Method and system for detecting obstacle in tunnel

Abstract

The present invention provides a method and system for detecting an obstacle in a tunnel. A sensing unit is provided beside a tunnel rail. The method comprises: acquiring train operation environment data by means of a sensing unit; analyzing the train operation environment data to obtain operation environment structured information and train operation state structured information, wherein the operation environment structured information comprises a rail area and a target feature of an obstacle, and the train operation state structured information comprises a train feature; processing the operation environment structured information and the train operation state structured information to obtain passing train position information and obstacle target information; on the basis of the passing train position information and the obstacle target information, determining whether there is an obstacle; and if there is an obstacle, triggering an alarm. The obstacle on the line can be effectively and accurately detected, thereby improving the detection efficiency, and effectively guaranteeing the safety of passengers.

Description

Obstacle detection method and system in tunnel Technical Field

[0001] The present invention relates to the field of rail transportation technology, and in particular to a method and system for detecting obstacles in tunnels. Background Art

[0002] Currently, obstacle detection is an important means to ensure driving safety on rail lines. Existing rail obstacle detection methods can be divided into contact detection methods and non-contact detection methods.

[0003] The contact detection method mainly installs a detection beam at the bottom of the train's front to detect targets on the track. When the detection beam collides with the target on the track, it notifies the train of the presence of an obstacle through mechanical and electronic conduction.

[0004] Non-contact detection methods are generally carried out on-board, with visual and radar sensors placed on the front of the train. Compared with contact detection methods, this method can detect obstacles ahead in real time and notify the train in real time. However, it also has shortcomings:

[0005] 1) Rail trains have large inertia and long braking distances. Even if they can detect obstacles, they still cannot effectively guarantee the safety of the train and passengers within a limited distance and time.

[0006] 2) When the train is moving, it is necessary to collect and analyze the environment and obstacles in front of the train in real time. This places high demands on the analysis algorithm and it is difficult to balance the real-time and accuracy of the detection.

[0007] Summary of the Invention

[0008] In view of this, the object of the present invention is to provide a method and system for detecting obstacles in a tunnel, which can effectively and accurately detect obstacles on the line, improve detection efficiency, and effectively ensure the safety of passengers.

[0009] In a first aspect, an embodiment of the present invention provides a method for detecting obstacles in a tunnel, wherein a sensing unit is provided beside a tunnel track, and the method comprises:

[0010] Collecting train operation environment data through the sensing unit;

[0011] Analyzing the train operating environment data to obtain operating environment structured information and train operating status structured information, wherein the operating environment structured information includes target features of the track area and the obstacle, and the train operating status structured information includes train features;

[0012] Processing the operating environment structured information and the train operating status structured information to obtain passing train position information and obstacle target information;

[0013] Determining whether the obstacle exists based on the passing train position information and the obstacle target information;

[0014] If it exists, an alarm is triggered.

[0015] Furthermore, collecting train operating environment data through the sensing unit includes:

[0016] When the perception unit is a visual sensor, the video image is collected by the visual sensor;

[0017] When the sensing unit is a laser radar, radar point cloud data is collected by the laser radar;

[0018] When the perception unit includes the visual sensor and the laser radar, the video image is collected by the visual sensor, and the radar point cloud data is collected by the laser radar.

[0019] Furthermore, the operating environment structured information and the train operating status structured information are processed to obtain the passing train position information and obstacle target information, including:

[0020] detecting first target information from the video image using an analysis algorithm;

[0021] or,

[0022] detecting second target information from the radar point cloud data using the analysis algorithm;

[0023] The first target information and the second target information both include tracks, trains and the obstacles.

[0024] Furthermore, the operating environment structured information and the train operating status structured information are processed to obtain the passing train position information and obstacle target information, including:

[0025] Fusing the first target information detected by the video image and the second target information detected by the radar point cloud data to obtain a fused target;

[0026] Making a judgment based on the category of the fused target and the degree of overlap of the target positions;

[0027] Fusing the video image and the radar point cloud data whose target position overlap is greater than a preset threshold into the same target, and setting a first confidence level;

[0028] Treating the video image and the radar point cloud data, whose target position overlap is less than the preset threshold, as two separate targets and setting a second confidence level; wherein the first confidence level has a higher priority than the second confidence level;

[0029] The position information of the passing train is calculated based on the depth information of the target, and combined with the constructed tunnel 3D space and the track area, it is determined whether the target is reasonable.

[0030] Furthermore, the first target information detected by the video image and the second target information detected by the radar point cloud data are fused to obtain a fused target, including:

[0031] Performing internal and external calibration on the laser radar and the visual sensor to obtain calibration parameters;

[0032] Synchronize the video image and the radar point cloud data, and match two frames of the video image and the radar point cloud data closest to the current time according to the timestamp as synchronization data;

[0033] The synchronized data is mapped according to the calibration parameters to obtain the fused target.

[0034] In a second aspect, an embodiment of the present invention provides an obstacle detection system in a tunnel, wherein a sensing unit is provided beside a tunnel track, and the system includes the sensing unit and a computing unit;

[0035] The sensing unit is used to collect train operation environment data;

[0036] The computing unit is configured to analyze the train operating environment data to obtain operating environment structured information and train operating status structured information, wherein the operating environment structured information includes target features of the track area and the obstacle, and the train operating status structured information includes train features; process the operating environment structured information and the train operating status structured information to obtain position information of a passing train and target information of an obstacle; determine whether the obstacle exists based on the position information of the passing train and the target information of the obstacle; and trigger an alarm if the obstacle exists.

[0037] Furthermore, the sensing unit is specifically configured to:

[0038] When the perception unit is a visual sensor, the video image is collected by the visual sensor;

[0039] When the sensing unit is a laser radar, radar point cloud data is collected by the laser radar;

[0040] When the perception unit includes the visual sensor and the laser radar, the video image is collected by the visual sensor, and the radar point cloud data is collected by the laser radar.

[0041] Furthermore, the computing unit is specifically configured to:

[0042] detecting first target information from the video image using an analysis algorithm;

[0043] or,

[0044] detecting second target information from the radar point cloud data using the analysis algorithm;

[0045] The first target information and the second target information both include tracks, trains and the obstacles.

[0046] In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the obstacle detection method in the tunnel as described above is implemented.

[0047] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having a non-volatile program code executable by a processor, wherein the program code enables the processor to execute the above-mentioned method for detecting obstacles in a tunnel.

[0048] An embodiment of the present invention provides an obstacle detection method and system in a tunnel, wherein a sensing unit is provided on the side of the tunnel track, including: collecting train operating environment data through the sensing unit; analyzing the train operating environment data to obtain operating environment structured information and train operating status structured information, wherein the operating environment structured information includes target features of the track area and the obstacle, and the train operating status structured information includes train features; processing the operating environment structured information and the train operating status structured information to obtain position information of a passing train and obstacle target information; judging whether an obstacle exists based on the passing train position information and the obstacle target information; and triggering an alarm if an obstacle exists. Obstacles on the line can be effectively and accurately detected, detection efficiency can be improved, and the safety of passengers can be effectively ensured.

[0049] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the present invention. The purposes and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] FIG1 is a flow chart of a method for detecting obstacles in a tunnel provided by a first embodiment of the present invention;

[0053] FIG2 is a schematic diagram of an obstacle detection system in a tunnel provided by a second embodiment of the present invention. DETAILED DESCRIPTION

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] To facilitate understanding of this embodiment, the embodiment of the present invention is described in detail below.

[0056] Example 1:

[0057] FIG1 is a flow chart of a method for detecting obstacles in a tunnel provided in a first embodiment of the present invention.

[0058] 1 , a sensing unit is provided beside the tunnel track. The method includes the following steps:

[0059] Step S101, collecting train operating environment data through a sensing unit;

[0060] Here, the train operation environment data includes passing trains, tracks, obstacles, and on-site tunnel environments. These targets are roughly 5-120 meters in front of the sensing unit, which is the effective analysis and monitoring range of the algorithm.

[0061] Step S102: Analyze the train operating environment data to obtain operating environment structured information and train operating status structured information. The operating environment structured information includes target characteristics of the track area and obstacles (category, size, distance, and whether they are intrusions), and the train operating status structured information includes train characteristics. The train characteristics may be train head characteristics.

[0062] Step S103: Processing the operating environment structured information and the train operating status structured information to obtain the passing train position information and obstacle target information;

[0063] Step S104, determining whether there is an obstacle based on the passing train position information and obstacle target information;

[0064] Step S105: If it exists, trigger an alarm.

[0065] This application realizes obstacle detection in tunnel sections by installing detection and analysis equipment on the trackside of tunnel sections, setting up visual sensors and lidars, and based on video image analysis and radar point cloud analysis algorithms.

[0066] Furthermore, step S101 includes the following steps:

[0067] Step S201: When the sensing unit is a visual sensor, a video image is captured by the visual sensor; wherein the visual sensor may be a camera, but is not limited to a camera, and may also be other devices;

[0068] Specifically, when collecting video images, the analysis algorithm detects targets such as tracks, trains, and obstacles from the video images, takes advantage of the stability of the trackside tunnel environment, fully learns its background image, and combines traditional image analysis algorithms with AI algorithms to improve the detection rate and real-time processing. Among them, traditional image analysis algorithms can use background subtraction, frame difference method, etc., and AI algorithms can select multiple different learning models for multiple detections.

[0069] Step S202: When the sensing unit is a laser radar, radar point cloud data is collected by the laser radar;

[0070] Specifically, when collecting radar point cloud data, the analysis algorithm detects targets such as tracks, trains, and obstacles from the radar point cloud. It uses the stability of the trackside tunnel environment and the three-dimensional information of the radar point cloud to fully learn its background point cloud and construct a 3D spatial model of the tunnel environment. It combines traditional point cloud algorithms with AI algorithms to improve the detection rate and real-time processing. Among them, traditional point cloud algorithms can use background subtraction, and AI algorithms can be based on 3D point cloud detection frameworks and learning models.

[0071] Step S203, when the perception unit includes a visual sensor and a laser radar, the video image is collected by the visual sensor, and the radar point cloud data is collected by the laser radar.

[0072] Furthermore, step S103 includes:

[0073] detecting first target information from the video image using an analysis algorithm;

[0074] or,

[0075] Detecting the second target information from the radar point cloud data using an analysis algorithm;

[0076] The first target information and the second target information both include tracks, trains and obstacles.

[0077] Furthermore, step S103 includes the following steps:

[0078] Step S301, fusing the first target information detected by the video image and the second target information detected by the radar point cloud data to obtain a fused target;

[0079] Step S301, judging based on the category of the fused target and the target position overlap;

[0080] Step S301: fusing the video image and radar point cloud data whose target position overlap is greater than a preset threshold into the same target, and setting a first confidence level;

[0081] Step S301: The video image and radar point cloud data, whose target position overlap is less than a preset threshold, are treated as two separate targets, and a second confidence level is set; wherein the first confidence level has a higher priority than the second confidence level, and the first confidence level is a higher confidence level, and the second confidence level is a lower confidence level;

[0082] Step S301 : Calculate the position information of the passing train based on the depth information of the target, and determine whether the target is reasonable by combining the constructed 3D space of the tunnel and the track area.

[0083] Here, if the target is unreasonable, it is eliminated.

[0084] Furthermore, step S301 includes the following steps:

[0085] Step S401, performing internal and external calibration on the laser radar and visual sensor to obtain calibration parameters;

[0086] Here, the fusion processing of video images and radar point clouds requires the calibration of the internal and external parameters between the lidar and the visual sensor. The lidar points are mapped to the video image through the calibrated internal parameters of the visual sensor and the external parameters between the lidar and the visual sensor.

[0087] Step S402: Synchronize the video image and the radar point cloud data, and match the two frames of video image and radar point cloud data closest to the current time according to the timestamp as synchronization data;

[0088] Step S403: Map the synchronized data according to the calibration parameters to obtain a fused target.

[0089] Specifically, when performing internal and external parameter calibration on the lidar and visual sensor, the video image and radar point cloud data have a time series, and the data of both need to be processed synchronously, that is, the two frames of video image and radar point cloud data closest to the current time are matched according to the timestamp as the synchronous data, and then the synchronous data are mapped according to the above calibration parameters to achieve data fusion.

[0090] In summary, if sensing units are deployed at intervals along the tunnel trackside line, the spacing between them is the effective detection distance of the corresponding analysis algorithm of the sensing unit, and obstacle detection covering the entire tunnel section line can be achieved. If the calculation unit detects an obstacle, it can be combined with the position information of the running train to alarm and trigger the braking of related trains, forming a closed loop of obstacle detection and alarm processing process, reducing dependence on external systems.

[0091] Among them, different types of perception units have different effective detection distances. For example, if the perception unit is a camera, it is arranged at a spacing of 60 meters; if the perception unit is a lidar, it is arranged at a spacing of 120 meters.

[0092] This application uses trackside sensors (visual sensors and lidar) to monitor passing trains and obstacles. Compared with existing vehicle-mounted methods, it can detect obstacles on the line earlier and alarm and brake related trains.

[0093] Based on video image / radar point cloud analysis algorithms and combined with the stability of the tunnel environment, obstacles on the line can be effectively and accurately detected. Compared with existing vehicle-mounted methods, this solution can better balance real-time detection and accuracy.

[0094] Example 2:

[0095] FIG2 is a schematic diagram of an obstacle detection system in a tunnel provided by a second embodiment of the present invention.

[0096] 2 , a sensing unit is provided at the tunnel trackside. The system includes a sensing unit and a computing unit. The computing unit is provided at the station and is connected to the sensing unit.

[0097] Sensing unit, used to collect train operation environment data;

[0098] The computing unit is used to analyze the train operating environment data to obtain operating environment structured information and train operating status structured information, wherein the operating environment structured information includes target features of the track area and obstacles, and the train operating status structured information includes train features; process the operating environment structured information and the train operating status structured information to obtain position information of passing trains and target information of obstacles; determine whether the obstacle exists based on the position information of passing trains and the target information of obstacles; and trigger an alarm if the obstacle exists.

[0099] Furthermore, the perception unit is specifically used to:

[0100] When the perception unit is a visual sensor, the video image is collected through the visual sensor;

[0101] When the sensing unit is a lidar, radar point cloud data is collected through the lidar;

[0102] When the perception unit includes a visual sensor and the laser radar, the video image is collected by the visual sensor, and the radar point cloud data is collected by the laser radar.

[0103] Furthermore, the computing unit is specifically used for:

[0104] detecting first target information from the video image using an analysis algorithm;

[0105] or,

[0106] Detecting the second target information from the radar point cloud data using an analysis algorithm;

[0107] The first target information and the second target information both include tracks, trains and obstacles.

[0108] An embodiment of the present invention provides an obstacle detection method and system in a tunnel, wherein a sensing unit is provided on the side of the tunnel track, including: collecting train operating environment data through the sensing unit; analyzing the train operating environment data to obtain operating environment structured information and train operating status structured information, wherein the operating environment structured information includes target features of the track area and the obstacle, and the train operating status structured information includes train features; processing the operating environment structured information and the train operating status structured information to obtain position information of a passing train and obstacle target information; judging whether an obstacle exists based on the passing train position information and the obstacle target information; and triggering an alarm if an obstacle exists. Obstacles on the line can be effectively and accurately detected, detection efficiency can be improved, and the safety of passengers can be effectively ensured.

[0109] An embodiment of the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the method for detecting obstacles in a tunnel provided in the above embodiment are implemented.

[0110] An embodiment of the present invention further provides a computer-readable medium having a non-volatile program code executable by a processor. The computer-readable medium stores a computer program. When the computer program is executed by the processor, the steps of the method for detecting obstacles in a tunnel of the above embodiment are executed.

[0111] The computer program product provided in the embodiments of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the previous method embodiments. For specific implementation, please refer to the method embodiments and will not be repeated here.

[0112] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems and devices can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0113] In addition, in the description of the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections; they may refer to direct connections or indirect connections through an intermediate medium; and they may refer to internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0114] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0115] In the description of the present invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0116] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for detecting obstacles in a tunnel, characterized in that: A sensing unit is provided beside the tunnel track, and the method comprises: Collecting train operation environment data through the sensing unit; Analyzing the train operation environment data to obtain operation environment structured information and train operation status structured information, wherein the operation environment structured information includes target features of the track area and the obstacles, and the train operation status structured information includes train features; Processing the operating environment structured information and the train operating status structured information to obtain passing train position information and obstacle target information; Determining whether the obstacle exists according to the passing train position information and the obstacle target information; If present, an alarm is triggered.

2. The method for detecting obstacles in a tunnel according to claim 1, characterized in that: Collecting train operation environment data through the sensing unit includes: When the perception unit is a visual sensor, the video image is collected by the visual sensor; When the sensing unit is a laser radar, radar point cloud data is collected by the laser radar; When the perception unit includes the visual sensor and the laser radar, the video image is collected by the visual sensor, and the radar point cloud data is collected by the laser radar.

3. The method for detecting obstacles in a tunnel according to claim 2, characterized in that: Processing the operating environment structured information and the train operating status structured information to obtain the passing train position information and obstacle target information, including: Detecting first target information from the video image using an analysis algorithm; or, Detecting second target information from the radar point cloud data using the analysis algorithm; Wherein, the first target information and the second target information both include tracks, trains and the obstacles.

4. The method for detecting obstacles in a tunnel according to claim 2, characterized in that: Processing the operating environment structured information and the train operating status structured information to obtain the passing train position information and obstacle target information, including: Fusing the first target information detected by the video image and the second target information detected by the radar point cloud data to obtain a fused target; Making a judgment based on the category of the fused target and the target position overlap; The video image and the radar point cloud data whose target position overlap is greater than a preset threshold are fused into the same target, and a first confidence level is set; The video image and the radar point cloud data whose target position overlap is less than the preset threshold are regarded as two separate targets, and a second confidence level is set; wherein the first confidence level has a higher priority than the second confidence level; The position information of the passing train is calculated according to the depth information of the target, and combined with the constructed tunnel 3D space and the track area, it is determined whether the target is reasonable.

5. The method for detecting obstacles in a tunnel according to claim 4, characterized in that: Fusing the first target information detected by the video image and the second target information detected by the radar point cloud data to obtain a fused target, including: Performing internal and external calibration on the laser radar and the visual sensor to obtain calibration parameters; Synchronously processing the video image and the radar point cloud data, and matching two frames of the video image and the radar point cloud data closest to the current time according to the timestamp as synchronization data; The synchronous data is mapped according to the calibration parameters to obtain the fused target.

6. An obstacle detection system in a tunnel, characterized in that: A sensing unit is arranged beside the tunnel track, and the system comprises the sensing unit and a computing unit; The sensing unit is used to collect train operation environment data; The computing unit is used to analyze the train running environment data to obtain running environment structured information and train running state structured information, wherein the running environment structured information includes target features of the track area and the obstacles, and the train running state structured information includes train features; Processing the operating environment structured information and the train operating status structured information to obtain passing train position information and obstacle target information; Determining whether the obstacle exists according to the passing train position information and the obstacle target information; If present, an alarm is triggered.

7. The obstacle detection system in a tunnel according to claim 6, characterized in that: The sensing unit is specifically used for: When the perception unit is a visual sensor, the video image is collected by the visual sensor; When the sensing unit is a laser radar, radar point cloud data is collected by the laser radar; When the perception unit includes the visual sensor and the laser radar, the video image is collected by the visual sensor, and the radar point cloud data is collected by the laser radar.

8. The obstacle detection system in a tunnel according to claim 7, characterized in that: The computing unit is specifically used for: Detecting first target information from the video image using an analysis algorithm; or, Detecting second target information from the radar point cloud data using the analysis algorithm; Wherein, the first target information and the second target information both include tracks, trains and the obstacles.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, wherein: When the processor executes the computer program, the method for detecting obstacles in a tunnel as described in any one of claims 1 to 5 is implemented.

10. A computer readable medium having a non-volatile program code executable by a processor, characterized in that: The program code enables the processor to execute the obstacle detection method in a tunnel according to any one of claims 1 to 5.

Images