Method for evaluating the accuracy of a system under test for determining the position of a rail vehicle

The method uses track signatures to enhance rail vehicle positioning accuracy by comparing them with a reference system, addressing inaccuracies in dynamic environments and tunnels, and providing real-time feedback, thereby improving localization systems.

DE102017114696B4Active Publication Date: 2024-11-07DEUTSCHES ZENTRUM FÜR LUFT UND RAUMFAHRT E V
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
DE102017114696
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2017-06-30
Publication Date
2024-11-07
Estimated Expiration
2037-06-30

AI Technical Summary

Technical Problem

Existing methods for evaluating the accuracy of rail vehicle positioning systems, particularly in dynamic environments, face challenges such as unobservable systematic errors, limited availability of GNSS measurements in tunnels, and insufficient consideration of realistic railway environments, leading to inaccurate track selection and positioning.

Method used

A method utilizing track signatures, including magnetic field, vibration, and track curvature, to determine the position of a rail vehicle, allowing for real-time or post-processing accuracy evaluation by comparing these signatures with a reference system, and providing a warning when thresholds are exceeded.

Benefits of technology

Enables highly accurate positioning in tunnels and dynamic environments, offering an independent reference to GNSS, and allowing for real-time assessment of virtual balise suitability, thus improving the reliability and accuracy of rail vehicle localization systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Method for evaluating the accuracy of a system under test for determining the track position of a railway vehicle, wherein the method - a track position determination reference system is used in which track signatures characterizing individual track sections of the railway with known track position are provided, which describe the location-dependent change of at least one physical quantity within the relevant track section to which the rail vehicle is exposed when traversing the track section, wherein the track position determination reference system determines the track signature of the relevant track section when currently traversing the track section for the purpose of evaluating the accuracy of the system to be tested and determines the current track position of the rail vehicle by comparison with the provided track signatures, - a system to be tested is provided which enables the position determination of a rail vehicle on the railway track without the use of track signatures or without the use of the track signature or group of track signatures on which the track position determination reference system is based, - the current track position of the rail vehicle is also determined using the system to be tested, - the result of the track position determination by the track position determination reference system is compared with the result of the track position determination by the system to be tested, - the accuracy of the track position determination by the system under test is evaluated based on the track position determination by the track position determination reference system and - the track position determination reference system as a permanent evaluation system (monitoring) of a train localization unit determines an accuracy and, if a threshold is exceeded, transmits a warning and / or message to the train driver and / or the train protection system and / or the control center.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a method for evaluating the accuracy of a system to be tested for determining the position of a rail vehicle, in particular for use in the approval of train localization systems and for use in tests of positioning systems which preferably operate on the basis of satellite navigation systems.

[0002] The invention particularly describes methods for evaluating methods for the precise track location of trains and for evaluating positions from GNSS measurements (Global Navigation Satellite System, such as GPS, GALILEO, GLONASS, BeiDou). Localization requires a map, and the result is determined using map coordination. Positioning involves determining a geoposition in a coordinate system, such as a position in the WGS84 Global Coordinate System (GPS standard).

[0003] According to the invention, track signatures are used for the reference system, in particular the magnetic field signature, the vibration signature, and the track curvature signature. A track signature is a location-dependent signal that is plotted over a traveled path (see also DE 10 2012 219 111 A1). Track signatures can be formed from measurements, e.g., the position and / or orientation (in space) of the rail vehicle, the magnetic field, the vibrations acting on the rail vehicle (in one or more spatial directions), the track curvatures, the curves (e.g., when turning at a switch), the "course angle," and / or changes in the aforementioned parameters.

[0004] The invention can be implemented in two variants as follows: 1. Reference for evaluating the localization result according to track number and position on the track of train localization procedures One challenge in evaluating the accuracy of train localization is the observability of the correct track, especially after crossing switches, and the position errors in the longitudinal direction of the track. Furthermore, systematic errors in the measurement systems are not observable or are only observable in certain scenarios, particularly for GNSS measurements. The localization result is now to be evaluated in real time or in post-processing. This evaluation can be used both in the evaluation of different train localization methods and in approval procedures. 2. Reference for the evaluation of dynamic GNSS measurements on moving trains Evaluating the accuracy of dynamic GNSS measurements on moving trains is of interest in the research and development of train localization methods and their approval. This requires either a high-precision reference system for position measurement in a dynamic environment or a high-precision rail map with geopositions and a localization on the map. With a virtual balise using GNSS, a crossing of a position marker is detected using GNSS, and a corresponding signal, telegram, or protocol (e.g., balise ID) is sent to the balise interface of an ETCS (European Train Control System). The virtual balise using GNSS is best used only at suitable locations where good reception conditions are expected. The locations for a virtual balise must be found and evaluated, particularly in approval procedures. The accuracy of a virtual balise or a potential position for a virtual balise can be evaluated using the method presented here, based on track signatures in a dynamic measurement from the train.

[0005] The state of the art includes: - the use of GNSS-based train localization for collision warning (DLR RCAS) as well as non-safety-critical applications without track selectivity (e.g. passenger information, position for rolling stock management); - Accuracy evaluations in current research of GNSS-based localization are in most cases carried out using simulators (e.g. Spirent GNSS simulator); - GNSS measurements and GNSS-based localization results are compared, among other things, with high-end GNSS measurements under good reception conditions. This method is used to evaluate simple and inexpensive GNSS receivers (mass market, low-cost). A geodetic GNSS receiver (high-end) or an RTK system serves as a reference. - Reference measurements of track positions using geodetic methods, using laser tachymeters with target tracking.

[0006] The disadvantages of the state of the art include the following: - In tunnels, no GNSS measurements are available, so track selectivity is not always possible due to GNSS accuracy and narrow track spacing. - A common assumption in simulators is ideal maps, simplified noise processes, and inadequate consideration of realistic railway environments with multipath propagation and the impact on phase resolution. In addition to simulation tests, accuracy studies from experiments or field tests are also required. - A GNSS reference requires favorable reception conditions, such as a continuous and unobstructed view of satellites and low multipath propagation. High-accuracy GNSS receivers use carrier phase measurements. These can easily be disrupted by a dynamically changing environment with trees, forests, bridges, or masts. - Commercial laser total stations have limited dynamic range and a range of a few hundred meters for dynamic measurements on moving trains. A line of sight is required, or the total station's position must be determined using GNSS or a similar system.

[0007] DE 10 2012 215 533 A1 discloses a method for determining the position of a rail vehicle. In this method, the track is optically recorded by a recording unit, recording landmarks along the track. These recordings are compared with previously stored landmark position data. If a recorded landmark matches a stored landmark, the position of the rail vehicle can be determined based on the detected landmark. This position determination based on the recording unit can be used in conjunction with an odometry of the rail vehicle, a global navigation satellite system, at least one additional sensor, and / or actively or passively communicating markers along the track to correct and / or validate the position of the rail vehicle.

[0008] DE 196 11 774 A1 describes a method for self-locating a track-guided vehicle and a device for implementing the method. The track-guided vehicle is equipped with two autonomous self-locating systems: a positioning system based on satellite navigation and a positioning system based on radar imaging of the track. Depending on which positioning system is available, this positioning system is used to determine the position of the track-guided vehicle. Thus, according to the known method, redundant positioning systems are used for self-locating.

[0009] The object of the invention is to be able to evaluate the accuracy of a system to be tested for (track or geo)positioning of a rail vehicle by means of a position-accurate reference system.

[0010] To achieve this object, the invention proposes a method for evaluating the accuracy of a system to be tested for determining the position of a rail vehicle, wherein according to a first variant of the method - a track position determination reference system is used in which, for individual track sections of the railway line with a known track position, track signatures characterising the relevant track section are provided, which describe the location-dependent change in at least one physical quantity within the relevant track section to which the rail vehicle is exposed when traveling over the track section, whereby the track position determination reference system determines the track signature of the relevant track section when the track section is currently being travelled over for the purpose of assessing the accuracy of the system to be tested and determines the current track position of the rail vehicle by comparing it with the provided track signatures, - a system to be tested is provided which enables the position of a railway vehicle on the railway track to be determined without the use of track signatures or without the use of the track signature or group of track signatures on which the track position determination reference system is based, - the system to be tested also determines the current track position of the rail vehicle, - the result of the track position determination by the track position determination reference system is compared with the result of the track position determination by the system to be tested, - the accuracy of the track position determination by the system under test is assessed based on the track position determination by the track position determination reference system and - the track position determination reference system, as a permanent evaluation system (monitoring) of a train localisation unit, determines an accuracy and, if a threshold value is exceeded, transmits a warning and / or message to the train driver and / or the train protection system and / or the control centre.

[0011] In this first variant, the track position determination reference system used according to the invention is operated using track signatures, whereby one or more track signatures can be used as defined above. The system under test does not operate with track signatures, or rather, does not operate with the track signature or group of track signatures used in the reference system.

[0012] Finally, within the scope of this variant of the invention, it is also possible for an automatic protocol of the deviations or a protocol of the deviations with times and / or positions to be created, stored and / or transmitted to a central location.

[0013] The method according to the invention provides highly accurate position determination of a rail vehicle on a railway track by linking track signatures. This allows positioning methods and systems based on other physical conditions to be tested and evaluated for their accuracy. How this track signature can be determined for a track section, i.e., its "fingerprint," is described, for example, in DE 10 2012 219 111 A1.

[0014] The advantages of the invention can be summarized as follows: - Localization using signatures (magnetic field, vibration, track curvature) enables longitudinal positioning even in tunnels as well as detection of the switch path for track-selective localization; - Measuring accuracy in real-world environments instead of simulation; - Reference by signatures is an independent reference to GNSS. This increases the significance of the accuracy results; - Localization reference using signatures enables dynamic measurements over long distances; - Evaluation of the suitability of potential positions for virtual balises based on GNSS, as well as real-time evaluation of a current virtual balise measurement using GNSS. Alternatively, the reference system can be used to verify the position and existence of radio balises (e.g., Eurobalise).

[0015] The invention is described below using two exemplary embodiments and with reference to the drawings. The invention can be seen in particular in the two variants shown below: a) Reference using signatures (e.g. magnetic field, vibration, track curvature) for an accuracy analysis of the localization result (track position) with regard to correct track and deviation in the longitudinal track direction of localization methods for rail vehicles for evaluation and assessment in real time or in post-processing. b) Reference using signatures (e.g., magnetic field, vibration, track curvature) for an analysis of the 2D or 3D accuracy of geographical positions from GNSS measurements in systems that operate with a high-precision map that stores geoposition data and reference signatures in a longitudinally parameterized form.

[0016] Example of implementation for a): Structure of the reference system for train localization with signatures: Method for evaluating the accuracy of train localization using a reference based on signatures, in particular the magnetic field signature, the vibration signature, and the track curvature signature. The structure of the accuracy evaluation of train localization using signatures is described in Fig. 1 shown schematically.

[0017] The signatures are generated from a measurement sequence of train-mounted sensors 3d and a sequence of speed values ​​3c, and then filtered (as described, for example, in DE 10 2012 219 111 A1). Using the speed, the time-sampled measurement signal is transformed into the spatial domain. The method requires speed values ​​derived from speed sensors (wheel rotation, Doppler radar) or from magnetic field or vibration measurements. A map 3b contains longitudinally parameterized reference signatures from transformed and filtered measurement data, which are stored together with a track identification number and the track length parameter (track position). The localization result with the track identification number and the track position from signatures is achieved by comparing reference signatures from the map and the last measured measurement signature (block 3a).

[0018] The accuracy evaluation is achieved by comparing the localization result of a measurement data signature 3e with a train localization without this measurement data signature 2e (e.g., using GNSS measurement). An evaluation 4 determines the longitudinal error on the track and evaluates whether the track was correctly or incorrectly identified. It can be used for testing purposes in field tests and for monitoring purposes during regular operation.

[0019] Example for b): Evaluation of dynamic GNSS measurements

[0020] The structure is similar to example 1, with the difference that a reference position in geocoordinates 3f is compared with the GNSS position result 2f, and an error 7 can be determined. The error is a length and can be represented as a simple Euclidean distance, and / or as a deviation in the longitudinal and transverse track directions and / or in north-east elevation.

[0021] For the evaluation of GNSS positions, map 3b also contains parameterized geopositions based on the track position. For the evaluation of further measurements, the map also contains additional measured features. Localization on the map is achieved using signature data (magnetic field, vibration, curvature), and the reference geoposition 3f can be determined using the map and the localization result. List of reference symbols 1 rail vehicle 2 Equipment Under Test (train tracking system or GNSS receiver alone) 2a Localization procedure 2b route map 2c GNSS receiver 2d Optional: additional sensors such as distance measurement, IMU, camera, radar, lidar 2nd localization result 2f GNSS receiver signal 3 Reference system with signature measurements 3a Localization procedure of the reference system with signature 3b Route map with signatures 3c Speed ​​information 3D sensor for signature measurement (e.g. magnetic field, vibration, curvature) 3e Reference result 3f Reference result in geocoordinates if necessary 4 Evaluation unit for accuracy 5 Accuracy result of localization 6 Evaluation unit for position accuracy (geocoordination) 7 Position accuracy result

Claims

[1] Method for evaluating the accuracy of a system to be tested for determining the track position of a rail vehicle, wherein the method - a track position determination reference system is used in which, for individual track sections of the railway line with a known track position, track signatures characterising the relevant track section are provided, which describe the location-dependent change in at least one physical quantity within the relevant track section to which the rail vehicle is exposed when traveling over the track section, whereby the track position determination reference system determines the track signature of the relevant track section when the track section is currently being travelled over for the purpose of assessing the accuracy of the system to be tested and determines the current track position of the rail vehicle by comparing it with the provided track signatures, - a system to be tested is provided which enables the position of a railway vehicle on the railway track to be determined without the use of track signatures or without the use of the track signature or group of track signatures on which the track position determination reference system is based, - the system to be tested also determines the current track position of the rail vehicle, - the result of the track position determination by the track position determination reference system is compared with the result of the track position determination by the system to be tested, - the accuracy of the track position determination by the system under test is assessed based on the track position determination by the track position determination reference system and - the track position determination reference system, as a permanent evaluation system (monitoring) of a train localisation unit, determines an accuracy and, if a threshold value is exceeded, transmits a warning and / or message to the train driver and / or the train protection system and / or the control centre. [2] Method according to claim 1, characterized by that an automatic log of deviations or a log of deviations with times and / or positions is created, stored and / or transmitted to a central location.

Citation Information

Patent Citations

  • Method and device for monitoring operation of a vehicle, in particular a rail vehicle

    DE102007038734A1

  • Method for determining position of e.g. suburban train, involves adjusting positions of rail vehicle in region of rail track from overrunning balise to successive balise based on exact position of overrunning balise

    DE102009042359A1

  • Method for determining position of rail vehicle by satellite navigation system, involves providing receiver unit in retainer at track, and comparing recorded data in retainer with stored data of landmarks

    DE102012215533A1

  • Method for localization of rail vehicle in rail network, involves matching corresponding period of reference signal of route section of rail network with selected portion of actual time signal

    DE102012219111A1

  • Method for the automatic detection of characteristic elements, in particular of a level crossing, and device for this purpose

    DE102013104088A1