed on the shuttle 9 of the intervention railway vehicle 3, at a distance and behind the system of transposition 16 in the direction of progression 100.
    The tracking system 12 comprises a linear camera 26 and an odometer 28 connected to a processing unit 30. The linear camera 26 can be made up, if necessary, of several sensor units, aligned along the same measurement line. The linear camera 26 is mounted on a tracking chassis 32 mounted under the main chassis 4 of the tracking railway vehicle 2. The odometer 28 is mounted on one of the running gear 11.1.
    As illustrated in FIG. 3, the tracking system 12 also comprises a man-machine interface 34 arranged in a control cabin 36 of the intervention machine 1. This man-machine interface 34 comprises a screen 38 and a man-machine input interface 40, which can be integrated into the screen if it is tactile, or constituted for example by a keyboard or a control lever.
    The transposition system 16, illustrated diagrammatically in FIG. 4, comprises a control unit 42, connected to at least one matrix transposition camera 44A and to a controller 46 for controlling the assembly of one or more tools 11. The matrix transposition camera 44A is here supported by the chassis 45 of the shuttle 9 of the intervention railway vehicle 3. If necessary, the control units 42, 30 of the transposition system 16 and of the tracking system 12 can be united in one unit. The set of one or more tools 11 can be of any type, in particular a tamping, tamping or bolting tool.
    By construction, the rails of the railroad track 22 locally define a reference plane, horizontal or inclined, depending on the inclination of the track. Insofar as the rail tracking vehicle 2 runs on the track 22, it is considered that the chassis 4 of the tracking device is parallel to this reference plane, which constitutes an acceptable approximation for the purposes of tracking the track. 22 and the control of the assembly of one or more tools 11. The photosensitive cells of the linear camera 26 are directed in a direction perpendicular to the reference plane.
    In the reference plane, and as illustrated in FIG. 5, the orientation of the measurement line 50 defined by the linear camera 26, is measured, in the reference plane of the channel, by an angle δ by with respect to the perpendicular 200 to a reference line 122A, which in this first embodiment is the neutral line of one of the rails 22A, 22B of the track 22 taken as the reference rail 22A. This orientation is considered unknown. It can vary as a function of the radius of curvature of the railroad track 22, of the positioning of the linear camera 26 with respect to the running gear of the tracking railway vehicle 2, and as a function of the yaw of the tracking railway vehicle 2 with respect to the track rail 22.
    At each pulse from the odometer 28, the curvilinear abscissa of the tracking system 12 is determined along the reference line 122A.
    At a given instant, the linear camera 26 captures instantaneous linear optical data constituting a measurement line, which covers the entire width of the railroad track 22, including where appropriate the width of the sleepers. This entry is repeated and successive measurement lines make it possible to construct a two-dimensional matrix image having a step which is a function of the distance traveled between two measurements.
    According to a first embodiment, the odometer 28 provides a pulse each time that a known elementary distance is crossed in the direction of progression 100 of the locating railway vehicle 2 on the track 22, and these successive pulses are used to trigger the linear camera 26. The spatial interval between two successive lines of the matrix image is then known and constant.
    According to another embodiment, the linear camera 26 is triggered at determined time intervals by dividing the time interval observed between two successive pulses of the odometer 28. This thus observes the time T elapsed between pulses 1-1 and I of the odometer, we divide this time T by a predetermined non-zero integer N, and we use, on the time interval separating pulse I and pulse 1 + 1 from the odometer, the period T / N as the time interval between two triggers of the linear camera. A posteriori, we observe the trigger between the pulses I and 1 + 1 of the odometer and we deduce the spatial step between two triggers of the linear camera between the pulses I and 1 + 1. If we consider that the speed of progression varies little on this scale, we can assume that the step is constant between two successive pulses. Other hypotheses are also possible, considering for example that the speed variation (acceleration or deceleration) is constant between two pulses, resulting in a linearly variable spatial step between two pulses I and 1 + 1.
    According to another embodiment, the linear camera 26 is triggered at regular intervals, and the visual image data are time-stamped. The pulses of the odometer 28 are also time-stamped, so that it is possible to determine by interpolation the path traveled between two successive lines of the matrix image, which may vary.
    According to a variant of this latter embodiment, the linear camera 26 is triggered at intervals which are not necessarily regular, this so as to densify the measurements in an area of interest where increased precision proves useful. , and to space them more in areas without interest, which allows an optimization of the volume of data.
    The matrix image resulting from successive captures of the linear camera 26 is distorted, due to the angle δ between the perpendicular to reference line 122A and the linear camera 26. If we consider, in the plane of reference, a Cartesian frame of reference whose abscissa axis is parallel to the reference line 122A and the ordinate axis perpendicular to the abscissa axis, it can be seen that two points PI, P2 of the railroad track 22 measured simultaneously by the linear camera 26 at a distance from each other, and which appear in the distorted matrix image as having the same abscissa, have in fact, in the Cartesian frame of reference defined above, different abscissas, the difference being proportional to the distance between the two points and at the sine of the angle δ. In addition, two points PI, P3 of the railway track 22 which have the same ordinate in the Cartesian frame of reference defined above, appear distant from each other in the distorted matrix image, their apparent distance being a function of the cosine of l 'angle δ.
    It is possible to "straighten" the distorted image by estimating the angle δ.
    A first estimate of the angle δ can be obtained by comparing, in a given measurement line of the linear camera, the distance measured between two known points of interest with a known spacing between these points. It is thus possible to evaluate in a given measurement line, the distance measured between two points located one at the center of the reference rail 22A and the other at the center of the other rail 22B of the railway track 22. The ratio between the spacing of the rails E and the measured distance D is equal in absolute value to the cosine of the angle δ:
    | · = | Cosd | But, apart from the fact that this estimate of the angle δ does not allow its sign to be determined, the precision of the estimate is low for low values of the angle δ, the derivative of the cosine function having a value close to zero. It also assumes that the actual spacing between the rails is constant and known with the desired precision.
    A second estimation method can be implemented from the raster images, by locating in the distorted raster image a predetermined object present on the track and whose outline or certain dimensions are known, and by comparing the outline apparent or the dimensions measured on the distorted matrix image with the known actual contour or dimensions. In practice, however, there are no objects that could serve as a comparison on a railway at sufficiently short intervals. In other words, using this method would lead to estimates of the angle δ too far apart from each other.
    To estimate the angle δ, it is therefore preferable to use an orientation device 52 linked to the tracking frame 32.
    According to one embodiment, the orientation device 52 comprises a probe fixed to the tracking frame.
    According to another embodiment, the orientation device 52 comprises at least one first matrix orientation camera 54A fixed to the locating frame 32 and having a direction of sight perpendicular to the reference plane of the railroad track 22, opposite one of the two rails 22A, 22B, for example the reference rail 22A. The orientation matrix image, delivered by the orientation matrix camera, makes it possible, by image processing, to identify the direction of the reference line 122A, and to determine its orientation in the orientation matrix image , which gives direct access to the orientation of the tracking frame 32, therefore to the angle δ of the linear camera 26.
    To limit the computing power necessary for processing the orientation image of the first matrix orientation camera 54A, it is possible to make the measurements of the angle δ only when necessary, especially when it is estimated or when it is determined that the angle δ is likely to have changed.
    In practice, it turns out that the variations observed in the angle δ are small over an inter-sleeper space. It is therefore advantageous to use pedometer 28 to trigger the first matrix orientation camera 54A each time a predetermined distance has been traveled on the reference rail 22A, this distance preferably being equal to the inter-sleeper space. It is also possible to use the image processing carried out moreover on the distorted matrix image delivered by the linear camera 26 and to trigger the shooting by the first matrix orientation camera 54A each time a new inter-sleeper space is detected on the distorted matrix image.
    One can also choose to trigger the first orientation matrix camera 54A when it is determined, using another measured data, that it is likely that the angle δ has been modified. One can for example use for this purpose an accelerometer positioned on the locating frame 32. One can also use the variations of the distance D mentioned above, measured between the rails 22A, 22B by the linear camera 26. It is naturally also possible to combine the triggering modes, for example by combining a systematic triggering by pedometer 28 and an additional triggering according to a data measured or monitored elsewhere.
    The amplitude and the sign of the angle δ being known, it is possible to "straighten" the distorted matrix image deduced from successive shots of the linear camera 26 and of the distance traveled measurements, delivered by the odometer 28.
    We can also limit this correction to a few points of interest identified on the distorted image on a given cross-space. It is, moreover, the preferred solution to limit the necessary computing power, insofar as the distortion observed on the distorted image does not prevent the identification of points of interest by image processing, on the distorted image not straightened.
    In practice, and assuming that a measurement of the angle δ has been triggered beforehand, a marker of spatial indexing with a predetermined signature is identified on the distorted matrix image, for example a center of a bolt 56 located near and on a predetermined side of the reference rail 22A, or any other predefined track component, for example a fastening element or a switching core. This identification can in particular be done by comparing the distorted matrix image to various predetermined shapes of the predefined channel component, distorted according to the angle δ, or using an artificial neural network or more generally an artificial intelligence unit. having received prior learning, in particular deep learning, for example by a pixel-by-pixel image segmentation technique (SegNet type), or by an object detection technique (RFCN type). Once this spatial indexing tag 56 has been identified, its apparent distance from the reference line 122A is read from the distorted image and the actual distance between the spatial indexing tag 56 and the reference line is calculated. 122A, measured perpendicular to the reference line 122A, as a function of the angle δ. We then have with the spatial indexing tag 56 and the reference line 122A (supposed to be rectilinear on the scale of an inter-sleeper space), a local two-dimensional reference frame for locating the cross-sleeper space considered. To fix the ideas, this referential can originate from the projection O of the spatial indexing tag on the reference line 122A, perpendicular to the reference line 122A, for the abscissa axis x the reference line 122A oriented in the direction progression 100 of vehicle 2 and for the ordinate axis the y axis perpendicular to the x axis, passing through the origin O (and by the spatial indexing tag 56).
    We are then interested in an area of interest delimited by the transverse edges 58, 60 of two successive crosspieces 62, 64 and the internal edges 66, 68 of the two rails 22A, 22B (whose possible curvature may be neglected on this scale). On the track, this zone constitutes a quadrilateral which can be defined by the intersections A, B, C, D between the edges of crosspieces 58, 60 and the internal edges 66, 68 of the two rails 22A, 22B. On the distorted matrix image, the image quadrilateral is itself distorted, but nevertheless identifiable by its vertices. It then suffices to identify the coordinates of the vertices in the distorted image, and to apply the necessary correction as a function of the angle δ to obtain the coordinates of the vertices A, B, C, D in the reference frame of reference.
    In practice, the contrast of the images is not always sufficient to make it possible to determine, on the distorted matrix image, the edges of the sleepers and of the rail in direct proximity to the points of intersection A, B, C, D According to a variant, it is therefore preferable to define the area of interest as a quadrilateral delimited by the intersections A ', B', C ', D', between neutral lines 162, 164 constructed for each of the crosspieces 62, 64 and neutral lines. 122A, 122B constructed for each of the rails 22A, 22B. Neutral lines are constructed by processing images across the entire cross-space.
    We are also interested in identifying possible obstacles in the quadrangular area of interest <A, B, C, D> or <A ', B', C ', D'>. The existence or absence of such obstacles makes it possible to qualify the area of interest as a potential intervention area (authorized area) or potential exclusion area (prohibited area). Regardless of the obstacle detection, the dimensions of the area of interest can be used to qualify the area of interest as a potential intervention area (if the inter-sleeper space is sufficient to allow subsequent intervention) or prohibited area.
    The above operations (calculation of the angle δ, identification of the spatial indexing tag 56 and calculation of its distance from the reference line 122A, identification of the quadrilateral of the points of interest <A, B, C, D> or <A ', B', C ', D'> and calculation of their coordinates in the reference frame O, x, y defined by the spatial indexing tag 56 and the reference line 122A, location possible obstacles) are performed cyclically for each inter-sleeper space, and initiated either by the odometer 28, or by the recognition of a cross-cut edge in the distorted matrix image. In practice, each cycle and each inter-transverse space is assigned a serial number.
    These treatments are performed by shape recognition algorithms. The man-machine interface 34 makes it possible, if necessary, to invalidate them (if it is considered that by default the automatic recognition is valid) or to validate them (at least in learning mode, as long as the level of confidence on the pattern recognition is insufficient). To this end, an operator can view the distorted image which scrolls on the control screen 38 of the man-machine interface 34. The quadrilateral <A, B, C, D> or <A ', B', C ', D'> is displayed on the screen as an overlay, for example by a colored quadrilateral, and any obstacles identified are, where appropriate, marked according to a predetermined visual convention (arrow, outline, etc.). We can predict that if it is in a validation mode, the operator clicks with a pointer in the area of interest to confirm its status. Conversely, we can predict that if it is in service mode, the operator clicks with a pointer in the areas of interest to deny its status. Naturally, a large number of variants can be provided for the interaction between the operator and the tracking system 12, depending on the ergonomics and the objectives sought.
    At the end of the locating procedure described above, there is for each inter-transverse space a locating, in a local frame of reference <0, x, y>, coordinates of points of interest A, B , C, D or A ', B', C ', D', delimiting an area of interest [A, B, C, D] or [A ', B', C ', D'], qualified as authorized or prohibited. We can also have other data such as an inter-sleeper line.
    These data are transmitted to the control unit 42 of the transposition system 16, so that it can take advantage of them when, following the continuous advance of the railway intervention vehicle 3, the system transposition 16 is at the height of a predetermined space previously identified by the tracking system 12. Due to the curvature of the track 22, the positioning of the shuttle 9 with respect to a given inter-sleeper space, when the 44A matrix camera of the transposition system 16 is located above said inter-transverse space, is different from the positioning previously taken by the chassis 4 of the tracking system 12, when the linear camera 26 was above the same inter-space sleepers.
    The transposition system 16 aims to allow a transposition, that is to say a change of Cartesian frame of reference, between the coordinates determined by the tracking system 12 in the tracking frame of reference <0, x, y> linked to a given inter-transverse space, in coordinates usable at the level of the transposition system 16, in particular for controlling the assembly of one or more tools 11.
    To do this, the matrix camera 44A of the transposition system 16 is arranged in front of the reference rail 22A, and has a sufficient width of field to capture both the reference rail 22A and the spatial indexing beacon 56, it being recalled that the latter was chosen close to the reference rail 22A.
    The transposition system 16 must first of all be able to determine what are the location data attributable to a given instant in an inter-sleeper space viewed by the matrix camera 44A of the transposition system 16.
    However, the knowledge of the geometry of the intervention machine 1 may prove to be insufficient to even roughly estimate the distance between the odometer 28 of the tracking system 12 and the matrix camera 44A of the transposition system 16, in the measurement where the shuttle 9 is assumed to be movable relative to the main chassis 6 of the intervention railway vehicle 3. An additional measurement is therefore used, which can be provided by a position sensor of the shuttle 9 relative to the main chassis 6 , or by an optional odometer 70 secured to an undercarriage 10 of the shuttle 2.
    The combined measurements of the shuttle 9 position sensor and the odometer 28 of the tracking system 12, or alternatively the measurements of the odometer 70 of the transposition system 16 make it possible to determine with a sufficient level of confidence, what are the tracking data attributable to a given instant in an inter-transverse space viewed by the matrix camera 44A of the transposition system 16.
    The matrix transposition camera 44A is linked to a fixed transposition reference system with respect to the shuttle 9. By processing the matrix image of the matrix transposition camera 44, the transposition unit 42 identifies the rail of reference 22A, constructs the neutral line 122A which constitutes the reference line, and determines its orientation in the matrix image, which gives direct access to an angle γ of orientation of the reference frame of reference O, x, y with respect to the frame of reference transposition. The transposition unit 42 also identifies the spatial indexing tag 56 and constructs the projection of the spatial indexing tag 56 on the reference rail 22A perpendicular to the latter, which defines the origin O of the reference frame of reference <0 , x, y> and its coordinates in the transposition repository. The transposition unit 42 can then transpose into the transposition repository the coordinates of the points of interest A, B, C, D that the tracking system 12 has transmitted to it in the tracking repository <0, x, y>.
    On this basis, it can transmit to the controller 46 for controlling the assembly of one or more tools 11 the transposed coordinates of the area of interest and its qualification (as a potential area of intervention). The controller 46 then generates commands which allow all of one or more tools 11 to intervene or not in the area of interest A, B, C, D thus delimited. Where appropriate, there may be one or more degrees of freedom of movement of the assembly of one or more tools 20 relative to the transposition frame 45 supporting the transposition camera or cameras 44A, 44B. The command may include a rotation of the assembly of one or more tools 20 about an axis perpendicular to the reference plane, or a translation of the assembly of one or more tools 20 in a transverse direction, to optimize the positioning of the assembly of one or more tools 20 relative to the inter-sleeper space. The order can also include lifting or diving the assembly of one or more tools 20 depending on the qualification of the area of interest, as a possible intervention area or as a prohibited area.
    Naturally, the examples shown in the figures and discussed above are given only by way of illustration and not by way of limitation. It is explicitly provided that the different embodiments illustrated can be combined with one another to propose others.
    The reference line chosen during the identification phase is not necessarily the line on which the estimate of the angle effectuée is carried out. If necessary, the neutral line 122B of the rail 22B can be chosen for the reference line, and the angle δ relative to the rail 22A can be measured.
    According to a variant, the reference line constructed by tracking system 12 is virtual, in the sense that it is not linked to a specific rail 122A, 122B. It may, for example, be a median line 222 of the railroad track constructed from neutral lines 122A and 122B of the rails 22A, 22B of the railroad track. This thus overcomes singularities, such as interruptions of one of the rails at the level of certain track switchgear.
    This center line can be constructed by processing the distorted matrix image, by searching point by point in each measurement line of the linear camera 26 the middle of the segment delimited by the centers of the two rails 22A, 22B, or, preferably, by first building the neutral line 122A, 122B of each rail 22A, 22B, then the line located midway between the neutral lines 122A, 122B.
    In this case of a virtual reference line constructed and used by the tracking system 12, the transposition system 16 must also be able to reconstruct the virtual reference line. To do this, the transposition system preferably comprises a second matrix transposition camera 44B disposed above and facing the second rail 22B.
    The second matrix transposition camera 44B is fixed to the transposition frame 45 of the shuttle 9 so that the relative positioning of the two matrix transposition cameras 44A, 44B is known and calibrated. Although each matrix transposition camera 44A, 44B has only a reduced field width allowing it to view only the rail 22A, 22B above which it is located, it is possible to determine the positioning of the center line 222 between the neutral lines 122A, 122B by determining the positioning of the neutral line 122A of the first rail 22A in the matrix image of the first transposition camera 44A, by determining the positioning of the neutral line 122B of the second rail 22B in the matrix image of the second transposition camera 44B and by calculating the center of the line segment between the two neutral lines 122A, 122B on the basis of these measurements and calibration data of the distance between the two transposition cameras 44A, 44B.
    According to a variant of the tracking system 12, the latter comprises a second matrix orientation camera 54B fixed to the chassis 4 of the tracking vehicle 4 and having an aiming direction perpendicular to the reference plane of the railroad track 22, in look of 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. Different algorithms can be implemented to exploit this data. We can for example assign a confidence index to each matrix image delivered by one or the other of the matrix orientation cameras, for example as a function of the contrast of the image, of the presence or not of the rail in the image or any other criterion, all to retain, for each inter-sleeper space, the measurement of the angle δ having the best confidence index. It is also possible to combine the measurements made to calculate an "average" angle δ. Since the distance between the two orientation matrix cameras 54A, 54B is known and calibrated, it is also possible to use these cameras to determine the virtual reference line 222 mentioned above. Finally, it is possible to envisage two independent measurement and calculation channels, one using the first matrix orientation camera 54A and the first matrix transposition camera 44A with respect to a first reference line 122A, and the other using the second matrix orientation camera 54B and the second matrix transposition camera 44B with respect to a first reference line 122B. If necessary, two odometers can also be provided at the monitoring system, one for each rail 22A, 22B, and each dedicated to one of the two measurement and calculation channels.
    In the identification phase, the identification of points of interest is not limited to the identification of vertices A, B, C, D or A ', B', C ', D' of a potential intervention quadrilateral. Other types of points of interest can be identified, for example coordinates of centers of screw heads to be screwed or changed. An intervention area is not necessarily a quadrilateral, but can be any polygon. We can also try to identify - instead of or in addition to points of interest A, B, C, D - lines of interest, for example the lines constituting the neutral lines 162, 164 of the crosspieces 62, 64, located each in the center of the corresponding crosspieces 62, 64, or the straight line 163 constituting the inter-sleeper line, which is the axis of symmetry for the preceding straight lines 162, 164, located halfway between two sleepers 62, 64. In the local repository <0, x, y> the coordinates of such straight lines can for example be those of two points belonging to them, or the coordinates of a point on the line and of an angle.
    Although the previous presentation was focused on the inter-sleeper space, it is also possible to analyze with the tracking system the track portions comprising the sleepers, in particular to detect there points of interest requiring an intervention.
    The processing of the pedometer 28 data, the linear camera 26 and the orientation device 52 by the tracking system 12 is done in real time, or very slightly deferred, so that it can be used by the tracking system. transposition 16 belonging to the same intervention machine 10 advancing continuously. The distance between the tracking system 12, located in a front zone 14 of the tracking vehicle 2 in the direction of progression 100 on the one hand, and the transposition system 16, located at a distance and towards the rear of the vehicle, d on the other hand, is used in particular to allow an operator to validate or invalidate points of interest A, B, C, D or their qualification.
    Alternatively, we can advance the intervention machine 10 in one direction for the location, then make him turn back to, during a second pass, perform the transposition and control of all or several tools. When transposing, the vehicle can move in the opposite direction to the direction of advance during tracking, or in the same direction.
    The railway intervention machine 10 in question may consist of one or more vehicles hinged together. Where appropriate, Thus, the tracking system 12, the transposition system 16 and the assembly of one or more tools 11 can be on a single vehicle. According to another embodiment, the tracking system 12 can be mounted on a carriage rolling on the track and articulated at the front of a rolling unit carrying the transposition system 16 and the assembly of one or more tools 11 .
    It is also conceivable that the tracking system 12 is mounted on an independent tracking vehicle independent of the intervention machine 10 carrying the assembly of one or more tools 11. In this latter hypothesis, it is advantageous to provide an absolute positioning unit on the locating vehicle, for example a GPS unit, in order to be able to assign coordinates to each absolute cross-reference space in the absolute positioning reference frame, the latter being sufficiently precise to discriminate between two successive inter-transverse spaces.
    If necessary, the assembly of one or more tools 20 can be fixed relative to the main chassis 6 of the intervention vehicle 3. It is then possible to roughly estimate the positioning of the assembly of a or several tools 20 from the data of the odometer 28 of the tracking system 12 alone, to determine which are, at a given instant, the tracking data relevant for the transposition.
    Although the description has more particularly focused on the use of the transposition procedure for the control of an intervention tool, the transposition can also be used to operate a precise auscultation of the path to using an auscultation instrument carried by a chassis distant from the locating chassis.
    The man-machine interface 34 can be remote from the vehicle 10, for example located in a remote control station.
    权利要求:
    Claims (15)
    [1" id="c-fr-0001]
    1. Method for locating a railroad track (22), executed by a rail locating system (12) progressing on the rail track (22) in a direction of progression (100), the locating process comprising the following actions:
    we acquire repeatedly, with a linear camera (26) of the railway tracking system (12) targeting the railway track (22), instantaneous linear optical data along an instantaneous measurement line (50), we acquire repetitively, with an orientation device (52) of the railway tracking system (12), orientation data of the railway tracking system (12) with respect to a reference line (22A) of the railway track (22 ), and by processing at least instantaneous linear optical data, a potentially distorted matrix image of an area of the surface of the railway track (22) is constructed, characterized in that:
    we identify points or lines of interest (A, B, C, D, A ', B', C ', D', 162,163, 164) in the potentially distorted matrix image, and we determine rectified coordinates of the points or lines of interest (A, B, C, D, A ', B', C ', D', 162, 163, 164), as a function of potentially distorted coordinates of the points or lines of interest (A, B , C, D, A ', B', C ', D', 162, 163, 164) in a repository of the potentially distorted matrix image and orientation data.
    [2" id="c-fr-0002]
    2. Method according to claim 1, characterized in that to acquire the orientation data, a first probe of the orientation device (52) detects an orientation of the tracking system (12) relative to a first rail (22A) of the railway, constituting a first orientation rail (22A), and preferably a second feeler of the orientation device (52) detects an orientation of the tracking system (12) relative to a second rail (22B ) of the railway, constituting a second orientation rail (22B).
    [3" id="c-fr-0003]
    3. Method according to any one of the preceding claims, characterized in that to acquire the orientation data, a first matrix orientation camera (54A) of the orientation device (52), disposed opposite a first rails of the railway constituting a first orientation rail (22A), takes pictures processed by the orientation device (52) to detect an orientation of the first orientation rail (22A) relative to a target of the first matrix orientation camera (54A), and preferably a second matrix orientation camera (54B) disposed opposite a second of the rails of the railway track, constituting a second orientation rail (22A ), takes shots and processes the shots, processed by the orientation device (52) to detect an orientation of the second orientation rail (22A) relative to a test pattern of the second camera orientation matrix (54B).
    [4" id="c-fr-0004]
    4. Method according to any one of the preceding claims, characterized in that the tracking system (12) detects areas of the surface of the railway track (22) comprising sleepers (62, 64) and inter-sleeper areas from the surface of the track (22), the orientation device delivering the orientation data only once for each of the inter-sleeper zones.
    [5" id="c-fr-0005]
    5. Method according to any one of the preceding claims, characterized in that the reference line (122A, 122B, 222) is a neutral line of one of the rails (22A, 22B) of the railway track (22), or a line constructed from the neutral lines (122A, 122B) of the rails (22A, 22B) of the railway track (22).
    [6" id="c-fr-0006]
    6. Method according to any one of the preceding claims, characterized in that the orientation data are used to construct the reference line (222) from the neutral lines (122A, 122B) of the rails (22A, 22B ) of the railway (22).
    [7" id="c-fr-0007]
    7. Method according to any one of the preceding claims, characterized in that one repetitively acquires, with one or more odometers (28) of the rail tracking system (12), progress data of the rail tracking system (12) on the track (22) in the direction of travel (100).
    [8" id="c-fr-0008]
    8. Method according to claim 7, characterized in that the acquisition of instantaneous linear optical data is triggered by the reception of progress data.
    [9" id="c-fr-0009]
    9. Method according to claim 7, characterized in that the instantaneous linear optical data and the progress data are acquired in a synchronized manner.
    [10" id="c-fr-0010]
    10. Method according to any one of claims 7 to 9, characterized in that the progress data and the instantaneous linear optical data are time stamped, the matrix image is preferably constructed as a function of time stamps.
    [11" id="c-fr-0011]
    11. Tracking method according to any one of claims 7 to 10, characterized in:
    by processing the potentially distorted matrix image we identify, in the potentially distorted matrix image, at least one spatial indexing tag (56) of predetermined signature, by processing the progress data and the orientation data, it is determined a curvilinear abscissa of the spatial indexing beacon (56) and a positioning of the spatial indexing beacon (56) relative to the reference line (122A, 122B, 222) of the railway track (22), and the coordinates adjusted points or lines of interest (A, B, C, D, A ', B', C ', D', 162, 163, 164) are determined in a frame of reference
    [12" id="c-fr-0012]
    12.
    [13" id="c-fr-0013]
    13.
    [14" id="c-fr-0014]
    14.
    [15" id="c-fr-0015]
    15.
    two-dimensional local tracking linked to the spatial indexing tag (56) and to the reference line (122A, 122B, 222).
    Method according to any one of the preceding claims, characterized in that the points or lines of interest (A, B, C, D, A ', B', C ', D', 162, 163, 164) constitute boundaries of an area of interest, preferably vertices of a quadrilateral constituting an area of interest.
    Tracking method according to any one of the preceding claims, characterized in that it comprises the reproduction of the matrix image on a display screen (38) of the tracking system (12).
    Locating method according to claim 13, characterized in that it includes the visual identification on the display screen (38) of the points or lines of interest (A, B, C, D, A ', B', C ', D', 162,163,164).
    Locating method according to any one of claims 13 to 14, in combination with claim 12, characterized in that it comprises a validation and / or an invalidation of at least some of the points or lines of interest or of the area of interest or of a qualification of the area of interest as a possible intervention or prohibited area, following a seizure on a man-machine input interface (40).
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3077553A1|2019-08-09|METHOD OF REPORTING POINTS OR LINES OF INTEREST ON A RAILWAY
    CN103578117B|2016-09-14|Determine the photographic head method relative to the attitude of environment
    CN105526882B|2018-03-02|Track switch Abrasion detecting system and its detection method based on structural light measurement
    EP2718730B1|2015-11-18|Device for measuring speed and position of a vehicle moving along a guidance track, method and computer program product corresponding thereto
    DE102009035336B3|2010-11-18|Device for optical scanning and measuring of environment, has optical measuring device for collection of ways as ensemble between different centers returning from laser scanner
    US6909514B2|2005-06-21|Wheel profile inspection apparatus and method
    CN106978774B|2019-02-22|A kind of road surface pit slot automatic testing method
    WO1997006409A1|1997-02-20|Process and device for the rapid detection of the position of a target marking
    EP3740616B1|2021-08-11|Method for controlling an assembly consisting of one or more intervention tools mounted on an intervention rail vehicle
    FR3077552A1|2019-08-09|METHODS OF REPORTING POINTS OR LINES OF INTEREST ON A RAILWAY, POSITIONING AND DRIVING A MACHINE OF INTERVENTION ON A RAILWAY
    FR3077582A1|2019-08-09|METHOD FOR CONTROLLING AN ASSEMBLY OF ONE OR MORE INTERVENTION TOOLS MOUNTED ON A RAIL VEHICLE OF INTERVENTION
    CN208458668U|2019-02-01|A kind of train bottom surface three-dimensional high-precision measuring system
    CN101793530B|2012-07-18|Method for measuring decoupling factor of optical fiber gyro
    EP3493153A1|2019-06-05|Method and system for detecting the faults in the geometry of a railway track
    KR102315405B1|2021-10-21|Apparatus and method for inspecting railway line
    EP2320404B1|2013-03-27|Method and apparatus for trajectory measuring of a passive mobile object
    CN107449373B|2021-04-13|High-speed structured light scanning method and system based on stereoscopic vision
    ABDEL-HAMEED2021|EVALUATION OF USING VIDEO CAMERA AND PHOTOGRAMMETRIC COMPUTER VISION TECHNIQUES FOR 3D RECONSTRUCTION AND TRACK DETECTION OF THE RAILWAY NETWORK IN EGYPT
    CN104359808B|2016-08-10|A kind of long distance wireless PM2.5 concentration detection method
    FR3106904A1|2021-08-06|Method for calibrating the extrinsic characteristics of a lidar
    CN113554697A|2021-10-26|Cabin section profile accurate measurement method based on line laser
    同族专利:
    公开号 | 公开日
    FR3077553B1|2020-02-28|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    FR2005890A1|1968-04-09|1969-12-19|Plasser Bahnbaumasch Franz|
    EP0115462A2|1983-01-31|1984-08-08|Commissariat A L'energie Atomique|Method for the automatic recognition of an image on the basis of a corresponding reference image|
    EP3138754A1|2015-09-03|2017-03-08|Rail Vision Europe Ltd|Rail track asset survey system|
    US20170069090A1|2015-09-07|2017-03-09|Kabushiki Kaisha Toshiba|Image processing device, image processing system, and image processing method|US10582187B2|2015-02-20|2020-03-03|Tetra Tech, Inc.|3D track assessment method|
    US10625760B2|2018-06-01|2020-04-21|Tetra Tech, Inc.|Apparatus and method for calculating wooden crosstie plate cut measurements and rail seat abrasion measurements based on rail head height|
    US10728988B2|2015-01-19|2020-07-28|Tetra Tech, Inc.|Light emission power control apparatus and method|
    US10730538B2|2018-06-01|2020-08-04|Tetra Tech, Inc.|Apparatus and method for calculating plate cut and rail seat abrasion based on measurements only of rail head elevation and crosstie surface elevation|
    US10807623B2|2018-06-01|2020-10-20|Tetra Tech, Inc.|Apparatus and method for gathering data from sensors oriented at an oblique angle relative to a railway track|
    US10908291B2|2019-05-16|2021-02-02|Tetra Tech, Inc.|System and method for generating and interpreting point clouds of a rail corridor along a survey path|
    法律状态:
    2019-08-09| PLSC| Publication of the preliminary search report|Effective date: 20190809 |
    2019-08-30| PLFP| Fee payment|Year of fee payment: 2 |
    2020-04-29| PLFP| Fee payment|Year of fee payment: 3 |
    2021-02-17| PLFP| Fee payment|Year of fee payment: 4 |
    2022-02-21| PLFP| Fee payment|Year of fee payment: 5 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1850961|2018-02-06|
    FR1850961A|FR3077553B1|2018-02-06|2018-02-06|METHOD FOR LOCATING POINTS OR LINES OF INTEREST ON A RAILWAY|FR1850961A| FR3077553B1|2018-02-06|2018-02-06|METHOD FOR LOCATING POINTS OR LINES OF INTEREST ON A RAILWAY|
    EP19702607.3A| EP3740410A1|2018-02-06|2019-01-31|Methods for locating points or lines of interest on a railway track, positioning and driving an intervention machine on a railway track|
    US16/968,072| US20210016811A1|2018-02-06|2019-01-31|Methods for locating points or lines of interest on a railway track, positioning and driving an intervention machine on a railway track|
    CN201980011978.9A| CN112105541A|2018-02-06|2019-01-31|Method for locating points or lines of interest on a railway track, locating and driving an intervention machine on a railway track|
    JP2020541546A| JP2021512242A|2018-02-06|2019-01-31|How to command a set of intervention tools attached to a railroad intervention vehicle|
    JP2020541511A| JP2021513481A|2018-02-06|2019-01-31|A method for locating points of interest or lines on railroad tracks, positioning and driving intervention machines on railroad tracks.|
    AU2019218476A| AU2019218476A1|2018-02-06|2019-01-31|Method for locating points or lines of interest on a railway track|
    PCT/EP2019/052441| WO2019154718A1|2018-02-06|2019-01-31|Methods for locating points or lines of interest on a railway track, positioning and driving an intervention machine on a railway track|
    CN201980011971.7A| CN112166064A|2018-02-06|2019-01-31|Method for locating points or lines of interest on a railway track|
    US16/968,089| US20210024108A1|2018-02-06|2019-01-31|Method for locating points or lines of interest on a railway track|
    PCT/EP2019/052444| WO2019154719A1|2018-02-06|2019-01-31|Method for locating points or lines of interest on a railway track|
    PCT/EP2019/052445| WO2019154720A1|2018-02-06|2019-01-31|Method for controlling an assembly consisting of one or more intervention tools mounted on an intervention rail vehicle|
    EP19702608.1A| EP3740616B1|2018-02-06|2019-01-31|Method for controlling an assembly consisting of one or more intervention tools mounted on an intervention rail vehicle|
    AU2019218475A| AU2019218475A1|2018-02-06|2019-01-31|Methods for locating points or lines of interest on a railway track, positioning and driving an intervention machine on a railway track|
    EP19706392.8A| EP3740411A1|2018-02-06|2019-01-31|Method for locating points or lines of interest on a railway track|
    JP2020541483A| JP2021512313A|2018-02-06|2019-01-31|How to locate a point of interest or line on a railroad track|
    AU2019218477A| AU2019218477A1|2018-02-06|2019-01-31|Method for commanding a set of one or more intervention tools mounted on a railway intervention vehicle|
    CN201980012054.0A| CN112119188A|2018-02-06|2019-01-31|Method for controlling a set of one or more intervention tools mounted on a railway intervention vehicle|
    US16/968,105| US20210024109A1|2018-02-06|2020-01-31|Method for commanding a set of one or more intervention tools mounted on a railway intervention vehicle|
    [返回顶部]