METHOD FOR IN-SITU AND REAL-TIME COLLECTION AND PROCESSING OF GEOMETRIC PARAMETERS OF RAILWAY LINES

Abstract

A method for in-situ and real-time collection and processing of geometric parameters of railway lines, in a particular but non-limiting manner to those related to the height and stagger of the contact wire in electrified lines and the gauges to specific elements of the infrastructure in any line, generated based on static measurements starting from two-dimensional scenes perpendicular to the track axis, by determining the number of angular positions per scene, determining the minimum number of passes in each position, obtaining the raw coordinates, applying an averaging algorithm, applying offset corrections, transforming coordinates and applying either the steps to salve for height and stagger of the overhead contact line, or applying the steps to salve for gauges to specific elements of the infrastructure. An optimized, efficient and simple method is achieved which enables the real-time management and processing of the data obtained from the railway infrastructure.

Claims

1. A method for in-situ and real-time collection and processing of geometric parameters of railway lines, comprising: determining a number of angular positions per scene necessary, wherein a minimum number of angular positions determines both a laser scanner model used and a configuration parameters thereof; determining a minimum number of passes per angular position, wherein the minimum number of passes per angular position is be set by a study of the evolution of the most representative centralization and dispersion statistics of the type of point clouds obtained in scenes of the railway environment; obtaining raw coordinates of each point of a scene according to a reference system used by a sensor for generating point clouds; obtaining an algorithm for averaging points in each angular position, eliminating both outliers and ghost points by: analyzing and determining a sampling distribution or distributions to which a data best fits in each angular position and determining the sampling distribution to which it fits with a highest quality, separating the sampling distributions found, and filtering any point located at a predetermined distance from a chosen centralization statistic, wherein said predetermined distance is defined based on a margin of +/−“n” times the dispersion statistic selected, “n” being an integer, calculating the sampling distribution that best fits a new filtered sample, setting minimum quality parameters of the fit in advance, checking a degree of fulfilment of the quality parameters from the previous fit and, when minimum requirements are exceeded, selecting an average value distance of this distribution together with the associated RSSI thereof as a representative point and removing the rest of the points; conducting constructive offset corrections of the equipment wherein the sensor for generating the point cloud is integrated, and coordinate transformation, wherein the X-axis is located on the above rail level (ARL), and the Y-axis starts from the axis of the track and is perpendicular to the ARL; applying either steps to salve for height and stagger of the overhead contact line, or applying the steps to salve for gauges to specific elements of the infrastructure.

2. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of seeking to salve for height and stagger of the overhead contact line, the following steps are applied: reducing the amount of points to be analyzed within a segment, restricting an area to one region of interest (ROi) defined dynamically according to a limit requirements for height and stagger established by each infrastructure manager; spatial grouping or clustering of the points that make up one or two contact wires, using an algorithm to do so that can be set depending on the physical features and constructive limits thereof; dividing the clusters from the previous step into subgroups of points that give rise to one or two contact wires, and establishing (i) a minimum and maximum threshold of points which define the potential thereof for representing one or two contact wires, and (ii) criteria for dividing the clusters from the previous step depending on the number of points making them up; spatial weighing of the points contained in the subgroups from the previous step based on the parameter indicating the power of the intensity reflected, and as a result, a point in a “virtual” position is obtained, which is weighted based on the RSSI of the original points; final filtering of virtual candidate points, selecting as contact wire or wires in the case of lines with double contact wire those which are the lowest and of highest stagger; synchronizing these specific and static measurements of the height and stagger of the contact wire or wires, with the rest of the parameters collected by the auxiliary sensors.

3. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the seeking to salve for gauges to specific elements of the infrastructure, the method comprises the steps of: reducing the amount of points to be analyzed within the segment, restricting the area to one region of interest (ROI) defined dynamically according to the geometric requirements of the specific element of the railway infrastructure; spatial grouping or clustering of the points that could potentially make up part of the specific element to be detected; determining the minimum distance from the origin of coordinates defined by the reference system used by the infrastructure manager; synchronizing these specific and static measurements of minimum distances gauges to specific elements of the infrastructure, with the rest of the parameters collected by the auxiliary sensors.

4. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein the minimum value of passes in each angular position is set by a study of the evolution of the average and the median as a centralization statistic, and of the deviation from the median as dispersion statistic that are most representative of the type of point clouds obtained in scenes of the railway environment.

5. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of obtaining the raw coordinates of each point of the scene, when the sensor for generating point clouds is a LIDAR, a set of points for each angular position is obtained which is defined by the distance thereof from the sensor and by a parameter indicative of the power of the intensity reflected, such as the RSSI (Received Signal Strength Indicator).

6. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of analyzing the distribution of the data in each angular position and determining the sampling distribution with the best fit, when a LIDAR is used to generate the point cloud, the samples for each angular position are fitted to a normal distribution, or to two partially-overlapping normal distributions.

7. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of separating the sampling distributions found, and filtering any point located at a predetermined distance from the chosen centralization statistic, when a LIDAR is used to generate the point cloud, all the points located outside of a margin of +/−3 times the deviation calculated with respect to the median are filtered.

8. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of calculating the sampling distribution that best fits this new filtered sample, a normal distribution is used.

9. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of synchronizing the specific and static measurements of the height and stagger of the contact wire or wires, with the rest of the parameters collected by the auxiliary sensors, wherein the parameters are: kilometer mark (KM) or distance travelled since the beginning of the sampling campaign, absolute referencing of the scene by means of GNSS coordinates, track gauge, cant and inclination of the track.

10. The method for in-situ and real-time collection and processing of geometric parameters of railway lines according to claim 1, wherein in the step of synchronizing the specific and static measurements of minimum distances to specific elements of the infrastructure, with the rest of the parameters collected by the auxiliary sensors, wherein the parameters are: kilometer mark (KM) or distance travelled since the beginning of the sampling campaign, absolute referencing of the scene by means of GNSS coordinates, track gauge, cant and inclination of the track.

Description

EXPLANATION OF THE FIGURES

[0060] In order to complement the description being made and with the object of helping to better understand the features of the invention, in accordance with a preferred practical exemplary embodiment thereof, said description is accompanied, as an integral part thereof, by a set of drawings where, in an illustrative and non-limiting manner, the following has been represented:

[0061] FIG. 1 shows a representation of the basic steps of the method object of the invention.

[0062] FIG. 2 shows a diagram with the steps that are carried out to salve for the height and stagger of an overhead contact line (6).

[0063] FIG. 3 shows the resolution of gauges to specific elements of the infrastructure (13).

PREFERRED EMBODIMENT OF THE INVENTION

[0064] In light of the figures, a preferred embodiment of the proposed invention is described below.

[0065] The method object of the invention for in-situ and real-time collection and processing of geometric parameters comprises the following steps: [0066] Determining the number of angular positions per scene (1) necessary to solve for the specific elements of the railway infrastructure based on a study of the casuistry of these environments. This minimum number of angular positions will determine both the laser scanner model used, as well as the configuration parameters thereof (in a particular but non-limiting manner: field of view, angular resolution and beam divergence). [0067] Determining the minimum number of passes per angular position (2) in order to ensure the repeatability of the measurements. This value will be set by a study of the evolution of the most representative centralisation and dispersion statistics of the type of point clouds obtained in scenes of the railway environment. In a particular but non-limiting manner: average and median as centralisation statistics; standard deviation and deviation from the median as dispersion statistics. [0068] Obtaining the raw coordinates of each point of the scene (3) according to the reference system used by the sensor for generating point clouds. In a particular but non-limiting manner, in one embodiment wherein the sensor for generating point clouds is a LIDAR, a set of points for each angular position is obtained which is defined by the distance thereof from the sensor and by a parameter indicative of the power of the intensity reflected, typically, the RSSI (Received Signal Strength Indicator). [0069] Algorithm for averaging points in each angular position, eliminating both outliers (4) and ghost points, consisting of the following steps: [0070] Analysing and determining the sampling distribution (or distributions) to which the data best fits in each angular position (4.1). In a particular but non-limiting manner, in one embodiment of the invention that uses a LIDAR to generate the point cloud, the samples for each angular position are fitted to a normal distribution, or to two partially-overlapping normal distributions. [0071] Separating (if applicable) the sampling distributions found, and filtering (4.2) (removing) any point located at a predetermined distance from the chosen centralisation statistic. Said distance is defined based on a margin of +/−“n” times the dispersion statistic selected, “n” being an integer. In a particular but non-limiting manner, in one embodiment of the invention that uses a LIDAR to generate the point cloud, all the points located outside of a margin of +/−3 times the deviation calculated with respect to the median are filtered (removed). [0072] Calculating the sampling distribution that best fits this new filtered sample (4.3), setting minimum quality parameters of the fit in advance. In a particular but non-limiting manner, the normal distribution with better fit. [0073] Checking the degree of fulfilment of the quality parameters from the previous fit (4.4) and, if the minimum requirements are exceeded, selecting the average value (distance) of this distribution (together with the associated RSSI thereof) and removing the rest of the points (4.5). [0074] Constructive offset corrections of the equipment wherein the sensor for generating the point cloud is integrated, and coordinate transformation (5). In other words, the presentation of the measurements depending on the reference system preferred by the infrastructure manager. Typically, a Cartesian coordinate system wherein the X-axis is located on the above rail level (ARL), and the Y-axis starts from the axis of the track and is perpendicular to the ARL.

[0075] Once the previous steps have been carried out, in the case of alternative 1 (solving for height and stagger of the overhead contact line (6)), the method would continue as follows, as shown in FIG. 2: [0076] Reducing the amount of points to be analysed within the segment (7), restricting the area to one region of interest (ROi) defined dynamically according to the limit requirements for height and stagger established by each infrastructure manager. [0077] Spatial grouping or clustering of the points that could potentially make up one (or two) contact wires (8), using an algorithm to do so that can be set depending on the physical features and constructive limits thereof (diameter of the wire, range of heights expected over the above rail level and separation between wires in the case of lines with double contact wire). Said parameters, together with those of the laser scanner, define a minimum and maximum number of expected impacts on an object of the “contact wire” type. [0078] Dividing the clusters from the previous step into subgroups of points (9) that could give rise to one or two contact wires. To do so, it establishes (i) a minimum and maximum threshold of points which define the potential thereof for representing one or two contact wires, and (ii) criteria for dividing the clusters from the previous step depending on the number of points making them up. [0079] Spatial weighting of the points contained in the subgroups (10) from the previous step based on the parameter indicating the power of the intensity reflected, typically, the RSSI. As a result of this weighting, a point in a “virtual” position is obtained, which is weighted based on the RSSI of the original points. This algorithm enables the precision of the measurements to be refined, since for laser scanner sensors, the higher the value of the reflected intensity, the greater the precision. These virtual points are defined by the coordinates thereof referring to the track system, for which reason the X-axis will correspond to the stagger of the candidate contact wire and the Y-axis to the height. [0080] Final filtering of virtual candidate points (11), selecting as contact wire or wires (in the case of lines with double contact wire) those which are the lowest and of highest stagger. With this, the height and stagger of the active contact wires can be solved for at all times, since situations can occur with the presence of more than two contact wires in one same scene (catenary section overlaps), wherein some of the wires of the scene are not in direct contact with the pantograph when the trains are circulating. [0081] Synchronising these specific and static measurements of the height and stagger of the contact wire or wires, with the rest of the parameters (12) collected by the auxiliary sensors, which complete the information required by the infrastructure manager. In a particular but non-limiting manner: kilometre mark (KM) or distance travelled since the beginning of the sampling campaign, absolute referencing of the scene by means of GNSS coordinates, track gauge, cant and inclination of the track.

[0082] Regarding the case of applying alternative 2 (solving for gauges to specific elements of the infrastructure (13)), the method would begin after the aforementioned step for transforming coordinates and applying offset corrections (5), and would continue as follows, as shown in FIG. 3. [0083] Reducing the amount of points to be analysed within the segment (14), restricting the area to one region of interest (ROI) defined dynamically according to the geometric requirements of the specific element of the railway infrastructure to be detected in each case (pole offset, electrical insulation clearances, etc.) [0084] Spatial grouping or clustering of the points that could potentially make up part of the specific element to be detected (15), based on defining patterns determined by the most representative geometric features thereof; in a particular but non-limiting manner, straight and curved ones. [0085] Determining the minimum distance from the origin of coordinates defined by the reference system used by the infrastructure manager (16) (typically the same one used for height and stagger of the contact wire or wires) to the points associated with a specific pattern in the previous step. [0086] Synchronising these specific and static measurements of minimum distances (gauges) to specific elements of the infrastructure (17), with the rest of the parameters collected by the auxiliary sensors, which complete the information required by the infrastructure manager. In a particular but non-limiting manner: kilometre mark (KM) or distance travelled since the beginning of the sampling campaign, absolute referencing of the scene by means of GNSS coordinates, track gauge, cant and inclination of the track.

[0087] Having sufficiently described the nature of the invention, as well as how to put it into practice, it must be noted that, within its essential nature, the invention may be carried out according to other embodiments differing in detail from that set out by way of example, which the protection sought would equally cover, provided that the fundamental principle thereof is not altered, changed or modified.