Method for determining a fault in a railway infrastructure by means of a detection device of a train, computer program product, computer-readable storage medium and detection device

Abstract

The invention relates to a method for determining a fault (18) in railway infrastructure by means of a detection device (12) of a train (10), comprising the steps of detecting the train's (10's) surroundings (14) by means of the detection device (12); determining an anomaly (16) associated with the railway infrastructure as a function of the detected surroundings (14) by means of an electronic computing unit (20) of the detection device (12); and determining the fault (18) as a function of the determined anomaly (16) by means of the electronic computing unit (20). The invention further relates to a computer program product, a computer-readable storage medium, and a detection device (12).

Description

[0001] The invention relates to a method for determining a disruption of a railway infrastructure by means of a detection device of a train according to the applicable patent claim 1. Furthermore, the invention relates to a computer program product, a computer-readable storage medium and a detection device.

[0002] Railway facilities and infrastructure require constant monitoring and control to ensure the operational safety of the railway infrastructure. This can provide clues to the presence of situations that could potentially cause disruptions. For example, noticeable damage to signaling systems, vegetation overgrowth on the track, signs of wear on the tracks and trackbed, and the like. With the use of assistance systems or automatic train control systems in railway technology, it has become possible to automatically record information about the surroundings of trains during travel using appropriate sensors or detection devices on the vehicles. This also records the surroundings of the railway facility on which trains equipped with such systems operate.These sensors or detection devices typically provide information directly used to control the train. However, as an additional benefit, they can also provide information that can be used to register safety-relevant conditions and, for example, centrally collect them via a data connection and then communicated to a control center.

[0003] The object of the present invention is to provide a method, a computer program product, a computer-readable storage medium and a detection device by means of which a disruption of a railway infrastructure can be determined in an improved manner by means of the detection device.

[0004] This object is achieved by a method, a computer program product, a computer-readable storage medium, and a detection device according to the independent patent claims. Advantageous embodiments are specified in the subclaims.

[0005] One aspect of the invention relates to a method for determining a disruption of a railway infrastructure using a detection device of a train. The surroundings of the train are detected using the detection device. An anomaly associated with the railway infrastructure is determined based on the detected surroundings using an electronic computing device of the detection device, and the disruption is determined based on the determined anomaly using the electronic computing device.

[0006] In particular, the invention takes advantage of the fact that detection devices already installed on the train can be used to detect and determine corresponding faults in the railway infrastructure.

[0007] Especially if the train is equipped with an automatic obstacle detection system, for example, a series of sensors or detection devices are present to perceive the train's surroundings. These provide image data, lidar, or radar data about the surroundings. In other words, the detection device can be designed as a camera, lidar sensor, radar sensor, or even an ultrasonic sensor.

[0008] Automatic obstacle detection evaluates sensor data for potential obstacles and hazards within the train's clearance. This data is also used to detect other situations that may affect rail operations.

[0009] These may include, for example, hazards on adjacent tracks, anomalies that require closer inspection, such as vegetation overgrowing the track, unexpected changes in the environment, or the like.

[0010] For this purpose, the relevant sensor data, particularly the exact location and time of origin, is stored and then either examined using anomaly detection methods, for example, based on an object recognition algorithm, or scanned for reasons for more detailed monitoring using detection mechanisms specifically tailored to specific situations. Optical or visual detection algorithms can be used for this purpose.

[0011] In particular, the determination of the fault may already include an analyzed classification of the fault, but raw data collected by the train's detection device on which the corresponding derivation is based may also be stored.

[0012] According to an advantageous embodiment, the anomaly is determined based on the current position of the train. In particular, the determined anomaly can thus be stored along with the position. For this purpose, corresponding virtual data or digital maps can be available, which, for example, contain relevant information regarding the objects expected on site. The anomaly can then be determined, particularly by comparing it with this map.

[0013] It is also advantageous if an expected detected environment is specified depending on the current position, and the expected detected environment is compared with the detected environment, and the anomaly is determined based on this. In particular, if the expected detected environment deviates from the actually detected environment, an anomaly can be determined. For this purpose, a database containing the expected detected environment can be available, allowing the detection device, in particular, to access this data and perform a corresponding evaluation.

[0014] It is also advantageous if, as railway infrastructure, a track and / or a track bed and / or a switch and / or a railway signal and / or a station is checked for a fault. In particular, other infrastructure objects can also be checked. This makes it possible, in particular, to check the entire railway infrastructure for a fault.

[0015] It is also advantageous if a specific fault is transmitted to a central electronic computing device. In particular, the central electronic computing device can be used, for example, to collect a large number of faults. This allows the central electronic computing device to be alerted that a corresponding fault has occurred. This can then initiate appropriate additional measures, such as initiating appropriate repair measures, so that the fault can be remedied in the future.

[0016] Furthermore, it has proven advantageous if the malfunction is verified by the central electronic computing device. This can be achieved, for example, by already having historical data on this anomaly. In particular, if historical data is already available, a corresponding malfunction can be verified. Only a verified malfunction leads to further measures. Especially since information obtained about the malfunction caused by a single train is usually incomplete, this information requires confirmation or verification in order to obtain specific, situation-specific data. This allows the malfunction to be identified with the utmost reliability.

[0017] It has also proven advantageous if a verification request for verifying the fault is transmitted to at least one other train via the central electronic computing device. For example, the other train can then also pass a corresponding location of the potentially determined fault and record it accordingly. The fault can be recorded using the same type of recording device or a different type of recording device. In particular, the verification request is thus generated by the central electronic computing device and transmitted to another train. This allows the other train to confirm the fault independently of the train. This allows the fault to be determined with the highest degree of reliability.

[0018] It is also advantageous if the next train is selected based on the route of the next train. For example, if the next train passes the anomaly, especially if a corresponding route plan is available that takes the train past the disruption, this train can be instructed to perform a verification check. This makes it easy to verify the disruption.

[0019] A further advantageous embodiment provides that the additional train is selected depending on the type of detection device present on the additional train. In particular, if, for example, a train does not have a detection device, it cannot be instructed to carry out the corresponding verification. In particular, the electronic computing device can decide which type of detection device is to be used to detect or verify the fault. For example, it can be provided that the central electronic computing device requires confirmation of a different detection type and thus selects a train that has a different detection type than the train that reported the fault. The fault can thus be reliably determined.

[0020] It is also advantageous if the next train is selected based on a potentially different detection angle for the potential disruption. For example, the next train can be selected based on an oncoming track or the oncoming direction of travel, allowing the disruption to be detected from a different angle. This allows for different viewing angles of the disruption, allowing for reliable verification of the disruption.

[0021] Furthermore, it has proven advantageous to select the next train depending on the time of day when the potential disruption is detected. For example, it can be specified that only trains passing the disruption during the day receive a corresponding verification order. This can prevent incorrect recordings or recordings with low data information density, for example, due to darkness.

[0022] According to a further advantageous embodiment, in the event of a specific fault, a warning message is generated for at least one other train. For example, the warning message can be generated by the train itself. If, for example, it is provided that the fault must first be verified by the central electronic computing device, the corresponding warning message can then be generated by the central electronic computing device. For example, it can then further be provided that, should the fault be appropriately verified, trains are instructed to travel past the fault at a reduced speed. Furthermore, the warning message can also be generated for appropriate personnel so that people can be dispatched to the fault location to repair it.

[0023] The presented method is, in particular, a computer-implemented method. Therefore, a further aspect of the invention relates to a computer program product with program code means that, when the program code means are processed by the electronic computing device, causes an electronic computing device to perform a method according to the preceding aspect.

[0024] Furthermore, the invention also relates to a computer-readable storage medium with the computer program product.

[0025] The invention also relates to a detection device for determining a disruption of a railway infrastructure, comprising at least one electronic computing device, wherein the detection device is designed to carry out a method according to the preceding aspect. In particular, the method is carried out by means of the detection device.

[0026] Yet another aspect of the invention relates to a train with a detection device according to the preceding aspect.

[0027] Advantageous embodiments of the method are to be regarded as advantageous embodiments of the computer program, the computer-readable storage medium, the detection device, and the train. The detection device and the train, in particular, have physical features for this purpose in order to be able to carry out corresponding method steps.

[0028] In the context of the present disclosure, an object detection algorithm can be understood as a computer algorithm capable of identifying and locating one or more objects within a provided input data set, for example, an input image, for example by defining corresponding bounding boxes or regions of interest (ROIs), and in particular by assigning a corresponding object class to each of the bounding boxes, wherein the object classes can be selected from a predefined set of object classes. The assignment of an object class to a bounding box can be understood as providing a corresponding confidence value or probability that the object identified within the bounding box belongs to the corresponding object class.For example, for a given bounding box, the algorithm may provide such a confidence value or probability for each of the object classes. Assigning the object class may, for example, involve selecting or providing the object class with the highest confidence value or probability. Alternatively, the algorithm may simply specify the bounding boxes without assigning a corresponding object class.

[0029] A computing unit / electronic computing device can be understood, in particular, as a data processing device that contains a processing circuit. The computing unit can therefore, in particular, process data to perform computing operations. This may also include operations for performing indexed access to a data structure, for example, a look-up table (LUT).

[0030] The computing unit may, in particular, contain one or more computers, one or more microcontrollers, and / or one or more integrated circuits, for example, one or more application-specific integrated circuits (ASICs), one or more field-programmable gate arrays (FPGAs), and / or one or more single-chip systems (SoCs). The computing unit may also contain one or more processors, for example, one or more microprocessors, one or more central processing units (CPUs), one or more graphics processing units (GPUs), and / or one or more signal processors, in particular one or more digital signal processors (DSPs). The computing unit may also include a physical or virtual network of computers or other of the aforementioned units.

[0031] In various embodiments, the computing unit includes one or more hardware and / or software interfaces and / or one or more memory units.

[0032] A memory unit can be a volatile data memory, for example a dynamic random access memory (DRAM) or a static random access memory (SRAM), or a non-volatile data memory, for example a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory or flash EEPROM, a ferroelectric random access memory (FRAM), a magnetoresistive random access memory,MRAM (magnetoresistive random access memory) or phase-change random access memory (PCRAM).

[0033] An environmental sensor system / detection device can be understood, for example, as a sensor system capable of generating sensor data or sensor signals that map, represent, or reproduce an environment. In particular, the ability to detect electromagnetic or other signals from the environment is not sufficient for a sensor system to be considered an environmental sensor system. For example, cameras, radar systems, lidar systems, or ultrasonic sensor systems can be considered environmental sensor systems.

[0034] Algorithms for automatic visual perception, which can also be referred to as computer vision algorithms, machine vision algorithms, or machine vision algorithms, can be considered computer algorithms for automatically performing a visual perception task. A visual perception task, also referred to as a computer vision task, can be understood, for example, as a task for extracting visual information from image data. In particular, the visual perception task can, in principle, in some cases, be performed by a human who is able to visually perceive an image corresponding to the image data. In the present context, however, visual perception tasks are also performed automatically, without the need for human assistance.

[0035] A computer vision algorithm, for example, to detect the disturbance, can, for example,

[0036] An image processing algorithm or an image analysis algorithm that is or was trained by machine learning and can be based, for example, on an artificial neural network, in particular a convolutional neural network. The computer vision algorithm can include, for example, an object detection algorithm, an obstacle detection algorithm, an object tracking algorithm, a classification algorithm, a semantic segmentation algorithm, and / or a depth estimation algorithm.

[0037] Corresponding algorithms can also be implemented analogously based on input data other than images visually perceivable by humans. For example, point clouds or images from infrared cameras, lidar systems, etc., can also be evaluated using appropriately adapted computer algorithms. Strictly speaking, the corresponding algorithms are not algorithms for visual perception, since the corresponding sensors can operate in ranges that are visually imperceptible to the human eye, for example, in the infrared range. Therefore, such algorithms are referred to as algorithms for automatic perception in the context of the present invention. Algorithms for automatic perception therefore include algorithms for automatic visual perception, but are not limited to these with regard to human perception.Consequently, an algorithm for automatic perception according to this understanding can include a computer algorithm for automatically performing a perception task, which is or has been trained, for example, through machine learning and can, in particular, be based on an artificial neural network. Such generalized algorithms for automatic perception can also include object detection algorithms, object tracking algorithms, classification algorithms, and / or segmentation algorithms, for example, semantic segmentation algorithms.

[0038] If an artificial neural network is used to implement an algorithm for automatic visual perception, a commonly used architecture is a convolutional neural network (CNN). In particular, a 2D CNN can be applied to corresponding 2D camera images. CNNs can also be used for other automatic perception algorithms. For example, 3D CNNs, 2D CNNs, or 1D CNNs can be applied to point clouds, depending on the spatial dimensions of the point cloud and the processing details.

[0039] The result or output of an automatic perception algorithm depends on the specific underlying perception task. For example, the output of an object detection algorithm may contain one or more bounding boxes defining a spatial position and optionally an orientation of one or more corresponding objects in the environment and / or corresponding object classes for the one or more objects. An output of a semantic segmentation algorithm applied to a camera image may contain a pixel-level class for each pixel of the camera image. Analogously, an output of a semantic segmentation algorithm applied to a point cloud may contain a corresponding point-level class for each of the points. The pixel-level or point-level classes, respectively, may define an object type to which the respective pixel or point belongs.

[0040] For applications or application situations that may arise in a method according to the invention and which are not explicitly described herein, it may be provided that, according to the method, an error message and / or a request to enter user feedback is output and / or a standard setting and / or a predetermined initial state is set.

[0041] Regardless of the grammatical gender of a particular term, persons with male, female or other gender identities are included.

[0042] Further features of the invention emerge from the claims, the figures and the description of the figures. The features and combinations of features mentioned above in the description as well as the features and combinations of features mentioned below in the description of the figures and / or shown in the figures can be encompassed by the invention not only in the respectively specified combination, but also in other combinations. In particular, the invention can also encompass embodiments and combinations of features that do not have all the features of an originally formulated claim. Furthermore, the invention can encompass embodiments and combinations of features that go beyond the combinations of features set out in the backreferences to the claims or deviate from them.

[0043] The sole FIG. shows a schematic block diagram according to an embodiment of the method.

[0044] The invention is explained in more detail below using specific embodiments and associated schematic drawings. In the figures, identical or functionally equivalent elements may be provided with the same reference numerals. The description of identical or functionally equivalent elements may not necessarily be repeated for different figures.

[0045] The figure shows a schematic block diagram according to one embodiment of a train 10 with one embodiment of a detection device 12. The detection device 12 is designed to detect a purely schematically represented environment 14 of the train 10. In the following exemplary embodiment, an anomaly 16 in a railway infrastructure can be detected, particularly in the environment 14.

[0046] According to one embodiment of the invention, the detection device 12 can be configured to determine a fault 18 in the railway infrastructure 16. For this purpose, the surroundings 14 of the train 10 are detected by the detection device 12. An anomaly 16 associated with the railway infrastructure is determined as a function of the detected surroundings 14 by means of an electronic computing device 20 of the detection device 12. The fault 18 is then determined as a function of the determined anomaly 16 by means of the electronic computing device 20.

[0047] In particular, it can be provided that the anomaly 16 is determined depending on the current position of the train 10. Furthermore, it can be provided that an expected detected environment 14 is specified depending on the current position and the expected detected environment 14 is compared with the detected environment 14, and the anomaly 16 is determined depending thereon.

[0048] As railway infrastructure, in particular a track and / or track bed and / or a switch and / or a railway track and / or a station can be checked for a fault 18.

[0049] The FIG. further shows that the determined disruption 18 can be transmitted to a central electronic computing device 22. In particular, it can be provided that the disruption 18 is verified by the central electronic computing device 22. For this purpose, it can be provided, in particular, that a verification order 24 is transmitted to at least one further train 26, 28 by means of the central electronic computing device 22. The further train 26, 28 can then be selected depending on a route of the further train 26, 28. For this purpose, the central electronic computing device 22 can communicate, in particular, with a route planner 30 which has the routes of the trains 10, 26, 28. Furthermore, it can be provided that the further train 26, 28 is selected depending on an existing detection device 32, 34 of the further train 26, 28.In the following exemplary embodiment, for example, a first additional train 26 has a first detection device 32, and a second additional train 28 has a second detection device 34. In particular, corresponding observations 36 of the additional trains 26, 28 are transferred to a situation assessment 38, which in turn is transmitted to the central electronic computing device 22.

[0050] It can also be provided that the further train 26, 28 is selected depending on a time of the potential detection of the potential disturbance 18.

[0051] It can also be provided that in the event of a specific fault 18, a warning message is generated for at least one other train 26, 28.

[0052] The central electronic computing device 22 can, for example, exchange potential train candidates 40 and location information 42 with the route planner 30.

[0053] In particular, if the train 10 is equipped for automatic obstacle detection, a series of sensors, in particular the detection device 12, are provided for perceiving the surroundings 14 of the train 10. These provide image data, lidar data, and radar data on the surroundings 14.

[0054] During automatic obstacle detection, the sensor data is evaluated with regard to possible obstacles and hazards in the clearance of train 10. In addition, this data is also used to detect other situations that affect rail operations.

[0055] These may include, for example, hazards on adjacent tracks, anomalies 16 that require closer inspection, such as vegetation on the track, unexpected changes in the environment, or the like.

[0056] For this purpose, sensor data about the exact location and time of origin are stored and then either examined using anomaly detection methods or searched for reasons for more detailed monitoring using detection mechanisms specifically tailored to specific situations.

[0057] If an indication for a more detailed investigation emerges, train 10 transmits this, in one embodiment, to the central electronic computing device 22, particularly in the form of a control center, in the form of an observation report. Based on the control center's decision, the other trains 26, 28, whose routes also pass the observation point, can then be requested to collect further observation data for the affected area and within the affected area and forward it to the central electronic computing device 22.

[0058] The observation report records the analyzed classification of the anomaly 16, but also the raw data collected by the detection device 12 of the train 10, on which the corresponding derivation is based.

[0059] Based on the observations of all trains 10, 26, and 28, a reliable situation assessment 38 can then be made, and the necessary action can be derived in an advantageous embodiment. This is then issued in the form of a message, for example, to the control center personnel or in combination with other electronic systems.

[0060] The situation assessment 38 takes into account in particular the available data on the normal condition of the line, observation data from the other trains 26, 28 in a closer temporal context and rules for assessing the situation which can be specified by the operations management.

[0061] In particular, the essential difference from the prior art lies in the integration of the observations from the "first-person" perspective of the train 10 equipped with the detection device 12 with the observations of other trains, in particular the additional trains 26, 28, including the possibility of active research through targeted evaluation of the sensor data by the additional trains 26, 28 in the event of anomalies 16. The additional use of the sensors present in the additional trains 26, 28 and intended for other purposes for the purposes described here is also characteristic of the invention and is not described in this way in the prior art.

Claims

1. A method for determining a disturbance (18) of a railway infrastructure by means of a detection device (12) of a train (10), comprising the steps of: - detecting an environment (14) of the train (10) by means of the detection device (12); - determining an anomaly (16) associated with the railway infrastructure as a function of the detected environment (14) by means of an electronic computing device (20) of the detection device (12); and - determining the disturbance (18) as a function of the determined anomaly (16) by means of the electronic computing device (20).

2. Method according to claim 1, characterized in that the anomaly (18) is determined depending on the current position of the train (10).

3. Method according to claim 2, characterized in thatan expected detected environment (14) is specified as a function of the current position and the expected detected environment (14) is compared with the detected environment (14) and the anomaly (16) is determined as a function thereof.

4. Method according to one of the preceding claims, characterized in that as railway infrastructure, a track and / or a track bed and / or a switch and / or a railway signal and / or a station is checked for a fault (18).

5. Method according to one of the preceding claims, characterized in that a specific fault (18) is transmitted to a central electronic computing device (22).

6. Method according to claim 5, characterized in that the fault (18) is verified by the central electronic computing device (22).

7. Method according to claim 5 or 6, characterized in thatby means of the central electronic computing device (22) a verification order (24) for verifying the fault (18) is transmitted to at least one further train (26, 28).

8. Method according to claim 7, characterized in that the further train (26, 28) is selected depending on a route of the further train (26, 28).

9. Method according to claim 7 or 8, characterized in that the further train (26, 28) is selected depending on a type of existing detection device (32, 34) of the further train (26, 28).

10. Method according to one of claims 7 to 9, characterized in that the further train (26, 28) is selected depending on a potential other detection angle on the potential disturbance (18).

11. Method according to one of claims 7 to 10, characterized in that the further train (26, 28) is selected depending on a time of potential detection of the potential disturbance (18).

12. Method according to one of claims 7 to 10, characterized in that in the event of a specific fault (18), a warning message is generated for at least one further train (26, 28).

13. Computer program product with program code means which causes an electronic computing device (20) to carry out a method according to one of claims 1 to 12 when the program code means are processed by the electronic computing device (20).

14. A computer-readable storage medium comprising at least one computer program product according to claim 13.

15. Detection device (12) for determining a fault (18) of a railway infrastructure, with at least one electronic computing device (20), wherein the detection device (12) is designed to carry out a method according to one of claims 1 to 12.

Images