Method for determining the speed of a vehicle, in particular a rail vehicle, on the basis of a transit time difference and speed error information

Abstract

The invention relates to a method for determining a speed of a vehicle (10), in particular a rail vehicle (10), during travel on a travel path (14), the vehicle (10) comprising two sensor units (16a; 16b) of the same design and the second sensor unit (16b) being mounted on the vehicle (10) at a distance along a direction of travel of the vehicle (10), and the method comprising the following steps: • reading in first and second sensor data captured by means of the first and second sensor units (16A; 16B) during the travel on the travel path (14), the first and second sensor data representing the same physical property of the travel path (14); • determining a transit time difference on the basis of the read-in first and second sensor data, which transit time difference has a value from a set of specified quantized values due to a quantization effect resulting from a specified capture rate of the sensor units (16a, 16b); • reading in specified speed error information which represents a statistical distribution of a speed error; and • determining a speed of the vehicle (10) on the basis of the determined transit time difference and the read-in speed error information by means of a computing unit (20) in order to provide a speed value or speed curve corrected in view of the quantization effect.

Description

[0001] Description

[0002] title

[0003] Method for determining the speed of a vehicle, in particular a rail vehicle, based on a running time difference and speed error information

[0004] The invention relates to a method for determining a speed of a vehicle, in particular a rail vehicle, when traveling on a track, a computing unit, a system and a rail vehicle, as well as a corresponding computer program and a storage medium.

[0005] State of the art

[0006] From DE 10 2022 123 270 A1 a transport system is known which comprises a rail system with pairs of rails which comprise a first and a second rail track, wherein the transport system comprises at least one rail-bound vehicle which is movable on the rail system, wherein the rail-bound vehicle comprises a group of at least three inductive sensors, wherein each inductive sensor of the group is connected to the rail-bound vehicle and is in a fixed spatial relationship to the other sensors of the group, wherein the transport system and / or the rail-bound vehicle comprises an evaluation module for detecting output signals of the inductive sensors of the group, wherein each inductive sensor of the group comprises at least one coil and is aligned with a rail track in such a way that the rail track influences the inductance of the coil in a way that can be detected by the evaluation module.Here, it is proposed to determine a movement variable, in particular a speed of the rail-bound vehicle, by forming a phase profile with support points corresponding to the number of inductive sensors in the group from the output signals of the inductive sensors in accordance with their spatial arrangement, generating a high-resolution phase profile by interpolation, reading in a location-dependent high-resolution phase profile from a map, and determining the movement variable, in particular the speed of the rail-bound vehicle, by comparing the high-resolution phase profile with the location-dependent high-resolution phase profile. This is done in particular by comparing the high-resolution phase profiles in such a way that a corresponding location-dependent high-resolution phase profile is determined for the high-resolution phase profile of the rail-bound vehicle.

[0007] To determine the speed of a vehicle, identical sensor units can be provided, which are arranged at a distance along a direction of travel on the vehicle and are designed to provide sensor data representing a physical property of the route. Based on the sensor data, a propagation time difference between the sensor data of the sensor units arranged at a distance along the direction of travel can be determined. Due to a predetermined sensor acquisition rate Fs, this determined propagation time difference is quantized to 8t. The minimum determinable propagation time difference is 8t. min= 1 / Fs. Accordingly, the speed determined based on the measured time difference is also quantized. The higher the vehicle speed and the smaller the spatial distance between the sensor units along the direction of travel, the more pronounced the quantization effect becomes, meaning the worse the speed resolution.

[0008] Disclosure of the invention

[0009] According to a first aspect, the subject of the present invention is a method, in particular a computer-implemented method, for determining a speed of a vehicle, in particular a rail vehicle, when traveling on a track, according to claim 1.

[0010] According to a second aspect, the subject of the present invention is a computing unit for determining a speed of a vehicle, in particular a rail vehicle, when traveling on a track, according to claim 10.

[0011] According to a third aspect, the present invention relates to a system for determining a speed of a vehicle, in particular a rail vehicle, according to claim 11. According to a fourth aspect, the present invention relates to a vehicle with a system for determining a speed of the vehicle according to claim 12.

[0012] According to a further aspect, the present invention relates to a computer program and a machine-readable storage medium.

[0013] The vehicle can be designed as a road vehicle, for example, as a passenger car or truck. Preferably, the vehicle is designed as a rail vehicle. Within the scope of the present invention, a rail vehicle or railway vehicle can be understood as a vehicle that can be driven on one or more rails. The rail vehicle can be designed, for example, as a locomotive, railcar, multiple unit, control car, power car, commuter train, light rail, subway, or tram. Accordingly, the track can be a road or a track.

[0014] The vehicle comprises a system for determining the speed of a vehicle, in particular a rail vehicle. The system comprises at least a first sensor unit and a second sensor unit, which are designed to provide sensor data when traveling along a track, each sensor data representing the same physical property of the track. Furthermore, the system comprises a computing unit for determining the speed of the vehicle. The system can be designed as a retrofit solution for an existing vehicle.

[0015] Advantageously, the sensor units are arranged on an underside of the vehicle facing the track traveled by the vehicle. For example, the sensor units can be arranged in the area of ​​a bogie of a rail vehicle. It is also conceivable for the sensor units to be arranged on a side oriented transversely to the direction of travel of the vehicle or on an upper side of the vehicle facing away from the track.

[0016] The sensor units are designed and arranged on the vehicle in such a way that the sensor data generated by the sensor units represent a physical quantity which has a temporally substantially constant value for each position or each point of the travel path. In particular, the physical quantity assumes substantially the same value when detected by the first sensor unit and when detected with a time delay by the second sensor unit, wherein a time offset depends on a speed of the vehicle and is less than or equal to 10 s, in particular less than or equal to 1 s, preferably less than or equal to 100 ms.

[0017] The physical property can be, for example, a magnetic field in the area of ​​the track, a curve radius of the track, an unevenness of the track that generates vibrations when traveling along the track, etc. Preferably, the sensor data are in the form of magnetic field data that represent a spatial progression of a magnetic field in the area of ​​the track.

[0018] Accordingly, the sensor unit is preferably designed as a magnetometer, in particular as a flux-gate magnetometer, and / or as an acceleration and / or rotation rate sensor, etc.

[0019] Preferably, the sensor units are configured and arranged on the vehicle in such a way that the magnetic field data acquired by the sensor unit represent the spatial distribution of the Earth's magnetic field along a track traveled by the vehicle, wherein the Earth's magnetic field is characteristically modified by the track and landmark units arranged in the area of ​​the track with one or more (ferro)magnetic landmark elements. In other words, the magnetic field data represent a characteristic spatial distribution, e.g., a magnetic field signature or a "fingerprint."

[0020] The method according to the first aspect of the invention comprises a step of reading in first sensor data acquired by the first sensor unit while traveling along the track and second sensor data acquired by the second sensor unit. The first and second sensor data represent the same physical property of the track.

[0021] The sensor data can represent, in particular comprise, a spatial progression of the physical quantity. For example, the sensor data can comprise values ​​of a magnetic field strength and / or flux density and / or values ​​representing the magnetic field strength and / or flux density, e.g. voltage, current or phase shift values. The sensor data can comprise a time series of sensor values ​​which were recorded by means of the sensor unit when traveling along the travel path. It is conceivable that the physical quantity is recorded periodically when traveling along the travel path, e.g. with a frequency or sampling rate greater than or equal to 10 Hz, preferably greater than or equal to 1 kHz, and less than or equal to 1 MHz. Accordingly, the recorded sensor values ​​can have a time interval of greater than or equal to 1 microsecond and less than or equal to 100 ms, preferably less than or equal to 1 ms. A time series can be a sequence of at least two, preferably more, at or.During a vehicle's journey, the time series refers to sensor values ​​recorded at different times and by the same sensor unit. Taking the vehicle's speed into account, the time series of sensor values ​​corresponds to the spatial progression of the physical quantity along the route.

[0022] Reading in the sensor data, in particular by means of a computing unit, can comprise receiving the sensor data, for example, from the sensor units and / or reading the sensor data from a storage medium, in particular a temporary one. The step of reading in the sensor data can be preceded by a step of capturing the sensor data using the vehicle's sensor units in order to provide the sensor data, for example, to the computing unit for determining the vehicle's speed.

[0023] The method according to the first aspect further comprises a step of determining, in particular calculating, a runtime difference based on the read-in first and second sensor data, in particular by means of the computing unit. Due to a quantization effect caused by a predetermined detection rate of the sensor units, the runtime difference has a value from a set of predetermined quantized values. In the context of the present invention, a runtime difference can be understood as a temporal difference or a time difference which extends between the detection of the physical quantity at an (arbitrary) position on the travel path by means of the first sensor unit and the detection of the same physical quantity at the same position on the travel path by means of the second sensor unit. The runtime difference can, for example,by means of (pairwise) cross-correlation, a least square method or a dynamic time warping (DTW) method based on the first and second sensor data.

[0024] The method according to the first aspect further comprises a step of reading in predefined speed error information, in particular by means of the computing unit. The speed error information represents a statistical distribution of a speed error. The statistical distribution can be designed as a normal distribution or as a distribution deviating from a normal distribution. In the case of a normal distribution, the distribution of the speed error can be represented by specifying a variance or a standard deviation, optionally also by specifying a mean value. In the case of a distribution deviating from a normal distribution, the distribution can be included in the speed error information, in particular can include values ​​of the speed error for one or more predefined speed values.

[0025] The method according to the first aspect further comprises a step of determining, in particular calculating, a speed of the vehicle based on the determined travel time difference and the read-in speed error information by means of the computing unit in order to provide a speed value or speed profile corrected taking into account the quantization effect. Preferably, determining the speed comprises correcting a provisional speed value determined based on the travel time difference, in particular predicted, based on the speed error information.

[0026] The method presented is preferably repeated, in particular periodically, for example once or several times per second, in particular with a frequency greater than or equal to 1 Hz and less than or equal to 100 Hz, during a journey of the rail vehicle, preferably in real time.

[0027] The computing unit is preferably arranged on the rail vehicle. It is conceivable that the computing unit is designed as a control unit of the rail vehicle assigned to the rail vehicle. It is also conceivable that the computing unit is arranged away from the rail vehicle, in particular as part of a cloud computing unit or a server backend. The computing unit can be connected to the sensor units via a wired or wireless communication link.

[0028] The method and devices according to the invention enable a runtime-based determination of the speed of a vehicle with particularly high accuracy. By taking into account the statistical distribution of the speed error, a quantization effect caused by the specified detection rate of the sensor units can be compensated, so that a smoothed speed, in particular a smoothed speed profile, can be provided. By reading in the specified speed error information, it is possible to dispense with determining the speed error information at runtime or while driving along the route, which means that the proposed approach is characterized by particularly short processing times and low computing resource requirements.

[0029] It is advantageous if the predetermined speed error information is embodied as characteristic curve information, which includes an assignment of speed values ​​to a respective variance and / or standard deviation of the corresponding speed error, in order to take the read-in characteristic curve information into account when determining the speed. The characteristic curve information preferably comprises a plurality of predetermined speeds and a respective variance and / or standard deviation of the statistical distribution of the speed error corresponding to the speed. This configuration requires particularly few memory resources to store the speed error information.

[0030] It is advantageous here if the speed is determined using a Kalman filter and the read-in characteristic curve information is taken into account in a correction step of the Kalman filter in order to correct a speed value predicted in a prediction step of the Kalman filter. In particular, the prediction step is independent of the first and second sensor data, in particular independent of data from the first and second sensor units. Preferably, a model for determining a predicted speed value is used in the prediction step, which model is independent of the first and second sensor data, in particular independent of data from the first and second sensor units. This configuration enables particularly efficient speed determination.

[0031] Alternatively, it is advantageous if the predetermined speed error information is embodied as characteristic map information, which includes an assignment of speed values ​​to a respective distribution of the corresponding speed error in order to take the read-in characteristic map information into account when determining the speed. The characteristic map information preferably comprises a plurality of predetermined speeds and a statistical distribution of the speed error corresponding to each respective speed. This configuration allows a particularly precise representation of the speed error to be taken into account when determining the speed, thereby further increasing accuracy.

[0032] It is advantageous here if the speed is determined using a Bayesian filter and the read-in characteristic map information is taken into account in a correction step of the Bayesian filter in order to correct a speed value predicted in a prediction step of the Bayesian filter. In particular, the prediction step is independent of the first and second sensor data, in particular independent of data from the first and second sensor units. Preferably, a model for determining a predicted speed value is used in the prediction step, which model is independent of the first and second sensor data, in particular independent of data from the first and second sensor units. This configuration enables particularly precise speed determination.

[0033] It is also advantageous if the assignment of speed values ​​to the respective variance and / or standard deviation and / or distribution of the speed error is determined before traveling along the route and is / was stored as characteristic curve information and / or characteristic map information. According to this embodiment, the assignment is determined, in particular the speed error information is determined, independently of the speed determination. In particular, there is no need to determine the speed error information while traveling along the route. When determining the assignment, a computing unit can be used whose computing resources are greater than the computing resources of the computing unit configured to execute the proposed method. This embodiment allows the method for determining the speed to be executed even on computing units or control units with limited computing resources.

[0034] It is advantageous here if the assignment is determined based on additional sensor data acquired by the first sensor unit and the second sensor unit and / or based on additional sensor data acquired by one or more further sensor units, wherein the additional sensor data represents the same physical property as the first and second sensor data. For example, the assignment is determined using a database in which a large number of sensor data items are stored, in particular from sensor units of different vehicles and / or from different journeys of the vehicle(s). This configuration makes it possible to provide speed error information that represents the static distribution of the speed error particularly precisely.

[0035] It is also advantageous if the method includes a step of outputting a signal depending on the determined speed of the vehicle, in particular by means of the computing unit, wherein the output signal is designed as an information signal representing the determined speed and / or as a control signal for controlling a unit arranged on or off the vehicle. The output signal can be a wirelessly or wired signal.

[0036] According to a further embodiment, in response to the output signal, an acoustic and / or optical and / or haptic warning is output to a driver and / or operator of the vehicle and / or to a dispatcher in order to inform the driver and / or operator and / or the dispatcher of the determined speed of the vehicle.

[0037] Alternatively or additionally, the unit can be controlled in response to the output signal, which is designed, for example, as a drive unit and / or brake unit of the vehicle, in particular to accelerate and / or decelerate the vehicle, for example, until the determined speed matches a predetermined speed. This configuration allows the determined speed to be transmitted to other (rail-

[0038] ) vehicles and / or an infrastructure facility and / or the driver or operator of the (rail) vehicle in order to further increase the safety during operation of the (rail) vehicle.

[0039] Also advantageous is a computer program product or computer program with program code that can be stored on a machine-readable carrier or storage medium such as a semiconductor memory, a hard disk memory or an optical memory and is used to carry out, implement and / or control the steps of the method according to one of the embodiments described above and below, in particular when the program product or program is executed on a computer or a computing unit.

[0040] In the following, the invention will be explained in more detail with reference to the drawings.

[0041] To show

[0042] Fig. 1 is a schematic representation of a vehicle according to an embodiment;

[0043] Fig. 2 is a flowchart of a method for determining a

[0044] Speed ​​of a rail vehicle;

[0045] Fig. 3 is a representation of a proposed

[0046] approach determined speed curve; and

[0047] Fig. 4 is a flowchart of a method for determining a

[0048] Speed ​​of a vehicle according to one embodiment.

[0049] Fig. 1 shows a vehicle 10 configured as a rail vehicle 10 with a system 12 for determining a speed of the vehicle 10 according to an embodiment of the present invention. The rail vehicle 10 travels along a track 14 comprising a track. The system 12 comprises a first sensor unit 16a, a second sensor unit 16b and a computing unit 18. The sensor units 16a, 16b are configured to measure a physical property of the

[0050] travel path 14 in order to provide the computing unit 18 with sensor data representing the physical property.

[0051] The sensor units 16a, 16b are arranged on the vehicle 10 at a distance from one another along a direction of travel of the vehicle 10. Furthermore, the sensor units 16a, 16b are arranged aligned with one another or one behind the other along a direction of travel of the vehicle 10. Here, the sensor units 16a, 16b are arranged on an underside of the vehicle 10 facing the track 14 traveled by the vehicle 10. For example, the sensor units 16a, 16b are arranged on a bogie 20 of the rail vehicle 10. According to this embodiment, the sensor units 16a, 16b are designed as magnetometers 16a, 16b. Accordingly, the sensor data is designed as magnetic field data, which represents a spatial profile of a magnetic field in the area of ​​the track.In other words, the sensor units 16a, 16b are designed and arranged on the rail vehicle 10 to acquire magnetic field data that are characteristically modified by the earth's magnetic field, e.g., rails or rail fastening means, present on the track 14 and in the vicinity of the track 14. The acquired magnetic field data thus represent a

[0052] A magnetic field signature or "fingerprint" based on which (ferro-)magnetic elements encompassed by the track 14 and / or arranged in the region of the track 14 can be identified. The magnetic field data preferably comprise a temporal or spatial profile of the magnetic flux density B along a direction of travel of the rail vehicle 10 while traveling along the track 14.

[0053] Due to the spaced and aligned arrangement of the sensor units 16a, 16b in the direction of travel of the rail vehicle 10, magnetic field signatures of the same area of ​​the track 14 are included in the magnetic field data of the sensor units 16a, 16b with a temporal offset. The sensor unit 16b, arranged behind the first sensor unit 16a in the direction of travel, is thus designed to detect magnetic field data of an area of ​​the track 14 with a delay relative to the sensor unit 16a. This delay or

[0054] The travel time difference is proportional to a spatial distance between the sensor units 16a, 16b along the direction of travel and inversely proportional to a speed of the vehicle 10. By determining the travel time difference, the speed of the vehicle 10 can be determined for a given distance between the sensor units 16a, 16b.

[0055] The propagation time difference or delay can be calculated, for example, using (pairwise) cross-correlation, a least squares method, or a dynamic time warping (DTW) method based on the first and second sensor data.

[0056] Such a method for determining a speed of the vehicle 10 from Fig. 1 is schematically illustrated in Fig. 2. The method is designated in its entirety by the reference numeral 100.

[0057] In step 110, the magnetic field is periodically recorded by the sensor units 16a, 16b while traveling along the track 14 in order to provide the computing unit 18 with corresponding first and second magnetic field data. In step 120, the magnetic field data are read in. The read-in magnetic field data represent a profile of the magnetic flux density B(t) along the direction of travel of the rail vehicle 10 as a function of time t. Here, a profile of the magnetic flux density B recorded by the second sensor unit 16b b (t) delayed in time compared to the course of the magnetic flux density B detected by the first sensor unit 16a a (t).

[0058] In step 130, a pairwise correlation is calculated between the magnetic flux densities B a (t), B b(t) is determined for each time t. The time difference between the first and second magnetic field data is determined as the temporal value θt for which the correlation function corr(t) reaches its maximum. In step 140, the speed of the vehicle 10 is determined as the quotient of the distance d and the time difference θt using the predetermined spatial distance d between the sensor units 16a, 16b.

[0059] The sensor units 16a, 16b have a predetermined detection rate or

[0060] Frame rate Fs, which is, for example, greater than or equal to 1 kHz and less than or equal to 30 kHz. Accordingly, the first and second sensor data have values ​​for discrete points in time at a distance equal to the reciprocal of the acquisition rate. This temporal discretization of the sensor data requires a minimum of 3t min= 1 / Fs for the temporal resolution of the determined time difference and thus results in a quantization of the time difference. In other words, due to the predetermined or defined acquisition rate of the sensor units, a quantization effect occurs when determining the time difference, so that the sound time difference can only take values ​​from a set of predetermined quantized values ​​{k * 8t mln | k = 1 , 2, ...}.

[0061] Determining the speed based on the quantized values ​​of the travel time difference accordingly results in a quantization of the determined speed values. Fig. 3 shows the speed values ​​determined according to method 100 from Fig. 2 as individual points, plotted over the measurement or travel time of the vehicle 10. The quantization effect is evident from the discretization of the speed values ​​along the vertical speed axis, with a quantization level or a distance between a plurality of identical speed values ​​increasing with increasing speed.

[0062] Fig. 3 further shows the speed curve determined in real time according to the proposed approach as a solid line. With the proposed approach, a speed value or speed curve corrected to take the quantization effect into account can be provided by the computing unit 18.

[0063] The computing unit 18 is configured to determine a speed of the vehicle 10 when traveling on the track 14.

[0064] For this purpose, the computing unit 18 preferably comprises a processor, a storage medium with a computer program, and at least one hardware and / or software interface. The computer program comprises instructions which, when executed by the processor, cause the speed of the vehicle 10 to be determined according to the method described below.

[0065] The computing unit 18 is configured to read in the first sensor data acquired by the first sensor unit 16a during travel along the track 14 and the second sensor data acquired by the second sensor unit 16b using the hardware and / or software interface. The computing unit 18 is connectable, in particular connected, to the sensor units 16a, 16b directly or indirectly for reading in the sensor data via the hardware and / or software interface.

[0066] The computing unit 18 is further configured to determine a propagation time difference based on the read-in first and second sensor data. Due to a quantization effect caused by a predetermined acquisition rate of the sensor units, the propagation time difference has a value from a set of predetermined quantized values.

[0067] Furthermore, the computing unit 18 is configured to obtain predetermined speed error information that represents a statistical distribution of a speed error. For example, the computing unit 18 is configured to read the speed error information from a memory module comprised by the computing unit 18 or associated with the computing unit. The speed error information is preferably configured as information determined before traveling along the track (14) and stored as characteristic curve information and / or characteristic map information.

[0068] According to one embodiment, the predetermined speed error information is embodied as characteristic curve information, which includes an assignment of speed values ​​to a respective variance and / or standard deviation of the corresponding speed error. Preferably, the computing unit 18 is configured to determine the speed of the vehicle 10 using a Kalman filter and to consider the read-in characteristic curve information in a correction step of the Kalman filter in order to correct a speed value predicted in a prediction step of the Kalman filter.

[0069] In this embodiment, the simplifying assumption is made that, for a fixed speed, the speeds determined based on the time difference are nearly normally distributed, with the mean corresponding to the current speed. Under this assumption, a standard deviation or variance can be calculated for each speed with respect to the speeds determined based on the time difference, which characterizes or represents a statistical distribution of the speed error or speed error values. The calculated characteristic curve (variance of the speed error distribution as a function of speed) is used in the Kalman filter as a state-dependent error to correct the predicted speed values ​​(the so-called covariance matrix of the measurement noise).

[0070] According to an alternative embodiment, the speed error information is embodied as characteristic map information, which includes an assignment of speed values ​​to a respective distribution of the corresponding speed error. The computing unit 18 is preferably configured to determine the speed using a Bayesian filter and to consider the read-in characteristic map information in a correction step of the Bayesian filter in order to correct a speed value predicted in a prediction step of the Bayesian filter.

[0071] In this embodiment, the simplifying assumption that, at a fixed speed, the speeds determined based on the travel time difference are almost normally distributed is omitted. Accordingly, the requirement of a normal distribution, which would be necessary for the use of a Kalman filter, is no longer met. In this case, a statistical distribution function can be calculated for each speed with respect to the speeds determined based on the travel time difference, which includes a statistical distribution of the speed error or speed error values. The calculated characteristic map (speed error distribution as a function of speed) is used in the Bayes filter as a state-dependent error to correct the predicted speed values, assuming a linear system model (analogous to the Kalman filter).

[0072] The computing unit 18 is configured to determine a speed of the vehicle based on the determined propagation time difference and the read-in speed error information in order to provide a speed value or speed profile corrected taking into account the quantization effect. Preferably, the computing unit 18 is additionally configured to output a signal depending on the determined speed of the vehicle 10 by means of the and / or another hardware and / or software interface.

[0073] According to this embodiment, the computing unit 18 is arranged on the vehicle 10. According to an alternative embodiment, the computing unit 18 can be arranged away from the vehicle 10 and, for example, be part of a server backend or a cloud computing unit. In this case, the computing unit 18 is connected to the sensor units 16a, 16b via a wireless communication link in order to receive the sensor data from the sensor units 16a, 16b.

[0074] Fig. 4 shows a flowchart of a method for determining the speed of a vehicle, in particular a rail vehicle, while traveling along a track. The vehicle has a first sensor unit and a second sensor unit, identical in construction to the first sensor unit and arranged on the vehicle at a distance along a direction of travel of the vehicle. The vehicle can, for example, be the rail vehicle shown in Fig. 1. The method is designated in its entirety by reference numeral 200.

[0075] Method 200 is executed when the vehicle travels along the track (14), for example, when the rail vehicle travels along a track. In particular, the method is executed repeatedly in the described sequence of method steps, preferably periodically or continuously.

[0076] In step 210, sensor data, each representing the same physical property of the track (14), is acquired by means of the vehicle's sensor units in order to provide the acquired sensor data to the computing unit. In step 210a, first sensor data is acquired by means of the first sensor unit. In step 210b, second sensor data is acquired by means of the second sensor unit. The first and second sensor data are preferably in the form of magnetic field data, which represent a spatial profile of a magnetic field in the region of the track. The acquired sensor data is provided to a computing unit assigned to the vehicle, preferably arranged on the vehicle, using a wireless or wired communication connection.

[0077] In step 220, the first and second sensor data are read in by means of the computing unit.

[0078] In step 230, a propagation time difference is determined by the computing unit based on the read-in first and second sensor data. The propagation time difference represents a temporal offset between the first and second sensor data, in particular a temporal offset between mutually corresponding parts of the first and second sensor data. The mutually corresponding parts of the first and second sensor data are, for example, signatures included in the sensor data or derived from the sensor data, which appear in the second sensor data with a time delay compared to the first sensor data when the first sensor unit is arranged in front of the second sensor unit in the direction of travel.

[0079] In step 240, speed error information is read in by the computing unit, which represents a statistical distribution of a speed error.

[0080] In step 250, the speed of the vehicle is determined based on the determined travel time difference and the read-in speed error information. Preferably, a preliminary speed is first determined based on the determined travel time difference, with the preliminary speed correspondingly comprising a speed value from a set of predetermined speed values ​​due to the quantization effect. Subsequently, the determined preliminary speed is corrected or smoothed based on the speed error information.

[0081] In step 260, a signal is output by the computing unit depending on the determined position of the vehicle. The output signal can, for example, be in the form of an information signal representing the determined speed.

Claims

Claims 1 . Method (200) for determining a speed of a vehicle (10), in particular a rail vehicle (10), when traveling along a track (14), wherein the vehicle (10) comprises a first sensor unit (16a) and a second sensor unit (16b) identical in construction to the first sensor unit (16a) and arranged on the vehicle (10) at a distance along a direction of travel of the vehicle (10), and the method (200) comprises the following steps: - reading (220) first sensor data acquired by the first sensor unit (16a) when traveling along the track (14) and second sensor data acquired by the second sensor unit (16b), wherein the first and second sensor data represent the same physical property of the track (14); - determining (230) a propagation time difference based on the read-in first and second sensor data, which has a value from a set of predetermined quantized values ​​due to a quantization effect caused by a predetermined detection rate of the sensor units (16a, 16b); - reading (240) a predetermined speed error information representing a statistical distribution of a speed error; and - Determining (250) a speed of the vehicle (10) based on the determined travel time difference and the read-in speed error information by means of a computing unit (20) in order to provide a speed value or speed curve corrected taking into account the quantization effect. Method (200) according to claim 1, characterized in that the predetermined speed error information is designed as characteristic curve information which comprises an assignment of speed values ​​to a respective variance and / or standard deviation of the corresponding speed error in order to take the read-in characteristic curve information into account when determining the speed.

3. Method (200) according to claim 2, characterized in that the speed is determined using a Kalman filter and the read-in characteristic information is taken into account in a correction step of the Kalman filter in order to correct a speed value predicted in a prediction step of the Kalman filter.

4. Method (200) according to claim 1, characterized in that the predetermined speed error information is designed as characteristic map information which comprises an assignment of speed values ​​to a respective distribution of the corresponding speed error in order to take the read-in characteristic map information into account when determining the speed.

5. The method (200) according to claim 4, characterized in that the speed is determined using a Bayesian filter and the read-in characteristic map information is taken into account in a correction step of the Bayesian filter in order to correct a speed value predicted in a prediction step of the Bayesian filter.

6. Method (200) according to one of claims 2 to 5, characterized in that the assignment of speed values ​​to the respective variance and / or standard deviation and / or distribution of the speed error is determined before traveling on the track (14) and is / was stored as characteristic curve information and / or characteristic map information.

7. The method (200) according to claim 6, characterized in that the assignment is determined based on further sensor data acquired by means of the first sensor unit (16a) and the second sensor unit (16b) and / or based on further sensor data acquired by means of one or more further sensor units, wherein the further sensor data represent the same physical property as the first and second sensor data.

8. Method (200) according to one of the preceding claims, characterized in that the vehicle (10) is designed as a rail vehicle (10) and the sensor data are designed as magnetic field data which represent a spatial course of a magnetic field in the region of the track (14).

9. Method (200) according to one of the preceding claims, characterized by a step of outputting (260) a signal as a function of the determined speed of the vehicle (10), wherein the output signal is designed as an information signal representing the determined speed and / or as a control signal for controlling a unit arranged on or off the vehicle (10).

10. A computing unit (20) for determining a speed of a vehicle (10), in particular a rail vehicle (10), when traveling along a track (14), wherein the vehicle (10) comprises a first sensor unit (16a) and a second sensor unit (16b) identical in construction to the first sensor unit (16a) and arranged at a distance along a direction of travel of the vehicle (10) on the vehicle (10), and the computing unit (20) is configured, - reading in first sensor data acquired by the first sensor unit (16a) and second sensor data acquired by the second sensor unit (16b) when traveling along the track (14), wherein the first and second sensor data represent the same physical property of the track (14), - to determine a propagation time difference based on the read-in first and second sensor data, which has a value from a set of predetermined quantized values ​​due to a quantization effect caused by a predetermined acquisition rate of the sensor units (16a, 16b), - read in a given speed error information which represents a statistical distribution of a speed error, and - to determine a speed of the vehicle (10) based on the determined travel time difference and the read-in speed error information in order to provide a speed value or speed curve corrected taking into account the quantization effect.

11. System (12) for determining a speed of a vehicle (10), in particular a rail vehicle (10), with - a first sensor unit (16a) which is designed to provide first sensor data when traveling along a travel path (14), - a second sensor unit (16b) identical in construction to the first sensor unit (16a) and designed to provide second sensor data when traveling along the track (14), - wherein the first and second sensor data represent the same physical property of the travel path (14), and a computing unit (20) according to claim 10.

12. Vehicle (10), in particular a rail vehicle (10), with a system (12) for determining a speed of the vehicle (10) according to claim 11, wherein the first sensor unit (16a) and the second sensor unit (16b) are arranged on the vehicle (10) at a distance from one another along a direction of travel of the vehicle (10).

13. Vehicle (10) according to claim 12, characterized in that the sensor units (16a, 16b) are arranged on an underside of the vehicle (10) facing the travel path (14) traveled by the vehicle (10).

14. A computer program comprising instructions which, when executed by a computing unit (20), cause the computing unit (20) to carry out the method (200) according to any one of claims 1 to 9.

15. A machine-readable storage medium on which the computer program according to claim 14 is stored.

Images