Railway vehicle stop platform identification method and system, electronic equipment and storage medium thereof

Abstract

The invention discloses a railway vehicle stop platform identification method and system, an electronic device and a storage medium thereof.The method comprises the steps that vehicle arrival information is obtained through a rail sensing camera mechanism, a 3D laser radar is scheduled to obtain rail area point cloud information in real time, the point cloud information is corrected and preprocessed, and rail vehicle feature information is extracted; and comparing the pre-stored operation information with the extracted feature information, and judging the real-time position of the vehicle at the station. According to the railway vehicle stop platform identification method and system, the electronic equipment and the storage medium thereof, the accuracy and reliability of railway vehicle stop platform identification can be effectively improved, and a powerful guarantee is provided for railway transportation safety and efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of railway track control, and in particular relates to a method, system, electronic equipment and storage medium thereof for identifying a railway vehicle stopping platform. Background Art

[0002] As one of the important modes of transportation in modern society, the safety and efficiency of railway transportation have a profound impact on social and economic development. With the continuous expansion of the railway network and the increase in train speed, the operation safety and efficiency of railway vehicles have become increasingly prominent. Especially in high-density and high-speed railway transportation systems, accurate train stops play a vital role in ensuring passenger safety, improving operational efficiency and reducing delays.

[0003] At present, the identification of railway vehicle stop platforms mainly relies on manual observation and some basic automation systems. Although the manual observation method is intuitive, it is inefficient and easily affected by human factors, such as fatigue and distraction, which may lead to misjudgment or missed judgment. In addition, with the increase in railway transportation volume, the limitations of manual monitoring have become more and more obvious, and it is difficult to meet the needs of modern railway transportation.

[0004] Although the existing automation system has reduced the burden of manual monitoring to a certain extent, its functions are relatively simple and mainly rely on fixed sensors and preset programs. These systems are unable to cope with complex situations. For example, in bad weather conditions (such as fog, heavy rain, etc.), the accuracy of sensors will be affected, resulting in a decrease in recognition accuracy. In addition, traditional automation systems also have deficiencies in data processing and real-time response, making it difficult to achieve rapid and accurate recognition of train stop status. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method, system, electronic device and storage medium for identifying railway vehicle stop platforms. By fusing lidar data and station control information and combining machine learning algorithms or Euclidean clustering adaptive algorithms, high-precision identification of the stop status of railway vehicles is achieved, the safety and efficiency of railway operations are improved, and the possibility of human error is reduced.

[0006] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0007] In a first aspect, the present invention provides a method for identifying a railway vehicle stopping at a platform, comprising the following steps:

[0008] Based on the pre-stored railway vehicle stop operation information, the track sensor camera mechanism corresponding to the stop of the dispatching vehicle is controlled to obtain the vehicle entry information;

[0009] According to the obtained vehicle entry information, the 3D laser radar is arranged in a linear array along the track direction to obtain the point cloud information of the track area in real time;

[0010] The acquired track area point cloud information is corrected and preprocessed, and the characteristic information of the track vehicle is extracted by combining the threshold-adjustable Euclidean clustering adaptive algorithm;

[0011] The pre-stored railway vehicle stop operation information is retrieved and compared with the extracted rail vehicle characteristic information to determine the real-time position result of the vehicle stop platform.

[0012] In this embodiment, constructing a track area point cloud extraction based on laser radar scan line features includes setting up a number of 3D laser radars evenly spaced along the track direction on the upper side of the track area close to the platform area; using the 3D laser radar to collect horizontal single scan line data along the track direction of the track area, and collecting vertical single scan line data of the track area perpendicular to the track direction in a fixed pitch angle resolution distribution mode.

[0013] In this embodiment, the point cloud scanning parameter control for obtaining the track area point cloud information and the platform area point cloud information is expressed by the formula:

[0014] k=T×f

[0015] φ=ω stop -ω start

[0016] δ pitch =φ / k;

[0017] Where k represents the number of scanning lines of a single frame of the point cloud image, T represents the scanning time, f represents the single scanning frequency, ω represents the pitch angle scanning range, and δ represents the pitch angle resolution.

[0018] In this embodiment, the point cloud information is corrected and preprocessed using the following method:

[0019] Input the point set P from the entire point cloud, determine the number of points m on a single scan line based on the known scan frequency f, use the original pulse distance R and device height H to obtain the transformation formula of the point cloud coordinates (x, y, z), and divide the point cloud into separate scan lines:

[0020] n=m*T / f;

[0021]

[0022] I i = {P j |P j ∈P m*i，m*(i+1)]}(I i ∈I, 0<i≤k);

[0023] In the formula, n represents the total number of points in a frame, m represents the number of scanning lines, and θ represents the horizontal angle of pulse scanning;

[0024] Randomly select a reference scan line, project it to the xoy plane and select a set of scan points, use the RANSAC algorithm to fit the scan line, obtain the spatial parameters and calculate the deviation angle, and calculate the second deviation angle by projecting the scan line to the yoz plane:

[0025] The deviation angle is obtained according to the fitting parameters of the scan line, and the angles αxoy and αyoz are used to construct the correction rotation matrices R1 and R2 represented by the scene point cloud. By applying the rotation matrices R1 and R2 to the original point cloud P, the corrected point cloud P' is obtained:

[0026]

[0027] P′=R1R2P.

[0028] In this embodiment, based on the traditional Euclidean clustering algorithm, it is determined that the central area of ​​the scan line at the starting pitch angle ω forms the area with the highest point cloud density, and an adjustable threshold strategy is formulated:

[0029] Cluster tracks, vehicles, and other objects according to the index input, and obtain the number of points in each cluster to build an n×3 point cloud matrix ζ;

[0030] Normalize the data of each point in the matrix ζ to its mean value, and use the normalized data to calculate the covariance matrix C. By solving equation E T CE=Λ to obtain three eigenvectors and corresponding eigenvalues;

[0031] The eigenvalues ​​are arranged in descending order as λ1>λ2>λ3, and the clusters are divided into track, vehicle or other object types according to the relationship between the eigenvalues;

[0032] Use a preset threshold μ 1 and μ 2 to determine the specific type of clustering. If λ1 / λ2> μ 1, then it is an orbit; if λ2 / λ3> μ 2, it is a vehicle; otherwise, it is another type of object.

[0033] In this embodiment, the adjustable threshold strategy is expressed by the formula:

[0034] ε adaptiυe =f(x,y,p i )ε;

[0035]

[0036] In the formula, the variable ε represents the initial input threshold in the traditional Euclidean clustering algorithm, f(x, y, p i ) represents the cluster center point p i The x- and y-direction associated modulation functions, δ horizontal Expressed as single scan horizontal angular resolution.

[0037] In this embodiment, the following steps are also included:

[0038] Retrieving pre-stored railway vehicle stop operation information to obtain the vehicle number, entry time, entry track, stop location identification, and departure time of the vehicle;

[0039] According to the vehicle speed obtained by the pre-processed track sensor camera mechanism, combined with the vehicle historical data and the preset vehicle deceleration mode, the change trend of the vehicle speed is judged in real time to determine whether the vehicle is temporarily stopping or decelerating to enter the station;

[0040] When it is determined to be a temporary stop or a reduced speed entry into the station, the vehicle's track stop position is obtained based on the rail vehicle feature information obtained and extracted in real time by the 3D lidar, and the position is matched with the stop position mark in the pre-stored railway vehicle stop operation information to determine the vehicle's stop status.

[0041] In a second aspect, the present invention provides a system for identifying and dispatching railway vehicles stopping at platforms, comprising the following modules:

[0042] An information acquisition module is used to control the track sensor camera mechanism corresponding to the stop of the dispatching vehicle based on the pre-stored railway vehicle stop operation information to obtain the vehicle entry information;

[0043] Point cloud acquisition module, used to control and dispatch 3D laser radars arranged in a linear array along the track direction to acquire point cloud information of the track area in real time according to the acquired vehicle entry information;

[0044] A feature acquisition module is used to perform correction preprocessing on the acquired track area point cloud information and extract rail vehicle feature information in combination with a threshold-adjustable Euclidean clustering adaptive algorithm;

[0045] The stop determination module is used to retrieve the pre-stored railway vehicle stop operation information and compare it with the extracted rail vehicle characteristic information to determine the real-time position result of the vehicle stop platform.

[0046] In a third aspect, the present invention provides an electronic device, comprising:

[0047] processor;

[0048] a memory for storing instructions executable by the processor;

[0049] The processor is configured to execute the instructions to implement the method for identifying a railway vehicle stopping platform as described in any one of the first aspects.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium, which, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute the method for identifying a railway vehicle stop platform as described in any one of the first aspects.

[0051] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the method for identifying railway vehicle stop platforms provided by the present invention utilizes a track sensing camera mechanism to obtain vehicle entry information, dispatches a 3D laser radar to obtain track area point cloud information in real time, performs correction preprocessing on the point cloud information, extracts rail vehicle feature information, compares pre-stored operating information and the extracted feature information, and determines the real-time position of the vehicle stop platform, which can effectively improve the accuracy and reliability of railway vehicle stop platform identification and provide strong guarantees for railway transportation safety and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The present invention provides a flow chart of a method for identifying a railway vehicle stopping at a platform according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0054] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0055] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0056] Example

[0057] like Figure 1 As shown, an embodiment of the present invention provides a method for identifying a railway vehicle stopping at a platform, comprising the following steps:

[0058] Based on the pre-stored railway vehicle stop operation information, the track sensor camera mechanism corresponding to the stop of the dispatching vehicle is controlled to obtain the vehicle entry information;

[0059] According to the obtained vehicle entry information, the 3D laser radar is arranged in a linear array along the track direction to obtain the point cloud information of the track area in real time;

[0060] The acquired track area point cloud information is corrected and preprocessed, and the characteristic information of the track vehicle is extracted by combining the threshold-adjustable Euclidean clustering adaptive algorithm;

[0061] The pre-stored railway vehicle stop operation information is retrieved and compared with the extracted rail vehicle characteristic information to determine the real-time position result of the vehicle stop platform.

[0062] In this embodiment, constructing a track area point cloud extraction based on laser radar scan line features includes setting up a number of 3D laser radars evenly spaced along the track direction on the upper side of the track area close to the platform area; using the 3D laser radar to collect horizontal single scan line data along the track direction of the track area, and collecting vertical single scan line data of the track area perpendicular to the track direction in a fixed pitch angle resolution distribution mode.

[0063] In this embodiment, the pre-stored railway vehicle stop operation information is retrieved to obtain the vehicle number, entry time, entry track, stop position identification, and departure time of the vehicle;

[0064] According to the vehicle speed obtained by the pre-processed track sensor camera mechanism, combined with the vehicle historical data and the preset vehicle deceleration mode, the change trend of the vehicle speed is judged in real time to determine whether the vehicle is temporarily stopping or decelerating to enter the station;

[0065] When it is determined to be a temporary stop or a reduced speed entry into the station, the vehicle's track stop position is obtained based on the rail vehicle feature information obtained and extracted in real time by the 3D lidar, and the position is matched with the stop position mark in the pre-stored railway vehicle stop operation information to determine the vehicle's stop status.

[0066] In this embodiment, the point cloud scanning parameter control for obtaining the track area point cloud information and the platform area point cloud information is expressed by the formula:

[0067] k=T×f

[0068] φ=ω stop -ω start

[0069] δ pitch =φ / k;

[0070] Where k represents the number of scanning lines of a single frame of the point cloud image, T represents the scanning time, f represents the single scanning frequency, ω represents the pitch angle scanning range, and δ represents the pitch angle resolution.

[0071] In this embodiment, the point cloud information is corrected and preprocessed using the following method:

[0072] Input the point set P from the entire point cloud, determine the number of points m on a single scan line based on the known scan frequency f, use the original pulse distance R and device height H to obtain the transformation formula of the point cloud coordinates (x, y, z), and divide the point cloud into separate scan lines:

[0073] n=m*T / f;

[0074]

[0075] I i = {P j |P j ∈P m*i，m*(i+1)]}(I i ∈I, 0<i≤k);

[0076] In the formula, n represents the total number of points in a frame, m represents the number of scanning lines, and θ represents the horizontal angle of pulse scanning;

[0077] Randomly select a reference scan line, project it to the xoy plane and select a set of scan points, use the RANSAC algorithm to fit the scan line, obtain the spatial parameters and calculate the deviation angle, and calculate the second deviation angle by projecting the scan line to the yoz plane:

[0078] The deviation angle is obtained according to the fitting parameters of the scan line, and the angles αxoy and αyoz are used to construct the correction rotation matrices R1 and R2 represented by the scene point cloud. By applying the rotation matrices R1 and R2 to the original point cloud P, the corrected point cloud P' is obtained:

[0079]

[0080] P′=R1R2P.

[0081] In this embodiment, based on the traditional Euclidean clustering algorithm, it is determined that the central area of ​​the scan line at the starting pitch angle ω forms the area with the highest point cloud density, and an adjustable threshold strategy is formulated:

[0082] Cluster tracks, vehicles, and other objects according to the index input, and obtain the number of points in each cluster to build an n×3 point cloud matrix ζ;

[0083] Normalize the data of each point in the matrix ζ to its mean value, and use the normalized data to calculate the covariance matrix C. By solving equation E T CE=Λ to obtain three eigenvectors and corresponding eigenvalues;

[0084] The eigenvalues ​​are arranged in descending order as λ1>λ2>λ3, and the clusters are divided into track, vehicle or other object types according to the relationship between the eigenvalues;

[0085] The preset thresholds μ1 and μ2 are used to determine the specific type of cluster. If λ1 / λ2>μ1, it is a track; if λ2 / λ3>μ2, it is a vehicle; otherwise, it is other object types.

[0086] In this embodiment, the adjustable threshold strategy is expressed by the formula:

[0087] ε adaptiυe =f(x,y,p i )ε;

[0088]

[0089] In the formula, the variable ε represents the initial input threshold in the traditional Euclidean clustering algorithm, f(x, y, p i ) represents the cluster center point p i The x- and y-direction associated modulation functions, δ horizontal Expressed as single scan horizontal angular resolution.

[0090] In a second aspect, the present invention provides a system for identifying and dispatching railway vehicles stopping at platforms, comprising the following modules:

[0091] An information acquisition module is used to control the track sensor camera mechanism corresponding to the stop of the dispatching vehicle based on the pre-stored railway vehicle stop operation information to obtain the vehicle entry information;

[0092] Point cloud acquisition module, used to control and dispatch 3D laser radars arranged in a linear array along the track direction to acquire point cloud information of the track area in real time according to the acquired vehicle entry information;

[0093] A feature acquisition module is used to perform correction preprocessing on the acquired track area point cloud information and extract rail vehicle feature information in combination with a threshold-adjustable Euclidean clustering adaptive algorithm;

[0094] The stop determination module is used to retrieve the pre-stored railway vehicle stop operation information and compare it with the extracted rail vehicle characteristic information to determine the real-time position result of the vehicle stop platform.

[0095] In a third aspect, the present invention provides an electronic device, comprising:

[0096] processor;

[0097] a memory for storing instructions executable by the processor;

[0098] The processor is configured to execute the instructions to implement the method for identifying a railway vehicle stopping platform as described in any one of the first aspects.

[0099] In a fourth aspect, the present invention provides a computer-readable storage medium, which, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to execute the method for identifying a railway vehicle stop platform as described in any one of the first aspects.

[0100] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for identifying a railway vehicle stopping at a platform, characterized in that: The steps include: Based on the pre-stored railway vehicle stop operation information, the track sensor camera mechanism corresponding to the stop of the dispatching vehicle is controlled to obtain the vehicle entry information; According to the obtained vehicle entry information, the 3D laser radar is arranged in a linear array along the track direction to obtain the point cloud information of the track area in real time; The acquired track area point cloud information is corrected and preprocessed, and the characteristic information of the track vehicle is extracted by combining the threshold-adjustable Euclidean clustering adaptive algorithm; The pre-stored railway vehicle stop operation information is retrieved and compared with the extracted rail vehicle characteristic information to determine the real-time position result of the vehicle stop platform.

2. The method for identifying a railway vehicle stopping platform according to claim 1, characterized in that: Constructing the track area point cloud extraction based on the laser radar scanning line characteristics includes setting up a number of 3D laser radars evenly spaced along the track direction on the upper side of the track area close to the platform area; using the 3D laser radar to collect horizontal single scanning line data along the track direction of the track area, and collecting vertical single scanning line data of the track area perpendicular to the track direction in a fixed pitch angle resolution distribution mode.

3. The method for identifying a railway vehicle stopping platform according to claim 2, characterized in that: The point cloud scanning parameter control for obtaining the point cloud information of the track area and the platform area is expressed by the formula: k=T×f φ = ω stop -oh start d pitch =φ / k; Where k represents the number of scanning lines of a single frame of the point cloud image, f represents the scanning time, f represents the single scanning frequency, ω represents the pitch angle scanning range, and δ represents the pitch angle resolution.

4. The method for identifying a railway vehicle stopping platform according to claim 3, characterized in that: The point cloud information is corrected and preprocessed using the following method: Input the point set P from the entire point cloud, determine the number of points m on a single scan line based on the known scan frequency f, use the original pulse distance R and device height H to obtain the transformation formula of the point cloud coordinates (x, y, z), and divide the point cloud into separate scan lines: n=m*T / f; I i ={P j |P j ∈P [m*i，m*(i+1] }(I i ∈I,0<i≤k); In the formula, n represents the total number of points in a frame, m represents the number of scanning lines, and θ represents the horizontal angle of pulse scanning; Randomly select a reference scan line, project it to the xoy plane and select a set of scan points, use the RANSAC algorithm to fit the scan line, obtain the spatial parameters and calculate the deviation angle, and calculate the second deviation angle by projecting the scan line to the yoz plane: The deviation angle is obtained according to the fitting parameters of the scan line, and the angles αxoy and αyoz are used to construct the correction rotation matrices R1 and R2 represented by the scene point cloud. By applying the rotation matrices R1 and R2 to the original point cloud P, the corrected point cloud P is obtained: P′=R1R2P.

5. The method for identifying a railway vehicle stopping platform according to claim 4, characterized in that: Based on the traditional Euclidean clustering algorithm, the central area of ​​the scan line at the starting pitch angle ω is determined to form the area with the highest point cloud density, and an adjustable threshold strategy is formulated: Cluster tracks, vehicles, and other objects according to the index input, and obtain the number of points in each cluster to build an n×3 point cloud matrix ζ; Normalize the data of each point in the matrix ζ to its mean value, and use the normalized data to calculate the covariance matrix C. By solving equation E T CE=Λ to obtain three eigenvectors and corresponding eigenvalues; The eigenvalues ​​are arranged in descending order as λ1>λ2>λ3, and the clusters are divided into track, vehicle or other object types according to the relationship between the eigenvalues; The preset thresholds μ1 and μ2 are used to determine the specific type of cluster. If λ1 / λ2>μ1, it is a track; if λ2 / λ3>μ2, it is a vehicle; otherwise, it is other object types.

6. The method for identifying a railway vehicle stopping platform according to claim 5, characterized in that: The adjustable threshold strategy is expressed as: e adaptiυe =f(x,y,p i )e; In the formula, the variable ε represents the initial input threshold in the traditional Euclidean clustering algorithm, f(x, y, p i ) represents the cluster center point p i The x- and y-direction associated modulation functions, δ horizontal Expressed as single scan horizontal angular resolution.

7. The method for identifying a railway vehicle stopping platform according to claim 2, characterized in that: The following steps are also included: Retrieving pre-stored railway vehicle stop operation information to obtain the vehicle number, entry time, entry track, stop location identification, and departure time of the vehicle; According to the vehicle speed obtained by the pre-processed track sensor camera mechanism, combined with the vehicle historical data and the preset vehicle deceleration mode, the change trend of the vehicle speed is judged in real time to determine whether the vehicle is temporarily stopping or decelerating to enter the station; When it is determined to be a temporary stop or a reduced speed entry into the station, the vehicle's track stop position is obtained based on the rail vehicle feature information obtained and extracted in real time by the 3D lidar, and the position is matched with the stop position mark in the pre-stored railway vehicle stop operation information to determine the vehicle's stop status.

8. A system for identifying and dispatching railway vehicles stopping at platforms, characterized in that: Includes the following modules: An information acquisition module is used to control the track sensor camera mechanism corresponding to the stop of the dispatching vehicle based on the pre-stored railway vehicle stop operation information to obtain the vehicle entry information; Point cloud acquisition module, used to control and dispatch 3D laser radars arranged in a linear array along the track direction to acquire point cloud information of the track area in real time according to the acquired vehicle entry information; A feature acquisition module is used to perform correction preprocessing on the acquired track area point cloud information and extract rail vehicle feature information in combination with a threshold-adjustable Euclidean clustering adaptive algorithm; The stop determination module is used to retrieve the pre-stored railway vehicle stop operation information and compare it with the extracted rail vehicle characteristic information to determine the real-time position result of the vehicle stop platform.

9. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method for identifying a railway vehicle stopping platform according to any one of claims 1 to 7.

10. A computer-readable storage medium, when instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the method for identifying a railway vehicle stop platform according to any one of claims 1 to 7.

Images