EP4286243A1 - Validation of a sensor unit of a rail vehicle for object localization - Google Patents

Abstract

An arrangement (1) for validating a function of a sensor unit (7) of a rail vehicle (6) for object localization is described. The arrangement (1) comprises a test rail section (2) and a test object (3) on the test rail section (2). The test object (3) comprises a first self-localization unit (4) for determining a reference position (RP_T) of the test object (3). The rail vehicle (6) comprises a sensor unit (7) for detecting a relative position (RLP) of the test object (3) to the sensor unit (7), a second self-localization unit (9) for determining an ego position (EP) of the rail vehicle (6) and a position determination unit (8) for determining an absolute position (AP) of the test object (3) based on the detected relative position (RLP) of the test object (3) to the sensor unit (7) and the determined ego position (EP) of the rail vehicle (6). Part of the arrangement (1) is also a validation unit (10) for determining a deviation (AW) of the determined absolute position (AP) of the test object (3) from the reference position (RP_T) of the test object (3) . A method for validating a function of a sensor unit (7) of a rail vehicle (6) for object localization is also described.

Description

The invention relates to an arrangement for validating a function of a sensor unit of a rail vehicle for object localization. The invention also relates to a method for validating a function of a sensor unit of a rail vehicle for object localization.

In rail traffic, it occasionally happens that objects, such as people or road vehicles, shopping carts that have been thrown onto the rails, or even boulders or fallen trees, end up on the track and therefore pose a risk to the safety of rail traffic and in the event of a fall of people and road vehicles themselves are at high risk due to the possibility of a collision with a moving rail vehicle. Therefore, such objects must be detected in time to initiate a braking process for an approaching rail vehicle, so that a collision between the rail vehicle and the object can be prevented. Furthermore, rail vehicles that are on the same track must also be recognized in order to avoid a collision between these rail vehicles. One of the critical tasks of a rail vehicle driver is to constantly monitor the rail area for possible obstacles or objects that could collide with the rail vehicle and to decide whether a reaction to this, such as a braking maneuver, needs to be initiated or not.

When driving vehicles without a driver, it is particularly important to have a system that automatically locates possible obstacles correctly and can decide whether they pose a danger to the vehicle itself or other road users, as this involves monitoring the obstacles By definition, a driver must forego the rail route.

Lidar systems and radar systems, which can detect the distances and directions in which objects are located, are suitable for such tasks.

Automated detection systems for potential obstacles have already been experimentally investigated and developed. Infrastructure-based solutions, in which a plurality of sensors are deployed to monitor a specific area of a railway line, are sometimes used to monitor areas where accidents often occur, such as railway crossings. However, it is not feasible to provide the entire rail network with infrastructure-based obstacle detection systems. Therefore, there is a need for an on-board system to fulfill a driver's task of observing and monitoring the rail route.

Most on-board systems include systems for measuring the position of detected obstacles. Knowing the exact position of a potential obstacle allows an assessment of the dangers posed by the obstacle. In particular, the position of the potential obstacle must be determined with sufficient accuracy in order to estimate how far away the potential obstacle is from a rail vehicle and whether or not the potential obstacle is in the loading gauge of the rail vehicle or may or may not be at a later point in time .

In order to meet safety requirements, the precision of the localization of potential obstacles must be checked or validated by the on-board systems used. To date, in order to prepare for such a validation, a test object is first positioned on a test rail section. Furthermore, position marks are determined and arranged on the test rail section. During validation, the positions marked by the defined position markers are approached by a rail vehicle with a sensor device to be validated. The position markers are usually determined individually by hand and are therefore quite time-consuming. In addition, when the rail vehicle approaches the individual position markers, inaccuracies can occur that have a negative impact on the reliability of the validation.

The task is therefore to provide an arrangement and a method for validating the function of an on-board sensor system for obstacle localization for rail vehicles, which works more reliably, more robustly and more precisely than previous approaches.

This object is achieved by an arrangement for validating a function of a sensor unit of a rail vehicle for object localization according to claim 1 and a method for validating a function of a sensor unit of a rail vehicle for object localization according to patent claim 13.

The arrangement according to the invention for validating a function of a sensor unit of a rail vehicle for object localization has a test rail section. As explained in detail later, the test rail section preferably comprises a shape which is designed such that a distance of an object from a center line of the test rail section can be easily determined. It should be emphasized at this point that the sensor unit preferably has a plurality of sensors in order to be able to localize an object. Typical methods of localization include methods of triangulation and/or transit time measurement. As at least partially mentioned at the beginning, sensor units for such tasks preferably include optical-based or infrared-based lidar systems and/or radar systems and/or 3D cameras.

The arrangement according to the invention also has a test object which is located on the test rail section. The test object includes an object to be detected or localized on the test rail section for test purposes. Such an object to be localized preferably comprises a silhouette of an object that typically occurs in the area of rail traffic, such as a silhouette of a human person or a vehicle, which is to be detected and localized in front of the sensor unit of the rail vehicle as part of the validation. Such a shape of the test object advantageously corresponds to an external shape of a potential obstacle that typically occurs during use, so that the test conditions approximately correspond to real conditions.

Part of the test object is also a first self-localization unit for determining a reference position of the test object. The first self-localization unit determines its own position, which essentially corresponds to the ego position of the test object. The ego position of the test object is used as the reference position of the test object. The ego position of the first self-localization unit can be converted into the reference position of the test object if the ego position of the first self-localization unit does not correspond exactly to the ego position or reference position of the test object. The ego position is determined on the basis of a recording of measurement data by the first self-localization unit, which is correlated with the ego position of the first self-localization unit. The measurement data can include time difference measurement data, in particular satellite navigation measurement data, odometric measurement data, distance measurement data, speed measurement data or acceleration measurement data.

The rail vehicle includes a sensor unit for detecting a relative position of the test object to the sensor unit. For this purpose, the sensor unit measures the direction and the distance at which the test object is located relative to the sensor unit. The relative position indicates the position of the test object in a coordinate system that moves with the rail vehicle or the sensor unit of the rail vehicle. The rail vehicle also includes a second self-localization unit for determining an ego position of the rail vehicle. More precisely, the second self-localization unit first determines its own position by measuring its own position, ie the ego position of the second self-localization unit, from which the ego position of the rail vehicle or the ego position of the sensor unit of the rail vehicle are determined differ minimally from each other, but can be easily converted into one another if the orientation of the rail vehicle is known. The second self-localization unit determines the ego position of the rail vehicle based on measurement data that is correlated with the position of the rail vehicle. The second self-localization unit is preferably set up to determine an ego pose of the rail vehicle. Such an ego pose shows not only the ego position of the rail vehicle but also its orientation. Advantageously, not only the distance between the rail vehicle and the test object, but also the direction in which the test object is located relative to the pose of the rail vehicle and thus the exact relative position of the test object to the rail vehicle can be determined on the basis of the ego pose. The alignment of the rail vehicle and thus also the alignment of the sensor unit in space can be determined both by a direct measurement of the orientation and by deriving the orientation based on the determined ego position of the rail vehicle and the rail course of the test rail route at the ego position, in particular determined by map data or satellite navigation data. If the course of the test rail route is known, the orientation usually results of the rail vehicle from its ego position. The orientation of the rail vehicle is position-independent, particularly when the test rail route runs in a straight line. In this variant, the orientation of the rail vehicle is advantageously fixed and no locally variable orientation of the rail vehicle has to be taken into account during validation, for example by using maps or by measuring the orientation of the rail vehicle and/or its sensor unit.

It should also be mentioned that the arrangement according to the invention can advantageously also be supplemented by an adjustment unit which, based on the validation, adjusts the position determination of the test object in order to improve or ensure the required accuracy of object localization by the sensors of a relevant rail vehicle.

The rail vehicle of the arrangement according to the invention also includes a position determination unit for determining an absolute position of the test object based on the detected relative position of the test object to the sensor unit and the determined ego position of the rail vehicle. Based on the knowledge of the route and the knowledge of the ego position of the rail vehicle on the test rail route, the position determination unit calculates a pose of the rail vehicle. Based on the knowledge of the pose of the rail vehicle and the relative position of the test object, the position determination unit calculates an absolute position in a stationary coordinate system, the position and orientation of which is independent of the pose and the current position of the rail vehicle.

The arrangement according to the invention also includes a validation unit for determining a deviation of the determined absolute position of the test object from the reference position of the test object. To determine the deviation, the validation unit carries out a comparison between the determined absolute position of the test object and the reference position of the test object. The validation achieves a positive result if the deviation falls below a threshold value. Validation can advantageously be automated. In particular, no markings for individual ego positions of the rail vehicle need to be arranged by hand on the test rail route. By determining a plurality of ego positions of the rail vehicle, a geographical map with ego positions can be created, as will be explained in detail later, the ego positions being determined by determining the ego position of the rail vehicle by the second self-localization unit of the rail vehicle be approached by the rail vehicle as part of a test run. As a basis for validation, an accurate determination of geodata to determine the ego position of a rail vehicle in any type of environment can advantageously be carried out without external services or prior determination of stationary measured infrastructure points. Based on this geodata, all features for validating an obstacle detection system in the railway environment can be derived, in particular a determination of a precision of a measurement of a distance of an object from the rail vehicle or the rail line.

In the method for validating a function of a sensor unit of a rail vehicle for object localization, a reference position of a test object is determined by a first self-localization unit arranged on the test object. The reference position can be determined, for example, by determining the geographical coordinates of the test object or a specific location on the test object, for example a geometric center of the test object.

Furthermore, a relative position of the test object to a sensor unit arranged on the rail vehicle is determined. This relative position results from the distance and direction measured by the sensor unit in which the test object is located relative to the rail vehicle or the sensor unit.

In addition, an ego position or preferably the ego pose of the rail vehicle is determined by a second self-localization unit arranged on the rail vehicle. As already mentioned, the ego pose not only indicates the exact ego position but also the orientation of the rail vehicle. The ego position is preferably given in geographical coordinates. The orientation of the rail vehicle can be described, for example, by an angle relative to a reference direction.

Based on the detected relative position of the test object to the sensor unit and the determined ego position of the rail vehicle, an absolute position of the test object is determined within the scope of the method according to the invention.

Finally, an object localization function of the sensor unit is validated by determining a deviation of the determined absolute position of the test object from the reference position of the test object. This means that it is determined whether a deviation in determining the absolute position of the test object falls below a predetermined threshold value or not. The method according to the invention shares the advantages of the arrangement according to the invention.

The validation method according to the invention can be used in particular for an adjustment method in which an adjustment of the position determination of the test object is carried out on the basis of the validation in order to improve or ensure the required accuracy of object localization by the sensors of a relevant rail vehicle. In such a method, based on the deviation measured during validation, a correction of the alignment of the sensor units and/or a computational correction is carried out in order to compensate for the deviation determined.

A majority of the aforementioned components of the arrangement according to the invention can be completely or partially in the form of Software modules are implemented in a processor of a corresponding computing system, for example in a control device or a computing system of the rail vehicle or a computing system of the test object or also an externally arranged computing system. This applies in particular to the position determination unit of the rail vehicle and the validation unit. A largely software-based implementation has the advantage that previously used computing systems can also be easily retrofitted by a software update in order to work in the manner according to the invention. In this respect, the task is also solved by a corresponding computer program product with a computer program which can be loaded directly into a computing system, with program sections to carry out the steps of the method according to the invention, at least the steps that can be carried out by a computer, in particular the steps of determining an absolute position of the test object and of validating the object localization or object localization function when the program is executed in the computing system. In addition to the computer program, such a computer program product may optionally contain additional components such as: B. documentation and/or additional components also include hardware components, such as hardware keys (dongles, etc.) for using the software.

For transport to the computing system or to the control device and/or for storage on or in the computing system or the control device, a computer-readable medium, e.g. a memory stick, a hard drive or another transportable or permanently installed data carrier, on which the data that can be read by a computing system can be used and executable program sections of the computer program are stored. For this purpose, the computing system can, for example, have one or more cooperating microprocessors or the like.

The dependent claims and the following description each contain particularly advantageous refinements and developments of the invention. In particular, the Claims of one claim category can also be developed analogously to the dependent claims of another claim category and their description parts. In addition, within the scope of the invention, the various features of different exemplary embodiments and claims can also be combined to form new exemplary embodiments.

In one embodiment of the arrangement according to the invention, the test object has a transmitting unit. The transmitting unit is preferably designed to transmit the determined reference position of the test object to the rail vehicle or another device set up for evaluation and validation. Such a transmitting unit enables the transmission of the information about the reference position of the test object to a device arranged remotely from the test object and also the processing of the information about the reference position of the test object at a position remote from the test object. For this purpose, a receiving unit is arranged at this remote position, which serves to receive the determined reference position of the test object. Such a remote position may include, for example, the position of the rail vehicle. Ie, the rail vehicle can have the receiving unit. This variant is particularly advantageous if the validation unit is also included in the rail vehicle. In this variant, the validation unit receives the information about the reference position of the test object via the receiving unit and carries out a comparison with the absolute position of the test object determined by the position determination unit of the rail vehicle. However, the validation unit can also be arranged spatially separate from the rail vehicle, for example stationary or on a mobile testing device. In this variant, the rail vehicle also includes a transmitting unit in order to transmit the determined absolute position of the test object to the mobile testing device. In this variant, validation takes place spatially separate from the rail vehicle. In this variant, only a single validation unit is advantageously required for a plurality of rail vehicles, which means resources can be saved.

In a variant of the arrangement according to the invention, the test object comprises the validation unit and a receiving unit for receiving the absolute position of the test object determined by the position determination unit. Furthermore, the rail vehicle comprises a transmitting unit for transmitting the absolute position of the test object determined by the position determination unit to the receiving unit of the test object or the validation unit of the test object. In this variant, the test data is evaluated in the test object itself. The rail vehicle only serves as a mobile sensor unit for measuring the ego position of the rail vehicle and for measuring the relative position of the test object to the rail vehicle. In this variant, the validation of the sensor function of different rail vehicles can be carried out with a single validation unit if the same test object is used to validate the sensor function of different rail vehicles.

In a variant of the arrangement according to the invention, it has an evaluation device that is separate from the rail vehicle and the test object. In this variant, the separate evaluation device comprises the validation unit and a receiving unit for receiving the absolute position of the test object determined by the position determination unit of the rail vehicle and for receiving the reference position of the test object from the test object. In this variant, the rail vehicle includes a transmission unit for transmitting the absolute position of the test object determined by the position determination unit to the separate evaluation device. This variant has the advantage that the evaluation device can now be used flexibly to validate the function of different rail vehicles and different test objects, with only a single one Validation unit is required for the use cases mentioned. In this way, resources can be saved. The evaluation device can be permanently installed on a specific test track; it can also be designed as a mobile or transportable device for validation on different test tracks. For transport, the evaluation device can be designed with a chassis and a traction unit for transport between different test routes. In the last variant, the scope of application of the evaluation device is expanded to include different test rail routes, so that resource efficiency is further increased through the versatile possible uses of one and the same validation unit.

• Satellitennavigationsdaten,
• IMU-Daten (IMU = inertial measurement unit = inertiale Messeinheit)
• Geschwindigkeitssensordaten,
• Odometriedaten,
• sogenannte RTK-Korrekturdaten.

The second self-localization unit in or on the rail vehicle of the arrangement according to the invention records measurement data for position estimation and orientation estimation for the position and, if applicable, the pose of the rail vehicle. These measurement data can be recorded using technically very different types of measuring devices. The measurement data mentioned may include the following types of measurement data:

• satellite navigation data,
• IMU data (IMU = inertial measurement unit)
• speed sensor data,
• odometry data,
• so-called RTK correction data.

RTK correction data (RTK = Real time kinematics) is used in geodesy to precisely determine position coordinates using satellite navigation methods. Accuracies of 1 to 2 cm are achieved. The coordinates of the positions are calculated in real time after initialization. The positions are determined using a reference antenna from a reference station and a second antenna positioned on a so-called rover. The position of the rover will be determined by three-dimensional polar attachment to the reference station using the baseline method. The rail vehicle preferably includes the second antenna, ie the second self-localization unit of the rail vehicle preferably includes this second antenna. The reference station can, for example, be part of the test object and in this case the first self-localization unit can include the reference antenna. However, the reference station can also be delivered from another or external facility. The RTK correction data can be provided in particular via the mobile radio network or via radio from a device that generates correction data for satellite navigation.

An inertial measuring unit measures acceleration and deceleration data and can be used to determine the orientation of a rail vehicle and thus to determine the pose of the rail vehicle.

Speed sensor data and odometry data can be used to determine the ego position of a rail vehicle.

• eine Navigationseinheit zum Empfangen eines Satellitennavigationssignals,
• eine Empfangseinheit zum Empfangen von RTK-Korrekturdaten,
• eine inertiale Messeinheit.

In a variant of the arrangement according to the invention, the first and/or the second self-localization unit comprises at least one of the following types of self-localization units:

• a navigation unit for receiving a satellite navigation signal,
• a receiving unit for receiving RTK correction data,
• an inertial measurement unit.

Satellite navigation can achieve high levels of accuracy for determining position or direction of movement in the context of differential satellite navigation by adding reference stations and evaluating the phase shift of the carrier wave. An inertial measurement unit can be used for orientation measurement. Speed data and odometry data can also be used to determine position. Different measuring methods can also be advantageously combined. Depending on the situation and the current applicability of the different methods, the measured values from different measuring methods can be combined, for example through data fusion, in particular through weighted averaging of the measured values.

The position determination unit of the arrangement according to the invention is preferably set up to determine a position and a course of a center line of the test rail route based on the ego position of the rail vehicle. If the lateral position of the second self-locating unit on the rail vehicle is known, an absolute position on a center line of the test rail section can be determined by traveling along the test rail section. This center line can advantageously be used as a reference line for a lateral position of the test object to be measured.

The position determination unit is preferably set up to determine a distance of the test object from the center line. A distance from a point to the center line can be found using linear algebra. The vector that runs through the point and is oriented perpendicular to the center line is determined. By knowing the distance of a point, in particular an object, from the center line, it can advantageously be determined whether the object is in the travel area of the rail vehicle or in the rail area or not.

The validation unit of the arrangement according to the invention is preferably set up to determine a deviation of the distance of the test object from the center line determined on the vehicle side from a reference distance of the test object to the center line. The reference distance of the test object to the center line can, for example, be based on the reference position of the Test object and the knowledge of the course of the center line can be calculated. The comparison results in a lateral deviation of the absolute position of the test object measured on the vehicle side.

The position determination unit is preferably set up to determine a distance between the rail vehicle and the test object, preferably in the longitudinal direction. A distance measurement can, for example, be carried out trigonometrically or as a transit time measurement.

The position determination unit is preferably set up to detect a deviation between the distance detected on the vehicle side, preferably in the longitudinal direction, between the rail vehicle and the test object and a reference distance between the rail vehicle and the test object determined on the basis of the reference position of the test object and the determined ego position of the rail vehicle, preferably in the longitudinal direction.

The test rail section preferably has a straight course. Advantageously, the pose of the rail vehicle does not change in this variant of the geometric shape of the test rail section depending on the position of the rail vehicle on the test rail section. The pose of the rail vehicle therefore does not have to be determined in order to determine the relative position or absolute position of the test object based on sensor data.

The test rail section preferably comprises a curvilinear course and a center line of the test rail section is approximated by straight-line sections. A distance from a point of an object to the partial distances can advantageously be calculated using simple linear algebraic methods.

• FIG 1 eine schematische Darstellung einer Anordnung zur Validierung einer Funktion einer Sensoreinheit eines Schienenfahrzeugs zur Objektlokalisierung gemäß einem Ausführungsbeispiel der Erfindung,
• FIG 2 ein Flussdiagramm, welches ein Verfahren zur Validierung einer Funktion einer Sensoreinheit eines Schienenfahrzeugs zur Objektlokalisierung gemäß einem Ausführungsbeispiel der Erfindung veranschaulicht,
• FIG 3 eine schematische Darstellung eines Vorgangs zur Erstellung einer Karte mit Ego-Positionen, die das Schienenfahrzeug im Rahmen der Validierung anfährt,
• FIG 4 eine Darstellung der möglichen Fehler bei der Ermittlung der Ego-Position eines Schienenfahrzeugs und deren Auswirkungen auf eine Objektlokalisierung,
• FIG 5 eine Darstellung von Positionen von Objekten mit unterschiedlichen lateralen Abständen zu einer Mittenlinie einer Schienenstrecke,
• FIG 6 ein Schaubild, welches eine Ermittlung des minimalen Abstandes zwischen einem Punkt und einer konstruierten Strecke veranschaulicht,
• FIG 7 ein Schaubild, welches eine Ermittlung eines longitudinalen Abstandes zwischen einem Punkt und einem Hilfspunkt in longitudinaler Richtung veranschaulicht,
• FIG 8 einen Kreisabstand auf einer Sphäre,
• FIG 9 eine Darstellung von Winkelbeziehungen in einem Kreisbogen,
• FIG 10 ein Schaubild, welches einen Validierungsprozess dokumentiert,
• FIG 11 eine schematische Darstellung einer Anordnung zur Validierung einer Funktion einer Sensoreinheit eines Schienenfahrzeugs zur Objektlokalisierung gemäß einem zweiten Ausführungsbeispiel der Erfindung,
• FIG 12 eine schematische Darstellung einer Anordnung zur Validierung einer Funktion einer Sensoreinheit eines Schienenfahrzeugs zur Objektlokalisierung gemäß einem dritten Ausführungsbeispiel der Erfindung.

The invention is explained in more detail below with reference to the attached figures using exemplary embodiments. Show it:

• FIG 1 a schematic representation of an arrangement for validating a function of a sensor unit of a rail vehicle for object localization according to an exemplary embodiment of the invention,
• FIG 2 a flowchart illustrating a method for validating a function of a sensor unit of a rail vehicle for object localization according to an exemplary embodiment of the invention,
• FIG 3 a schematic representation of a process for creating a map with ego positions that the rail vehicle approaches as part of the validation,
• FIG 4 a representation of the possible errors when determining the ego position of a rail vehicle and their effects on object localization,
• FIG 5 a representation of positions of objects with different lateral distances from a center line of a rail route,
• FIG 6 a diagram illustrating a determination of the minimum distance between a point and a constructed route,
• FIG 7 a diagram illustrating a determination of a longitudinal distance between a point and an auxiliary point in the longitudinal direction,
• FIG 8 a circle distance on a sphere,
• FIG 9 a representation of angular relationships in a circular arc,
• FIG 10 a diagram that documents a validation process,
• FIG 11 a schematic representation of an arrangement for validating a function of a sensor unit of a rail vehicle for object localization according to a second exemplary embodiment of the invention,
• FIG 12 a schematic representation of an arrangement for validating a function of a sensor unit of a rail vehicle for object localization according to a third exemplary embodiment of the invention.

In FIG 1 a schematic representation of an arrangement 1 for validating a function of a sensor unit 7 of a rail vehicle 6 for object localization is shown. The arrangement 1 comprises a straight test rail section 2, on which there is a test object 3, which is in FIG 1 is only shown schematically on the right. Furthermore, there is a rail vehicle 6 on the test rail route 2, which is in FIG 1 is drawn on the left side.

The test object 3 is arranged at a distance d _r from the center line 2a of the test rail section 2. The test object 3 includes a first self-localization unit 4, in this exemplary embodiment a satellite navigation unit, for determining a reference position RP _T of the test object 3. For this purpose, the first self-localization unit 4 receives a satellite signal, on the basis of which it determines the reference position RP _T of the test object 3. The test object 3 further includes a transmission unit 5 for transmitting the determined reference position RP _T of the test object 3 to the rail vehicle 6 by radio.

The rail vehicle 6 includes the already mentioned sensor unit 7, in this exemplary embodiment a lidar unit, for detection and localization, ie for determining a relative position RLP of the test object 3 to the sensor unit 7. For this purpose, the sensor unit 7 measures the longitudinal distance d _l of the test object 3 to the sensor unit 7 and the lateral distance d _r of the test object 3 to the center line 2a of the test rail section 2. The position and the course of the center line 2a can be determined in advance by traveling along the test rail section 2 and detecting the ego position EP of the rail vehicle 6 at different points on the test rail route 2 are determined. To determine the ego position EP, the rail vehicle 6 includes a second self-localization unit 9.

The second self-localization unit 9 includes in the in FIG 1 shown embodiment also a satellite navigation unit for receiving a satellite signal to determine the ego position EP of the rail vehicle 6. Furthermore, the rail vehicle 6 also includes a position determination unit 8, which determines an absolute position AP of the test object 3 based on the detected relative position RLP of the test object 3 to the sensor unit 7 and determined based on the determined ego position EP of the rail vehicle 6. The absolute position AP determined by the position determination unit 8 is also transmitted to a validation unit 10, which is also part of the rail vehicle 6. The validation unit 10 carries out a comparison between the determined absolute position AP of the test object 3 and the reference position RP _T of the test object 3 determined by the test object 3 itself. For this purpose, the rail vehicle 6 includes a receiving unit 11 with which the information about the reference position RP _T from the test object 3 is received by radio.

In FIG 2 a flowchart 200 is shown, which illustrates a method for validating a function of a sensor unit 7 of a rail vehicle 6 for object localization according to an exemplary embodiment of the invention.

In step 2.I, a reference position RP _T of a test object 3 is arranged in the test object 3 first self-localization unit 4 determined. For this purpose, a satellite signal PS is received by the first self-localization unit 4, in this exemplary embodiment a satellite navigation unit.

In step 2.II, the determined reference position RP _T of the test object 3 is transmitted to a receiving unit 11 of the rail vehicle 6 by a transmitting unit 5 arranged on the test object 3.

In step 2.III, a relative position RLP of the test object 3 to a sensor unit 7 arranged on the rail vehicle 6 is detected. For this purpose, the sensor unit 7 measures in FIG 2 illustrated embodiment a lidar unit, a distance and an azimuth of the test object 3 relative to the orientation of the sensor unit 7.

In step 2.IV, an ego position EP of the rail vehicle 6 is additionally determined by a second self-localization unit 9 arranged on the rail vehicle 6. The orientation of the rail vehicle 6 is also known based on the known geometry of the test rail section 2, for example the test rail section 2 runs in a straight line.

In step 2.V, an absolute position AP of the test object 3 is determined based on the detected relative position RLP of the test object 3 to the sensor unit 7 and the determined ego position EP as well as the known orientation of the rail vehicle 6.

In step 2.VI, the object localization function of the sensor unit 7 is validated by determining a deviation AW of the determined absolute position AP of the test object 3 from the reference position RP _T of the test object 3.

FIG 3 shows a schematic representation of a process for creating a map with ego positions EP1, ..., EP8, the the rail vehicle 6 (see FIG 1 ) approaches as part of the validation. The ego positions EP1, EP2 ..., EP8 are at different positions on the test rail section 2. The distances d _l are therefore (see FIG 1 ) between the test object 3 and the rail vehicle 6 at the different ego positions EP1, ..., EP8 are also different. Based on the measurement data at different measurement positions, the accuracy of the measurement can be determined depending on distances or distances d _l and it can be determined up to which distance the sensor unit 7 meets the minimum requirements for measurement accuracy.

In FIG 4 is a representation of the possible errors when determining the ego position EP of a rail vehicle 6 and their effects on object localization. An inaccurate determination or a deviation δL of the ego position EP of the rail vehicle 6 in the longitudinal L and a deviation δLat in the lateral Lat can lead to false positive or false negative results. In FIG 4 An example is shown in which an object O is actually standing on a rail line, but is incorrectly located outside the rail area due to a deviation in the determination of the ego position EP of the rail vehicle 6. This shift results from the deviation δL of the measured ego position EP of the rail vehicle 6 from its true ego position. In addition to the correct determination of the ego position EP of the rail vehicle 6, the correct classification of the detected object O also depends on the correct determination of the relative position RLP of the object O of the rail vehicle 6. Both deviations are ultimately included in the validation, since both the ego position EP of the rail vehicle 6 and the relative position RLP correspond to the true position of the object O; in the case of validation with the test object 3, this is the reference position RP _T of the test object 3 (see FIG 1 ), influence the absolute position AP of the object O to be compared.

FIG 5 shows a representation of positions of objects O with different lateral distances from a center line 2a of a rail route 6. The center line 2a is composed of straight line sections STA, which approximate the true course of the route characterized by curves.

dargestellte Strecke mit den Punkten Q und Q', wird durch den Vektor r repräsentiert. Die kleinste Distanz d_r zwischen dem Punkt P und der Geraden $y=-\frac{a}{b}x-\frac{c}{b}$

ist eine konstruierte Strecke, welche im rechten Winkel zur Geraden y verläuft und durch den Vektor v repräsentiert wird. Damit r orthogonal zur Geraden y verläuft, muss das Skalarprodukt aus dem Vektor r und dem in Richtung der Gerade y verlaufenden Vektor in Summe Null ergeben.In FIG 6 is a diagram which shows a determination of the minimum distance between a point P with the coordinates (x ₀ , y ₀ ) and a constructed route STA with the points Q and Q' with the coordinates (x ₁ , y ₁ ) and (x ₂ , y ₂ ) illustrated. The distance d between the point P and the straight line $y=-\frac{a}{b}x-\frac{c}{b}$

The route shown with the points Q and Q' is represented by the vector r represented. The smallest distance d _r between the point P and the straight line $y=-\frac{a}{b}x-\frac{c}{b}$

is a constructed line that runs at right angles to the line y and through the vector v is represented. With it r orthogonal to the line y, the dot product must be the vector r and the vector running in the direction of the straight line y sum to zero.

The vector v , which is perpendicular to the line y, is then calculated as: $$\overrightarrow{v}=\left(\begin{array}{c}a\\ b\end{array}\right)$$

ergibt sich nach kurzer Rechnung somit zu $$d_r=\frac{\left|\left(x_2-x_1\right)\left(y_1-y_0\right)-\left(\left(x_1-x_0\right)\right)\left(y_2-y_1\right)\right|}{\sqrt{{\left(x_2-x_1\right)}^2+{\left(y_2-y_1\right)}^2}}$$

The smallest distance d _r results as a projection of the vector r on the vector v . The distance d _r between the point P and the straight line $y=-\frac{a}{b}x-\frac{c}{b}$

After a short calculation, this results in: $$d_r=\frac{\left|\left(x_2-x_1\right)\left(y_1-{\mathrm{<figure-callout\; id="0"\; label="y"\; filenames="imgf0004.png,imgf0006.png"\; state="\{\{state\}\}">y</figure-callout>}}_0\right)-\left(\left(x_1-x_0\right)\right)\left({\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="imgf0002.png"\; state="\{\{state\}\}">y</figure-callout>}}_2-y_1\right)\right|}{\sqrt{{\left(x_2-x_1\right)}^2+{\left({\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="imgf0002.png"\; state="\{\{state\}\}">y</figure-callout>}}_2-y_1\right)}^2}}$$

In FIG 7 a diagram is shown which shows a determination of a longitudinal distance d _l between a point P ₀ with the coordinates (x ₀ , y ₀ ) and an auxiliary point P 'with the coordinates (x _r ', y _r ') on the constructed route STA illustrated in the longitudinal direction. Using the lateral distance d _r calculated perpendicular to the line y Between the point P with the coordinates (x _r , y _r ) and the straight line y, the spanning hypotenuse d ₀ is first calculated to calculate the longitudinal distance d _l . The longitudinal distance d _l then results as the leg of the triangle from the other two known distances d _r , d ₀ .

In FIG 8 a circle distance on a sphere between a point P and P ₀ is shown. Using the so-called Haversine formula, the distance between two given points on a sphere can be calculated. This means that the distance between two points whose geocoordinates are known can be easily calculated.

wobei $$\mathit{hav}\left(\theta \right)=\mathit{hav}\left({\mathit{lat}}_2-{\mathit{lat}}_1\right)+\mathit{cos}\left({\mathit{lat}}_1\right)\cdot \mathit{cos}\left({\mathit{lat}}_2\right)\cdot \mathit{hav}\left({\mathit{lon}}_2-{\mathit{lon}}_1\right).$$

und Gleichung (4) in Gleichung (3) eingesetzt ergibt schließlich den Abstand von d₀, wobei r den Erdradius repräsentiert.In FIG 9 A representation of angular relationships in a unit circle is shown. The points A, B and the center O' form an isosceles triangle whose legs have a length of 1. One of the legs, the height h of the isosceles triangle and half the base of the isosceles triangle, form a right triangle with the leg of unit length 1 as the hypotenuse and the angle θ between the leg of the isosceles triangle and the height h. The base of the isosceles triangle is divided into two equal halves at the base point C. If the height h is extended in the x direction, the extended half-line intersects the unit circle at point D. The function value sinθ is formed by half the base of the isosceles triangle. The function value cosθ is formed by the height h of the isosceles triangle and the function value havθ (hav = abbreviation for "haversine") is formed by the distance between points C and D. Should now be in FIG 7 shown hypotenuse d ₀ from the coordinates (lat ₁ , lon ₁ ), (lat ₂ , lon ₂ ) of the two points P, P ₀ , where "lat" stands for latitude and "lon" for longitude and "1", "2 " the two different points P, P ₀ symbolize on the globe are calculated, the distance between the points P and P ₀ results, ie the distance d ₀ to: $${\mathrm{<figure-callout\; id="0"\; label="d"\; filenames="imgf0004.png,imgf0006.png"\; state="\{\{state\}\}">d</figure-callout>}}_0=r\mathit{archaeological}\left(\mathit{hav}\left(\theta \right)\right),$$

where $$\mathit{hav}\left(\theta \right)=\mathit{<figure-callout\; id="2"\; label="hav\; lat"\; filenames="imgf0002.png"\; state="\{\{state\}\}">hav</figure-callout>}\left({\mathit{lat}}_{}\right)\left({\mathit{}}_2-{\mathit{lat}}_1\right)+\mathit{cos}\left({\mathit{lat}}_1\right)\cdot \mathit{cos}\left({\mathit{lat}}_2\right)\cdot \mathit{<figure-callout\; id="2"\; label="hav\; lon"\; filenames="imgf0002.png"\; state="\{\{state\}\}">hav</figure-callout>}\left({\mathit{lon}}_{}\right)\left({\mathit{}}_2-{\mathit{lon}}_1\right).$$

and equation (4) inserted into equation (3) finally gives the distance of d ₀ , where r represents the earth's radius.

As in connection with FIG 7 explained, the opposite side d _l and thus the longitudinal distance d _l can be calculated from the hypotenuse d ₀ and the lateral distance d _r .

In FIG 10 A diagram is shown that documents a validation process. In the diagram, measured values, in this case measured longitudinal distances d _l in meters (abbreviated as "m") between a rail vehicle 6 or a sensor unit 7 arranged at the front of the rail vehicle 6 and a test object 3 are shown as a function with a solid staircase-shaped line from the measurement time t (the time is given in coordinated universal time, "UTC"). Furthermore, in the diagram, reference values d _lref for a longitudinal distance between the rail vehicle 6 or the sensor unit 7 and the test object 3, which are determined by measuring the self-localization units 4, 9 of the test object 3 and the rail vehicle 6, are also shown with a solid line represents. Furthermore, in the diagram in FIG 10 a tolerance range T with a maximum deviation of the measured longitudinal distances d _l of 10% from the respective reference value d _lref is marked with dashed lines.

In FIG 11 is a schematic representation of an arrangement 1 for validating a function of a sensor unit 7 of a rail vehicle 6 for object localization according to a second exemplary embodiment of the invention. At the in FIG 11 Arrangement 1 shown does not have the rail vehicle 6, but rather the test object 3 as the validation unit 10 on. In this variant, the test object 3 comprises differently than in the case in FIG 1 In the first exemplary embodiment shown, a receiving unit 5 'for receiving the absolute position AP of the test object 3 determined by the position determination unit 8. Furthermore, in contrast to that in FIG 1 In the first exemplary embodiment shown, the rail vehicle 6 has a transmitting unit 11 'for transmitting the absolute position AP of the test object 3 determined by the position determination unit 8 to the test object 3. In this variant, a single validation unit 10 can be used to validate different rail vehicles, thereby achieving a resource saving effect. The test object 3 also has the same as in FIG 1 shown first embodiment, a first self-localization unit 4, which transmits a determined reference position RP _T of the test object 3 directly to the validation unit 10 encompassed by the test object 3.

In FIG 12 is a schematic representation of an arrangement 1 for validating a function of a sensor unit 7 of a rail vehicle 6 for object localization according to a third exemplary embodiment of the invention.

In the third exemplary embodiment, the arrangement 1 has a stationary evaluation device 12 that is separate from the rail vehicle 6 and the test object 3. In this case, the stationary evaluation device 12 comprises the validation unit 10 and a receiving unit 5" for receiving the absolute position AP of the test object 3 determined by the position determination unit 8 of the rail vehicle 6. In the third exemplary embodiment, the receiving unit 5" is also set up to determine the reference position RP _T of the test object 3 from the test object 3 and to pass on the received information AP, RP _T to the validation unit 10. The stationary evaluation device 12 can also be designed as a so-called remote computer and can be connected to the test object 3 and the rail vehicle 6 via a communication network, for example the Internet, in order to receive the measurement data or information AP, RP _T determined by the test object 3 and the rail vehicle 6. The rail vehicle 6 points in the in FIG 12 shown third embodiment, similar to the second embodiment, a transmission unit 11 'for transmitting the absolute position AP of the test object 3 determined by the position determination unit 8 to the evaluation device 12. The test object 3 in the third exemplary embodiment is constructed analogously to the test object 3 of the first exemplary embodiment. However, in the third exemplary embodiment, the transmitting unit 5 of the test object 3 does not transmit the information about the reference position RP _T of the test object 3 to the rail vehicle 6, but to the separate stationary evaluation device 12.

Finally, it should be pointed out once again that the methods and devices described above are merely preferred exemplary embodiments of the invention and that the invention can be varied by a person skilled in the art without departing from the scope of the invention, insofar as it is specified by the claims. For the sake of completeness, it should also be noted that the use of the indefinite articles “a” or “an” does not exclude the fact that the characteristics in question can be present multiple times. Likewise, the term “unit” does not exclude the fact that it consists of several components, which may also be spatially distributed.

Claims (15)

Arrangement (1) for validating a function of a sensor unit (7) of a rail vehicle (6) for object localization, comprising: - a test rail section (2), - a test object (3) on the test rail section (2), comprising a first self-localization unit (4) for determining a reference position (RP _T ) of the test object (3), - the rail vehicle (6), comprising: - a sensor unit (7) for detecting a relative position (RLP) of the test object (3) to the sensor unit (7), - a second self-localization unit (9) for determining an ego position (EP) of the rail vehicle (6), - a position determination unit (8) for determining an absolute position (AP) of the test object (3) based on the detected relative position (RLP) of the test object (3) to the sensor unit (7) and the determined ego position (EP) of the rail vehicle (6) , - a validation unit (10) for determining a deviation (AW) of the determined absolute position (AP) of the test object (3) from the reference position (RP _T ) of the test object (3). Arrangement according to claim 1, wherein the test object (3) has a transmission unit (5) for transmitting the determined reference position (RP _T ) of the test object (3) to the validation unit (10). Arrangement according to claim 1 or 2, wherein the rail vehicle (6) comprises a receiving unit (11) for receiving the determined reference position (RP _T ) of the test object (3). Arrangement according to one of the preceding claims, wherein the rail vehicle (6) comprises the validation unit (10). Arrangement according to claim 1, wherein - the test object (3) - the validation unit (10) and - a receiving unit (5') for receiving the absolute position (AP) of the test object (3) determined by the position determination unit (8). and - the rail vehicle (6) has a transmitting unit (11') for transmitting the absolute position (AP) of the test object (3) determined by the position determination unit (8). Arrangement according to one of claims 1 or 2, having an evaluation device (12) separate from the rail vehicle (6) and the test object (3), comprising: - the validation unit (10), - a receiving unit (5") for receiving the absolute position (AP) of the test object (3) determined by the position determination unit (8) and the reference position (RP _T ) of the test object (3), wherein the rail vehicle (6) has a transmission unit (11') for transmitting the absolute position (AP) of the test object (3) determined by the position determination unit (8). Arrangement according to one of the preceding claims, wherein the first and/or the second self-localization unit (4, 9) comprises at least one of the following types of self-localization units: - a navigation unit for receiving a satellite navigation signal, - a receiving unit for receiving RTK correction data, - an inertial measurement unit. Arrangement according to one of the preceding claims, wherein the position determination unit (8) is set up to determine a path of a center line (2a) of the test rail section (2) based on the ego position (EP) of the rail vehicle (6). Arrangement according to claim 8, wherein - the position determination unit (8) is set up to determine a distance (d _r ) of the test object (3) from the center line (2a), and - the validation unit (10) is set up to determine a deviation (AW) of the determined distance (d _r ) of the test object (3) from the center line (2a) from a reference distance of the test object (3) to the center line (2a). Arrangement according to one of the preceding claims, wherein - the position determination unit (8) is set up to determine a longitudinal distance (d _l ) between the rail vehicle (6) and the test object (3), and - the validation unit (10) is set up to measure a deviation (AW) between the detected longitudinal distance (d _l ) between the rail vehicle (6) and the test object (3) and a deviation based on the reference position (RP _T ) of the test object (3 ) and the determined ego position (EP) of the rail vehicle (6) to determine the longitudinal reference distance (d _lref ) between the rail vehicle (6) and the test object (3). Arrangement according to one of the preceding claims, wherein the test rail section (2) has a straight course. Arrangement according to one of the preceding claims, wherein the test rail section (2) comprises a curvilinear course and a center line (2a) of the test rail section (2) is approximated by rectilinear sections. Method for validating a function of a sensor unit (7) of a rail vehicle (6) for object localization, comprising the steps: - Determining a reference position (RP _T ) of a test object (3) by a first self-localization unit (4) arranged on the test object (3), - detecting a relative position (RLP) of the test object (3) to a sensor unit (7) arranged on the rail vehicle (6), - Determining an ego position (EP) of the rail vehicle (6) by a second self-localization unit (9) arranged on the rail vehicle (6), - Determining an absolute position (AP) of the test object (3) based on the detected relative position (RLP) of the test object (3) to the sensor unit (7) and the determined ego position (EP) of the rail vehicle (6), - Validating an object localization of the sensor unit (7) by determining a deviation (AW) of the determined absolute position (AP) of the test object (3) from the reference position (RP _T ) of the test object (3). Computer program product, comprising commands which, when the program is executed by a computer, cause it to carry out the steps of determining an absolute position (AP) of the test object (3) and validating the object localization of the method according to claim 13. Computer-readable storage medium comprising instructions which, when executed by a computer, cause it to carry out the steps of determining an absolute position (AP) of the test object (3) and validating the object localization of the method according to claim 13.

Images