Method for monitoring a rail segment

Abstract

The invention relates to a method for monitoring obstacles on a track section, in which obstacles are detected by means of at least one first sensor running on a vehicle and at least one second sensor running on the track section in a stationary manner. The sensors are oriented towards the track section during operation such that each location of the track section is present in a first image of the at least one first sensor and in a second image of the at least one second sensor. The first image and the second image are analyzed for an obstacle. If an obstacle is detected in both the at least one first image and the at least one second image by the analysis, a signal indicating the presence of the obstacle is generated. If an obstacle is detected only in the at least one first image or only in the at least one second image by the analysis, a first signal indicating a fault is generated. The invention also relates to a railway installation, a computer program product and a computer-readable storage medium.

Description

TECHNICAL FIELD

[0001] The invention relates to a method for monitoring a track section. Furthermore, the invention also relates to a railway installation having a platform and a track section abutting the platform. The invention also relates to a computer program product comprising program instructions. The invention also relates to a computer-readable storage medium comprising data. BACKGROUND

[0002] According to the prior art, it is known that a platform can be monitored by installing a camera on the platform or on a train. For example, an operator of a railway operator monitors the camera images in order to be able to interrupt a train operation when an obstacle is found on a track section extending along the platform. An obstacle is an object or a living being whose dimensions can lead to a collision with an incoming train. Here, it is a foreign object on the track section that usually enters the track section from the platform. Similar situations also occur at dangerous areas of railway crossings.

[0003] As can be seen from the prior art described above, the manual monitoring of track sections, in particular platforms and railway crossings, requires a large amount of manpower and is therefore cost-intensive. In addition, the possibility of false decisions by the staff cannot be ruled out. However, the problem with monitoring automation is that the tasks to be processed relate to safety-relevant issues (safety in the description of the invention means operational safety, also referred to as safety) and therefore the automation must meet the safety requirements of railway operations.

[0004] In order to increase the reliability of obstacle detection in railway traffic, various approaches have been taken in the past. For this purpose, according to the documents EP 4124542 A1 and EP 4299411 A1, for example, markers are used which carry information in coded form in order to be detected as objects. These markers are intended to more reliably improve obstacle detection at railway crossings and the like, since the detection of all markers proves that no obstacle is obscuring the markers. The document EP 4389558 A1 also proposes that obstacle detection can be tested during the operation of the track vehicle in order to increase the reliability of obstacle detection. SUMMARY

[0005] The technical problem addressed by the invention is to eliminate the problems described in the prior art. The technical problem addressed by the invention is in particular to provide a method for monitoring a track section by means of which monitoring can be automated while achieving a high level of safety. Furthermore, the technical problem addressed by the invention is also to provide a vehicle, a computer program product and a computer-readable storage medium in order to be able to implement the improved method.

[0006] According to a first aspect of the present application, a method for monitoring a track section is described, wherein the track section is detected by at least one first sensor for imaging and at least one second sensor for imaging in order to monitor the track section.

[0007] The sensors for imaging are sensors whose measured values can be converted into an image of the environment to be monitored, primarily an image of the track section and the platform edge. This is, for example, an optical camera, radar, lidar, etc.

[0008] In the present description, the track section refers to a track facility, which comprises an upper structure consisting primarily of a track and sleepers (in addition, for example, fastening means for the rails, etc.), as well as a lower structure, which provides a foundation for the sleepers, for example, ballast or a concrete base. It is apparent that the aforementioned components of the track section form part of the visible surface and are thus also imaged in the imaging process. However, if these components are obscured by objects, this can be determined by analyzing the images generated in the imaging process in a manner known per se. The method can use artificial intelligence generated with computer assistance for this purpose.

[0009] The device is a computer-aided or computer-implemented device if it has a computing environment, or the method is a computer-aided or computer-implemented method if the computing environment performs at least one step of the method.

[0010] The computing environment refers to an IT infrastructure consisting of functional components, such as processors, storage units, programs, and data processed by the programs, which is used to execute at least one application program that needs to be completed. Other functional components can include sensors and actuators, which can enable the computing environment to interact with the outside world. The IT infrastructure can also be organized as a network of the aforementioned functional components.

[0011] The computing mechanism constitutes functional units in the computing environment, which can be assigned to the application programs, for example, consisting of a plurality of program modules, and can run the application programs. When running the application programs, these functional units constitute a closed system, which is physical (for example, computer, processor) and / or virtual (for example, program module).

[0012] The computer is an electronic device with data processing properties consisting of a plurality of functional components. The computer can be a client, a server, a handheld computer, a communication device, and other electronic devices for data processing, which can be equipped with a processor and a storage unit, and can also be connected to a network via an interface.

[0013] The processor can be a transducer, a sensor or an electronic circuit for generating a measurement signal. The processor can be a central processing unit (CPU), a microprocessor, a microcontroller or a digital signal processor, which can be used in combination with a storage unit for storing program instructions and data. The processor can also be understood as a virtualized processor or a soft CPU.

[0014] The storage unit can be designed as a computer-readable memory in the form of a working memory (random access memory, RAM) or a data memory (hard disk or data carrier).

[0015] The program modules are individual software functional units which can implement the method steps in the program flow according to the application. These software functional units can be implemented in a single computer program or in a plurality of computer programs which communicate with one another. The interfaces which are implemented here can be implemented in software technology in a single processor or in hardware technology when a plurality of processors are used.

[0016] The interfaces can be implemented in hardware technology, for example wired or wireless connections, or in software technology, for example the interaction between individual program modules of one or a plurality of computer programs, and preferably in the form of digital data sets or analog signals for exchanging data.

[0017] To avoid misunderstandings, it is noted here that the individual claim features are numbered with lower-case Latin letters, irrespective of the claim number. This means that each letter only occurs once in the entire claim group, so that the relevant claim feature can be specified unambiguously without reference to the claim number. The order of the letters is therefore not important.

[0018] According to the application provision is made for

[0019] a) the at least one first sensor is operated on a vehicle and the at least one second sensor is operated position-f fixedly on a track section, wherein the sensors are oriented in operation towards the track section, so that each location of the track section is present in the first image of the at least one first sensor and in the second image of the at least one second sensor respectively,

[0020] b) the first and second images are computer-aided analyzed for obstacles in the track section,

[0021] c) if an obstacle is identified in both the at least one first image and the at least one second image by the analysis, a signal is generated which indicates the presence of an obstacle,

[0022] d) if an obstacle is identified in the at least one first image only or in the at least one second image only by the analysis, a first signal is generated which indicates a fault.

[0023] The terms used in the present description have the following meanings.

[0024] In the present description, when referring to a first sensor and a second sensor, the first sensor shall be associated with the vehicle and thus shall be mobile on the vehicle, while the second sensor shall be associated with the track section and thus shall be stationary in order to be able to continuously operate on the relevant track section. In contrast thereto, the first sensor can only operate on the track section while the vehicle is passing the relevant track section. If multiple sensors are used on the vehicle for the purpose of the present invention, all of these sensors are first sensors. If multiple stationary second sensors are used on the track section for the purpose of the present invention, all of these sensors are second sensors.

[0025] In the present description, when referring to detection by generating images, it is meant that the sensor used for imaging scans the track section or measures the radiation emitted by the track section. This can be achieved for example optically using an image sensor or for example by radar scanning, wherein the image can also be generated by radar scanning.

[0026] When talking about image overlap, it is meant that the overlap area, which preferably can cover 100% of both images, is captured by both the first sensor and the second sensor. In other words, the image is redundantly captured by two imaging systems, namely a stationary imaging system on the line side and a mobile imaging system of the vehicle. This results in an advantageous redundancy, wherein this redundancy can be used to detect a failure of both systems (more on this below). Thus, both systems support each other in order to continuously check their functional safety (safety).

[0027] In the sense of the present invention, an obstacle is meant to be all objects that are detected on the track section but shall not or shall not be allowed to be present there. For example, a person, an animal or a larger inanimate object. Plants or parts of plants also belong to inanimate objects as they are not able to move on their own. Objects that belong to the broad scope of railway infrastructure and cover the track section, such as switch boxes, are not considered obstacles.

[0028] The signal indicating an obstacle shall, due to its characteristics, be interpreted as a foreign object being identified in the track section. In the following, this will be referred to as obstacle signal. According to its characteristics, the signal indicating a failure shall be interpreted as a failure existing that causes the foreign object to be identified only once. Since the overlap of the generated images is sufficient for each position of the track section to appear in at least two images, a failure must exist. But this also means that the foreign object identified influences the generation of at least two images. Thus, if the foreign object is identified only once, there is a failure with respect to the normal operation of the program, which can have various causes.

[0029] 1) The foreign object can be located at the edge of the image, so that it is not completely displayed and thus not recognized.

[0030] 2) The image recognition can not have been successful, although the foreign object is completely located in the image.

[0031] 3) The sensor used to generate the image can have failed, for example completely.

[0032] The fault signal can thus have various interpretations, which require different measures. It can be checked whether the hardware and software involved in the method are functioning properly (for example, maintenance is required). Furthermore, there can be an obstacle, so that in this case an obstacle signal is also generated according to step c). In other words, even if this case cannot be ruled out, the fact that a fault signal is generated in step d) does not mean that the obstacle signal generated in step c) is incorrect. In order to ensure operational safety (safety), the obstacle signal should be considered to be generated correctly in case of doubt.

[0033] The invention has the advantage that, by generating the obstacle signal redundantly, additional fault detection of the hardware and software involved in the method is reliably achieved in the method flow. In other words, if the method is functioning properly, an obstacle signal must be generated twice for the relevant foreign object, i.e. redundantly. If this does not occur, one of the above-mentioned faults 1) to 3) is indicated. Thus, a partial failure of the system does not affect its recognition of foreign objects in the track section. This is because, as described above, an obstacle signal is triggered even if only one foreign object is recognized as an obstacle. At the same time, the method can be repaired in time, so that a complete failure usually does not occur. For example, a damaged camera can be replaced. This increases the operational safety of the method, so that the method can also be used for safety-relevant applications, for example the monitoring of a platform track section as mentioned here. In the ideal case, the safety levels SIL-1 and SIL-2 applicable for such tasks in railway operations can be achieved.

[0034] The requirements relating to the authentication of safety-relevant applications are very strict, for example, in the field of railway technology. According to the international standard IEC 61508 or the European standard EN 50129, which is specific to the field of railways, safety functions are classified into four safety integrity levels (SIL) or safety requirement levels in order to meet the required functional safety (safety) requirements. Here, safety integrity level 4 represents the highest safety integrity level, while safety integrity level 1 represents the lowest safety integrity level. The respective safety integrity level influences the confidence interval of the measured values, i.e. the higher the safety integrity level that the respective device must achieve, the smaller the confidence interval. The functional safety of different safety integrity levels can be described by the expected failure frequency of the safety-relevant system MTBF (mean time between failures) in years (a). The range for SIL-1 is 10... 100 a, for SIL-2 100... 1000 a, for SIL-3 1000... 10000 a and for SIL-4 10000... 100000 a.

[0035] According to another aspect of the application, a railway facility having a track section is described, wherein a plurality of sensors for imaging for monitoring the track section are installed in the railway facility. The railway facility can in particular comprise a railway crossing or a platform, which are arranged directly adjacent in the relevant track section.

[0036] According to the application, the railway facility has a computing environment, which is provided for implementing the method according to the preceding description. The advantages associated with this aspect of the application have already been described above and are hereby incorporated by reference.

[0037] According to another aspect of the application, a computer program product is described, which contains program instructions, which can be executed by the computing environment. According to the application, at least steps b), c) and d) of the method according to the preceding description are implemented.

[0038] According to the application, a computer program product is described, which contains program modules, which can be executed in the same computing unit or in a plurality of computing units of the computing environment. By means of the computer program product, which can contain one or a plurality of computer programs, the method according to the application and / or embodiments thereof can be executed and by means of the execution the above-mentioned advantages can be achieved.

[0039] According to another aspect of the application, a computer-readable storage medium is described, which comprises data, which is stored as a data group by the storage medium. According to the application, the data group, which enables the execution of the computer program product according to the preceding description, is implemented.

[0040] Furthermore, a providing device for storing and / or providing a computer program is described, which providing device is present in the form of a computer-readable storage medium. The providing device is, for example, a storage unit for storing a computer program and for providing a call. Alternatively or additionally, the providing device can also be a web service, a computer system, a server system, in particular a distributed, for example cloud-based, computer system or a virtual computer system, which stores a computer program on a computer-readable storage medium and provides it, preferably in the form of a data stream.

[0041] This providing takes place in the form of a program data set which describes the program modules, for example as a file, in particular a download file, of a computer program product, or as a data stream, in particular a download data stream. For example, the computer program product is transmitted using the providing device to a computing environment, so that the method according to the application can be executed in one or more computing units of the computing environment.

[0042] Designs of the application

[0043] The variant embodiments which describe improvements of the application are set out below, but do not limit the basic idea of the application.

[0044] According to one variant, the described aspect of the application is specified in such a way that each second sensor belongs to a second sensor group which consists of a plurality of second sensors which are arranged at a distance from one another on the track section, wherein the images produced by the second sensor group each adjoin one another in the border region with the adjacent images, that is to say the mutual distance of the sensors is chosen in such a way that an overlap or seamless connection of the individual images is achieved, taking into account the image section which can be imaged by the sensors.

[0045] In the present description, when it is stated that images adjoin one another, this means that the content of the adjacent images matches one another, so that a seamless overall image is formed. The focal point of this overall image is the track section which is to be monitored, which in this way is completely and seamlessly imaged in each group of images. The border region of the images is therefore located at the edge of the images. In other words, the border region of the images is the connecting region between the adjacent images, by means of which the content of the images can be seamlessly connected together. Computer-aided algorithms enable the individual images to be combined into an overall image, which algorithms are known per se and therefore do not need to be explained in detail here.

[0046] The border region can thus mean the edge region of the adjacent images, which adjoin one another in this edge region. Here, an overlap region can also occur in the edge region for practical reasons (orientation of the cameras in the station). Basically, however, the design of the application is intended to avoid using such an overlap region for generating redundant pixels in the overall image of the set of images, i.e. the set of images generated by the sensors. Rather, for generating redundancy, a first sensor or a plurality of first sensors on the vehicle are to be used. It is noted here that, when the vehicle is moving, a complete image of the track section can also be generated even with only one first sensor, since the first sensor used for imaging can move along with the vehicle along the track section. However, a plurality of first sensors can also be used on the vehicle, which form a first sensor group.

[0047] Since there are also two sets of images corresponding to the first and second sensors, respectively, each of which itself contains a seamless detection of the track section, it can be ensured that the imaging redundancy of the above-mentioned individual image elements, i.e. once in the first set and once in the second set, is ensured. Of course, this does not exclude that a third set or more sets of images can also be generated, resulting in a triple or even multiple redundancy.

[0048] The advantage of this variant is that by assigning the sensors to the first set (or using a single first sensor) and the second set, respectively, the degree of redundancy in the imaging of objects in the images can be well controlled. The imaging of these objects occurs only once in each set of images, i.e. only twice when applying the first and second set. This simplifies the method of determining the necessity of a fault signal. In addition, it is also possible to advantageously determine the occurrence of a fault relatively reliably.

[0049] According to one variant, the above-mentioned aspects of the application are specified in such a way that the monitoring of the track section is checked by the at least one first sensor in that:

[0050] e) a first test structure is arranged in the track section as an obstacle, a collision-free driving of the vehicle over the first test structure is not possible, and the position of the first test structure in the track section is used as a position data set in the method,

[0051] f) the first images are computer-aided analyzed for obstacles in the track section,

[0052] g) a second signal is generated which prompts a fault if no obstacle is identified in the position specified by the position data set by the analysis.

[0053] Test structures which do not pose a danger when the vehicle passes through need to have spatial dimensions which do not lead to a collision when the vehicle passes through the test structure. This can be, for example, a three-dimensional model of a small object. The advantage of test structures which do not pose a danger when passing through is that these test structures can be permanently installed in the track bed so that the functionality of the vehicle-side system can be checked using the computing environment for obstacle recognition, including the vehicle-side sensors. This advantageously increases the probability of fault exposure and thus reduces the probability of undiscovered faults endangering the functional safety of the system.

[0054] Since the position of the test structure can be known by evaluating the position data set, the position and the time at which the obstacle constituting the test structure must be detected can be predicted in combination with the position of the vehicle, which is determined in a known manner in accordance with the conventional functional safety (safety) of the railway traffic. If the expected recognition does not occur, a fault is indicated, and a second signal can be generated. In this way, fault disclosure of the used method and the used components can be advantageously achieved.

[0055] According to one variant, the above-mentioned aspects of the application are specified in such a way that the signal indicating the obstacle produced according to step c) described above is marked as a test signal.

[0056] Marking the signal indicating the obstacle as a test signal can advantageously prevent the signal from being interpreted by a computer as a signal indicating an actual obstacle. Of course, it can also be alternatively achieved by introducing a test procedure in the method according to the application which is related to the recognition of the test structure, and obstacles recognized during this time are not considered to be real obstacles. Marking the corresponding signal as a test signal can achieve additional safety with respect to fault interpretation, wherein the test procedure in the method does not need to be marked as a test procedure. In this way, real obstacles can be detected during the test, which do not produce the signal marked as a test signal.

[0057] According to one variant, the above-mentioned aspects of the application are specified in such a way that the test structure is constituted by obstacles which are displayed two-dimensionally in an image plane, wherein

[0058] h) the image plane is oriented parallel to the base of the track section or at an angle of at most 30°, preferably at most 10°, with respect to the base of the track section,

[0059] i) the obstacles in the image plane are displayed distortedly, wherein the obstacles are displayed undistortedly with respect to the image axis of the first sensor.

[0060] In other words, the obstacles displayed in the image create optical illusions for the imaging sensor. This can also advantageously display larger obstacles than are required for the vehicle to pass without a collision. This makes the test procedure of the application also suitable for these larger obstacles, whereby a more reliable fault detection can be carried out.

[0061] According to one variant, the above-mentioned aspects of the application are specified in such a way that the monitoring of the track section for obstacles by the at least one second sensor, when the vehicle is located on the track section, is checked in such a way that:

[0062] j) the current position of the vehicle is provided to the position data set,

[0063] k) the second image is computer-aided analyzed for obstacles in the track section,

[0064] l) a third signal is generated which indicates a fault, if no obstacle is identified in the current position indicated by the position data set by means of the analysis.

[0065] The advantage of this variant is that the test procedure can be carried out at regular intervals on the track section, on which the vehicle is regularly used on the platform in order to allow passengers to get on and off. The test procedure can be carried out at this time. The vehicle must be identified at this time. This allows the function of the method and the operating condition of the hardware components involved in the execution of the method, such as sensors, to be checked at regular intervals. This contributes to an increased safety of the execution of the method, while the test procedure also allows faults of the components involved in the execution of the method to be detected.

[0066] According to one variant, the above-mentioned aspects of the application are specified in such a way that, as soon as the first signal or the second signal or the third signal is evaluated, the monitoring of the track section for obstacles is required to be carried out manually by a railway staff.

[0067] The variant of the application advantageously takes into account that, as soon as a fault signal (first signal, second signal or third signal, which indicates the presence of a fault) is generated, the safe computer-aided obstacle detection can no longer be ensured. In this case, a manual obstacle detection by the station staff or the train driver can be relied on. In order to ensure a reliable manual monitoring of the track, it can also be necessary to reduce the maximum permissible speed. Advantageously, however, the train operation does not have to be completely stopped, since this is only an over-monitoring of the track for obstacles, while the train operation itself can still be carried out reliably. BRIEF DESCRIPTION OF DRAWINGS

[0068] Further details of the application are described below with reference to the drawings. Identical or corresponding elements are provided with the same reference signs in the various figures, and are only explained several times if there are differences between the figures.

[0069] The embodiments described below are preferred embodiments of the application. Parts of the embodiments described in the examples represent, respectively, variants of the application which are independent of one another and which can also improve the application independently of one another and thus can be considered as an integral part of the application, either alone or in a combination different from that shown. Furthermore, the components described can also be combined with the aforementioned variants of the application.

[0070] Figure 1 An embodiment of the device of the railway installation according to the application is shown, as well as the interaction between the functional components used.

[0071] Figure 2 An embodiment of the computing environment of the device according to Figure 1 is shown as a block diagram of the individual functional components and the interfaces formed between them, in which the individual computing units execute program modules which can be run in one or more exemplary computers, and in which the interfaces shown can be implemented in the computers by software technology or between different computers by hardware technology.

[0072] Figure 3 and Figure 4 A schematic example of a track segment image of a prepared track segment is shown (for example on a platform).

[0073] Figure 5 An embodiment of the method according to the application is shown in the form of a flowchart, in which the method steps shown can be implemented individually or in groups by program modules, and in which the computing units and interfaces are according to Figure 2 are shown by way of example. DETAILED DESCRIPTION

[0074] A platform BS and a track GL are shown according to Figure 1 on which a vehicle FZ is driving into the platform BS. Furthermore, a control center LZ is also provided, in which, for example, the current actual operating plan can be monitored. Communication between the control center LZ, the vehicle FZ and the platform BS can take place via a fifth interface S5, a sixth interface S6 and a seventh interface S7 by means of antennas AT.

[0075] The platform BS is equipped with four first sensors SN11, SN12, SN13, SN14, which are monitoring cameras in the example according to Figure 1 They monitor the track segment STA on the track GL in front of the platform BS. If a living obstacle HD1 in the form of a human being or a non-living obstacle HD2 in the form of an object appears on the track GL, this is detected by the sensors.

[0076] At the same time, the vehicle FZ is also provided with a second sensor SN2, which is aligned with the track GL in the direction of travel FR with an image axis and can therefore detect living and non-living obstacles HD2 at a distance, respectively. There is therefore a redundancy in the imaging detection system on the vehicle FZ and on the platform BS in terms of obstacle detection.

[0077] From Figure 1 and Figure 2 it can be seen that the method according to the application is run in a computing environment RU. The computing organs and functional components used are integrated with one another via interfaces. It can be seen that a first computer CP1 (see Figure 2 ) is in communication with the first sensors SN11, SN12, SN13, SN14 on the platform BS. A third computer CP3 is used in the control center LZ and a second computer CP2 is provided in the vehicle FZ, which is in communication with the second sensor SN2 via an eighth interface S8. Of course, the computing environment RU, which consists of the control center LZ, the platform BS and the vehicle FZ, can also contain a plurality of computers, Figure 2 the computers shown in

[0078] According to Figure 2 , the computers and sensors that make up the computing organs are shown in more detail, respectively. In the first computer CP1, a first processor PR1 is connected to a first storage unit SE1 via an eleventh interface S11. In the second computer CP2, a second processor PR2 is connected to a second storage unit SE2 via a twelfth interface S12. In the third computer CP3, a third processor PR3 is connected to a third storage unit SE3 via a thirteenth interface S13. In the sensor S11, a fifth processor PR5 is connected to a fifth storage unit SE5 via a fifteenth interface S15. In the sensor S12, a sixth processor PR6 is connected to a sixth storage unit SE6 via a sixteenth interface S16. In the sensor S13, a seventh processor PR7 is connected to a seventh storage unit SE7 via a seventeenth interface S17. In the sensor S14, an eighth processor PR8 is connected to an eighth storage unit SE8 via an eighteenth interface S18. In the second sensor SN2, a ninth processor PR9 is connected to a ninth storage unit SE9 via a nineteenth interface S19.

[0079] From Figure 1 and Figure 2The following conclusions can be drawn from the combination of the above. The fifth processor PR5 of the first sensor SN11, SN12, SN13, SN14 is interconnected with the first processor PR1 of the first computer CP1 via the first interface S1. The sixth processor PR6 of the second sensor SN2 is interconnected with the first processor PR1 of the first computer CP1 via the second interface S2. The seventh processor PR7 of the third sensor is interconnected with the first processor PR1 of the first computer CP1 via the third interface S3. The eighth processor PR8 of the fourth sensor is interconnected with the first processor PR1 of the first computer CP1 via the fourth interface S4. The first processor PR1 is interconnected with the second processor PR2 via the fifth interface S5. The second processor PR2 is interconnected with the third processor PR3 via the sixth interface S6. The first processor PR1 is interconnected with the second processor PR2 via the seventh interface S7.

[0080] If in the description of the application only a computer, a processor, a storage unit, a sensor or an interface is mentioned, the content is generally applicable to all computers, processors, storage units and other functional components listed above, which are connected via interfaces to form a computing environment RU.

[0081] Figure 3 In the figure, a vehicle FZ on the track GL, a platform BS and a railway crossing BU on the track GL are also shown. The first sensors SN11, SN12 and the second sensor SN2 are also used. In the figure, the first test structure TS1 and the second test structure TS2 are shown in a perspective view, as they are arranged on the track GL. Figure 3 In the figure, the track GL is folded onto the drawing plane for the display of the first test structure TS1 and the second test structure TS2 arranged parallel to the ground. In this case, the perspective view is involved as shown in the figure. Figure 1 The distorted representation of the moving obstacle is shown. In reality, however, the test structures are arranged very flat in the ballast, so that the vehicle FZ can drive over these test structures without collision.

[0082] Figure 4 In the figure, the appearance of the second test structure TS2 in the images taken by the first sensors SN11, SN12 and the second sensor SN2 is shown. Apparently due to optical illusions, the impression is created that a living obstacle is standing on the track GL. In the figure, a first image B1 and a second image B12 are shown, wherein the first image B1 is taken by one of the first sensors SN11, SN12, SN13, SN14 and the second image B12 is taken by one of the second sensors SN2.

[0083] In addition to the test procedure described in more detail below in Figure 5 In addition to the test procedure described in more detail below in Figure 4The double arrow shown indicates that, if the second test structure TS2 is photographed by one of the first sensors SN11, SN12 and one of the second sensors SN2, an image comparison can also be made in the test program in the sense of redundant obstacle recognition.

[0084] The method according to the application is explained step by step below according to the flow chart in Figure 5

[0085] The method is started in a first step 1 (START).

[0086] In a second step 2, the track section STA is detected using the first sensor SN12 and / or the second sensor SN2 (SCN). Here, images are generated which can identify obstacles (see Figure 1 ).

[0087] In a third step 3, the images generated by the sensors are analyzed (ANL) in order to identify obstacles on the track section STA. The obstacle recognition itself takes place by means of known methods of analysis of computer-aided generated images. In order to identify obstacles, the images are subjected to computer-aided image processing in a known manner and the image content is analyzed in order to identify obstacles (known image processing algorithms are applied).

[0088] In a fourth step 4, it is asked whether an obstacle has been identified on the track section STA (OBS). If no obstacle has been identified, recursion takes place and step 2 is repeated. If an obstacle has been identified, step 5 is continued.

[0089] In a fifth step 5, it is further asked whether the obstacle has been identified in regular operation (REG?). If the obstacle has been identified, step 6 is continued; if the obstacle has not been identified, step 7 is continued, in which case a training program of the imaging sensors is initiated.

[0090] In a sixth step 6, a first obstacle warning signal is generated and output or further processed in the computer-aided flow (WRN1). Here, the signal will lead to a disruption of the train traffic since there is an immediate danger of collision. The warning signal can also be issued to the driver of an incoming train in the running state, for example, in order to prepare for braking.

[0091] In a seventh step 7, it is further asked whether the obstacle identified is a train (TRN?). Trains are generally not considered obstacles since the stopping of a train at a platform BS is part of normal railway operation. In this case, step 11 is continued.

[0092] ​In a tenth step 11, a second test procedure (abbreviated: TST2) is executed. This test procedure serves to detect a fault in the first sensor SN11, SN12, SN13, SN14, thereby advantageously increasing the safety level when the method is executed. In this way, a fault during the execution of the method can be detected early and maintenance measures (abbreviated: MSR) can be taken in accordance with step 12.

[0093] In order to ensure that the train can drive into the BS platform as planned, a comparison with the train schedule, in particular the current actual train schedule, can be used, which is not shown in Figure 5 the figure, for example in the control center LZ via the first interface S1. If the train cannot arrive at the BS platform as planned, an obstacle signal can be generated, since an unplanned train can constitute an obstacle for a train running as planned.

[0094] If the train cannot be detected, it means that there is another obstacle. However, since this is a test structure, no emergency collision with the obstacle occurs. In this case, the first test procedure (abbreviated: TST1) is continued in accordance with the eighth step 8. This test procedure serves to detect a fault in the second sensor SN2, thereby advantageously increasing the safety level when the method is executed. In a ninth step 9, a second signal is generated as a warning, which can be output or computer-aided (abbreviated: WRN2).

[0095] In a tenth step 10, safety measures (abbreviated: MSR) are executed. Appropriate measures must be taken, for example manual monitoring of the track section STA by a railway employee. However, if the fault is only temporary and the subsequent recursive cycle brings the automatic obstacle monitoring function back into operation again, it is possible to return to step 2.

[0096] List of reference signs

[0097] AT antenna

[0098] B1 first image

[0099] B12 second image

[0100] BS platform

[0101] BU railway crossing

[0102] CP1 first computer

[0103] CP2 second computer

[0104] CP3 third computer

[0105] FR direction of travel

[0106] FZ vehicle

[0107] GL track

[0108] HD1 living obstacle

[0109] HD2 non-living obstacle

[0110] LZ control center

[0111] PR1 first processor

[0112] PR2 second processor

[0113] PR3 third processor

[0114] PR5 fifth processor

[0115] PR6 sixth processor

[0116] PR7 seventh processor

[0117] PR8 eighth processor

[0118] PR9 ninth processor

[0119] RU computing environment

[0120] S1 first interface

[0121] S11 eleventh interface

[0122] S12 twelfth interface

[0123] S13 thirteenth interface

[0124] S15 fifteenth interface

[0125] S16 sixteenth interface

[0126] S17 seventeenth interface

[0127] S18 eighteenth interface

[0128] S19 nineteenth interface

[0129] S2 second interface

[0130] S3 third interface

[0131] S4 fourth interface

[0132] S5 fifth interface

[0133] S6 sixth interface

[0134] S7 seventh interface

[0135] S8 eighth interface

[0136] SE1 first storage unit

[0137] SE2 second storage unit

[0138] SE3 third storage unit

[0139] SE4 fourth storage unit

[0140] SE5 fifth storage unit

[0141] SE6 sixth storage unit

[0142] SE7 seventh storage unit

[0143] SE8 eighth storage unit

[0144] SE9 ninth storage unit

[0145] SN11, SN12, SN13, SN14 first sensor

[0146] SN2 second sensor

[0147] STA track segment

[0148] TS1 first test structure

[0149] TS2 second test structure

Claims

1. A method for monitoring obstacles in a track segment (STA), wherein the track segment (STA) obstacles are detected by means of at least one first sensor for imaging (B1) and at least one second sensor for imaging (B12), characterized in that, a) The at least one first sensor (SN11, SN12, SN13, SN14) operates on the vehicle (FZ), and the at least one second sensor (SN2) operates at a fixed position on the track segment (STA), wherein these sensors are oriented toward the track segment (STA) during operation such that each location on the track segment (STA) is respectively present in the first image (B1) of the at least one first sensor (SN11, SN12, SN13, SN14) and in the second image (B12) of the at least one second sensor (SN2). b) Computer-aided analysis of the first image (B1) and the second image (B12) for obstacles in the track segment (STA). c) If an obstacle is identified in at least one first image (B1) and at least one second image (B12) through the analysis, a signal indicating the presence of an obstacle is generated. d) If an obstacle is identified by the analysis only in the at least one first image (B1) or only in the at least one second image (B12), a first signal indicating a fault is generated.

2. The method according to claim 1, characterized in that, Each second sensor (SN2) belongs to a second sensor group consisting of multiple second sensors (SN2) arranged at intervals on the track segment (STA), wherein the images generated by the second sensor group are respectively adjacent to each other in the boundary region.

3. The method according to claim 1 or 2, characterized in that, The functionality of the track segment (STA) is checked by at least one first sensor (SN11, SN12, SN13, SN14) in the following manner: e) A first test structure (TS1) is arranged as an obstacle in the track segment (STA), such that a vehicle (FZ) can drive over the first test structure without a collision, and the position of the first test structure in the track segment (STA) is used as a location data set in the method. f) Computer-aided analysis of the first image (B1) for obstacles in the track segment (STA). g) If no obstacle is identified in the location indicated by the location data set through the analysis, a second fault signal is generated.

4. The method according to claim 3, characterized in that, The signal indicating an obstacle generated in step c) of claim 1 is marked as a test signal.

5. The method according to claim 3 or 4, characterized in that, The test structure consists of obstacles displayed in two dimensions on an image plane, wherein, h) The image plane is parallel to the base of the track segment (STA) or oriented at an angle of up to 30°, preferably up to 10°, relative to the base of the track segment (STA). i) Obstacles in the image plane are displayed in a distorted manner, wherein the obstacles are displayed without distortion with reference to the image axes about the first sensors (SN11, SN12, SN13, SN14).

6. The method according to any one of the preceding claims, characterized in that, When the vehicle (FZ) is on the track segment (STA), the functionality of the track segment (STA) is monitored by at least one second sensor (SN2) in the following manner: j) Provide the current location of the vehicle (FZ) to the location data group. k) The second image (B12) is analyzed by computer-aided means to identify obstacles in the track segment (STA). l) If no obstacle is identified in the current location indicated by the location data set through the analysis, a third signal indicating a fault is generated.

7. The method according to claim 6, characterized in that, Once the assessment determines the first, second, or third signal, it requires railway personnel to manually monitor obstacles on the track section (STA).

8. A railway facility having a track section (STA), wherein, The railway facility is equipped with a plurality of imaging sensors for monitoring track sections (STA), characterized in that the railway facility has a computing environment (RU) configured to implement the method according to any one of the preceding claims.

9. A computer program product comprising program instructions executable by the computing environment (RU) to at least implement steps b), c) and d) of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium comprising data, said data being stored as a data set by said storage medium thereby enabling the execution of the data set of a computer program product according to claim 9.

Images