Method for determining position of track-guided vehicle

Abstract

The invention relates to a method for determining the position of a track-guided vehicle. A sensor on the vehicle side generates a measurement result while the vehicle travels, which measurement result represents a track over which the vehicle travels, and determines a current position of the vehicle on the track in a computer-aided manner on the basis of the measurement result. The measurements include a pitch feature that describes a variation in the pitch of the sensor from a superstructure having a track and a ballast bed thereof based on a travel traveled by the vehicle. From a known reference position, the vehicle is relatively positioned in a computer-assisted manner. Models repeating after the passage of a predetermined travel section are identified and counted in the pitch feature in a computer-aided manner. In order to determine the position of each identified model, the respective length of the stroke segment is added to a reference position.

Description

Technical Field

[0001] This invention includes a method for determining the position of a track-guided vehicle. It also includes a vehicle for track-guided traffic. Furthermore, it includes a computer program product containing program instructions. Finally, it includes a computer-readable storage medium containing data. Background Technology

[0002] For smooth operation, railway operators require platforms equipped with platform screen doors (also known as PSDs) to have a large number of stopping points (e.g., more than 7 stopping points), high stopping accuracy (e.g., less than 30 centimeters), and high operational flexibility in both directions of travel.

[0003] To approach each stopping position, the vehicle's control, such as the ATO (Automatic Train Operation) component (which operates, for example, based on the principles of the European Train Control System, or CBTC), needs to read data from multiple fixed transponders (hereinafter referred to as transponders) laid in the track. These transponders must be installed in fixed windows in the space preceding the stopping point in the track. As the vehicle's transponder antenna passes the transponders, the vehicle's absolute position is constantly determined, and (e.g., via odometers, radar) the measurement error relative to the measured position is reset based on the distance traveled. If all the windows for the stops to be configured are considered, there will be a large number of transponders at each station. This results in a considerably high project processing cost.

[0004] To eliminate the need for transponders, it is known to locate vehicles on a track segment by scanning the track segment with sensors, such as radar or optical imaging devices, and then comparing the sensor results with a reference result that has been pre-recorded and stored to identify the relevant location and thus locate the vehicle. For example, document EP 795455A1 describes that, to perform this method at a station, the track bed can be scanned and the track bed image obtained in this way can be compared with a reference image. However, a drawback of this method is that the risk of altering the current track bed image is particularly high, especially in station areas, because, for example, passengers may throw garbage into the track bed. Therefore, there is a risk of location failure using this method. Although it is suggested, according to document DE19529986 A1, that additional, particularly prominent imaging structures outside the track, including the station, can be used for location, this method is also difficult to implement, especially on platforms, because such objects may be obscured by passengers waiting on the platform.

[0005] Furthermore, existing methods involve relatively high equipment costs and computational demands. It must be considered that if station positioning is to be implemented for all vehicles, each vehicle must be equipped with the necessary sensors and computing power for the process. This expense contrasts sharply with the cost of equipping all stations with transponders, thus quickly rendering even state-of-the-art solutions uneconomical. Summary of the Invention

[0006] The technical problem to be solved by this invention is to eliminate the problems described in the prior art. This technical problem is solved in particular by a method that allows for the positioning of track-guided vehicles, especially in station areas, with reasonable hardware and computing power expenditure and sufficient stopping accuracy. The stopping accuracy should be particularly sufficient for stopping on platforms with platform doors. Furthermore, according to the invention, the technical problem is also solved by a vehicle, a computer program product, and a computer-readable storage medium, which allow for the execution of the improved method.

[0007] According to a first aspect of the invention, a method for determining the position of a track-guided vehicle is described, wherein,

[0008] a) Sensors on the vehicle side generate measurement results as the vehicle moves, and these measurement results represent the track traversed by the vehicle.

[0009] b) Based on the measurement results, the computer-aided determination of the vehicle's current position on the track.

[0010] This is a positioning method that the vehicle can autonomously perform using sensors on the vehicle side. Therefore, it does not rely on data connections to track-side infrastructure to perform the positioning method.

[0011] If the device has at least one computer or processor, then the device is computer-aided or computer-implemented; or if at least one computer or processor performs at least one method step of the method, then the method is computer-aided or computer-implemented.

[0012] A computing environment is a type of IT infrastructure that consists of functional components such as processors, storage units, and programs, as well as data to be processed by those programs, used to execute at least one application that must complete a task. Other functional components may include sensors and actuators, enabling the computing environment to interact with the outside world. IT infrastructure can also be organized as a network of these functional components.

[0013] Computational instances form functional units within a computing environment. These functional units can be associated with applications (e.g., provided by multiple program modules) and can be executed. When the application is executed, these functional units form physically (e.g., computer, processor) and / or virtually (e.g., program modules) independent systems.

[0014] A computer is an electronic device with data processing capabilities, composed of multiple functional components. A computer can be, for example, a client, server, handheld computer, communication device, and other electronic devices for data processing. It may have a processor and storage unit and may also be connected to a network via an interface.

[0015] A processor can be, for example, a converter, a sensor for generating measurement signals, or an electronic circuit. A processor can be a main processor (central processing unit, CPU), a microprocessor, a microcontroller, or a digital signal processor, possibly combined with storage units for storing program instructions and data. A processor can also be understood as a virtualized processor or a soft CPU.

[0016] Storage units can be designed as computer-readable storage in the form of working memory (random access memory, RAM) or data storage (hard disk drive or data carrier).

[0017] A program module is a separate software functional unit capable of implementing a sequence of program steps according to the method steps of the invention. These software functional units may be implemented in a single computer program or in multiple computer programs communicating with each other. The interfaces implemented herein may be implemented in software within a single processor, or in hardware if multiple processors are used.

[0018] Interfaces can be implemented in hardware, such as wired or wireless connections, or in software, such as as interactions between program modules of one or more computer programs.

[0019] To avoid misunderstanding, it should be noted that the features of each claim are numbered with lowercase Latin letters, regardless of the claim number. This means that each letter appears only once in the entire claim set, thus allowing the relevant claim features to be clearly addressed without specifying claim numbers. However, this is precisely why the order of the letters is not important.

[0020] According to the present invention,

[0021] c) The measurement results include spacing characteristics (also known as height characteristic distribution), which describe the change in the distance between the sensor (which forms a fixed point on the vehicle) and the superstructure having the track and its track bed, based on the distance traveled by the vehicle.

[0022] d) And starting from a known reference position, the vehicle is relatively positioned with computer assistance, wherein the computer assists in identifying and counting repeating models after passing a predetermined travel segment in the spacing feature, and in order to determine the position of each identified model, the corresponding length of the travel segment is added to the reference position.

[0023] In other words, the present invention alternatively proposes a method for determining the relative vehicle position between a sensor for distance measurement, precisely fixed to the vehicle, and the superstructure via continuous, for example, laser-based, distance measurements. A height characteristic distribution of distance variation is generated based on the distance measurements. This height characteristic distribution is two-dimensional and is created based on one-dimensional measurements, i.e., distance values ​​extending along a route in the vehicle's direction of travel.

[0024] The track is the runway for rail vehicles. A track consists of two steel rails, sleepers that support the rails and connect them as direct support (also called the ballast), and fasteners securing the rails to the ballast. Together with the lower supports and the ballast bed, the track forms the superstructure known as the depot.

[0025] In particular, starting from a known reference position, the vehicle is relatively positioned with computer assistance, thereby identifying and counting repeating models in the spacing features after passing through a predetermined travel segment of constant length (at least in a certain route segment), and in order to determine the position (at least in the route segment), the number of identified models is multiplied by the length of the route segment and added to the reference position.

[0026] Typically, the travel segment has a constant length. This is, for example, defined by the sleeper spacing of the track (described in more detail below). If the travel segment has a constant length, only the length of the relevant station needs to be known in order to pre-set it in the method. However, different lengths of travel segments may also need to be considered. This is the case, for example, if the platform is first on both sides of a straight section and then a curve in the track, where the length of the track segment in the curve is different from the length of the track segment in the straight section. In this exemplary case, both different lengths of travel segments must be considered, along with the location where the travel segment length changes, so that this location can also be pre-set in the method.

[0027] The advantage of this invention is that the distance sensor uses an inexpensive sensor because it only needs to measure a one-dimensional distance between the vehicle and the superstructure. Consequently, the method for determining the height characteristics also advantageously requires very little computational power, as it can be determined by considering the sampling frequency and the temporal order of the various distance values ​​for vehicle speed.

[0028] It should be noted that distance measurement need not be performed in a safety-related manner, only for ATO. Safety-related position detection for ATP (Automatic Train Protection) is performed using methods already implemented in safety-related train control, but with lower positioning accuracy requirements compared to stopping at platform doors. Door release continues based on ATP position generation. Positioning via the track's height characteristics allows the doors to be aligned with the PSD as precisely as possible. When safety is mentioned in the description of this invention, it means functional safety (also known as security), which describes the reliability of the railway system during operation. This does not imply safety against unauthorized attacks (also known as security).

[0029] Due to varying passenger numbers, the vehicle may deflect by an offset (in vehicles without automatic load compensation). However, the offset only results in a constant change in the distance to the vehicle. The height of the rail fasteners relative to the substructure is unaffected by the profile and can still be identified as a model without being affected.

[0030] The advantages of this invention can be summarized as follows:

[0031] The number of transponders in a station can be significantly reduced. By using multiple stop windows, the number of transponders in the station area can be reduced from >25 to 1-2, ensuring safe positioning of the ATP.

[0032] - The additional onboard hardware required is limited to sensors with electronic evaluation units located downstream.

[0033] - Due to the sensor's clear identification of the model, the measurement inaccuracies of the last-pass transponder as the basis do not add any further significant inaccuracies.

[0034] - The relatively small distance between the components on each side of the line (approximately 60 cm) means that measurements taken using odometers and radar in parallel will produce almost no relative measurement error on this section of the road.

[0035] - Sensor-based measurements are independent of wheel slippage during braking / acceleration.

[0036] - The height change caused by the vehicle "deflection" when fully loaded with passengers is robust.

[0037] According to another aspect of the invention, a vehicle for rail-guided transportation is described, the vehicle having a computing environment configured to support vehicle control. According to the aspects described herein,

[0038] e) A distance sensor is mounted on the vehicle, particularly on the underside of the vehicle, and the distance sensor is configured to measure the distance between the sensor and a tracked superstructure below the vehicle.

[0039] f) Configure the computing environment to perform at least steps b) and d) of the aforementioned methods.

[0040] The advantages relating to this aspect of the invention have been explained above, and reference has been made to these advantages.

[0041] According to another aspect of the present invention, a computer program product is described, comprising program instructions executable by a computing environment. As specified in the aspects described herein, at least steps b) and d) of the foregoing method are performed.

[0042] According to the present invention, a computer program product comprising program modules is therefore described by program instructions, whereby the program modules can run on the same or multiple processors. By means of a computer program product comprising one or more computer programs, the method according to the present invention and / or exemplary embodiments thereof can be performed, and the aforementioned advantages are achieved through this performance.

[0043] According to another aspect of the invention, a computer-readable storage medium is described, which contains data stored as a dataset by said storage medium. According to the described aspect of the invention, the dataset is capable of running computer program products of the aforementioned type.

[0044] Therefore, a provisioning device is described for storing and / or providing computer programs in the form of a computer-readable storage medium. The provisioning device is, for example, a storage unit that stores computer programs and makes them retrievable. Alternatively or additionally, the provisioning device is a network service, a computer system, a server system, particularly a distributed system, such as a cloud-based computer system or a virtual computer system, which stores computer programs on a computer-readable storage medium and preferably provides them in the form of a data stream.

[0045] This is provided as a computer program file, particularly as a downloadable file, or as a data stream, particularly as a downloadable data stream, in the form of a program dataset describing program modules. For example, the computer program is transmitted to a computing environment using the providing device, enabling the method according to the invention to be executed in one or more computing instances within that computing environment.

[0046] Design scheme of the present invention

[0047] The following description explains and illustrates further variations of the invention, without limiting the basic ideas of the invention.

[0048] According to one variant, the above-mentioned aspects of the invention are specified such that the spacing variation curve when the track passes over the sleepers and / or rail fastening devices on the roadbed formed by the superstructure is identified as a repeating model.

[0049] Therefore, in the sense described in this invention, the spacing variation curve should be understood as a curve of variation in spacing characteristics, such as those produced when a rail passes over a sleeper or rail fastening device. Thus, the spacing variation curve is part of the spacing characteristics (also called height characteristics), which are repetitive and can be identified by the method according to the invention.

[0050] The advantage of this variant is that the mentioned spacing variation curves repeat regularly; that is, the sleepers or fasteners fixed to this or another subgrade support the rails at regular intervals, and are therefore very suitable for distance measurement according to the invention. These structures are also large compared to the size of the rails, making them easier to identify even at higher speeds. It must be considered that trains can enter stations at speeds up to 80 km / h, therefore reliable model identification must be guaranteed according to the permitted entry speed.

[0051] Upon entering a station, starting from the last read absolute position, the spacing characteristics are measured directly alongside the rails and continuously compared with the track characteristics stored inside the vehicle. Each passing rail fastener and sleeper (if present) represents a significant change in distance, which is identified as a model. The precise position of each rail fastener / sleeper is stored in the track characteristics on the vehicle side (at least if they are not constant) and transmitted to the ATO (Automatic Train Operation System) upon detection. However, it can also be assumed, for example, that the distance between rail fasteners / sleepers (corresponding to the length of the travel segment) is approximately 60 cm. The number of identified models must then be multiplied by this length, making the calculation process particularly fast and consuming particularly low computational and storage resources. To determine positioning, the vehicle can also use position information provided by the odometer, information based on height characteristics (derived from the positions of detected rail fasteners and compared with the expected positions), and the distance between the current and last rail fasteners (if they are not constant) which can be considered separately.

[0052] According to one variant, the above-mentioned aspects of the invention are specified such that the sensor repeatedly or continuously measures the distance from the measurement point to the upper structure below the vehicle, wherein the measurement result includes a two-dimensional, oriented spacing feature of multiple measurement points along the vehicle's driving direction.

[0053] The advantage of this variant is that it establishes a closed spacing characteristic representing the travel distance. This spacing characteristic is obtained by interpolating the measurement points and is therefore almost continuous. The measurement points must be close enough that the resolution of the spacing characteristic allows the model to be imaged in a way that can be recognized by a computer. The sensor sampling rate must be selected to take into account a sufficient density of measurement points and must be in the range >10kHz to provide sufficient resolution for trackside components (rail fasteners, sleepers), with typical entry speeds up to 80 km / h.

[0054] The advantage lies in using machine learning algorithms to optimize the stored route features. Repeatedly driving through the same route segments generates training data. The result of this training is a more accurate and less prone-to-fault location detection function.

[0055] According to one variation, the above-mentioned aspects of the invention are specified such that the sensor measures the spacing relative to one of the rails with a constant lateral offset, such that the measured spacing characteristics include, as a model, measurement points for rail fastening devices on the roadbed.

[0056] In the context of this invention, lateral offset should be understood as the distance from the rail, which is measured horizontally and perpendicular to the vehicle's direction of travel. This means that the lateral offset remains substantially constant as the vehicle moves on the track. Furthermore, the lateral offset is chosen to cover the rail fastening devices on the roadbed during measurement. Therefore, these are included as a model in the formed spacing features. This model is advantageously particularly characteristic. It typically includes supports for the fastening clamps for the rails and associated fastening screws, which quickly produce several feature jumps in the height features. Each time this feature is identified, the travel segment extending from the sleeper to the sleeper is identified.

[0057] One or more distance sensors are required as hardware on the bogie, with their measurement direction pointing vertically downwards. It is advantageous to install them near the wheels to ensure they are positioned above the rail fasteners on the track side, especially in curves.

[0058] According to one variation, the above aspects of the invention are specified such that the method is performed while the vehicle is entering the station until it stops, particularly when the vehicle is simultaneously aligned with platform screen doors installed on the station platform.

[0059] The advantage of this variant is that the positioning accuracy of the train in the station must be exceptionally precise compared to an idle track. This is the only way to ensure that the carriage stops at the position expected by passengers during passenger transport. This is especially important when using platform screen doors, as the train's side doors must be aligned with the platform screen doors. Therefore, the ATO (Automatic Train Protection) supports the precise stopping position of the train on the platform, while the ATP (Automatic Train Protection) remains responsible for door release.

[0060] According to one variation, the above aspects of the invention are specified such that the reference position is determined by the vehicle and / or transmitted to the vehicle before the vehicle enters the station.

[0061] The advantage of this variant is that the closer the reference position for precise positioning is to the station, the more accurate the positioning on the platform. Positioning can be performed externally to the vehicle, for example, using a satellite (GNNS system). Another possibility is via a transponder, which sends a telegram for positioning. In these cases, the reference position, or at least one identifier used to determine the reference position, is sent to the vehicle. If the vehicle uses its own sensors to determine its absolute position, such as by detecting road signs, the reference position is determined by the vehicle itself.

[0062] According to one variant, the above-mentioned aspects of the invention are specified such that the length of the travel segment is stored in the vehicle's storage unit together with the lengths of other track segments for stations passed through and / or travel segments for other stations, and the length of the travel segment is selected before entering the station, taking route planning into account, or the length of the travel segment is transmitted to the vehicle via an external interface before entering the station.

[0063] The advantage of this variant is that reliable positioning is possible even at stations where the travel segments do not conform to standard lengths. This is applicable, for example, to uneven sleeper spacing if the travel segments are specified by sleepers or elements on the sleepers, such as screws or fasteners. In extreme cases, even a separate length can be preserved for each travel segment. Another application arises when travel segments of varying lengths are used along the platform, where the travel segments within the track have the same length. This might be the case, for example, if there are straight-line segments and curved segments on the platform serving as track segments. In this case, each track segment only needs to preserve the uniform length of the relevant travel segment.

[0064] According to one variant, the above aspects of the invention are specified such that the reference position and / or the length of the travel segment are transmitted to the vehicle via a transponder installed in the track within the station's entry area.

[0065] The advantage of this variant is that standardized line elements can be used with transponders (especially Eurobalises), allowing communication with the vehicle to utilize existing hardware. Positioning using transponders can also be performed with high precision. Vehicles without stored travel segment lengths can also advantageously operate using the method according to the invention if the length of the travel segment is also transmitted by the transponder. These are only transmitted upon entering the station. This advantageously makes the method applicable to all vehicles equipped with the hardware (vehicle-side sensors) or software (computer program) according to the invention, without local limitations.

[0066] According to one variant, the above aspects of the invention are specified such that machine learning is used to set up and / or optimize the method during vehicle operation, and the model is created and / or optimized in the machine learning.

[0067] The advantage of this variant is that the process can be optimized if the vehicle repeatedly travels to certain stations. When the vehicle repeatedly approaches these stations, the specific characteristics of the travel segment lengths to those stations can be taken into account.

[0068] In the context of this invention, artificial intelligence (hereinafter also abbreviated as KI), or artificial intelligence in a narrow sense, is understood as the ability of computer-aided machine learning, also known as machine learning (hereinafter also abbreviated as ML). This involves statistical learning with algorithmic parameterization, and is preferably applicable to very complex applications. Using ML, a system identifies and learns models and patterns in recorded process data based on previously input learning data. With appropriate algorithms, ML can be used to independently find solutions to emerging problems. ML is divided into three domains—supervised learning, unsupervised learning, and reinforcement learning—with more specific applications such as regression and classification, structure recognition and prediction, data generation (sampling), or autonomous behavior.

[0069] In supervised learning, the system is trained by connecting known input data with relevant outputs, thereby learning an approximate functional relationship. Here, the availability of suitable and sufficient data is crucial, as the system will learn incorrect functional relationships if trained with inappropriate (e.g., unrepresentative) data. In unsupervised learning, the system is also trained using sample data, but only the input data, independent of known outputs. It learns how to form and expand data sets and where biases or anomalies occur, which is typical for relevant applications. This allows it to describe the application and identify error states. In reinforcement learning, the system learns through trial and error, proposing solutions to a given problem and receiving positive or negative ratings for the suggestions through a feedback function. Based on the reward mechanism, the AI ​​system learns to perform appropriate functions.

[0070] According to one variation, the above aspects of the invention are specified such that an optical distance sensor, particularly a laser-based distance sensor, is installed as a sensor.

[0071] The advantages of this variant are that the optical sensors are inexpensive to purchase and reliable in operation. In particular, using laser-based distance sensors allows for the generation of measurements at the desired frequency, thus enabling the generation of measurement points at the desired density, as already explained.

[0072] According to a variant, the above aspects of the invention are specified such that the computing environment of the vehicle has a computing instance configured to perform automatic train operation (hereinafter also referred to as ATO), and the computing instance is further configured to perform at least steps b) and d) of the aforementioned method.

[0073] The advantage of this variant is that the ATO can locate the vehicle with sufficiently high accuracy to stop it on the platform. The device with vehicle-side sensors according to the invention can advantageously support the ATO in accomplishing this task by performing the method according to the invention. Attached Figure Description

[0074] Further details of the invention are described below with reference to the accompanying drawings. Identical or corresponding elements have the same reference numerals in the various drawings and are interpreted multiple times only where there are differences between the various drawings.

[0075] The exemplary embodiments described below are preferred embodiments of the invention. In the exemplary embodiments, each component of the described embodiments represents a different variant of the invention that can be viewed independently of each other, and which also independently develops the invention and is therefore considered part of the invention individually or in combinations different from those shown. Furthermore, the described components can also be combined with the above-described variants of the invention.

[0076] Figure 1 The exemplary embodiments of the device according to the invention and the causal relationships between the functional components used are illustrated schematically.

[0077] Figure 2 An example of the travel segment of the superstructure and the spacing characteristics obtained from the measurement results as an exemplary embodiment of the method according to the invention are shown in the longitudinal sectional view.

[0078] Figure 3 Showing according to Figure 1An exemplary embodiment of the computing environment of the device is shown as a block diagram of the various functional components and the interfaces formed therebetween, wherein each computing instance executes a program module, each program module may run on one or more computers as an example, and wherein the interface shown may be implemented accordingly using software on one computer or using hardware across different computers.

[0079] Figure 4 The flowchart illustrates an exemplary embodiment of the method according to the invention, wherein the illustrated method steps can be implemented individually or in groups by program modules, and wherein, according to Figure 3 The computational instances and interfaces are illustrated in the examples. Detailed Implementation

[0080] Figure 1 The vehicle FZ is shown, which travels or remains stationary on track GL via wheels RD. Track GL consists of rails SC, which are secured to sleepers SW by fastening bolts BS. Track GL is also housed in a ballast stock ST that forms the track bed, wherein track GL and ballast stock ST form the superstructure OB.

[0081] As vehicle FZ travels on track GL, the distance 'a' to the surface of the superstructure OB is measured by sensor SN. This distance 'a' is measured perpendicularly along the z-direction and indicated by a coordinate system with its origin located within sensor SN. The train travels along the x-direction, which is perpendicular to the coordinate system defined by sensor SN. Figure 1 The driving direction is the same in the example. Additionally, the sensor SN is along... Figure 1 The y-direction shown is mounted on the vehicle's FZ, and... Figure 1 The track SC shown is laterally offset.

[0082] The sleepers SW each have a first length L1 defining a first travel segment WS1, a second length L2 defining a second travel segment WS2, a third length L3 defining a third travel segment WS3, and a fourth length L4 defining a fourth travel segment WS4. To understand the positioning process according to the invention, it is necessary to understand the relationship between the travel segments and the lengths L (the sum of the lengths is independent of whether there is only one or more lengths on the platform). Once the vehicle FZ passes the transponder BL, the transponder BL is used to perform high-precision absolute positioning, i.e., the determination of the reference position RPOS. Subsequently, each passing sleeper SC is identified by the sensor SN, and thus the first travel segment WS1, second travel segment WS2, third travel segment WS3, and fourth travel segment WS4 are recorded sequentially. To perform relative positioning, after passing the first travel segment WS1, the first length L1 is added to the absolute position determined by the transponder BL; after passing the second travel segment WS2, the second length L2 is added; after passing the third travel segment WS3, the third length L3 is added; after passing the fourth travel segment WS4, the fourth length L4 is added, and so on.

[0083] POS=RPOS+L1+L2+L3+L4 (1)

[0084] If a separate length L exists for each travel segment, this example of the processing order according to the invention is used, where the length L is stored, for example, in the vehicle FZ. When calculating the current position of the vehicle FZ, a different length L is invoked individually. As described, the length L can be determined by machine learning or by precise measurement of the superstructure OB.

[0085] Instead of the individualized length L of the travel segment, a uniform (e.g., average) length L of the travel segment can also be used to calculate the relative position. In the example shown, this is the average sleeper distance SW. In this case, the method is simplified because after sensor SN detection, only the crossed sleepers SW need to be counted and multiplied by the average length L of the travel segment. The result is then added to the absolute reference position RPOS determined by crossing the transponder BL in the manner described.

[0086] POS = RPOS + N*L (2)

[0087] Figure 2 A more detailed cross-section of the superstructure OB is shown. The sleeper SW can be seen, and the rail SC is secured to the sleeper SW with fastening screws BS. Fastening clamps BK are also used here for clarity. Figure 1 Not shown in the image.

[0088] As can be seen, the spacing feature AP is shown at the same scale as the cross-section of the superstructure OB. This spacing feature is determined by the sensor SN, as shown below. Figure 1As shown. Here, the average length L of the travel segment is as described in more detail above. Repeating models MT can be seen in the spacing feature AP, which correspond to the corresponding spacing feature AP (i.e., height feature) of the sleeper SW, including the clamping clamp BK and the fastening screw BS. Due to their unique characteristics, these MT models can be easily identified with computer assistance. It can also be seen that the spacing feature AP of the stack ST is non-uniform, resulting in no repeatable variation curves for the spacing feature AP appearing here.

[0089] The computing environment RU that runs the method according to the invention can be referenced in conjunction with [other methods]. Figure 1 and Figure 3 The Automatic Train Protection (ATP), Automatic Train Operation (ATO), and Sensor Module (SNM), as well as the computing instances used in the functional components, interact with each other through interfaces. The vehicle-side sensor SN and the fourth processor PR4 are connected to each other via the first interface S1. The fourth processor PR4 and the first processor PR1 are connected to each other via the second interface S2. The third processor PR3 and the first processor PR1 are connected to each other via the third interface S3. The first processor PR1 and the second processor PR2 are connected to each other via the fourth interface S4. Through the fifth interface S5, when vehicle FZ passes transponder BL, the third processor PR3 and transponder BL can connect to each other via a radio interface. For this purpose, vehicle FZ and transponder BL are equipped with antennas AT capable of enabling this transmission.

[0090] according to Figure 3 The computer forming the computing instance is shown in more detail. In the first computer CP1, the first processor PR1 is connected to the first storage unit SE1 via the eleventh interface S11. In the second computer CP2, the second processor PR2 is connected to the second storage unit SE2 via the twelfth interface S12. When referring to a computer, processor, memory unit, or interface in the context of this invention description, this information generally refers to all the computers, processors, memory units, and other functional components detailed above, which are connected via interfaces to facilitate the formation of the computing environment RU.

[0091] The method according to the present invention will be explained step by step by example, as follows: Figure 4 The flowchart is shown. Figure 4 The box also indicates an example, where according to Figure 1 and Figure 3 The functional components and computing instances can perform the individual steps. Computer-aided steps occur in a processor not shown in detail. As an example, reading data and storing it in a storage unit is shown. This is based on... Figure 1 and Figure 2 In terms of interfaces, they are also Figure 3 The middle mark.

[0092] In the first step 1, the method begins (abbreviated as: start).

[0093] In the second step, 2, the vehicle FZ (LOC) is located. Here, the reference position RPOS is determined and stored in the first storage unit SE1 via the eleventh interface S11. As described, the reference position RPOS can be determined by passing the transponder BL.

[0094] In step 3, the distance a (abbreviated as MSR) from the sensor SN to the upper structure OB is measured. As described, the sum of the measured distances a generates a distance feature AP, which serves as the basis for further methods. The distance feature AP can be generated, for example, in a fourth processor PR4 and transmitted to the first processor PR1 via a second interface S2 in a manner not shown.

[0095] In step 4, the first processor PR1 queries whether the model MT (abbreviated as MT?) was identified in the previous spacing feature AP. To this end, model recognition is performed in the first processor PR1. For this purpose, known computer-aided methods for model recognition are used, and model recognition can preferably be trained using machine learning through artificial intelligence. Training can be performed according to... Figure 4 The method is performed beforehand and / or improved during its execution to optimize model recognition of MT. Furthermore, when using a route map atlas, machine learning can be used to learn the individual lengths L of the travel segments at each station, for example, the length L preset by the sleepers SW of the track GL. If the model MT is identified, proceed to step 7. If the model MT is not identified, proceed to step 5.

[0096] In step 5, it asks whether the method should be stopped (STP?). This is the case, for example, if vehicle FZ has already stopped. If necessary, the method will begin again after step 1 when vehicle FZ continues moving. If the method is not stopped, recursion occurs and step 3 continues. If the method is stopped, step 6 (STOP?) is performed. This step stops the method.

[0097] In step 7, inquire whether the personalized length Li can be used for a single travel segment or a group of travel segments (abbreviated as Li?). If so, proceed to step 10. If not, proceed to step 8.

[0098] In step 8, the count variable N is increased by one (abbreviated as N = N + 1). If there are no individual lengths Li for each travel segment, and only a standardized length L is assumed, then this count variable should be used. This standardized length L represents, for example, the average length L of the travel segment or a precise length L within a tolerance range. For example, this length L could correspond to a uniform sleeper distance SW on the rail.

[0099] In step 9, the current position is calculated in the calculation step according to equation (2) (abbreviated as CLC). For the purpose of calculation, the normalized length L and the applicable position are read out, that is, the reference position RPOS from the first storage cell SE1 via the eleventh interface S11 (if this step is performed for the first time) or the last calculated position (if this step is performed repeatedly).

[0100] In step 10, alternatively (i.e., in another case according to the query in step 7), the current position (CLC for short) is calculated in the calculation step according to equation (1). For the purpose of calculation, the current length Ln and the applicable position are read, i.e., the reference position RPOS from the first storage unit SE1 via the eleventh interface S11 (if the step is performed for the first time) or the last calculated position (if the step is performed repeatedly).

[0101] In step 11, the calculated position (OT-POS) is output. The data record containing the position can be transmitted, for example, via the fourth interface S4 to the second processor PR2 to ensure the functionality of the ATP in the vehicle FZ. Then, proceed to step 5 as described above.

[0102] List of reference numerals

[0103] a Spacing

[0104] AP Spacing Features

[0105] AT antenna

[0106] ATO (Automatic Train Operation)

[0107] ATP (Automatic Train Protection)

[0108] BK Fastening Clip

[0109] BL transponder

[0110] BS fastening screws

[0111] CP1 First Computer

[0112] CP2 Second Computer

[0113] FZ vehicles

[0114] GL orbit

[0115] L length

[0116] L1 First Length

[0117] L2 Second Length

[0118] L3 Third Length

[0119] L4 Fourth Length

[0120] MT model

[0121] OB superstructure

[0122] POS location

[0123] PR1 First Processor

[0124] PR2 Second Processor

[0125] PR3 Third Processor

[0126] PR4 Fourth Processor

[0127] RD wheels

[0128] RPOS Reference Location

[0129] RU computing environment

[0130] S1 First Interface

[0131] S11 Eleventh Interface

[0132] S12 Twelfth Interface

[0133] S2 Second Interface

[0134] S3 Third Interface

[0135] S4 fourth interface

[0136] S5 fifth interface

[0137] SC rails

[0138] SE1 First Storage Unit

[0139] SE2 Second Storage Unit

[0140] SN sensor

[0141] SNM sensor module

[0142] ST stack

[0143] SW sleepers

[0144] WS1 First travel segment

[0145] WS2 second stroke segment

[0146] WS3 third stroke segment

[0147] WS4 fourth stroke segment

Claims

1. A method for determining the position of a track-guided vehicle (FZ), wherein, a) Sensors (SN) on the vehicle side (FZ) generate measurement results as the vehicle (FZ) moves, the measurement results representing the track (GL) traversed by the vehicle (FZ). b) Based on the measurement results, the computer-aided determination of the current position of the vehicle (FZ) on the track (GL) is performed. Its features are, c) The measurement results include a spacing feature (AP), which describes the change in the distance (A) between the sensor (SN) and the superstructure (OB) having the track (GL) and its track bed, based on the distance traveled by the vehicle (FZ). d) Starting from a known reference position (RPOS), the vehicle (FZ) is relatively positioned with computer assistance, wherein the model (MT) that repeats after a predetermined travel segment is identified and counted in the spacing feature (AP), and in order to determine the position of each identified model (MT), the corresponding length (L) of the travel segment is added to the reference position (RPOS).

2. The method according to claim 1, characterized in that, The spacing variation curves as the track passes over the sleepers (SC) and / or rail (SC) fastening devices on the roadbed formed by the superstructure (OB) are identified as repeating models (MT).

3. The method according to claim 1 or 2, characterized in that, The sensor (SN) repeatedly or continuously measures the distance (A) from the measurement point to the superstructure (OB) below the vehicle (FZ), wherein the measurement result includes a two-dimensional spacing feature (AP) of multiple measurement points pointing in the direction of travel of the vehicle (FZ).

4. The method according to claim 3, characterized in that, The sensor (SN) measures the spacing (A) relative to one of the rails (SC) with a constant lateral offset, such that the measured spacing feature (AP) contains, as a model (MT), the measurement points for the rail (SC) fastening devices on the roadbed.

5. The method according to any one of the preceding claims, characterized in that, The method is performed while the vehicle (FZ) enters the station until the vehicle (FZ) stops, especially when the vehicle (FZ) is simultaneously aligned with the platform screen doors installed on the station platform.

6. The method according to claim 5, characterized in that, The reference position (RPOS) is determined by the vehicle (FZ) and / or transmitted to the vehicle (FZ) before the vehicle (FZ) enters the station.

7. The method according to any one of claims 5 or 6, characterized in that, The length (L) of the travel segment is stored in the storage unit of the vehicle (FZ) together with the lengths (GL) of other track segments for stations passed through and / or the lengths (L) of travel segments for other stations, and the length of the travel segment is selected before entering the station with route planning in mind, or the length of the travel segment is transmitted to the vehicle (FZ) via an interface outside the vehicle (FZ) before entering the station.

8. The method according to claim 5, characterized in that, The reference position (RPOS) and / or the length (L) of the travel segment are transmitted to the vehicle (FZ) via a transponder (BL) installed on the track (GL) within the station's entry area.

9. The method according to any one of the preceding claims, characterized in that, Machine learning is used to set up and / or optimize the method while the vehicle (FZ) is running, and the model (MT) is created and / or optimized in the machine learning.

10. A vehicle for rail-guided transportation, the vehicle having a computing environment (RU) configured to support vehicle (FZ) control, characterized in that: e) A distance sensor is mounted on the vehicle (FZ), particularly on the underside of the vehicle, and is configured to measure the distance (A) between the sensor (SN) and the superstructure (OB) with rails (GL) below the vehicle (FZ). f) The computing environment (RU) is configured to perform at least steps b) and d) of the method according to any one of claims 1 to 9.

11. The vehicle according to claim 10, characterized in that, Optical distance sensors, especially laser-based distance sensors, are installed as sensors (SNs).

12. The vehicle according to claim 10 or 11, characterized in that, The computing environment (RU) of the vehicle (FZ) has a computing instance configured to perform Automatic Train Operation (ATO), and the computing instance is further configured to perform at least steps b) and d) of the method according to any one of claims 1 to 9.

13. A computer program product comprising program instructions executable by a computing environment (RU) such that at least steps b) and d) of the method according to any one of claims 1 to 9 are performed.

14. A computer-readable storage medium comprising data stored in the storage medium as a dataset, such that the dataset is capable of running a computer program product according to claim 13.

Images