Train running detection method and system based on multi-feature fusion

By using multi-feature fusion detection from radar and cameras, the accuracy and reliability issues of trains stopping at maintenance stations have been resolved, enabling precise and coarse monitoring at different distances and ensuring accurate train stops.

CN116811962BActive Publication Date: 2026-03-24HUIZHIAN INFORMATION TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-09
Publication Date
2026-03-24

AI Technical Summary

Technical Problem

In existing technologies, the real-time relative position relationship of trains cannot be accurately tracked when they stop at maintenance stations, resulting in insufficient accuracy and reliability of stopping control, and radar equipment is prone to false alarms.

Method used

By combining radar equipment and cameras for multi-feature fusion detection, radar scanning is used to detect the train's driving status information. Combined with image analysis, the train's attitude and distance are adjusted, and braking control commands are sent to ensure accurate stopping.

Benefits of technology

It improves the accuracy and reliability of train stopping control at maintenance stations. Through multi-feature monitoring using radar and cameras, it achieves precise and coarse monitoring at different distances, ensuring that trains can stop in a timely and accurate manner.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116811962B_ABST
    Figure CN116811962B_ABST
Patent Text Reader

Abstract

The application provides a train driving detection method and system based on multi-feature fusion, which utilizes radar equipment and a camera to perform multi-feature monitoring on a train in the case that the train is far away from a maintenance station and close to the maintenance station, performs rough monitoring on the train in the case that the train is far away from the maintenance station in a radar scanning mode, and performs accurate monitoring on the train in the case that the train is close to the maintenance station in an image shooting mode, so that appropriate brake control instructions can be sent to the train in time, the train can be ensured to be parked in a corresponding platform area of the maintenance station, and the control accuracy and reliability of the train parked in the maintenance station are improved.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of train driving monitoring, in particular to a train driving detection method and system based on multi-feature fusion. BACKGROUND

[0002] Trains need to enter the station for maintenance after a certain mileage, at which time the train needs to be parked at the corresponding maintenance station for inspection. In order to ensure that the train can be parked at the maintenance station, a radar device is usually arranged outside the maintenance station to detect the train by radar scanning, so as to determine whether the train is approaching the maintenance station. However, the above method is only used to roughly judge the relative motion relationship between the train and the maintenance station, and cannot accurately track the real-time relative position relationship between the train and the maintenance station, and only relying on the radar device to dynamically monitor the train can easily cause false positives, thereby reducing the control accuracy and reliability of the train parked at the maintenance station. SUMMARY

[0003] In view of the defects of the prior art, the present application provides a train driving detection method and system based on multi-feature fusion, which uses a radar device to detect the train by radar scanning to obtain the driving state information of the train, so as to determine whether the train enters the monitoring area range and estimate the driving duration of the train in the monitoring area range; during the driving duration, the train driving image is shot and analyzed to obtain the driving posture information of the train and the actual distance value between the train and the maintenance station, so as to adjust the driving state of the train; according to the actual distance value, a brake control instruction is sent to the driving control platform of the train, so as to instruct the train to be parked at the platform area corresponding to the maintenance station, which uses the radar device and the camera to monitor the train in the case of a large distance and a small distance between the train and the maintenance station, respectively, and uses the radar scanning method to roughly monitor the train when the distance is large, and uses the image shooting method to accurately monitor the train when the distance is small, so as to send appropriate brake control instructions to the train in time, so as to ensure that the train can be parked at the platform area corresponding to the maintenance station, thereby improving the control accuracy and reliability of the train parked at the maintenance station.

[0004] The present application provides a train driving detection method based on multi-feature fusion, which comprises the following steps:

[0005] Step S1, instructing the radar device at a preset distance value from the maintenance station to detect the train by radar scanning to obtain the driving state information of the train; according to the train driving state information, determining whether the train enters the monitoring area range of the maintenance station; and according to the train driving state information, estimating the driving duration of the train in the monitoring area range;

[0006] Step S2, instructing the camera located at the maintenance station to take pictures of the train within a time period corresponding to the driving duration, to obtain train driving images; analyzing and processing the train driving images to obtain driving posture information of the train and an actual distance value between the train and the maintenance station; and adjusting the driving state of the train according to the driving posture information and the actual distance value.

[0007] Step S3, sending a brake control instruction to a driving control platform of the train according to the actual distance value, so as to instruct the train to stop at a platform area corresponding to the maintenance station.

[0008] Further, in the step S1, the radar device at a preset distance value from the maintenance station is instructed to perform radar scanning detection on the train, so as to obtain train driving state information; it is judged whether the train enters a monitoring area range of the maintenance station according to the train driving state information; and the driving duration of the train in the monitoring area range is estimated according to the train driving state information, which specifically includes:

[0009] The radar device at a preset distance value from the maintenance station is periodically instructed to outwardly scan and emit radar signals, so as to perform radar scanning detection on the train near the radar device; and the distance between the train and the radar device and the driving speed of the train are obtained according to the radar signals emitted outwardly by the radar device and the radar signals reflected back by the train.

[0010] It is judged whether the train enters the monitoring area range of the maintenance station according to the distance and the radius of the monitoring area range of the maintenance station; if yes, the driving duration of the train in the monitoring area range is estimated according to the driving speed; and if no, the radar device is continuously instructed to periodically outwardly scan and emit radar signals.

[0011] Further, in the step S2, the camera located at the maintenance station is instructed to take pictures of the train within a time period corresponding to the driving duration, to obtain train driving images; the train driving images are analyzed and processed to obtain driving posture information of the train and an actual distance value between the train and the maintenance station; and the driving state of the train is adjusted according to the driving posture information and the actual distance value, which specifically includes:

[0012] The camera located at the maintenance station is instructed to take dynamic pictures of the train within a time period corresponding to the driving duration, to obtain train driving dynamic images; and the train driving dynamic images are frame-processed to obtain a plurality of train driving image frames.

[0013] The contour information of the train is extracted from each train driving image frame, and the driving posture information of the train is obtained according to the contour information corresponding to the train driving image frame; wherein the driving posture information includes an inclination angle of the train body relative to a plane on which the track is located during train driving.

[0014] identify the indication information of the preset interval sign on the track from each train running image frame, so as to obtain the actual distance value between the train and the maintenance station;

[0015] According to the running posture information, it is judged whether the train is in a running unstable state during running. If the train is in a running unstable state and the actual distance value is greater than a preset distance threshold, the train is instructed to run at a predetermined speed. If the train is in a running stable state and the actual distance value is greater than a preset distance threshold, the train is instructed to enter an acceleration running mode with a predetermined acceleration. If the actual distance value is less than or equal to a preset distance threshold, the train is instructed to enter a deceleration running mode.

[0016] Further, in the step S3, according to the actual distance value, a brake control instruction is sent to the driving control platform of the train, so as to instruct the train to stop at the platform area corresponding to the maintenance station, which specifically comprises:

[0017] If the actual distance value is less than or equal to a preset distance threshold, the running distance value between the train and the platform area to be stopped on the track is determined according to the actual distance value. According to the running distance value and the real-time running speed of the train, a deceleration acceleration value corresponding to the deceleration running mode is determined, so as to send a deceleration brake control instruction to the driving control platform of the train, thereby instructing the train to stop at the platform area corresponding to the maintenance station.

[0018] The present application also provides a train running detection system based on multi-feature fusion, which comprises:

[0019] A radar scanning control module is used to instruct the radar equipment with a preset distance value from the maintenance station to detect the train by radar scanning, so as to obtain the running state information of the train.

[0020] A train running analysis module is used to judge whether the train enters the monitoring area range of the maintenance station according to the train running state information, and estimate the running duration of the train in the monitoring area range according to the train running state information.

[0021] An image shooting control and analysis module is used to instruct the camera located at the maintenance station to shoot the train in the corresponding time period of the running duration, so as to obtain the train running image. The train running image is analyzed and processed to obtain the running posture information of the train and the actual distance value between the train and the maintenance station.

[0022] A train running adjustment module is used to adjust the running state of the train according to the running posture information and the actual distance value.

[0023] A train stopping control module is configured to send a brake control instruction to a driving control platform of the train according to the actual distance value, so as to instruct the train to stop at a platform area corresponding to the maintenance station.

[0024] Further, the radar scanning control module instructs a radar device with a preset distance value from the maintenance station to perform radar scanning detection on the train, so as to obtain train running state information, which specifically includes:

[0025] The radar device with the preset distance value from the maintenance station periodically scans outward to emit a radar signal, so as to perform radar scanning detection on the train near the radar device; and according to the radar signal emitted outward by the radar device and the radar signal reflected back by the train, a distance between the train and the radar device and a running speed of the train are obtained.

[0026] The train running analysis module is configured to determine whether the train enters a monitoring area range of the maintenance station according to the train running state information, and estimate a running duration of the train in the monitoring area range according to the train running state information, which specifically includes:

[0027] According to the distance and a radius of the monitoring area range of the maintenance station, it is determined whether the train enters the monitoring area range of the maintenance station; if so, the running duration of the train in the monitoring area range is estimated according to the running speed.

[0028] Further, the image shooting control and analysis module instructs a camera located at the maintenance station to shoot the train in a time period corresponding to the running duration, so as to obtain train running images; and the train running images are analyzed and processed to obtain train running posture information and an actual distance value between the train and the maintenance station, which specifically includes:

[0029] The camera located at the maintenance station is instructed to dynamically shoot the train in a time period corresponding to the running duration, so as to obtain train running dynamic images; and the train running dynamic images are frame-processed to obtain a plurality of train running image frames.

[0030] Train contour information is extracted from each train running image frame, and the train running posture information is obtained according to the contour information corresponding to the train running image frame; wherein the running posture information includes an inclination angle of a vehicle body relative to a plane on which a track is located during train running.

[0031] From each train running image frame, indication information of a preset interval indicator on the track is recognized, so as to obtain the actual distance value between the train and the maintenance station.

[0032] The train running adjustment module is configured to adjust a running state of the train according to the running posture information and the actual distance value, which specifically includes:

[0033] According to the driving posture information, it is judged whether the train is in a driving unstable state during driving, if the train is in the driving unstable state and the actual distance value is greater than a preset distance threshold, the train is instructed to drive at a predetermined speed, if the train is in a driving stable state and the actual distance value is greater than the preset distance threshold, the train is instructed to enter an acceleration driving mode with a predetermined acceleration, and if the actual distance value is less than or equal to the preset distance threshold, the train is instructed to enter a deceleration driving mode.

[0034] Further, the train stopping control module sends a brake control instruction to a driving control platform of the train according to the actual distance value, so as to instruct the train to stop at a platform area corresponding to the maintenance station, specifically comprising:

[0035] If the actual distance value is less than or equal to the preset distance threshold, a driving distance value between the train and the platform area to be stopped on the track is determined according to the actual distance value, a deceleration acceleration value corresponding to the deceleration driving mode is determined according to the driving distance value and a real-time driving speed of the train, a deceleration brake control instruction is sent to the driving control platform of the train, so as to instruct the train to stop at the platform area corresponding to the maintenance station.

[0036] Compared with the prior art, the train driving detection method and system based on multi-feature fusion utilize a radar device to perform radar scanning detection on the train to obtain driving state information of the train, so as to judge whether the train enters a monitoring area range and estimate a driving duration of the train in the monitoring area range, during the driving duration, train driving images are shot and analyzed to obtain driving posture information of the train and an actual distance value between the train and the maintenance station, so as to adjust the driving state of the train, according to the actual distance value, a brake control instruction is sent to a driving control platform of the train, so as to instruct the train to stop at a platform area corresponding to the maintenance station, in the case that the train is far away from the maintenance station and close to the maintenance station, the radar device and the camera are respectively utilized to perform multi-feature monitoring on the train, in the case that the train is far away from the maintenance station, the train is roughly monitored in a radar scanning mode, and in the case that the train is close to the maintenance station, the train is accurately monitored in an image shooting mode, so as to timely send a suitable brake control instruction to the train, so as to ensure that the train can be aligned and stopped at the platform area corresponding to the maintenance station, thereby improving control accuracy and reliability of the train stopping at the maintenance station.

[0037] Additional features and advantages of the application will be set forth in the description that follows, and in part will be apparent from the description, or can be learned by practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0038] The technical solutions of the embodiments of the present application will be further described below with reference to the drawings and embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings described below are only some of the embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0040] Figure 1 A flowchart of a train running detection method based on multi-feature fusion provided by the present application.

[0041] Figure 2 A structural diagram of a train running detection system based on multi-feature fusion provided by the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort belong to the scope of protection of the present application.

[0043] Reference Figure 1 A flowchart of a train running detection method based on multi-feature fusion provided by the embodiments of the present application. The train running detection method based on multi-feature fusion includes the following steps:

[0044] Step S1, a radar device indicating a preset distance value from a maintenance station performs radar scanning detection on a train, so as to obtain train running state information; according to the train running state information, it is judged whether the train enters a monitoring area range of the maintenance station; and according to the train running state information, the running duration of the train in the monitoring area range is estimated;

[0045] Step S2, a camera located at the maintenance station photographs the train in a time period corresponding to the running duration, so as to obtain train running images; the train running images are analyzed and processed, so as to obtain train running posture information and an actual distance value from the maintenance station; according to the running posture information and the actual distance value, the running state of the train is adjusted;

[0046] Step S3, according to the actual distance value, a brake control instruction is sent to a driving control platform of the train, so as to instruct the train to stop at a platform area corresponding to the maintenance station.

[0047] The train driving detection method based on multi-feature fusion has the advantages that: the radar equipment is used to perform radar scanning detection on the train to obtain the driving state information of the train, so as to determine whether the train enters the monitoring area range of the repair station and estimate the driving duration of the train in the monitoring area range; the train driving image is shot and analyzed during the driving duration to obtain the driving posture information of the train and the actual distance value between the train and the repair station, so as to adjust the driving state of the train; according to the actual distance value, a brake control instruction is sent to the driving control platform of the train, so as to instruct the train to stop at the platform area corresponding to the repair station; in the case that the train is far away from the repair station and close to the repair station, the radar equipment and the camera are respectively used to perform multi-feature monitoring on the train, the train is roughly monitored by the radar scanning mode when the train is far away from the repair station, and the train is accurately monitored by the image shooting mode when the train is close to the repair station, so that the appropriate brake control instruction can be sent to the train in time, the train can be aligned and stopped at the platform area corresponding to the repair station, and thus the control accuracy and reliability of the train stopping at the repair station are improved.

[0048] Preferably, in the step S1, the radar equipment indicating the preset distance value from the repair station performs radar scanning detection on the train, so as to obtain the driving state information of the train; according to the driving state information of the train, it is determined whether the train enters the monitoring area range of the repair station; and according to the driving state information of the train, the driving duration of the train in the monitoring area range is estimated.

[0049] The radar equipment indicating the preset distance value from the repair station periodically scans outward to emit a radar signal, so as to perform radar scanning detection on the train near the radar equipment; according to the radar signal emitted outward by the radar equipment and the radar signal reflected back by the train, the distance between the train and the radar equipment and the driving speed of the train are obtained.

[0050] According to the distance and the radius of the monitoring area range of the repair station, it is determined whether the train enters the monitoring area range of the repair station; if yes, the driving duration of the train in the monitoring area range is estimated according to the driving speed; if not, the radar equipment continues to periodically scan outward to emit a radar signal.

[0051] The beneficial effects of the above technical solutions are: the radar device is erected at a position separated from the maintenance station by a preset distance value, so that the radar device serves as the first monitoring checkpoint for the train approaching the maintenance station. The radar device periodically emits radar signals outward, and when the radar signals reach the surface of the train and are reflected back, the radar device performs difference processing on the strength or time of the outward emitted radar signals and the received reflected radar signals, i.e., obtains the separation distance between the train and the radar device and the running speed of the train, thereby quantitatively calibrating the running state of the train. In addition, the separation distance and the monitoring area range radius of the maintenance station are compared, and if the separation distance is less than or equal to the monitoring area range radius, it is determined that the train has entered the monitoring area range of the maintenance station, at which time the subsequent timely indication of the camera to dynamically photograph the train is facilitated, and the switching from rough monitoring by the radar device to precise monitoring by the camera is achieved. According to the monitoring area range radius and the running speed, the duration of the train running inside the monitoring area range is determined, which facilitates the control of the duration of the dynamic photographing of the train by the camera.

[0052] Preferably, in this step S2, the camera located at the maintenance station is instructed to photograph the train in the time period corresponding to the running duration, obtaining train running images; the train running images are analyzed and processed to obtain the running posture information of the train and the actual distance value between the train and the maintenance station; and the running state of the train is adjusted according to the running posture information and the actual distance value, specifically including:

[0053] The camera located at the maintenance station is instructed to dynamically photograph the train in the time period corresponding to the running duration, obtaining train running dynamic images; and the train running dynamic images are frame-processed to obtain a plurality of train running image frames.

[0054] The contour information of the train is extracted from each train running image frame, and the running posture information of the train is obtained according to the contour information corresponding to the train running image frame; wherein the running posture information includes the inclination angle of the train body relative to the plane of the track during the running of the train.

[0055] The indication information of the preset interval indicator on the track is identified from each train running image frame, thereby obtaining the actual distance value between the train and the maintenance station.

[0056] According to the driving posture information, it is judged whether the train is in a driving unstable state during driving; if the train is in the driving unstable state and the actual distance value is greater than the preset distance threshold, the train is instructed to drive at a predetermined speed, so that the train can be driven stably and side-tilting accidents can be avoided; if the train is in a driving stable state and the actual distance value is greater than the preset distance threshold, the train is instructed to enter an acceleration driving mode at a predetermined acceleration, so that the train can be quickly driven to the maintenance station while being driven stably; and if the actual distance value is less than or equal to the preset distance threshold, the train is instructed to enter a deceleration driving mode, so that the train can be accurately parked after entering the platform area of the maintenance station.

[0057] The beneficial effects of the above technical solution are that when the train enters the monitoring area of the maintenance station, the camera pre-installed at the maintenance station is instructed to dynamically photograph the train, and the train driving dynamic image obtained by photographing is divided into a plurality of train driving image frames. The driving posture information of the train and the actual distance value between the train and the maintenance station are obtained by identifying the train contour and the indication information of the preset indicator of the track for each train driving image frame, so that the driving state of the train itself and the relative position relationship between the train and the maintenance station can be quantitatively calibrated.

[0058] Preferably, in the step S3, a brake control instruction is sent to the driving control platform of the train according to the actual distance value, so as to instruct the train to be parked at the platform area corresponding to the maintenance station, which specifically includes:

[0059] If the actual distance value is less than or equal to the preset distance threshold, the driving distance value between the train and the platform area to be parked on the track is determined according to the actual distance value; the deceleration acceleration value corresponding to the deceleration driving mode is determined according to the driving distance value and the real-time driving speed of the train, so as to send a deceleration brake control instruction to the driving control platform of the train, thereby instructing the train to be parked at the platform area corresponding to the maintenance station.

[0060] The beneficial effects of the above technical solution are that by the above-mentioned manner, the train is accurately controlled to decelerate and brake according to the driving distance value between the train and the platform area to be parked on the track and the real-time driving speed of the train, so that the train can be accurately parked at the corresponding platform area of the maintenance station, and the parking reliability of the train is improved.

[0061] Referring to Figure 2 The structure schematic diagram of the train driving detection system based on multi-feature fusion provided by the embodiment of the present application is shown. The train driving detection system based on multi-feature fusion includes:

[0062] a radar scanning control module, configured to instruct a radar device with a preset distance value from the inspection station to perform radar scanning detection on the train, so as to obtain train running state information;

[0063] a train running analysis module, configured to determine whether the train enters a monitoring area range of the inspection station according to the train running state information, and estimate a running duration of the train in the monitoring area range according to the train running state information;

[0064] an image shooting control and analysis module, configured to instruct a camera located at the inspection station to shoot the train in a time period corresponding to the running duration, so as to obtain train running images, and analyze and process the train running images, so as to obtain train running posture information and an actual distance value between the train and the inspection station;

[0065] a train running adjustment module, configured to adjust a running state of the train according to the running posture information and the actual distance value;

[0066] a train parking control module, configured to send a brake control instruction to a driving control platform of the train according to the actual distance value, so as to instruct the train to park at a platform area corresponding to the inspection station.

[0067] The train running detection system based on multi-feature fusion has the following beneficial effects: the radar device is used to perform radar scanning detection on the train, so as to obtain the train running state information, and determine whether the train enters the monitoring area range and estimate the running duration of the train in the monitoring area range; the train running images are shot and analyzed during the running duration, so as to obtain the train running posture information and the actual distance value between the train and the inspection station, and adjust the running state of the train; the brake control instruction is sent to the driving control platform of the train according to the actual distance value, so as to instruct the train to park at the platform area corresponding to the inspection station; in the case that the train is far away from the inspection station or is close to the inspection station, the radar device and the camera are respectively used to perform multi-feature monitoring on the train, the train is coarsely monitored by the radar scanning method when the train is far away from the inspection station, and the train is accurately monitored by the image shooting method when the train is close to the inspection station, so that appropriate brake control instructions can be sent to the train in time, and the train can be parked at the platform area corresponding to the inspection station, thereby improving the control accuracy and reliability of the train parked at the inspection station.

[0068] Preferably, the radar scanning control module instructs the radar device with the preset distance value from the inspection station to perform radar scanning detection on the train, so as to obtain the train running state information, specifically including:

[0069] The radar device periodically scans outward to emit radar signals to detect the train near the radar device; the distance between the train and the radar device and the running speed of the train are obtained according to the radar signals emitted outward by the radar device and the radar signals reflected back by the train;

[0070] The train running analysis module judges whether the train enters the monitoring area range of the maintenance station according to the train running state information; and estimates the running duration of the train in the monitoring area range according to the train running state information, specifically including:

[0071] The train is judged to enter the monitoring area range of the maintenance station according to the distance and the radius of the monitoring area range of the maintenance station; if yes, the running duration of the train in the monitoring area range is estimated according to the running speed.

[0072] The beneficial effects of the above technical solutions are that the radar device is erected at a position apart from the maintenance station by a preset distance value, so that the radar device serves as the first monitoring checkpoint for the train running close to the maintenance station. The radar device periodically scans outward to emit radar signals, when the radar signals reach the surface of the train and are reflected back, the radar device performs difference processing on the intensity or time of the radar signals emitted outward and the radar signals reflected back, that is, the distance between the train and the radar device and the running speed of the train are obtained, so that the running state of the train is quantitatively calibrated. In addition, the distance and the radius of the monitoring area range of the maintenance station are compared, if the distance is less than or equal to the radius of the monitoring area range, it is determined that the train has entered the monitoring area range of the maintenance station, at this time, the subsequent dynamic shooting of the train by the camera is facilitated, and the switching from the rough monitoring by the radar device to the accurate monitoring by the camera is realized. According to the radius of the monitoring area range and the running speed, the running duration of the train in the monitoring area range is determined, so as to facilitate the control of the dynamic shooting duration of the train by the camera.

[0073] Preferably, the image shooting control and analysis module instructs the camera located at the maintenance station to shoot the train in a time period corresponding to the running duration, to obtain train running images; the train running images are analyzed and processed to obtain the running posture information of the train and the actual distance value between the train and the maintenance station, specifically including:

[0074] The camera located at the maintenance station is instructed to dynamically shoot the train in a time period corresponding to the running duration, to obtain train running dynamic images; the train running dynamic images are frame-processed to obtain a plurality of train running image frames;

[0075] The shape contour information of the train is extracted from each train running image frame, and the running posture information of the train is obtained according to the shape contour information corresponding to the train running image frame; wherein the running posture information includes the inclination angle of the train body relative to the plane of the track during the train running process;

[0076] The indication information of the preset distance indicator on the track is recognized from each train running image frame, so as to obtain the actual distance value between the train and the maintenance station;

[0077] The train running adjustment module adjusts the running state of the train according to the running posture information and the actual distance value, and specifically includes:

[0078] According to the running posture information, it is judged whether the train is in a running unstable state during the running process; if the train is in a running unstable state and the actual distance value is greater than a preset distance threshold, the train is instructed to run at a predetermined speed; if the train is in a running stable state and the actual distance value is greater than a preset distance threshold, the train is instructed to enter an acceleration running mode with a predetermined acceleration; if the actual distance value is less than or equal to a preset distance threshold, the train is instructed to enter a deceleration running mode.

[0079] The beneficial effects of the above technical solution are: when the train enters the monitoring area range of the maintenance station, the camera pre-installed at the maintenance station is instructed to dynamically shoot the train, and the train running dynamic image obtained by shooting is divided into a plurality of train running image frames. The train shape contour and the indication information of the preset indicator on the track are identified for each train running image frame, so as to obtain the running posture information of the train and the actual distance value between the train and the maintenance station, so as to quantitatively calibrate the running state of the train itself and the relative position relationship with the maintenance station.

[0080] Preferably, the train parking control module sends a brake control instruction to the driving control platform of the train according to the actual distance value, so as to instruct the train to park in the corresponding platform area of the maintenance station, and specifically includes:

[0081] If the actual distance value is less than or equal to a preset distance threshold, the running distance value between the train and the platform area to be parked on the track is determined according to the actual distance value; the deceleration acceleration value corresponding to the deceleration running mode is determined according to the running distance value and the real-time running speed of the train, so as to send a deceleration brake control instruction to the driving control platform of the train, so as to instruct the train to park in the corresponding platform area of the maintenance station.

[0082] The beneficial effects of the above technical solutions are: through the above manner, the train is accurately controlled by deceleration braking based on the running distance value of the train on the track between the current train and the platform area to be stopped and the real-time running speed of the train, so as to ensure that the train can be aligned and stopped at the corresponding platform area of the maintenance station, and improve the stopping reliability of the train.

[0083] From the above embodiment, it can be seen that the train running detection method and system based on multi-feature fusion utilizes the radar equipment to perform radar scanning detection on the train to obtain the running state information of the train, so as to determine whether the train enters the monitoring area range and estimate the running duration of the train in the monitoring area range; the train running image is shot and analyzed during the running duration to obtain the running posture information of the train and the actual distance value between the train and the maintenance station, so as to adjust the running state of the train; according to the actual distance value, a brake control instruction is sent to the driving control platform of the train, so as to instruct the train to stop at the corresponding platform area of the maintenance station; in the case that the train is far away from the maintenance station and close to the maintenance station, the radar equipment and the camera are respectively utilized to perform multi-feature monitoring on the train, the train is roughly monitored by the radar scanning manner when far away from the maintenance station, and the train is accurately monitored by the image shooting manner when close to the maintenance station, so as to timely send appropriate brake control instructions to the train, ensure that the train can be aligned and stopped at the corresponding platform area of the maintenance station, and thus improve the control accuracy and reliability of the train stopping at the maintenance station.

[0084] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A train movement detection method based on multi-feature fusion, characterized in that, It includes the following steps: Step S1: Instruct the radar equipment at a preset distance from the maintenance station to perform radar scanning detection on the train to obtain the train's driving status information; based on the train's driving status information, determine whether the train has entered the monitoring area of ​​the maintenance station; and based on the train's driving status information, estimate the duration of the train's travel within the monitoring area. Step S2: Instruct the camera located at the maintenance station to take pictures of the train during the time period corresponding to the duration of the journey, and obtain train travel images; analyze and process the train travel images to obtain the train's travel posture information and the actual distance value between the train and the maintenance station; The train's driving status is adjusted based on the driving posture information and the actual distance value; Step S3: Based on the actual distance value, send a braking control command to the train's driving control platform to instruct the train to stop at the platform area corresponding to the maintenance station. In step S2, the camera located at the maintenance station is instructed to capture images of the train during the time period corresponding to the duration of the journey, thereby obtaining train travel images. The train travel images are then analyzed and processed to obtain the train's travel posture information and the actual distance between the train and the maintenance station. Adjusting the train's travel status based on the travel posture information and the actual distance specifically includes: The camera located at the maintenance station is instructed to dynamically capture images of the train during the time period corresponding to the duration of the journey, thereby obtaining dynamic images of the train's journey; the dynamic images of the train's journey are then processed into frames to obtain several train image frames. The train's outline information is extracted from each train travel image frame. Based on the outline information corresponding to the train travel image frame, the train's travel posture information is obtained. The travel posture information includes the tilt angle of the train body relative to the plane where the track is located during the train's travel. The indication information of the preset spacing indicator on the track is identified from each train travel image frame to obtain the actual distance value between the train and the maintenance station. Based on the driving posture information, it is determined whether the train is in an unstable driving state during operation; if the train is in an unstable driving state and the actual distance value is greater than a preset distance threshold, the train is instructed to drive at a constant speed at a predetermined speed; if the train is in a stable driving state and the actual distance value is greater than the preset distance threshold, the train is instructed to enter an acceleration driving mode with a predetermined acceleration; if the actual distance value is less than or equal to the preset distance threshold, the train is instructed to enter a deceleration driving mode. Specifically, in step S3, sending a braking control command to the train's driving control platform based on the actual distance value to instruct the train to stop at the platform area corresponding to the maintenance station includes: If the actual distance value is less than or equal to a preset distance threshold, the distance traveled by the train on the track between the current station and the platform area where it needs to stop is determined based on the actual distance value; based on the distance traveled and the real-time speed of the train, the deceleration acceleration value corresponding to the deceleration mode is determined, and a deceleration and braking control command is sent to the train's driving control platform to instruct the train to stop at the platform area corresponding to the maintenance station.

2. The train movement detection method based on multi-feature fusion as described in claim 1, characterized in that: In step S1, the radar equipment, which is positioned at a preset distance from the maintenance station, performs a radar scan to detect the train, thereby obtaining the train's driving status information. Based on the train's driving status information, it is determined whether the train has entered the monitoring area of ​​the maintenance station. Furthermore, based on the train's driving status information, the estimated duration of the train's travel within the monitoring area specifically includes: The radar equipment, which is positioned at a preset distance from the maintenance station, periodically scans and transmits radar signals to detect trains near the equipment. Based on the radar signals emitted by the equipment and the radar signals reflected back from the train, the distance between the train and the radar equipment, as well as the train's speed, are obtained. Based on the distance between the two sides and the radius of the monitoring area of ​​the maintenance station, it is determined whether the train has entered the monitoring area of ​​the maintenance station; if so, the duration of the train's travel within the monitoring area is estimated based on the travel speed; if not, the radar equipment is instructed to periodically scan and transmit radar signals outward.

3. A train operation detection system based on multi-feature fusion, characterized in that, It includes: The radar scanning control module is used to instruct radar equipment at a preset distance from the maintenance station to perform radar scanning detection on the train in order to obtain the train's running status information. The train travel analysis module is used to determine whether the train has entered the monitoring area of ​​the maintenance station based on the train travel status information; and to estimate the duration of the train's travel within the monitoring area based on the train travel status information. The image capture control and analysis module is used to instruct the camera located at the maintenance station to capture images of the train during the time period corresponding to the continuous travel, thereby obtaining train travel images; and to analyze and process the train travel images to obtain the train's travel posture information and the actual distance value between the train and the maintenance station. The train driving adjustment module is used to adjust the train's driving state based on the driving posture information and the actual distance value. The train stopping control module is used to send braking control commands to the train's driving control platform based on the actual distance value, thereby instructing the train to stop at the platform area corresponding to the maintenance station. The image capture control and analysis module instructs the camera located at the maintenance station to capture images of the train during the time period corresponding to the duration of the journey, obtaining train driving images. The analysis and processing of the train driving images to obtain the train's driving posture information and the actual distance value between the train and the maintenance station specifically includes: The camera located at the maintenance station is instructed to dynamically capture images of the train during the time period corresponding to the duration of the journey, thereby obtaining dynamic images of the train's journey; the dynamic images of the train's journey are then processed into frames to obtain several train image frames. The train's outline information is extracted from each train travel image frame. Based on the outline information corresponding to the train travel image frame, the train's travel posture information is obtained. The travel posture information includes the tilt angle of the train body relative to the plane where the track is located during the train's travel. The indication information of the preset spacing indicator on the track is identified from each train travel image frame to obtain the actual distance value between the train and the maintenance station. The train driving adjustment module adjusts the train's driving state based on the driving posture information and the actual distance value, specifically including: Based on the driving posture information, it is determined whether the train is in an unstable driving state during operation; if the train is in an unstable driving state and the actual distance value is greater than a preset distance threshold, the train is instructed to drive at a constant speed at a predetermined speed; if the train is in a stable driving state and the actual distance value is greater than the preset distance threshold, the train is instructed to enter an acceleration driving mode with a predetermined acceleration; if the actual distance value is less than or equal to the preset distance threshold, the train is instructed to enter a deceleration driving mode. Specifically, the train stopping control module sends a braking control command to the train's driving control platform based on the actual distance value, thereby instructing the train to stop at the platform area corresponding to the maintenance station. If the actual distance value is less than or equal to a preset distance threshold, the distance traveled by the train on the track between the current station and the platform area where it needs to stop is determined based on the actual distance value; based on the distance traveled and the real-time speed of the train, the deceleration acceleration value corresponding to the deceleration mode is determined, and a deceleration and braking control command is sent to the train's driving control platform to instruct the train to stop at the platform area corresponding to the maintenance station.

4. The train operation detection system based on multi-feature fusion as described in claim 3, characterized in that: The radar scanning control module instructs radar equipment at a preset distance from the maintenance station to perform radar scanning detection on the train, thereby obtaining the train's running status information, specifically including: The radar equipment, which is positioned at a preset distance from the maintenance station, periodically scans and transmits radar signals to detect trains near the equipment. Based on the radar signals emitted by the equipment and the radar signals reflected back from the train, the distance between the train and the radar equipment, as well as the train's speed, are obtained. The train travel analysis module determines whether the train has entered the monitoring area of ​​the maintenance station based on the train travel status information; and estimates the duration of the train's travel within the monitoring area based on the train travel status information, specifically including: Based on the distance between the two sides and the radius of the monitoring area of ​​the maintenance station, it is determined whether the train has entered the monitoring area of ​​the maintenance station; if so, the duration of the train's travel within the monitoring area is estimated based on the travel speed.

Citation Information

Patent Citations

  • Subway train inbound speed control method and positioning monitoring system

    CN110562303A

  • Train and automatic stopping control method and device thereof

    CN112441078A

  • Intelligent monitoring equipment for road transportation

    CN113706885A