Ambient sensing device for a rail vehicle, comprising an adaptable, particularly lidar-based, sensing device

Abstract

The invention relates to an ambient sensing device (16) for a rail vehicle (10). The invention also relates to a method for operating the ambient sensing device (16) for a rail vehicle (10). The invention also relates to a rail vehicle (10) having such an ambient sensing device (16).

Description

[0001] Title: Environment detection device for a

[0002] Rail vehicle with an adaptable, in particular LiDAR-based, detection device

[0003] Description

[0004] The invention relates to an environment detection device for a rail vehicle. The invention also relates to a method for operating the environment detection device for a rail vehicle. Furthermore, the invention relates to a rail vehicle with such an environment detection device.

[0005] In rail transport, it is common for a train driver to survey the track ahead and react accordingly to visually detectable obstacles, hazards, and signals. To increase safety in rail transport or, if necessary, to enable assisted driving or even autonomous driving of rail vehicles, it would be desirable to automate such functions.

[0006] LiDAR-based detection devices, for example, are known for detecting an environment. LiDAR (abbreviation for Light detection and ranging or Light imaging, detection and ranging) is a method for optical distance and speed measurement. It is a form of three-dimensional laser scanning, whereby LiDAR enables three-dimensional pixel-by-pixel scanning of an environment, for example of objects in the environment. LiDAR-based detection devices emit laser pulses and detect the backscattered light. The distance to the location of the scattering is calculated from the light propagation time of the detected signals. This allows detection and, if necessary, identification of objects in the environment based on the detected signals.Using suitable computing devices and / or computing methods, the following data types, for example, can be determined based on detected signals: a relative speed between a detected object and the rail vehicle, a distance between a detected object and the rail vehicle, and an ego position of the rail vehicle. The corresponding speed values ​​can be used to classify and identify a detected object. Knowledge of highly accurate speed information can also be used to support or improve the accuracy of determining the ego position of a rail vehicle.

[0007] Due to the long braking distances required in rail transport, assisted driving or even autonomous driving of rail vehicles requires sensor systems that can detect an area far away from the rail vehicle, particularly in the vicinity of the track traveled by the rail vehicle. However, the sensors used to date generally have only a low resolution, so that classification or recognition of different objects over long distances is not possible.

[0008] The object of the present invention is to enable improved environmental detection in rail vehicles.

[0009] This object is achieved by an environment detection device for a rail vehicle according to claim 1, a method for operating the environment detection device according to claim 6 and a rail vehicle according to claim 11.

[0010] The environment detection device according to the invention for a rail vehicle has at least one detection device based on the detection of electromagnetic radiation, in particular a LiDAR-based detection device, which is designed to detect an environment of the rail vehicle in a field of view which is spanned by a horizontal image angle and a vertical image angle, wherein the detection device is designed such that a plurality of fields of view, at least two fields of view, can be detected with the detection device, wherein a respective field of view is spanned by a respective horizontal image angle and a respective vertical image angle,wherein a respective horizontal angle of view of a respective field of view deviates from the horizontal angle of view of a respective other field of view and / or wherein the vertical angle of view of a respective field of view deviates from the vertical angle of view of a respective other field of view, and wherein the detection device is switchable between detection of the respective fields of view.

[0011] For example, the detection device is set up in such a way that a first field of view, which is spanned by a first horizontal image angle and a first vertical image angle, can be detected at least temporarily with the detection device, and wherein at least one further, second field of view, which is spanned by a second horizontal image angle and a second vertical image angle, can be detected at least temporarily, wherein the first horizontal image angle deviates from the second horizontal image angle and / or wherein the first vertical image angle deviates from the second vertical image angle, and wherein the detection device can be switched between detecting the first field of view and detecting the at least one further, second field of view.

[0012] The first horizontal angle of view of the first field of view is, for example, larger than the second horizontal angle of view of the second field of view. The first vertical angle of view of the first field of view is, for example, larger than the second vertical angle of view of the second field of view. However, it is sufficient if either the horizontal angles of view or the vertical angles of view, preferably the horizontal angles of view, differ.

[0013] The detection device is thus designed to detect at least two fields of view of different sizes, for example a first field of view and a second field of view, wherein the first field of view is larger, in particular comprises a larger horizontal angle of view and thus a larger horizontal extent, than the second field of view.

[0014] The number of two fields of view is merely an example. The detection device can be set up such that three or four or five or more, in particular any desired number of fields of view can be detected, wherein the fields of view differ in their horizontal and / or vertical angle of view and thus in their vertical and / or horizontal extent. A first field of view that can be detected with the detection device is spanned, for example, by a first horizontal angle of view, which is, for example, 50° to 130°, preferably 60° to 120°, in particular 60°, and by a first vertical angle of view, which is, for example, 15° to 50°, preferably 20° to 40°, in particular 30°.

[0015] A further field of view, for example a second field of view, which can be detected by the detection device, is spanned by a further, for example a second, horizontal angle of view, which is for example 40° to 125°, preferably 50° to 115°, in particular 50° to 55°, and by a further, for example second, vertical angle of view, which is for example 10° to 45°, preferably 15° to 35°, in particular 30 to 35°.

[0016] The further horizontal angle of view and / or the further vertical angle of view deviates from the first horizontal angle of view and / or from the first vertical angle of view by 5° to 10°, in particular by 5° or 6° or 8° or 10°, and is in particular smaller by 5° to 10°, in particular by 5° or 6° or 8° or 10°.

[0017] According to the invention, it is provided, for example, that a respective horizontal image angle of a respective field of view deviates from the horizontal image angle of a respective other field of view by 5° to 10°, in particular by 5° or 6° or 8° or 10°, and in particular by 5° to 10°, and / or a respective vertical image angle of a respective field of view deviates from the vertical image angle of a respective other field of view by 5° to 10°, in particular by 5° or 6° or 8° or 10°, and in particular by 5° to 10°. According to the invention, a plurality of fields of view can therefore be detected with the detection device, wherein the plurality of fields of view differ from one another in that the respective horizontal image angles and / or the respective vertical image angles are different.For example, a respective further horizontal angle of view and / or a respective further vertical angle of view of a further field of view is 5° to 10° smaller than the horizontal angle of view and / or the vertical angle of view of a respective other field of view.

[0018] The number of pixels, for example, 2048x1024, remains the same for each field of view when the detection device scans the surroundings pixel-by-pixel in the field of view, meaning it is not changed. This allows for a higher resolution when scanning a smaller field of view than when scanning a larger field of view. A higher resolution enables better detection and, if necessary, better classification of objects.

[0019] According to one embodiment, the orientation of a respective field of view can differ from the orientation of another field of view. The orientation can also be specified individually. For example, a field of view can be aligned centrally to a central longitudinal axis of the rail vehicle or deviating from the central orientation. A first field of view is, for example, the largest possible detectable field of view in a basic orientation aligned centrally to the central longitudinal axis. The orientation of a respective further field of view can, for example, be individually specified deviating from the central orientation to the central longitudinal axis within the extent of the first field of view in the basic orientation.

[0020] According to one embodiment, the surroundings detection device comprises a control device for controlling the detection device, wherein the control device is designed to control the detection device in such a way that a respective field of view can be detected with the detection device and / or to switch between the detection of a respective field of view and the detection of a respective other field of view. The control device therefore controls the detection device in such a way that fields of view of different sizes, i.e. fields of view with a different horizontal and / or vertical extent, can be detected. The control device therefore controls the detection device, for example, in such a way that a first field of view is initially detected and then switching is carried out to detect a second field of view and the second field of view is detected.In the following, a large or comparatively larger field of view is understood to mean a field of view with a larger horizontal and / or vertical extent than a small or comparatively smaller field of view. Accordingly, a small or comparatively smaller field of view is understood to mean a field of view with a smaller horizontal and / or vertical extent.

[0021] The control device can also be configured to control the detection device in such a way that an orientation of a respective field of view can be specified.

[0022] According to one embodiment, it is provided that the environment detection device comprises a computing device or that a computing device is assigned to the environment detection device, wherein the computing device is designed to detect and / or classify an object in a field of view of the detection device based on the detection of the environment of the rail vehicle in the field of view of the detection device. The detection and / or classification takes place, for example, by means of appropriate software, in particular an algorithm for evaluating the detected signals and for image recognition, in particular also using a neural network.

[0023] Based on the evaluation of the detected signals, in particular based on the detection and / or classification of objects in the surroundings of the rail vehicle in the field of view of the detection device, corresponding messages can be output, for example in the form of information to a central control center and / or in the form of warning signals to a train driver. Control signals for controlling the rail vehicle, for example for braking the rail vehicle and / or changing a speed of the rail vehicle, can also be generated and / or output. In particular, the rail vehicle can also be controlled accordingly.

[0024] According to one embodiment, it is provided that the environment detection device is set up so that the control of the detection device by means of the control device or the switching between a detection of a respective field of view and a detection of a respective, different field of view takes place depending on one or more of the following factors: a) a speed of the rail vehicle, b) a position of the rail vehicle, c) a position of the rail vehicle in relation to a region of interest (ROI), d) an object detected in a field of view, e) an object classified in a field of view. It can be provided, for example, that a comparatively smaller field of view is detected at a higher speed of the rail vehicle, and a comparatively larger field of view is detected at a comparatively lower speed of the rail vehicle.It can be provided, for example, that at a certain position of the rail vehicle, for example outside of town, a smaller field of view is recorded, and at a different position of the rail vehicle, for example in town, in particular initially, a comparatively larger field of view is recorded.

[0025] A region of interest (ROI) is, for example, a system-inherently defined area, such as a train station, a railroad crossing, a bridge crossing, a tunnel entrance, a tunnel exit, etc. A region of interest (ROI) can also be defined, particularly depending on a situation or event, particularly by external triggering. Defining an ROI based on an external trigger can be advantageous, for example, if people or unwanted objects are detected or suspected in a specific area.

[0026] It can be provided, for example, that when the rail vehicle approaches such an area of ​​interest, or is located in such an area of ​​interest, in particular initially, a larger field of view is detected. It can be provided, for example, that when an object is detectable in a field of view, the system switches to detecting a comparatively smaller field of view in order to be able to detect the object at a higher resolution. A higher resolution can potentially result in improved or reliable classification of the object. It can be provided, for example, that when an object has been successfully or reliably classified in a field of view, the system switches to detecting a comparatively larger field of view in order to enlarge the detectable area again.

[0027] According to one embodiment, it is provided that the environment detection device is configured such that the control of the detection device by means of the control device or the switching between a detection of a respective field of view and a detection of a respective, different field of view takes place in such a way that the respective field of view is first detected by means of the detection device in an N-stage detection with N=(2, 3, 4, ...) in a respective n-th with level n = ( 1 to N) an n-th field of view with an n-th horizontal angle of view and an n-th vertical angle of view can be detected, wherein the next level of the N-level detection can be carried out if in a respective n-th level in the n-th field of view an object is detectable but not successfully classified, wherein the n-th horizontal angle of view of the n-th field of view of the n-th level is greater than the (n+l)-th horizontal angle of view of the (n+l)-th field of view of the (n+l)-th level and / or the n-th vertical angle of view of the n-th field of view of the n-th level is greater than the (n+l)-th vertical angle of view of the (n+l)-th field of view of the (n+l)-th level.It can therefore be provided that a larger field of view is captured first, and if an object is detectable in the field of view, the next stage of capture is carried out in which a comparatively smaller field of view is captured in order to be able to capture the object at a higher resolution. A higher resolution can possibly achieve an improved or reliable classification of the object. It may be that capturing the comparatively smaller field of view does not yet allow a reliable or successful classification of the object. In this case, a further stage of capture can be carried out in which an even smaller field of view is captured in order to be able to capture the object at an even higher resolution, and so on. The execution of the N-stage capture can be ended or aborted if a successful orReliable classification of the object has been achieved and / or when no object is detected anymore and / or after a predefined maximum number of stages has been performed. After completing the N-stage detection, it is possible to switch back to detecting a comparatively large field of view.

[0028] Further embodiments relate to a method for operating an environment detection device for a rail vehicle, according to the described

[0029] Embodiments. The method according to the invention thus has the advantages of the environment detection device according to the invention.

[0030] The method comprises: operating the detection device in such a way that the detection device at least temporarily detects a respective field of view and at least temporarily detects a respective other field of view, and switching between the detection of the respective field of view and the detection of the respective other field of view.

[0031] According to one embodiment, the method comprises: controlling the detection device by means of a control device in such a way that the detection device either detects a respective field of view, or at least detects a respective other field of view, and / or switches between the detection of the respective field of view and the detection of the respective other field of view.

[0032] According to one embodiment, the method comprises: detecting and / or classifying an object in a field of view of the detection device based on the detection of the surroundings of the rail vehicle in the field of view of the detection device by means of a computing device.

[0033] According to one embodiment, it is provided that the operation and / or control of the detection device by means of the control device or a switching between a detection of the respective field of view and a detection of the respective, other field of view takes place depending on one or more of the following factors: a) a speed of the rail vehicle, b) a position of the rail vehicle, c) a position of the rail vehicle in relation to a region of interest (ROI), d) an object detected in a field of view, e) an object classified in a field of view.

[0034] According to one embodiment, it is provided that the operation and / or control of the detection device by means of the control device or a switching between a detection of a respective field of view and a detection of a respective, different field of view takes place in such a way that by means of the detection device, in an N-stage detection with N=(2, 3, 4, ...) in a respective n-th with stage n = ( 1 to N) an n-th field of view with an n-th horizontal angle of view and an n-th vertical angle of view is detected, wherein the next stage of the N-stage detection is carried out if in a respective n-th stage in the n-th field of view an object is detected but not successfully classified, wherein the n-th horizontal angle of view of the n-th field of view of the n-th stage is greater than the (n+l)-th horizontal angle of view of the (n+l)-th field of view of the (n+l)-th stage and / or the n-th vertical angle of view of the n-th field of view of the n-th stage is greater than the (n+l)-th vertical angle of view of the (n+l)-th field of view of the (n+l)-th stage.

[0035] Further embodiments relate to a rail vehicle comprising an environment detection device according to the described embodiments. The rail vehicle according to the invention has the environment detection device according to the invention. The rail vehicle according to the invention thus has the advantages of the environment detection device according to the invention. Some components of the environment detection device according to the invention can, if necessary after supplementing with hardware systems, such as a sensor unit and mechanical components such as holding devices, be partially or predominantly designed in the form of software components. This particularly applies to parts of the control device and the computing device.

[0036] In principle, these components can also be partially implemented, particularly for performing fast and powerful computing steps, in the form of software-supported hardware, such as FPGAs or the like. Likewise, required interfaces, for example, for transferring data from other software components, can be designed as software interfaces. Interfaces can also be implemented as hardware interfaces that are controlled by suitable software.

[0037] A largely software-based implementation can be advantageous since it means that computer systems already present in a rail vehicle can be easily retrofitted with a software update after any addition of additional hardware elements, such as additional sensor units, in order to be operated in the manner according to the invention. In this respect, the object is also achieved by a corresponding computer program product with a computer program which can be loaded directly into a memory device of a control device and / or a computing device of a rail vehicle, with program sections in order to carry out the steps of the method according to the invention which can be carried out by software when the computer program is executed in the control device and / or in the computing device.

[0038] Such a computer program product may, in addition to the computer program, include additional components such as documentation and / or additional components, including hardware components for using the software.

[0039] A computer-readable medium, such as a memory stick, a hard disk, or another portable or permanently installed data storage device, can be used for transport to the storage device and / or for storage, on which the program sections of the computer program that can be read and executed by the control device and / or the computing device are stored. For this purpose, the control device and / or the computing device can, for example, have one or more cooperating microprocessors, graphics processors, or the like.

[0040] Further advantages will become apparent from the description and the accompanying drawings. Exemplary embodiments of the invention are illustrated in the drawings and explained in more detail in the following description. Like reference symbols in different figures designate like elements or at least elements that are functionally comparable. When describing individual figures, reference may also be made to elements from other figures. These show, in schematic form:

[0041] Fig. 1 is a schematic view of a rail vehicle with an environment detection device in a view from above (a)) and in a view from the side (b)) according to a first embodiment;

[0042] Fig. 2 is a schematic view of a rail vehicle with an environment detection device in a view from above according to a further embodiment (a)) and according to a further embodiment (b));

[0043] Fig. 3 shows a detailed view of the environment detection device in a view from above according to the first embodiment (a), according to a further embodiment (b) and according to a further embodiment (c);

[0044] Fig. 4 is a schematic view of a rail vehicle with an environment detection device in a view from above in a first position at a time t1 and in a second position at a time t2.

[0045] In the following, the function and structure of an environment detection device according to the invention and a method for operating an environment detection device are explained with reference to the figures.

[0046] Figure 1 shows a schematic view of a rail vehicle, which is provided with the reference number 10 in its entirety. Fig. 1 a) shows the rail vehicle 10 in a view from above and Fig. 1 b) in a view from the side. The rail vehicle 10 is located on a rail line 12. The rail vehicle 10 comprises an environment detection device 14. By means of the environment detection device 14, the environment of the rail vehicle 10 is detected, in particular comprising a part of the rail line 12 in front of the rail vehicle 10 and an area around the rail line 12.

[0047] The surroundings detection device 14 comprises at least one detection device based on the detection of electromagnetic radiation, in particular a LiDAR-based detection device 16. In the following, the invention is explained using a LiDAR-based detection device 16 as an example. However, it can also be another detection device based on the detection of electromagnetic radiation. The LiDAR-based detection device 16 is configured to detect the surroundings of the rail vehicle 10 in a field of view 18. The field of view 18 is spanned by a horizontal image angle a_h and a vertical image angle a_v.

[0048] The LiDAR-based detection device 16 enables three-dimensional pixel-by-pixel scanning of the surroundings within the field of view 18. The LiDAR-based detection device 16 emits laser pulses and detects the backscattered light. The distance to the location of the scattering can then be calculated from the light propagation time of the detected signals. Thus, detection and, if necessary, identification of objects in the surroundings can be carried out on the basis of the detected signals. For this purpose, the surroundings detection device 14 comprises a computing device 20. According to a further embodiment, the computing device 20 can also be provided not as part of the surroundings detection device 14, but rather assigned to it. The computing device 20 is designed to detect and / or classify an object in a field of view 18 of the LiDAR-based detection device 16 based on the detected signals.The detection and / or classification is carried out, for example, by means of appropriate software, in particular an algorithm for evaluating the detected signals and for image recognition, in particular using a neural network.

[0049] Based on the evaluation of the detected signals, in particular based on the detection and / or classification of objects in the surroundings of the rail vehicle in the field of view of the LiDAR-based detection device, corresponding messages can be output, for example in the form of information to a central control center and / or in the form of warning signals to a train driver. Control signals for controlling the rail vehicle 10, for example for braking the rail vehicle 10 and / or changing a speed of the rail vehicle 10, can also be generated and / or output. In particular, the rail vehicle 10 can also be controlled accordingly.

[0050] Figure 2 shows a schematic view of the rail vehicle 10 in a view from above according to two further embodiments, cf. Fig. 2 a) and Fig. 2 b). The LiDAR-based detection device 16 is configured so that a plurality of different fields of view 18 can be detected with the LiDAR-based detection device. In the example, two fields of view 18-1, 18-2 (cf. Fig. 2 a) and 18-1, 18-2' (cf. Fig. 2 b)) are shown. The fields of view 18-1, 18-2 and 18-1, 18-2' are not detected simultaneously, but one after the other. Thus, switching occurs from the detection of the field of view 18-1 to the detection of the field of view 18-2 or 18-2' or vice versa.

[0051] In the example, the fields of view 18-1 and 18-2, or 18-1 and 18-2', differ in that the horizontal angle of view a_h_l of the field of view 18-1 differs from the horizontal angle of view a_h_2 or a_h_02 ' of the field of view 18-2 or 18-2'. In the example, the horizontal angle of view a_h_2 or a_h_02 ' is smaller than the horizontal angle of view a_h_l . Accordingly, the field of view 18-2 or 18-2' covers a smaller horizontal extent than the field of view 18-1. In comparison, the fields of view 18-2 or 18-2' are therefore smaller than the field of view 18-1.

[0052] According to Fig. 2 a), the field of view 18-1 and the field of view 18-2 are aligned centrally with respect to a central longitudinal axis Z of the rail vehicle 10. According to Fig. 2 b), the field of view 18-1 is aligned centrally with respect to the central longitudinal axis Z of the rail vehicle 10. The field of view 18-2' is aligned deviating, in the example rotated to the right, from the central longitudinal axis Z of the rail vehicle 10. The field of view 18-1 is, for example, the largest possible detectable field of view in a basic orientation aligned centrally with respect to the central longitudinal axis Z. For a respective field of view 18-2, 18-2', an orientation deviating from a central orientation with respect to the central longitudinal axis Z within the extent of the field of view 18-1 in the basic orientation can therefore be specified.

[0053] In the example, the surroundings detection device comprises a control device 22 for controlling the LiDAR-based detection device 16. The control device is configured to control the LiDAR-based detection device 16 in such a way that a respective field of view 18, 18-1, 18-2, 18-2' can be detected with the LiDAR-based detection device 16, and / or to switch between detecting a respective field of view 18, 18-1 and detecting a respective other field of view 18, 18-2, 18-2'. The control device 22 can, for example, also assign an orientation of the respective field of view with respect to the central longitudinal axis Z of the rail vehicle 10 and control the LiDAR-based detection device 16 accordingly.

[0054] The control of the LiDAR-based detection device 16 by means of the control device 22 or the switching between a detection of a respective field of view 18, 18-1 and a detection of a respective, different field of view 18, 18-2, 18-2' takes place, for example, depending on one or more of the following factors: a) a speed of the rail vehicle, b) a position of the rail vehicle, c) a position of the rail vehicle in relation to a region of interest (ROI), d) an object detected in a field of view, e) an object classified in a field of view. Regarding a): It can be provided, for example, that at a higher speed of the rail vehicle 10 a smaller field of view 18-2, 18-2' is detected, and at a comparatively lower speed of the rail vehicle a comparatively larger field of view 18-1 is detected.Regarding b): It can be provided, for example, that for a certain position of the rail vehicle, for example outside of town, a smaller field of view 18-2, 18-2' is detected, and for a different position of the rail vehicle, for example in town, in particular initially, a comparatively larger field of view 18-1 is detected. Regarding d) and e): It can be provided, for example, that if an object is detectable in a field of view 18-1, the detection of a comparatively smaller field of view 18-2, 18-2' is switched over in order to be able to detect the object in a higher resolution. A higher resolution can possibly lead to an improved or reliable classification of the object. It can be provided, for example, that if an object is successfully or unsuccessfully detected in a field of view.has been reliably classified, switching to the detection of a comparatively larger field of view 18-1 is carried out in order to enlarge the detectable area again. The number of pixels in the pixel-by-pixel scanning of the surroundings in the field of view 18, 18-1, 18-2, and 18-2' by means of the LiDAR-based detection device 16 remains the same for each field of view 18, 18-1, 18-2, and 18-2', i.e., it is not changed. As a result, a higher resolution can be achieved when scanning in a smaller field of view 18-2, 18-2' than when scanning in a larger field of view 18-1. A higher resolution enables better detection and, if necessary, better classification of objects.

[0055] With reference to Fig. 3, for example, the detection and capture of an object is explained. First, for example, at a time t1, the surroundings are captured in a field of view 18-1. The field of view 18-1 encompasses a horizontal angle a_h_l. An object 0 is detected in the field of view 18-1. However, the detected signals are not yet sufficient to reliably classify and thus identify the object 0.

[0056] In the example, the LiDAR-based detection device 16 is controlled by the control device 22 to switch from detecting the field of view 18-1 to detecting a field of view 18-2. At time t2, in the example, the surroundings are encompassed in the field of view 18-2. The field of view 18-2 encompasses a horizontal angle a_h_2. The horizontal angle a_h_2 is smaller than the horizontal angle a_h_l. Object 0 is also detected in the field of view 18-2. The number of pixels in the pixel-by-pixel scanning of the surroundings in the fields of view 18-1 and 18-2 remains the same. This allows object 0 in the field of view 18-2 to be detected with a slightly higher sampling rate. For example, the detected signals are not yet sufficient to reliably classify and thus identify object 0.

[0057] In the example, the LiDAR-based detection device 16 is controlled by the control device 22 to switch from detecting the field of view 18-2 to detecting a field of view 18-3. At time t3 in the example, the surroundings are encompassed in the field of view 18-3. The field of view 18-3 encompasses a horizontal angle ah 3. The horizontal angle a_h_3 is smaller than the horizontal angle a_h_2. Object 0 is also detected in the field of view 18-3. The number of pixels in the pixel-by-pixel scanning of the surroundings in the fields of view 18-1, 18-2, and 18-3 remains the same. This allows object 0 in the field of view 18-3 to be detected with a slightly higher sampling rate. In the example, object 0 in the field of view 18-3 can be detected more accurately due to the higher resolution and can thus be reliably classified and detected.

[0058] For example, after the successful classification of object 0, the system switches back to capturing a field of view 18-1 in order to be able to capture a wider area again.

[0059] For example, with reference to Fig. 4, the acquisition of a field of view is explained when the rail vehicle 10 approaches a region of interest, ROI, or is located in a region of interest.

[0060] An area or region of interest (ROI) is, for example, a train station, a level crossing, a bridge passage, a tunnel entrance, a tunnel exit, etc. In areas of train stations or level crossings, for example, there is an increased number of pedestrians and / or cars. For example, stone deposits may be more common in the area of ​​tunnel entrances or exits along the tracks. Depending on the ROI, it can be provided, for example, that a larger or smaller field of view is recorded when approaching the ROI. In the example, the rail vehicle is moving over land outside of populated areas. For example, the rail vehicle 10 is traveling at a higher speed and initially a smaller field of view 18-2 is recorded at time t=1. At time t=2, the rail vehicle 10 has approached the ROI. The ROI is, for example, a level crossing.At time t=2, a larger field of view 18-1 is acquired to enlarge the detectable area. Depending on the detection of an object in the field of view 18-1, it is also possible to switch back to the acquisition of a field of view 18-2 to improve the resolution, see Fig. 3.

[0061] In a further development of the invention, it can be provided that a siding is at least partially detected during the detection of the surroundings. This allows the siding to be monitored as well. If an object is detected in the area of ​​the siding and, if necessary, also classified, a message can be sent, for example, to a central control center or to an oncoming rail vehicle.

[0062] In a further development of the invention it can be provided that when the surroundings are detected it is at least partially detected whether the width of a route has changed, for example it has become narrower. A narrowing of the route can be due, for example, to undercutting, rock avalanches, snow avalanches or debris avalanches. In a further development of the invention it can be provided that the detection of a respective field of view or the switching of the detection to another field of view is predetermined depending on the course of the route. For example the field of view can be widened in the area of ​​a curve and / or the alignment of the field of view in relation to the central longitudinal axis of the rail vehicle can be adapted to the course of the route, for example the course of a curve.

[0063] In a further development of the invention, it can be provided that the environment detection device 16 or the method for operating the environment detection device 16 exchanges information and data with other systems of the rail vehicle 10 and / or accesses other systems of the rail vehicle to obtain information or data. For example, data from a navigation device or stored route map of the rail vehicle can be used to determine a position of the rail vehicle, in particular also with regard to ROIs, or to improve the determination of the position. A navigation device uses, for example, suitable navigation satellite systems such as GPS and / or Galileo to determine the location. Such navigation satellite systems are combined with an inertial measurement unit (IMU), for example to bridge their function, particularly in tunnels.For example, current driving data, such as speed values, from a control device of the rail vehicle can be used to determine a distance or a time of reaching a position or in any case to improve the determination.

[0064] In a further development of the invention, it can be provided that the environment detection device 16 or the method for operating the environment detection device 16 comprises further detection devices, for example a RADAR-based detection device, an ultrasonic sensor system, a camera-based, preferably stereo camera-based detection device, etc., or receives data and information from such detection devices. Based on data or information from the further detection device, for example, the detection and / or classification of an object can be improved. The data or information from the further detection device can also be used to specify the detection of a respective field of view.

Claims

Patent claims 1. Environment detection device (14) for a rail vehicle (10), comprising at least one detection device based on the detection of electromagnetic radiation, in particular a LiDAR-based detection device, which is configured to detect an environment of the rail vehicle (10) in a field of view spanned by a horizontal image angle (a_h) and a vertical image angle (a_v), wherein the detection device is configured such that a plurality of fields of view (18, 18-1, 18-2, 18-2', 18-3), at least two fields of view, can be detected with the detection device, wherein a respective field of view (18) is spanned by a respective horizontal image angle (a_h) and a respective vertical image angle (a_v),wherein a respective horizontal image angle (a_h) of a respective field of view deviates from the horizontal image angle (a_h) of a respective other field of view (18) and / or wherein the vertical image angle (a_v) of a respective field of view (18) deviates from the vertical image angle (a_v) of a respective other field of view (18), and wherein the detection device (16) is switchable between detecting the respective fields of view (18, 18-1, 18-2, 18-2', 18-3).

2. Environment detection device (14) according to claim 1, wherein the environment detection device (14) comprises a control device (22) for controlling the detection device (16), wherein the control device (22) is configured to control the detection device (16) in such a way that a respective field of view (18, 18-1, 18-2, 18-2', 18-3) can be detected with the detection device (16), and / or can be switched between the detection of a respective field of view (18-1, 18-2, 18-2', 18-3) and the detection of a respective other field of view (18-1, 18-2, 18-2', 18-3).

3. Environment detection device (14) according to one of claims 1 or 2, wherein the environment detection device (14) comprises a computing device (20) or wherein the environment detection device (14) is assigned a computing device (20), wherein the computing device (20) is designed to detect and / or classify an object (0) in a field of view (18) of the detection device (16) based on the detection of the environment of the rail vehicle (10) in the field of view (18) of the detection device (16).

4. Environment detection device (14) according to one of the preceding claims, wherein the environment detection device (14) is designed such that the control of the detection device (16) by means of the control device (22) or the switching between a detection of a respective field of view (18-1, 18-2, 18-2', 18-3) and a detection of a respective, other field of view (18-1, 18-2, 18-2', 18-3) takes place depending on one or more of the following factors: a) a speed of the rail vehicle (10), b) a position of the rail vehicle (10), c) a position of the rail vehicle (10) in relation to an area of ​​interest, engl. Region of Interest, ROI, d) of an object (0) detected in a field of view (18-1, 18-2, 18-2', 18-3), e) of an object (0) classified in a field of view (18-1, 18-2, 18-2', 18-3). Surroundings detection device (14) according to one of the preceding claims, wherein the surroundings detection device (14) is configured such that the control of the detection device (16) by means of the control device (22) or the switching between a detection of a respective field of view (18-1, 18-2, 18-2', 18-3) and a detection of a respective, other field of view (18-1, 18-2, 18-2', 18-3) is carried out in such a way that by means of the detection device (16) first the respective field of view (18-1, 18-2, 18-2', 18-3) that by means of the detection device (16) in an N-stage detection with N=(2, 3, 4, ... ) in a respective n-th with stage n=(1 to N) an n-th field of view (18-n) with an n-th horizontal angle of view () and a n-th vertical angle of view, wherein the next stage of the N-stage detection can be carried out if in a respective n-th stage in the n-th field of view (18-n) an object (0) is detectable but not successfully classified, wherein the n-th horizontal angle of view (a_h_n) of the n-th field of view (18-n) of the n-th stage is greater than the (n+l)-th horizontal angle of view (a_h_n+l) of the (n+l)-th field of view of the (n+l)-th stage and / or the n-th vertical angle of view (a_v_n) of the n-th field of view (18—n) of the n-th stage is greater than the (n+l)-th vertical angle of view (a_v_n+l) of the (n+l)-th field of view (18-n+l) of the (n+l)-th stage.

6. Method for operating an environment detection device (14) for a rail vehicle (10) according to one of claims 1 to 5, the method comprising operating the detection device (16) in such a way that the detection device (16) at least temporarily detects a respective field of view (18) and at least temporarily a respective other field of view (18), and switching between the detection of the respective field of view (18) and the detection of the respective other field of view (18).

7. The method according to claim 6, wherein the method comprises: controlling the detection device (16) by means of a control device (22) such that with the detection device (16) either a respective field of view (18) is detected, or at least one respective other field of view (18) is detected, and / or switching is made between the detection of the respective field of view (18) and the detection of the respective other field of view (18).

8. The method according to one of claims 6 or 7, wherein the method comprises: detecting and / or classifying an object (0) in a field of view (18) of the detection device (16) based on the detection of the surroundings of the rail vehicle (10) in the field of view (18) of the detection device (16) by means of a computing device (20).

9. The method according to one of claims 6 to 8, wherein the operation and / or control of the detection device (16) by means of the control device (22) or a switching between a detection of the respective field of view and a detection of the respective other field of view takes place depending on one or more of the following factors: a) a speed of the rail vehicle (10), b) a position of the rail vehicle (10), c) a position of the rail vehicle (10) in relation to a region of interest, ROI, d) an object (0) detected in a field of view, e) an object (0) classified in a field of view (18).

10. Method according to one of claims 6 to 9, wherein the operation and / or control of the detection device (16) by means of the control device (22) or a switching between a detection of a respective field of view (18) and a detection of a respective, other field of view (18) is carried out in such a way that by means of the detection device (16) in an N-stage detection with N=(2, 3, 4, ...) in a respective n-th with stage n = ( 1 to N) an n-th field of view with an n-th horizontal angle of view and an n-th vertical angle of view is detected, wherein the next stage of the N-stage detection is carried out if in a respective n-th stage in the n-th field of view an object is detected but not successfully classified, wherein the n-th horizontal angle of view of the n-th field of view of the n-th stage is greater than the (n + l)-th horizontal angle of view of the (n + l)-th field of view of the (n + l)-th stage and / or the n-th vertical angle of view of the n-th field of view of the n-th stage is greater than the (n + l)-th vertical angle of view of the (n + l)-th field of view of the (n + l)-th stage.

11. Rail vehicle (10) comprising a Environment detection device (14) according to one of the Claims 1 to 5.

12. Computer program product with a computer program which is stored in a memory unit of a control device (22) of a rail vehicle (10) is loadable, with program sections to carry out the method according to one of claims 6 to 10 when the computer program is executed in the control device (22).

13. Computer-readable medium on which Computer unit executable program sections are stored in order to carry out the method according to one of claims 6 to 10 when the program sections are executed by the computer unit.

Images