Initial positioning method and system for wakeup process of rail train and storage medium

Abstract

The invention provides an initial positioning method and system for a wakeup process of a rail train and a storage medium. The method comprises the following steps: acquiring a train wakeup instruction of a target train; acquiring a train initial position of the target train through the laser radar based on the train wake-up command; judging whether the target train is in a sleep awakening track or not according to the initial position of the train; if the target train is located in the sleep awakening track, whether the target train is located between the two indication boards or not is judged through the camera device; if the target train is located between the two indication boards, the indication boards are scanned through the laser radar, and a distance scanning result is obtained; and calculating according to the distance scanning result to obtain the train accurate position of the target train. The method has the effect that when the dormant train is awakened, the initial positioning of the train can be accurately obtained even if unexpected displacement occurs.

Description

Technical Field

[0001] The present invention belongs to the technical field of train control management, and in particular relates to an initial positioning method, system and storage medium for a railway train wake-up process. Background Art

[0002] With the development of rail transit, rail transit control systems have experienced manual driving, semi-automatic driving and fully automatic driving, and their safety and automation levels have been continuously improved. In the communication-based train operation control system, the train's safety protection distance, movement authorization, operation interval, stop accuracy, transportation efficiency, etc. are all related to the real-time position of the train, especially for the fully automatic operation system FAO (Fully Automatic Operation) that realizes the functional requirements of GoA3 and GoA4 (where GoA is the automation level, the full name is Grade of Automation). Only after the train wakes up the initial positioning successfully, can it continue to execute the subsequent wake-up process and complete the automatic depot and online operation of the train.

[0003] Before non-fully automatic driving trains depart from the depot, they need to be manually powered on and daily inspections completed. However, this manual operation consumes a lot of manpower, reduces efficiency, and may cause energy waste and equipment wear. To solve this problem, in fully automatic driving, the train is controlled by the ground control center (ZC) in the depot and automatically powered on to wake up. In the existing train wake-up positioning method, the train position is usually recorded by the on-board ATP before the train goes to sleep, and after the train wakes up and powers on, the position of the awakened train is verified by reading the positioning transponder information on the track. Only when the sleep position is consistent with the position shown by the positioning transponder, the train wake-up positioning is judged to be successful.

[0004] However, a dormant train may experience unexpected collisions during the dynamic testing of the train, including collisions with adjacent vehicles or buildings, resulting in unexpected position movements of the dormant train. The dormant train cannot detect unexpected position movements. Therefore, when the train is awakened according to the memorized position data, the initial positioning of the train obtained is inaccurate, which in turn causes the dormant train to be unable to execute subsequent wake-up processes. Summary of the invention

[0005] The present invention provides an initial positioning method, system and storage medium for a railway train awakening process, so as to solve the problem that when a dormant train is awakened, the initial positioning of the train may be inaccurate due to unexpected displacement during the dormant process.

[0006] In a first aspect, the present invention provides an initial positioning method for a railway train wake-up process, the method comprising the following steps:

[0007] Obtaining a train wake-up instruction of a target train, wherein the front and rear of the target train are both equipped with a sensing system, and the sensing system includes a laser radar and a camera device;

[0008] Based on the train wake-up command and using the laser radar, an initial train position of the target train is acquired;

[0009] Determine whether the target train is in a sleep-wake-up track according to the initial position of the train, wherein two indicator signs are preset in the sleep-wake-up track, and a distance between the two indicator signs is greater than the train length of the target train;

[0010] If the target train is in the sleep-wake-up track, determining whether the target train is between the two signboards by using the camera device;

[0011] If the target train is between the two signboards, the laser radar is used to scan the signboards to obtain a distance scanning result;

[0012] The precise train position of the target train is calculated based on the distance scanning result.

[0013] Optionally, the acquiring the train initial position of the target train by the laser radar based on the train wake-up command comprises the following steps:

[0014] Powering on and initializing the sensing system based on the train wake-up command;

[0015] Scanning the surrounding track environment of the target train multiple times by the laser radar, and constructing a target track map according to the three-dimensional point cloud information obtained by the scanning;

[0016] The target track map is compared with a preset track line map to determine the initial train position of the target train.

[0017] Optionally, comparing the target track map with a preset track line map to determine the initial train position of the target train comprises the following steps:

[0018] Transmitting the target track map to an onboard ATP system of the target train, wherein the onboard ATP system is preset with a track line map;

[0019] The target track map is compared with the track line map by the on-board ATP system to determine the initial train position of the target train.

[0020] Optionally, scanning the surrounding track environment of the target train multiple times by the laser radar and constructing a target track map according to the three-dimensional point cloud information obtained by the scanning comprises the following steps:

[0021] Scanning the surrounding track environment of the target train for the first time by the laser radar to obtain initial three-dimensional point cloud information and determine a closed-loop path;

[0022] Determine an initial radar position of the laser radar based on the initial three-dimensional point cloud information, and generate a plurality of feature points;

[0023] Continue to scan the surrounding track environment of the target train by the laser radar, and perform loopback detection based on the closed-loop path to obtain a plurality of three-dimensional point cloud information;

[0024] Superimposing the plurality of three-dimensional point cloud information onto the initial three-dimensional point cloud information to obtain target three-dimensional point cloud information, and correcting the initial radar position according to the positions of the plurality of feature points;

[0025] A target track map is constructed based on the target three-dimensional point cloud information and according to the corrected initial radar position.

[0026] Optionally, the determining whether the target train is between the two signboards by the camera device comprises the following steps:

[0027] Taking pictures of the front and rear of the target train by the camera device;

[0028] Preprocessing the front direction image and the rear direction image;

[0029] Based on a preset sign shape descriptor set and using an edge detection algorithm, determining whether both the vehicle front direction image and the vehicle rear direction image have sign feature areas;

[0030] If the sign feature area does not exist in any one or both of the front direction picture or the rear direction picture, then the judgment result is outputted that the target train is not between the two signboards;

[0031] If the sign feature area exists in both the front direction image and the rear direction image, extracting the color feature of the sign feature area;

[0032] Determining whether the color feature belongs to a preset color feature interval;

[0033] If the color feature belongs to the color feature interval, the output judgment result is that the target train is between the two signs;

[0034] If the color feature does not belong to the color feature interval, the output judgment result is that the target train is not between the two signs.

[0035] Optionally, the distance scanning result includes a front scanning result and a rear scanning result of the vehicle, and the step of scanning the sign with the laser radar to obtain the distance scanning result includes the following steps:

[0036] Scanning the sign located at the front of the target train using the laser radar configured at the front of the target train to obtain the front scanning result;

[0037] The laser radar disposed at the rear of the target train is used to scan the sign located at the rear of the target train to obtain the rear scanning result.

[0038] Optionally, the calculating the precise train position of the target train according to the distance scanning result comprises the following steps:

[0039] Taking pictures of the front and rear of the target train by the camera device;

[0040] Extracting a first sign image in the vehicle front direction picture and a second sign image in the vehicle rear direction picture using an image recognition model;

[0041] Using OCR technology to identify the first signboard image and the second signboard image, and adding signboard features to the first signboard image and the second signboard image respectively according to the identification results;

[0042] The precise train position of the target train is calculated by combining the sign features and the distance scanning result.

[0043] Optionally, the sign features include a front sign and a rear sign, the train precise position includes a front precise position and a rear precise position, and the calculating the train precise position of the target train by combining the sign features and the distance scanning result includes the following steps:

[0044] identifying the sign features of the first sign image and the second sign image;

[0045] If the sign feature of the first sign image is the front sign, and the sign feature of the second sign image is the rear sign, the precise position of the front of the vehicle is calculated based on the front scanning result, and the precise position of the rear of the vehicle is calculated based on the rear scanning result;

[0046] If the sign feature of the first sign image is the rear sign, and the sign feature of the second sign image is the front sign, the precise position of the rear of the vehicle is calculated based on the front scanning result, and the precise position of the front of the vehicle is calculated based on the rear scanning result.

[0047] In a second aspect, the present invention also provides an initial positioning system for a rail train wake-up process, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the computer program.

[0048] In a third aspect, the present invention further provides a computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method described in the first aspect.

[0049] The beneficial effects of the present invention are:

[0050] The initial positioning method of the rail train wake-up process adopted by the present invention includes the following steps: obtaining a train wake-up instruction of a target train; obtaining the initial train position of the target train based on the train wake-up command and through the laser radar; judging whether the target train is in a sleep wake-up track according to the initial train position; if the target train is in the sleep wake-up track, judging whether the target train is between two signs through the camera device; if the target train is between the two signs, scanning the signs with the laser radar to obtain a distance scanning result; and calculating the precise train position of the target train according to the distance scanning result.

[0051] By setting up signboards and executing the above steps, the sensing system can be used to scan the signboards to achieve the effect of obtaining the precise position of the target train in real time. Compared with the initial position acquisition method of reading the memorized position before sleep, it can avoid the problem of inaccurate initial positioning caused by the actual position being different from the memorized position due to unpredicted displacement of the train during sleep. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of the flow of the initial positioning method of the rail train wake-up process in the present invention.

[0053] Figure 2 It is a system structure diagram of the target train perception system in the present invention and a schematic diagram of the location of the signboard. DETAILED DESCRIPTION

[0054] The invention discloses an initial positioning method for a rail train wake-up process.

[0055] Reference Figure 1 The initial positioning method of the rail train wake-up process specifically includes the following steps:

[0056] S101. Obtain a train wake-up instruction of a target train.

[0057] Among them, refer to Figure 2 The front and rear of the target train are equipped with a perception system, which includes a laser radar, a camera device and a perception host. The camera device can be a high-definition camera. When the train wake-up command of the target train is obtained, the perception host will control the laser radar and the camera device to start.

[0058] S102. Based on the train wake-up command, the initial train position of the target train is obtained through the laser radar.

[0059] S103. Determine whether the target train is in the sleep-wake-up track according to the initial position of the train. If the target train is in the sleep-wake-up track, execute step S104.

[0060] If the target train is not in the sleep wake-up track, it means that the target train should not perform the train wake-up step at this time, so the train wake-up step is terminated. Figure 2 ,There are two signboards preset in the sleep-wake-up track, and the distance between the two signboards is greater than the train length of the target train.

[0061] S104. Determine whether the target train is between the two signboards through the camera device. If the target train is between the two signboards, execute step S105.

[0062] If the target train is not between the two signs, the train wake-up step is terminated.

[0063] S105. Use laser radar to scan the sign to obtain a distance scanning result.

[0064] Among them, LiDAR is a sensor that can calculate the distance of a target by emitting a laser beam and measuring the time difference between the laser beam and the target. LiDAR emits a laser beam that intersects with the target on the sign. When the laser beam intersects with the target, the LiDAR receives the reflected laser beam. By measuring the time difference between the emission and reception of the laser beam, the round-trip time of the laser beam can be calculated. Based on the speed of light and the round-trip time, the distance between the target and the LiDAR can be calculated, and the calculated distance is the distance scan result.

[0065] S106. Calculate the precise train position of the target train based on the distance scanning result.

[0066] The implementation principle of this embodiment is:

[0067] Obtain the train wake-up instruction of the target train; obtain the initial train position of the target train through the laser radar based on the train wake-up command; determine whether the target train is in the sleep wake-up track according to the initial train position; if the target train is in the sleep wake-up track, determine whether the target train is between two signboards through the camera device; if the target train is between the two signboards, scan the signboards with the laser radar to obtain the distance scanning result; calculate the precise train position of the target train according to the distance scanning result. Through the setting of the signboard and the execution of the above steps, the signboard can be scanned by the perception system to achieve the effect of obtaining the precise position of the target train in real time. Compared with the initial position acquisition method of reading the memory position before sleep, it can avoid the problem of inaccurate initial positioning caused by the train's actual position being different from the memory position due to unpredicted displacement during sleep.

[0068] In one embodiment, step S102, i.e., obtaining the initial train position of the target train through a laser radar based on the train wake-up command, specifically includes the following steps:

[0069] Power on and initialize the sensing system based on the train wake-up command;

[0070] The laser radar is used to scan the surrounding track environment of the target train multiple times, and a target track map is constructed based on the three-dimensional point cloud information obtained by the scan;

[0071] The target track map is compared with the preset track line map to determine the initial train position of the target train.

[0072] In this embodiment, refer to Figure 2 , comparing the target track map with the preset track line map to determine the initial train position of the target train. This step specifically includes the following steps: transmitting the target track map to the on-board ATP system of the target train, the on-board ATP system being preset with the track line map; comparing the target track map with the track line map through the on-board ATP system, finding the target map area in the track line map that is the same as the target track map, and obtaining the regional coordinates of the target map area as the initial train position of the target train.

[0073] In one embodiment, scanning the surrounding track environment of the target train multiple times by a laser radar, and constructing a target track map based on the three-dimensional point cloud information obtained by the scanning specifically includes the following steps:

[0074] The laser radar is used to scan the surrounding track environment of the target train for the first time to obtain initial 3D point cloud information and determine the closed-loop path;

[0075] Determine the initial radar position of the laser radar based on the initial three-dimensional point cloud information and generate multiple feature points;

[0076] Continue to scan the surrounding track environment of the target train through the laser radar, and perform loopback detection based on the closed-loop path to obtain multiple three-dimensional point cloud information;

[0077] The multiple three-dimensional point cloud information is superimposed on the initial three-dimensional point cloud information to obtain the target three-dimensional point cloud information, and the initial radar position is corrected according to the positions of the multiple feature points;

[0078] The target track map is constructed based on the target 3D point cloud information and the corrected initial radar position.

[0079] In this embodiment, the laser radar emits a laser beam, scans the surrounding environment and receives the reflected laser beam. Based on the round-trip time of the laser and the speed of light, the distance between the laser beam and the object in the environment is calculated. The distance and scanning angle information are converted into three-dimensional coordinates to form initial three-dimensional point cloud information. Based on the object features in the point cloud information, such as tracks, buildings, etc., the closed-loop path is determined. Using the initial three-dimensional point cloud information, the initial radar position of the laser radar is calculated through a point cloud registration algorithm. Multiple feature points, such as corner points, edge points, etc., are extracted from the point cloud for subsequent loop detection and position correction.

[0080] The laser radar continues to scan the environment around the target train to obtain new three-dimensional point cloud information. Using the loop detection algorithm, the new point cloud information is matched with the initial point cloud information to determine the closed-loop path. Multiple point cloud information is superimposed on the initial point cloud information to form the target three-dimensional point cloud information. Using the position information of the feature points, the initial radar position is corrected through the position correction algorithm to improve the accuracy of positioning. Finally, using the target three-dimensional point cloud information, the map construction algorithm is used to extract information such as tracks and obstacles, and a three-dimensional target track map is created based on the continuously accumulated point cloud data. Through the above steps, compared with directly accumulating point cloud data and creating a three-dimensional map, the problem of error accumulation can be avoided.

[0081] In one embodiment, the step of determining whether the target train is between two signs by using a camera device in step S104 specifically includes the following steps:

[0082] Taking pictures of the front and rear of the target train by using a camera device;

[0083] Pre-process the front and rear direction images;

[0084] Based on the preset sign shape descriptor set and using the edge detection algorithm, it is determined whether both the front and rear direction images have sign feature areas;

[0085] If there is no sign feature area in either the front direction image or the rear direction image or both, the output judgment result is that the target train is not between the two signboards;

[0086] If there is a sign feature area in both the front direction image and the rear direction image, extract the color features of the sign feature area;

[0087] Determine whether the color feature belongs to a preset color feature interval;

[0088] If the color feature belongs to the color feature interval, the output judgment result is that the target train is between the two signs;

[0089] If the color feature does not belong to the color feature interval, the output judgment result is that the target train is not between the two signs.

[0090] In this embodiment, the shapes of the two signs are both pre-set fixed shapes, such as round signs, and the surface color of the signs is also pre-set fixed color, and the color is more special than the ambient color, such as yellow. Therefore, when judging whether the target train is between the two signs by the camera device, only a simple image processing method can be used to complete the recognition of the signs in the image, without using complex methods such as image recognition models, which is conducive to improving the processing speed of the computer.

[0091] Specifically, firstly, the front and rear images of the target train are taken by a camera device, and then the front and rear images are preprocessed by denoising and graying. After the image is preprocessed, all edge shape features in the image are extracted by an edge detection algorithm, and then the sign feature area is captured from the edge shape features by a sign shape descriptor set corresponding to the fixed shape pre-set by the sign. The sign shape descriptor set contains shape descriptors corresponding to the fixed shape of the sign at multiple angles. The shape descriptor can be a Hu moment or a Zernike moment, which is used to describe the shape characteristics of the target object. The shape descriptor can capture the contour shape of the target object, and recognition is performed by comparing the similarity between shape descriptors.

[0092] If both the front and rear images can capture the characteristic area of ​​the sign, further confirmation is required based on the pre-set fixed color of the sign to avoid interference from obstacles in the image that are the same shape as the sign. The specific process is to extract the regional color of the characteristic area of ​​the sign. Since the preset fixed color of the sign may change slightly due to different light intensities at different angles, it is necessary to generate a color feature interval based on the fixed color in advance, which includes all changes in the fixed color under various angles of illumination, and then determine whether the color feature belongs to the preset color feature interval. If the color feature belongs to the color feature interval, it means that the target train is between the two signs; otherwise, it means that the target train is not between the two signs.

[0093] In one embodiment, the distance scanning result includes a front scanning result and a rear scanning result. Step S105 is to use a laser radar to scan the sign, and obtaining the distance scanning result specifically includes the following steps:

[0094] Using a laser radar disposed at the front of the target train to scan the sign located at the front of the target train, a front scanning result is obtained;

[0095] The laser radar disposed at the rear of the target train is used to scan the sign located at the rear of the target train to obtain the rear scanning result.

[0096] In this embodiment, step S106, i.e., calculating the precise train position of the target train according to the distance scanning result, specifically includes the following steps:

[0097] Taking pictures of the front and rear of the target train by using a camera device;

[0098] Extracting a first sign image in the front direction picture and a second sign image in the rear direction picture using an image recognition model;

[0099] Using OCR technology to recognize the first signboard image and the second signboard image, and adding signboard features to the first signboard image and the second signboard image respectively according to the recognition result;

[0100] The precise train position of the target train is calculated by combining the sign features and the distance scanning results.

[0101] In this embodiment, the specific steps of extracting the first sign image in the front direction picture and the second sign image in the rear direction picture using the image recognition model are as follows:

[0102] Collect a dataset of images containing signage. Make sure the dataset includes signage images in a variety of scenes, angles, and lighting conditions, and annotate each image with the location and category of the signage.

[0103] Preprocess the front and rear images, including image resizing and pixel value normalization. Make sure the input image size matches the image size used during training.

[0104] Download or train a YOLOv4 model, using the pre-trained weights as a starting point for faster convergence and improved performance.

[0105] According to the characteristics of the sign recognition task, adjust the configuration file of the YOLOv4 model. You need to adjust parameters such as the number of categories, anchor box, and input image size.

[0106] Train the model using the prepared dataset. Make sure the input images and annotation data are correctly passed to the model and optimized according to the loss function.

[0107] Use the validation set to evaluate the model performance and perform optimization based on the validation results. Adjust hyperparameters, training rounds, etc. to improve the model's performance on the sign recognition task.

[0108] Use the test set to evaluate the performance of the trained model on new data. Calculate indicators such as recognition accuracy and recall to evaluate the effectiveness of the model.

[0109] The front direction picture and the rear direction picture are respectively input into the trained model to obtain the first sign image and the second sign image, and then non-maximum suppression (NMS) is performed on the first sign image and the second sign image to remove overlapping detection frames in the image.

[0110] In this embodiment, the specific steps of using OCR technology to recognize the first signboard image and the second signboard image, and adding signboard features to the first signboard image and the second signboard image respectively according to the recognition results are as follows:

[0111] Collect a dataset of images containing sign text. These images can be close-up images of the sign, ensuring that the text is clearly visible. At the same time, each image needs to be correctly annotated with text for supervised learning.

[0112] Preprocess the collected images, including image resizing, pixel value normalization, etc. In order to improve the generalization ability of the model, data enhancement may also be required, such as random rotation, translation, scaling, etc.

[0113] The CRNN (Convolutional Recurrent Neural Network) model is selected for the text recognition task.

[0114] Use the prepared dataset to train the OCR model. The input is the image and the label is the corresponding text. The model will learn how to recognize and extract text information from the image. The training process needs to be optimized using the loss function of the OCR model.

[0115] Use the validation set to evaluate the performance of the OCR model and tune the model based on indicators such as recognition accuracy and character error rate. You may need to adjust hyperparameters, training rounds, etc. to improve model performance.

[0116] Use the test set to evaluate the performance of the trained OCR model on new data.

[0117] The text output by the OCR model may contain some errors or inaccuracies, and post-processing is performed, such as text error correction and removal of meaningless characters, to improve the accuracy of the recognition results.

[0118] In the overall process of sign recognition, the first sign image and the second sign image are input into the trained OCR model for text recognition, and sign features are added to the first sign image and the second sign image respectively according to the recognition results to identify whether the sign in the first sign image is the front sign or the rear sign.

[0119] In this embodiment, the sign features include a front sign and a rear sign, and the precise position of the train includes a front precise position and a rear precise position. The steps of calculating the precise position of the target train by combining the sign features and the distance scanning result specifically include the following steps:

[0120] identifying sign features of the first sign image and the second sign image;

[0121] If the sign feature of the first sign image is a front sign, and the sign feature of the second sign image is a rear sign, the precise position of the front of the vehicle is calculated based on the front scanning result, and the precise position of the rear of the vehicle is calculated based on the rear scanning result;

[0122] If the sign feature of the first sign image is a rear sign, and the sign feature of the second sign image is a front sign, the precise position of the rear of the vehicle is calculated based on the front scanning result, and the precise position of the front of the vehicle is calculated based on the rear scanning result.

[0123] In this embodiment, since the precise position of the train includes the precise position of the front and the rear, it is necessary to determine the logical direction of the target train in the map before determining the precise position of the train. Since the two signs are in a long-term fixed state, it is possible to combine image recognition and OCR technology to determine whether the front of the target train and the front sign are on the same side. If the sign feature of the first sign image is the front sign, it means that the front of the target train and the front sign are on the same side. At this time, the scanning result of the front is converted into point cloud data, and the distance value of each measurement point is converted into a coordinate point in three-dimensional space. The point cloud data is filtered and denoised to remove invalid or abnormal measurement points. According to the known position and geometry of the sign, the position and posture information of the front of the target train relative to the front sign is calculated through the valid measurement points in the point cloud data, so as to finally determine the precise position of the front. The calculation method of the precise position of the rear is similar.

[0124] The present invention also discloses an initial positioning system for a rail train wake-up process, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, an initial positioning method for a rail train wake-up process as described in any one of the above-mentioned embodiments is implemented.

[0125] The implementation principle of this embodiment is:

[0126] By calling the program, the following steps are performed: obtaining the train wake-up instruction of the target train; obtaining the initial train position of the target train through the laser radar based on the train wake-up command; judging whether the target train is in the sleep wake-up track according to the initial train position; if the target train is in the sleep wake-up track, judging whether the target train is between two signboards through the camera device; if the target train is between two signboards, scanning the signboard with the laser radar to obtain the distance scanning result; calculating the precise train position of the target train according to the distance scanning result. Through the setting of the signboard and the execution of the above steps, the signboard can be scanned by the perception system to achieve the effect of obtaining the precise position of the target train in real time. Compared with the initial position acquisition method of reading the memory position before sleep, the problem of inaccurate initial positioning caused by the unpredicted displacement of the train during sleep, which makes the actual position different from the memory position, can be avoided.

[0127] The present invention also discloses a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the initial positioning method for a rail train wake-up process as described in any one of the above-mentioned embodiments are implemented.

[0128] The implementation principle of this embodiment is:

[0129] By calling the program, the following steps are performed: obtaining the train wake-up instruction of the target train; obtaining the initial train position of the target train through the laser radar based on the train wake-up command; judging whether the target train is in the sleep wake-up track according to the initial train position; if the target train is in the sleep wake-up track, judging whether the target train is between two signboards through the camera device; if the target train is between two signboards, scanning the signboard with the laser radar to obtain the distance scanning result; calculating the precise train position of the target train according to the distance scanning result. Through the setting of the signboard and the execution of the above steps, the signboard can be scanned by the perception system to achieve the effect of obtaining the precise position of the target train in real time. Compared with the initial position acquisition method of reading the memory position before sleep, the problem of inaccurate initial positioning caused by the unpredicted displacement of the train during sleep, which makes the actual position different from the memory position, can be avoided.

[0130] A person skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of protection of the present application is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present application as above, which are not provided in detail for the sake of simplicity.

[0131] One or more embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the present application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present application should be included in the protection scope of the present application.

Claims

1. An initial positioning method for a rail train wake-up process, characterized in that: The steps include: Obtaining a train wake-up instruction of a target train, wherein the front and rear of the target train are both equipped with a sensing system, and the sensing system includes a laser radar and a camera device; Based on the train wake-up command and using the laser radar, an initial train position of the target train is acquired; Determine whether the target train is in a sleep-wake-up track according to the initial position of the train, wherein two indicator signs are preset in the sleep-wake-up track, and a distance between the two indicator signs is greater than the train length of the target train; If the target train is in the sleep-wake-up track, determining whether the target train is between the two signboards by using the camera device; If the target train is between the two signboards, the laser radar is used to scan the signboards to obtain a distance scanning result; The precise train position of the target train is calculated based on the distance scanning result.

2. The initial positioning method for a rail train wake-up process according to claim 1, characterized in that: The method of obtaining the initial train position of the target train based on the train wake-up command and using the laser radar comprises the following steps: Powering on and initializing the sensing system based on the train wake-up command; Scanning the surrounding track environment of the target train multiple times by the laser radar, and constructing a target track map according to the three-dimensional point cloud information obtained by the scanning; The target track map is compared with a preset track line map to determine the initial train position of the target train.

3. The initial positioning method for a rail train wake-up process according to claim 2, characterized in that: The step of comparing the target track map with a preset track line map to determine the initial train position of the target train comprises the following steps: Transmitting the target track map to an onboard ATP system of the target train, wherein the onboard ATP system is preset with a track line map; The target track map is compared with the track line map by the on-board ATP system to determine the initial train position of the target train.

4. The initial positioning method for a rail train wake-up process according to claim 2, characterized in that: The method of scanning the surrounding track environment of the target train multiple times by the laser radar and constructing a target track map according to the three-dimensional point cloud information obtained by the scanning comprises the following steps: Scanning the surrounding track environment of the target train for the first time by the laser radar to obtain initial three-dimensional point cloud information and determine a closed-loop path; Determine an initial radar position of the laser radar based on the initial three-dimensional point cloud information, and generate a plurality of feature points; Continue to scan the surrounding track environment of the target train by the laser radar, and perform loopback detection based on the closed-loop path to obtain a plurality of three-dimensional point cloud information; Superimposing the plurality of three-dimensional point cloud information onto the initial three-dimensional point cloud information to obtain target three-dimensional point cloud information, and correcting the initial radar position according to the positions of the plurality of feature points; A target track map is constructed based on the target three-dimensional point cloud information and according to the corrected initial radar position.

5. The initial positioning method for a rail train wake-up process according to claim 1, characterized in that: The step of determining whether the target train is between the two signboards by the camera device comprises the following steps: Taking pictures of the front and rear of the target train by the camera device; Preprocessing the front direction image and the rear direction image; Based on a preset sign shape descriptor set and using an edge detection algorithm, determining whether both the vehicle front direction image and the vehicle rear direction image have sign feature areas; If the sign feature area does not exist in any one or both of the front direction picture or the rear direction picture, then the judgment result is outputted that the target train is not between the two signboards; If the sign feature area exists in both the front direction image and the rear direction image, extracting the color feature of the sign feature area; Determining whether the color feature belongs to a preset color feature interval; If the color feature belongs to the color feature interval, the output judgment result is that the target train is between the two signs; If the color feature does not belong to the color feature interval, the output judgment result is that the target train is not between the two signs.

6. The initial positioning method for a rail train wake-up process according to claim 1, characterized in that: The distance scanning result includes a front scanning result and a rear scanning result of the vehicle. The use of the laser radar to scan the sign to obtain the distance scanning result includes the following steps: Scanning the sign located at the front of the target train using the laser radar configured at the front of the target train to obtain the front scanning result; The laser radar disposed at the rear of the target train is used to scan the sign located at the rear of the target train to obtain the rear scanning result.

7. The initial positioning method for a railway train wake-up process according to claim 6, characterized in that: The step of calculating the precise train position of the target train according to the distance scanning result comprises the following steps: Taking pictures of the front and rear of the target train by the camera device; Extracting a first sign image in the vehicle front direction picture and a second sign image in the vehicle rear direction picture using an image recognition model; Using OCR technology to identify the first signboard image and the second signboard image, and adding signboard features to the first signboard image and the second signboard image respectively according to the identification results; The precise train position of the target train is calculated by combining the sign features and the distance scanning result.

8. The initial positioning method for a railway train wake-up process according to claim 7, characterized in that: The sign features include a front sign and a rear sign, the train precise position includes a front precise position and a rear precise position, and the step of combining the sign features and the distance scanning result to calculate the train precise position of the target train includes the following steps: identifying the sign features of the first sign image and the second sign image; If the sign feature of the first sign image is the front sign, and the sign feature of the second sign image is the rear sign, the precise position of the front of the vehicle is calculated based on the front scanning result, and the precise position of the rear of the vehicle is calculated based on the rear scanning result; If the sign feature of the first sign image is the rear sign, and the sign feature of the second sign image is the front sign, the precise position of the rear of the vehicle is calculated based on the front scanning result, and the precise position of the front of the vehicle is calculated based on the rear scanning result.

9. An initial positioning system for a railway train wake-up process, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images