Visual positioning apparatus and method for train in tunnel, device, and medium

Abstract

A visual positioning apparatus and method for a train in a tunnel, a device, and a medium. The apparatus comprises a vehicle-mounted controller (50), a vehicle-mounted visual odometer (400), a train surface identification code (40), a trackside visual beacon (100), a trackside visual axle counter (300), and feature points (200) arranged inside the tunnel according to a set distance. The vehicle-mounted visual odometer (400) acquires the trackside visual beacon (100) and the feature points (200), and the trackside visual axle counter (300) acquires the train surface identification code (40). The vehicle-mounted controller (50) obtains a current speed and a relative displacement of a train from the vehicle-mounted visual odometer (400), calculates a current location of the train in the route on the basis of the trackside visual beacon (100) and performs calibration and positioning, and simultaneously obtains a current location of the train in the route from the trackside visual axle counter (300) and performs calibration and positioning, so as to implement continuous speed measurement and positioning of the train. According to the visual positioning apparatus for a train in a tunnel, the redundancy and availability of a positioning system can be effectively improved.

Description

A visual positioning device, method, equipment and medium for trains in tunnels Technical Field

[0001] This invention relates to rail transit signaling systems, and more particularly to a visual positioning device, method, equipment, and medium for trains in tunnels. Background Technology

[0002] Currently, there are two commonly used speed measurement and positioning methods in the rail transit signaling industry. One method uses photoelectric / Hall speed sensors combined with ground beacons, and the other uses satellite virtual transponders combined with speed sensors. The former's positioning accuracy depends on the density of beacon deployment, resulting in higher system costs. The latter's satellite signals are easily affected by tunnels, buildings, etc., and have high requirements for terrain, resulting in lower universality.

[0003] A search revealed Chinese Patent Publication No. CN111114593A, which discloses a metro train autonomous positioning device, system, and method based on multi-source information fusion. Specifically, it discloses an inertial navigation system (INS) for autonomous navigation and positioning of the metro train, calculating INS positioning information; a vision device for monitoring image beacons placed at fixed locations along the metro track, calculating visual positioning information; and a data processing device for receiving information reported by the INS and vision devices, fusing the information, and obtaining fused positioning information. Although this patent uses a vision device for positioning, the vision device is used primarily to measure the distance between the train's front and landmarks, and the information collected by the vision device is only one source of fused information, with its weight not clearly defined. Furthermore, it does not address operations when the onboard autonomous positioning system fails.

[0004] Meanwhile, Chinese Patent Publication No. CN116946214A discloses a train positioning and speed measurement system, method, device, and medium based on visual recognition. Specifically, it includes: mileage markers with QR codes, placed along the track at predetermined intervals; a camera for capturing images of the QR codes on the mileage markers; and a safety host connected to the camera for parsing the QR code images captured by the camera, identifying the mileage markers ahead, and using the positioning pattern on the QR code, calculating the distance from the vehicle to the next mileage marker using a distance measurement algorithm. Although this patent uses a visual device for positioning, the visual device is used for distance measurement and requires the presence of kilometer markers within the field of view, making it a discontinuous distance measurement method.

[0005] Furthermore, Chinese Patent Publication No. CN116136404A discloses a low-cost, all-terrain train control onboard equipment speed measurement and positioning method and device. Specifically, if the satellite signal status is invalid, a multi-sensor fusion method is used to establish an inertial sensor error model. The difference between the inertial sensor and other sensor measurements is used to estimate the inertial sensor error and perform feedback correction to obtain the final train status information. Although this patent uses a vision device for positioning, the specific positioning process of the vision device is not described, and the information collected by the vision device is only one of the fusion information sources, and its weight is not clearly defined.

[0006] It is evident that the speed measurement and positioning methods commonly used in existing technologies cannot balance cost and universality in tunnel environments. Furthermore, the publicly available visual positioning schemes cannot be used alone (due to the problem of discontinuous positioning), requiring the assistance of other types of speed measurement and positioning equipment, and lacking contingency plans after failure. Summary of the Invention

[0007] The purpose of this invention is to overcome the defects of the prior art by providing a visual positioning device, method, equipment and medium for trains in tunnels.

[0008] The objective of this invention can be achieved through the following technical solutions:

[0009] According to a first aspect of the present invention, a train visual positioning device in a tunnel is provided, the device comprising an on-board controller, an on-board visual odometer, a train surface marking code, a trackside visual beacon, a trackside visual meter axis, and feature points arranged at predetermined distances inside the tunnel.

[0010] The on-board visual odometer collects trackside visual beacons and feature points, while the trackside visual odometer axis collects the train surface identification code. The on-board controller obtains the current speed and relative displacement from the on-board visual odometer, calculates the current position on the track based on the trackside visual beacons, and performs calibration positioning. At the same time, it obtains the current position on the track from the trackside visual odometer axis and performs calibration positioning, thereby realizing continuous train speed measurement and positioning.

[0011] As a preferred technical solution, the on-board controller includes a first computing unit and a first storage device. The first computing unit obtains an electronic route map containing trackside visual beacon information from the first storage device. The first storage device is used to store the electronic route map and train parameters.

[0012] As a preferred technical solution, the vehicle-mounted visual odometer includes a first camera and a first image processing unit. The first camera acquires trackside images and trackside visual beacons and sends them to the first image processing unit. The first image processing unit calculates the relative running distance of the train based on the acquired trackside images and performs positioning calibration of the train based on the trackside visual beacons.

[0013] As a preferred technical solution, the train surface identification code includes a first QR code and a first barcode, both of which correspond to the train's unique identification code.

[0014] As a preferred technical solution, the trackside visual beacon includes up and down direction indicators, a second QR code, and a second barcode, wherein the second QR code and the second barcode each correspond to a unique identifier of the visual beacon in the electronic route map.

[0015] As a preferred technical solution, the trackside visual measuring axis includes a second camera, a second image processing unit, and a communication device. The second camera collects the train surface identification code and sends it to the second image processing unit. The second image processing unit identifies the train, calculates the train's position, and sends a position message to the train through the communication device, while simultaneously sending a train position message to the control center.

[0016] As a preferred technical solution, the feature points arranged at set intervals inside the tunnel include reflective letter and number patterns, and the patterns are not repeated along the entire line.

[0017] As a preferred technical solution, the distance between any two feature points is determined by the field of view of the camera of the on-board visual odometer, ensuring that at least one feature point is included in the image collected in each cycle when the train is at a set speed.

[0018] According to a second aspect of the present invention, a positioning method using the aforementioned train visual positioning device in a tunnel is provided, the method comprising the following steps:

[0019] Step S1: The on-board controller pre-stores an electronic route map containing trackside visual beacon information, while the trackside visual axle pre-stores the identification codes of all possible trains running on the line.

[0020] Step S2: The train initializes its positioning based on the trackside visual beacons.

[0021] Step S3: The on-board visual odometer periodically acquires trackside images and calculates the relative running distance, and the on-board controller accumulates the running distance.

[0022] Step S4: When a trackside visual beacon is detected, the onboard visual odometry performs positioning calibration.

[0023] Step S5: When the trackside visual axle detects the passing of a train, it identifies the train and continuously calculates the train's position, then sends a position message to the train and simultaneously sends a train position message to the control center.

[0024] Step S6: When the train experiences a communication failure, a failure of the onboard controller, or a failure of the onboard visual odometer, the trackside visual odometer detects the train's position and reports the faulty train's position to the control center.

[0025] As a preferred technical solution, the trackside visual beacon information in step S1 includes a unique identifier for each visual beacon and its location on the track.

[0026] As a preferred technical solution, the specific process of positioning initialization in step S2 is as follows:

[0027] The onboard visual odometer detects a trackside visual beacon, identifies the beacon's unique identifier based on the second QR code and second barcode on the beacon, obtains the basic position based on the unique identifier's location on the route map, calculates the offset of the basic position based on the distance difference between the focal center of the first camera and the beacon's center, and obtains the running direction based on the up and down direction markings on the beacon and the displacement direction relative to the beacon in the previous and next cycles, thereby calculating the train's initial positioning.

[0028] As a preferred technical solution, the calculation of the relative running distance in step S3 is specifically as follows:

[0029] Step S301: Train the static image data of feature points inside the tunnel to obtain the initial model;

[0030] Step S302: Combine the dynamic image data captured at low, medium, and high vehicle speeds to continue training the initialization model and obtain the basic model;

[0031] Step S304: Add noise data to the dynamic image data, train the base model, and obtain the application model;

[0032] Step S305: In field application, the application model is used to identify feature points in the acquired images, and the relative distance of the same feature point in two consecutive cycles is calculated to obtain the relative running distance of the train.

[0033] As a preferred technical solution, the positioning calibration in step S4 specifically includes:

[0034] Step S401: The on-board controller calculates the train's position one based on the position of the visual beacon, which is the same as the positioning initialization process in step S2.

[0035] Step S402: The on-board controller accumulates the relative displacement fed back by the visual odometer in each cycle to obtain the second train position.

[0036] Step S403: Correct the train's position 2 based on the train's position 1 to obtain the train's position 3, that is, obtain a more accurate train position.

[0037] As a preferred technical solution, the process of identifying the train and continuously calculating its position in step S5 specifically involves:

[0038] Step S501: Train a model for recognizing trains using trackside visual axle counting.

[0039] Step S502: Pre-define and mark absolute position points within the field of view of the second camera of the trackside visual measuring axis, as reference points for locating the train position in the subsequent process;

[0040] Step S503: When a train is detected passing by (the train appears in the frame from not appearing in the frame to appearing in the frame in two consecutive cycles), the train's identification code is identified (and cross-verified) through the QR code and barcode on the train body to determine the train's identity, and the absolute position of the train is calculated through the pre-calibrated reference points.

[0041] Step S504: Continuously update the position of the train (update in the same way as calculating the relative running distance in step S3) until the train leaves the camera's field of view.

[0042] According to a third aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described thereon.

[0043] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] 1) The train visual positioning technology of this invention only uses onboard visual sensors and trackside visual tags and sensors. Compared with traditional beacons, odometers, and axle counters, the equipment is simpler and has lower deployment, operation, and maintenance costs.

[0046] 2) This invention fully considers the failure situation of on-board equipment and provides a backup positioning mode through trackside visual axis measurement, thereby enhancing the redundancy and availability of the positioning system;

[0047] 3) This invention uses a visual sensor to achieve the positioning process. Compared with the slippage error that is easily introduced by traditional speed measurement and positioning devices, and the problem that satellite positioning technology is prone to signal interruption inside tunnels, this invention can achieve better positioning accuracy and improve positioning availability. Attached Figure Description

[0048] Figure 1 is a flowchart of a train visual positioning method in a tunnel provided by an embodiment of the present invention;

[0049] Figure 2 is a structural diagram of a train visual positioning device in a tunnel provided by an embodiment of the present invention;

[0050] Figure 3 is a schematic diagram of the deployment of a train visual positioning device in a tunnel according to an embodiment of the present invention;

[0051] Figure 4 is an initialization positioning flowchart of a train visual positioning method in a tunnel provided by an embodiment of the present invention;

[0052] Figure 5 is a flowchart of the trackside visual axle counting detection train positioning method in a tunnel provided by an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0054] Figure 1 is a flowchart of a train visual positioning method in a tunnel according to an embodiment of the present invention. The specific process is as follows:

[0055] Step S1: The on-board controller pre-stores an electronic route map containing trackside visual beacon information, and the trackside visual beacon pre-stores the identification codes of all possible trains running on the line. The trackside visual beacon information includes the unique identification code of each visual beacon and its location on the line.

[0056] Step S2: The train initializes its positioning based on the trackside visual beacons;

[0057] The initial positioning process includes the following steps: the onboard visual odometer detects a trackside visual beacon; the unique identifier of the beacon is identified based on the QR code and barcode on the beacon; the basic position is obtained based on the location of the unique identifier on the route map; the relative offset is obtained based on the distance difference between the focal center of the camera and the center of the beacon; and the running direction is obtained based on the up and down direction markings on the beacon and the displacement direction relative to the beacon in the previous and next cycles. Thus, the initial positioning of the train is calculated.

[0058] Step S3: The train's visual odometer periodically acquires trackside images and calculates the relative running distance, while the onboard controller accumulates the running distance;

[0059] The process of calculating the relative running distance includes: first, training the static image data of feature points in the tunnel to obtain an initial model; then, combining the dynamic image data taken at low, medium, and high vehicle speeds to further train the initial model to obtain a basic model; finally, adding noise data to the dynamic image data to train the basic model to obtain an application model; in field application, the application model is used to identify feature points in the acquired images, and the relative distance of the same feature point in two consecutive cycles is calculated to obtain the relative running distance of the train.

[0060] Step S4: When a trackside visual beacon is detected, the train's visual odometer performs positioning calibration;

[0061] The positioning calibration process includes: first, the on-board controller calculates the train's position 1 based on the position of the visual beacon, similar to the positioning initialization process in step S2; second, the on-board controller also accumulates the relative displacement fed back by the visual odometer every cycle to obtain the train's position 2; finally, the train's position 2 is corrected based on the train's position 1 to obtain the train's position 3, that is, to obtain a more accurate train position.

[0062] Step S5: When the trackside visual axle detects the passing of a train, it identifies the train, continuously calculates the train's position, and sends a position message to the train, while simultaneously sending a train position message to the control center.

[0063] The process of detecting, identifying, and continuously updating the train's position includes: training a model for identifying trains beforehand for the trackside axle counter, with the training process being the same as in step S3; pre-marking and calibrating absolute position points within the field of view of the trackside visual axle counter camera as reference points for locating the train's position in subsequent processes; in formal engineering applications, when a train is detected passing by (from not appearing in the frame to appearing in the frame in two consecutive cycles), the train's identification code is identified (and cross-verified) through the QR code and barcode on the train body to determine the train's identity, and the absolute position of the train is calculated using the pre-marked reference points; the train's position is continuously updated (the update method is the same as calculating the relative running distance in step 3) until the train leaves the camera's field of view.

[0064] Step S6: When the train experiences a communication failure, or the train's onboard controller fails, or the train's visual odometer device fails, the trackside visual odometer axle detects the train's position and reports the faulty train's position to the control center.

[0065] Figure 4 is an initialization positioning flowchart of a train visual positioning method in a tunnel according to an embodiment of the present invention. During train movement, the visual odometer detects a trackside visual beacon; by recognizing a QR code and barcode, the unique identifier of the beacon is identified; it is determined whether the beacon identifier corresponding to the QR code and the beacon identifier corresponding to the barcode are consistent. If they are inconsistent, the visual beacon information is discarded; if they are consistent, further processing is performed; the legality of the visual beacon is determined by combining the stored electronic route map and the beacon identifier naming rules. If it is invalid, the visual beacon is discarded; if it is valid, further processing is performed; the position of the visual beacon in the electronic route map is located; the offset between the camera center and the visual beacon center is calculated; the train's running direction is determined based on the up / down markers of the visual beacon and the movement direction of the previous and next two cycles; the distance traveled by the train from capturing the visual beacon to completing the calculation is calculated; and the final initial positioning of the train is obtained.

[0066] Figure 5 is a flowchart of the trackside visual axle counting detection train positioning method in a tunnel according to an embodiment of the present invention. The trackside visual axle counting detects the passing of a train, i.e., the train moves from not appearing in the frame to appearing in the frame; it identifies the QR code and barcode on the train surface and identifies the unique train identification code corresponding to each; it determines whether the beacon identification code corresponding to the QR code and the beacon identification code corresponding to the barcode are consistent. If they are inconsistent, an alarm is sent to the control center and an image is uploaded; if they are consistent, further processing is performed; it combines the stored identification codes of all trains that may appear on the line and the train identification code naming rules to determine the legality of the train identification code. If it is invalid, an alarm is sent to the control center and an image is uploaded; if it is valid, further processing is performed; it combines the position of a pre-calibrated reference point to calculate the offset between the reference point and the position of the train identification code, thereby calculating the absolute position of the train; and it reports the train's position to the control center.

[0067] The above is an introduction to the method embodiments. The following describes the solution of the present invention further through device embodiments.

[0068] Figure 2 is a structural diagram of a train visual positioning device in a tunnel according to an embodiment of the present invention. The device includes an on-board controller 50, an on-board visual odometer 400, a train surface marking code 40, a trackside visual beacon 100, a trackside visual measuring axis 300, and feature points 200 arranged at set intervals inside the tunnel. The on-board controller 50 includes a first processing unit 501 and a first storage device 502. The on-board visual odometer 400 includes a first camera 30 and a first image processing unit 60. The trackside visual measuring axis 300 includes a second camera 70, a second image processing unit 80, and a communication device. The trackside visual beacon 100 includes up and down markers 20 and QR codes and barcodes 10.

[0069] Figure 3 is a deployment diagram of a train visual positioning device in a tunnel according to an embodiment of the present invention. The onboard controller 50 and the onboard visual odometer 400 are installed on the train. Train surface identification codes 40 are sprayed onto the train surface. Trackside visual beacons 100, trackside visual meter axes 300, and feature points 200 arranged at predetermined intervals inside the tunnel are arranged on both sides of the trackside inside the tunnel. The first processing unit in the onboard controller calculates the train's positioning information based on the relative displacement obtained from the train visual odometer, the identification code information of the visual beacons, the position calibration information obtained from the control center, and the electronic map, visual beacon identification codes, and position information obtained from the first storage device. The image processing unit and communication device of the trackside visual meter axis, based on the train identification code information captured by the camera and combined with all possible train identification codes on the pre-stored line, pre-delineate and mark the absolute position points, calculate the position of the passing train, and send it to the control center 500.

[0070] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0071] This invention also provides an electronic device including a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0072] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0073] The processing unit executes the various methods and processes described above, such as methods S1 to S6. For example, in some embodiments, methods S1 to S6 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S6 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S6 by any other suitable means (e.g., by means of firmware).

[0074] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0075] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0076] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0077] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A train visual positioning device for use in tunnels, characterized in that, The device includes an onboard controller (50), an onboard visual odometer (400), a train surface marking code (40), a trackside visual beacon (100), a trackside visual meter axis (300), and feature points arranged at set intervals inside the tunnel. The on-board visual odometer (400) collects trackside visual beacons (100) and feature points, and the trackside visual meter axis (300) collects the train surface identification code (40). The on-board controller (50) obtains the current speed and relative displacement from the on-board visual odometer (400), calculates the current position on the line according to the trackside visual beacons (100) and performs calibration positioning, and at the same time obtains the current position on the line from the trackside visual meter axis (300) and performs calibration positioning, thereby realizing continuous speed measurement and positioning of the train.

2. The train visual positioning device in a tunnel according to claim 1, characterized in that, The on-board controller (50) includes a first computing unit (501) and a first storage device (502). The first computing unit (501) obtains an electronic route map containing trackside visual beacon information from the first storage device (502). The first storage device (502) is used to store the electronic route map and train parameters.

3. The train visual positioning device in a tunnel according to claim 1, characterized in that, The vehicle-mounted visual odometer (400) includes a first camera (30) and a first image processing unit (60). The first camera (30) acquires trackside images and trackside visual beacons (100) and sends them to the first image processing unit (60). The first image processing unit (60) calculates the relative running distance of the train based on the acquired trackside images and performs positioning calibration of the train based on the trackside visual beacons (100).

4. The train visual positioning device in a tunnel according to claim 1, characterized in that, The train surface identification code (40) includes a first QR code and a first barcode, each corresponding to a unique train identification code.

5. The train visual positioning device in a tunnel according to claim 1, characterized in that, The trackside visual beacon (100) includes up and down direction indicators (20), a second QR code, and a second barcode (10). The second QR code and the second barcode (10) each correspond to the unique identification code of the visual beacon in the electronic route map.

6. The train visual positioning device in a tunnel according to claim 1, characterized in that, The trackside visual measuring axis (300) includes a second camera (70), a second image processing unit (80), and a communication device. The second camera (70) collects the train surface identification code (40) and sends it to the second image processing unit (80). The second image processing unit (80) identifies the train, calculates the train position, and sends a position message to the train through the communication device, while also sending a train position message to the control center.

7. The train visual positioning device in a tunnel according to claim 1, characterized in that, The feature points arranged at set intervals inside the tunnel include reflective letter and number patterns, and none of them are repeated along the entire line.

8. The train visual positioning device in a tunnel according to claim 7, characterized in that, The distance between any two of the feature points is determined by the field of view of the camera of the onboard visual odometer (400), ensuring that at least one feature point is included in the image acquired in each cycle when the train is at a set speed.

9. A positioning method using the train visual positioning device in a tunnel as described in claim 1, characterized in that, The method includes the following steps: Step S1, the on-board controller (50) pre-stores an electronic route map containing trackside visual beacon information, while the trackside visual axle counter (300) pre-stores the identification codes of all possible trains running on the line; Step S2: The train initializes its positioning based on the trackside visual beacon (100); Step S3: The on-board visual odometer (400) periodically acquires trackside images and calculates the relative running distance, and the on-board controller (50) accumulates the running distance; Step S4: When a trackside visual beacon (100) is detected, the on-board visual odometer (400) performs positioning calibration. Step S5: When the trackside visual measuring axis (300) detects the passing of a train, it identifies the train and continuously calculates the train's position, then sends a position message to the train and simultaneously sends a train position message to the control center. Step S6: When the train experiences a communication failure, or the onboard controller (50) fails, or the onboard visual odometer (400) fails, the trackside visual odometer axis (300) detects the train's position and reports the faulty train's position to the control center.

10. The positioning method according to claim 9, characterized in that, The trackside visual beacon information in step S1 includes a unique identifier for each visual beacon and its location on the track.

11. The positioning method according to claim 9, characterized in that, The specific process of positioning initialization in step S2 is as follows: The onboard visual odometer (400) detects the trackside visual beacon (100), identifies the beacon's unique identifier based on the second QR code and second barcode on the trackside visual beacon (100), obtains the basic position based on the unique identifier's position on the route map, obtains the offset of the basic position based on the distance difference between the focal center of the first camera (30) and the beacon's center, and obtains the running direction based on the up and down direction markings on the beacon and the displacement direction relative to the beacon in the previous and next cycles, thereby calculating the train's initial positioning.

12. The positioning method according to claim 9, characterized in that, The calculation of the relative running distance in step S3 is specifically as follows: Step S301: Train the static image data of feature points inside the tunnel to obtain the initial model; Step S302: Combine the dynamic image data captured at low, medium, and high vehicle speeds to continue training the initialization model and obtain the basic model; Step S304: Add noise data to the dynamic image data, train the base model, and obtain the application model; Step S305: In field application, the application model is used to identify feature points in the acquired images, and the relative distance of the same feature point in two consecutive cycles is calculated to obtain the relative running distance of the train.

13. The positioning method according to claim 9, characterized in that, The positioning calibration in step S4 specifically involves: Step S401: The on-board controller calculates the train's position one based on the position of the visual beacon; Step S402: The on-board controller accumulates the relative displacement fed back by the visual odometer in each cycle to obtain the second train position. Step S403: Correct the train's position 2 based on the train's position 1 to obtain the train's position 3.

14. The positioning method according to claim 9, characterized in that, The specific steps in step S5, namely identifying the train and continuously calculating its position, are as follows: Step S501: Train a model for recognizing trains for the trackside visual axis counting (300); Step S502: Within the field of view of the second camera of the trackside visual measuring axis (300), absolute position points are pre-marked and marked as reference points for locating the train position in the subsequent process. Step S503: When a train is detected passing by, the train's identification code is identified by the QR code and barcode on the train body to determine the train's identity, and the absolute position of the train is calculated by using pre-marked reference points. Step S504: Continuously update the position of the train until the train leaves the camera's field of view.

15. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 9 to 14.

16. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 9 to 14.

Images