Tunnel limit determination method, device, electronic device and storage medium

Abstract

The present disclosure provides a method, device, electronic device and storage medium for determining tunnel limits. The method comprises: obtaining first point cloud data corresponding to the area ahead of the running rail vehicle; filtering the first point cloud data to obtain second point cloud data; wherein the filtering process is used to remove the parts corresponding to the top and bottom of the tunnel in the first point cloud data; rasterizing the second point cloud data to obtain a two-dimensional point cloud image corresponding to the second point cloud data; the two-dimensional point cloud image includes a plurality of point cloud points; segmenting the two-dimensional point cloud image to obtain point cloud points corresponding to the tunnel wall; in the running direction of the rail vehicle, dividing the point cloud points corresponding to the tunnel wall into a plurality of point cloud point groups; for each point cloud point group, performing straight line fitting on the point cloud points included therein to obtain a plurality of fitting straight lines corresponding one to one to the plurality of point cloud point groups; performing smoothing on the plurality of fitting straight lines to obtain the tunnel limit.

Description

Technical Field

[0001] The present disclosure relates to the field of rail transit technology, and in particular to a method for determining tunnel limits, an apparatus, electronic equipment and a storage medium. Background Art

[0002] As an important part of the transportation system, rail transit plays an important role in people's daily production and life. Rail vehicles will pass through tunnels during operation. In order to ensure the safety of driving in tunnels, laser point clouds of tunnel walls are generally collected by installed laser radars, and tunnel limits are formed accordingly. The tunnel limits are used to assist rail vehicles in detecting and identifying obstacles to achieve safe driving. However, when the existing technology forms tunnel limits based on the laser point clouds of tunnel walls, there are generally problems such as large data processing volume, low data processing efficiency, and insufficient accuracy. Summary of the invention

[0003] In view of this, the purpose of the present disclosure is to provide a tunnel delimitation determination method, device electronic device and storage medium.

[0004] Based on the above objectives, the present disclosure provides a method for determining a tunnel boundary, comprising:

[0005] Acquire first point cloud data corresponding to the area ahead of the rail vehicle;

[0006] Performing filtering processing on the first point cloud data to obtain second point cloud data; wherein the filtering processing is used to remove the parts corresponding to the top and bottom of the tunnel in the first point cloud data;

[0007] Performing rasterization processing on the second point cloud data to obtain a two-dimensional point cloud image corresponding to the second point cloud data; the two-dimensional point cloud image includes a plurality of point cloud points;

[0008] Segmenting the two-dimensional point cloud image to obtain point cloud points corresponding to the tunnel wall;

[0009] In the running direction of the rail vehicle, the point cloud points corresponding to the tunnel wall are divided into several point cloud point groups;

[0010] For each of the point cloud point groups, performing straight line fitting on the point cloud points included therein to obtain a number of fitting straight lines corresponding one-to-one to the point cloud point groups;

[0011] A plurality of the fitting straight lines are smoothed to obtain the tunnel limit.

[0012] Based on the same inventive concept, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the methods described above is implemented.

[0013] Based on the same inventive concept, the present disclosure also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute any of the above methods.

[0014] From the above, it can be seen that the tunnel limit determination method, device electronic device and storage medium provided by the present disclosure effectively remove useless parts in the point cloud data through filtering processing, and then determine the point cloud points corresponding to the tunnel wall through simple and efficient segmentation processing after rasterization processing, and then through the straight line fitting method, the tunnel limit is determined with a small amount of calculation. It can be seen that the tunnel limit determination scheme disclosed in the present disclosure has the advantages of high accuracy and data processing efficiency and small amount of data calculation, and can better realize the safe driving assistance for rail vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A flow chart of a method for determining tunnel boundaries according to an embodiment of the present disclosure;

[0017] Figure 2 is a schematic diagram of a three-dimensional rectangular coordinate system in an embodiment of the present disclosure;

[0018] Figure 3 is a schematic diagram of the filtering process in the embodiment of the present disclosure;

[0019] Figure 4 A schematic diagram of rasterization and segmentation processing in an embodiment of the present disclosure;

[0020] Figure 5 This is a schematic diagram of a fitted straight line obtained by straight line fitting in an embodiment of the present disclosure;

[0021] Figure 6 A schematic diagram of tunnel boundaries in an embodiment of the present disclosure;

[0022] Figure 7 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0023] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0024] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0025] As described in the background art, when the prior art forms tunnel boundaries based on the laser point cloud of the tunnel wall, there are generally problems such as large data processing volume, low data processing efficiency and insufficient accuracy.

[0026] In view of the problems existing in the above-mentioned prior art, the present disclosure provides a method for determining tunnel limits, an electronic device and a storage medium, which effectively removes useless parts of the point cloud data through filtering processing, and then determines the point cloud points corresponding to the tunnel wall through simple and efficient segmentation processing after rasterization processing, and then realizes the determination of tunnel limits through a small amount of calculation by linear fitting. It can be seen that the tunnel limit determination scheme disclosed in the present disclosure has the advantages of high accuracy and data processing efficiency, and small amount of data calculation, and can better realize the safe driving assistance of rail vehicles.

[0027] The solutions of the present disclosure are further illustrated below through specific examples.

[0028] First, the embodiment of the present disclosure provides a method for determining tunnel boundaries. Figure 1 The tunnel limit determination method comprises the following steps:

[0029] Step S101: Acquire first point cloud data corresponding to the area ahead of the rail vehicle.

[0030] When a laser beam hits the surface of an object, the reflected laser will carry information such as direction and distance. If the laser beam is scanned along a certain trajectory, the reflected laser point information will be recorded while scanning. Since the scanning is extremely fine, a large number of laser points can be obtained, thus forming a laser point cloud. The laser point cloud is the point cloud data described in this embodiment. In this embodiment, the first point cloud data is the laser point cloud corresponding to the area in front of the rail vehicle obtained by scanning the laser radar on the rail vehicle.

[0031] The data included in the first point cloud data is based on a coordinate system. Figure 2 , a three-dimensional rectangular coordinate system is predefined. In the three-dimensional rectangular coordinate system, the running direction of the rail vehicle is the X-axis, the width direction of the tunnel is the Y-axis, and the height direction of the tunnel is the Z-axis. The data in the first point cloud data are all represented by the coordinates of the three-dimensional rectangular coordinate system to represent their position information.

[0032] Among them, since the density of the point cloud data collected by the laser radar is relatively high and there will be certain noise data, the first point cloud data can also be preprocessed in this embodiment. The preprocessing includes downsampling the voxels of the first point cloud data to reduce the point cloud density, which can reduce the amount of calculation while ensuring that the geometric structure of the first point cloud data is not destroyed. In addition, the preprocessing can also include statistical filtering of the first point cloud data to filter out noise data such as outliers.

[0033] Step S102: filtering the first point cloud data to obtain second point cloud data; wherein the filtering is used to remove the parts corresponding to the top and bottom of the tunnel in the first point cloud data.

[0034] In this step, the first point cloud data is filtered. The filtering process refers to removing a portion of the first point cloud data according to certain filtering conditions. Specifically, the filtering process removes the portion corresponding to the top and bottom of the tunnel in the first point cloud data. In this embodiment, the first point cloud data is obtained by collecting data from the area in front of the rail vehicle, which includes the portion corresponding to the top and bottom of the tunnel. When determining the tunnel limit, the portion of the tunnel corresponding to the tunnel limit is the tunnel wall, so the portion corresponding to the top and bottom of the tunnel in the first point cloud data is useless for determining the tunnel limit. Therefore, in order to remove useless data, reduce the amount of data and improve accuracy, filtering is performed in this step.

[0035] In some embodiments, filtering the first point cloud data can be achieved by the following steps: obtaining a predetermined coordinate value interval; removing the portion of the Z-axis coordinate value in the first point cloud data that does not fall within the coordinate value interval. The predetermined coordinate value interval needs to be determined according to the tunnel specifications in the actual scene. For example, for the tunnel specifications used by common rail vehicles, the coordinate value interval can be selected as (-0.2, 4.5).

[0036] refer to Figure 3, the Z-axis coordinate values ​​corresponding to the two dotted lines are the predetermined coordinate value intervals. According to the correspondence between the three-dimensional rectangular coordinate system and the tunnel space, the top and bottom of the tunnel correspond to the positions with larger coordinate values ​​in the positive and negative directions of the Z-axis. Based on the coordinate value interval, the data in the first point cloud data is filtered and processed, and the part of the Z-axis coordinate value that does not fall into the coordinate value interval is removed. The remaining part after filtering is referred to as the second point cloud data in this embodiment.

[0037] Step S103: rasterize the second point cloud data to obtain a two-dimensional point cloud image corresponding to the second point cloud data; the two-dimensional point cloud image includes a plurality of point cloud points.

[0038] In this step, the second point cloud data is rasterized. The rasterization process refers to converting the second point cloud data in the form of a vector image into a two-dimensional point cloud image in the form of a bitmap. In this embodiment, the rasterization process renders the second point cloud data into a two-dimensional plane formed by the XY axis. Figure 4 , a two-dimensional point cloud image includes several point cloud points (such as Figure 4 ). The two-dimensional point cloud image is divided into a number of grid cells arranged in an array, wherein at least some of the grid cells include point cloud points; that is, in general, some grid cells include point cloud points, while some grid cells do not. For a grid cell, there is generally corresponding geometric position information (generally the center coordinates of the grid cell), basic information such as the number of point cloud points, and some other statistical information such as normal vectors and gradients that may be used.

[0039] Step S104: segment the two-dimensional point cloud image to obtain point cloud points corresponding to the tunnel wall.

[0040] In this step, the two-dimensional point cloud image is segmented, which means segmenting the two-dimensional point cloud image to separate the point cloud points corresponding to the tunnel wall in the two-dimensional point cloud image. Specifically, considering that the laser emitted by the laser radar cannot penetrate the tunnel wall when collecting point cloud data; therefore, in the two-dimensional point cloud image, the point cloud points corresponding to the tunnel wall should be distributed at positions with larger coordinate values ​​in the positive and negative directions of the Y axis; that is, the point cloud points corresponding to the tunnel wall are distributed in the grid cells on the two sides of the Y axis.

[0041] In some embodiments, segmentation processing is performed on the two-dimensional point cloud image to obtain point cloud points corresponding to the tunnel wall, which can be achieved through the following steps: determining the center coordinates of the grid unit; selecting a number of target grid units from a number of grid units; the target grid unit includes a point cloud point, and the absolute value of the Y-axis coordinate value of its center coordinate is the largest; and the point cloud points included in the target grid unit are used as the point cloud points corresponding to the tunnel wall.

[0042] Among them, for each grid cell, its center coordinate is determined. Then, based on the center coordinate of the grid cell, the target grid cell is selected. The target grid cell must meet the following conditions: the absolute value of the Y-axis coordinate value of the center coordinate of the grid cell is the largest, and the grid cell contains point cloud points. Finally, all the point cloud points included in the target grid cell are determined as the point cloud points corresponding to the tunnel wall. The results of the segmentation process can be referred to Figure 4 As shown (the target grid cell is as Figure 4 The grid cells included in the elliptical dotted frame).

[0043] It can be understood that the tunnel structure has two tunnel walls on both sides; based on the running direction of the rail vehicle, the two tunnel walls can be the left tunnel wall and the right tunnel wall respectively. Therefore, the point cloud points corresponding to the tunnel walls can specifically include: point cloud points corresponding to the left tunnel wall, and point cloud points corresponding to the right tunnel wall.

[0044] Step S105: In the running direction of the rail vehicle, the point cloud points corresponding to the tunnel wall are divided into a plurality of point cloud point groups.

[0045] In this step, in the running direction of the rail vehicle, that is, in the X-axis direction, the point cloud points corresponding to the tunnel wall are divided into a plurality of point cloud point groups. Specifically, in the X-axis direction, according to the X-axis coordinates of the point cloud points, the point cloud points corresponding to the tunnel wall are divided into different point cloud point groups, and each point cloud point group includes a plurality of point cloud points.

[0046] In some embodiments, in the direction of rail vehicle operation, the point cloud points corresponding to the tunnel wall are divided into several point cloud point groups, which can be achieved through the following steps: determining the radar detection distance used when collecting the first point cloud data; determining the segmentation distance based on the radar detection distance; and in the X-axis direction, dividing the point cloud point groups in sequence according to the segmentation distance.

[0047] The specific value of the segment distance is positively correlated with the detection distance of the laser radar. That is, the longer the detection distance of the laser radar, the larger the segment distance. For example, if the effective detection distance of a certain type of radar tunnel wall is 150 meters, the segment distance is divided into (0,40], (41,90], (91,150], and the point cloud in the distance is sparse, so the interval is relatively longer.

[0048] It can be understood that the point cloud points corresponding to the left tunnel wall and the point cloud points corresponding to the right tunnel wall are divided into point cloud groups respectively. That is, the point cloud groups obtained after the division may include: the left point cloud group corresponding to the left tunnel wall and the right point cloud group corresponding to the right tunnel wall; and there is a one-to-one correspondence between the left point cloud group and the right point cloud group. Figure 5As shown, the dotted lines in the figure indicate different segment distances. For the same segment distance, it corresponds to a left point cloud point group and a right point cloud point group.

[0049] Step S106: for each of the point cloud point groups, perform straight line fitting on the point cloud points included therein to obtain a number of fitting straight lines corresponding one-to-one to the point cloud point groups.

[0050] In this step, the point cloud points included in the point cloud point group are fitted with a straight line. After the straight line fitting, each point cloud point will get a corresponding fitting straight line. Among them, when fitting the point cloud points with a straight line, the least squares method, the Ransac algorithm (Random Sample Consensus), the gradient descent algorithm, etc. can be used.

[0051] In some embodiments, the point cloud points are fitted with a straight line using a Ransac algorithm, specifically including:

[0052] For each pair of left point cloud point group and right point cloud point group, the following processing is performed:

[0053] S1. Determine the equation of the first straight line of the left point cloud point group, y 1 =kx+b 1 ; Determine the equation of the second straight line of the right point cloud point group, y 2 =kx+b 2 .The slope of the first straight line equation is the same as that of the second straight line equation, that is, their slope k is preset to the same value.

[0054] S2. Randomly select three point cloud points from the point cloud point group corresponding to the left tunnel wall and the point cloud point group corresponding to the right tunnel wall. The variance of the Y-axis coordinate values ​​of the three selected point cloud points needs to be greater than a certain threshold and the positive and negative signs of the coordinate Y values ​​of the three point cloud points cannot be all the same, so as to ensure that two of the three point cloud points come from the point cloud point group corresponding to the left tunnel wall and the point cloud point group corresponding to the right tunnel wall, respectively.

[0055] S3. Determine the candidate fitting lines corresponding to the left point cloud point group and the right point cloud point group respectively through the selected three point cloud points. The candidate fitting lines are obtained by determining the parameters of the line equation after performing line fitting on the three point cloud points.

[0056] S4. According to a predetermined inlier determination threshold, determine the number of inliers from the left point cloud point group to the corresponding candidate fitting line, and determine the number of inliers from the right point cloud point group to the corresponding candidate fitting line. The inliers refer to point cloud points within a certain distance to the candidate fitting line. The distance to the candidate fitting line is the inlier determination threshold.

[0057] S5, repeating the above steps S1 to S4 until a predetermined number of iterations is reached to obtain the fitting lines corresponding to the pair of left point cloud point groups and right point cloud point groups, wherein the pair of candidate fitting lines with the largest number of inliers is determined as the fitting lines corresponding to the pair of left point cloud point groups and right point cloud point groups.

[0058] After the above-mentioned Ransac algorithm is used to perform straight line fitting for each pair of left point cloud point group and right point cloud point group, a number of fitting straight lines corresponding to the point cloud point groups are obtained. For details, please refer to Figure 5 As shown (the fitting straight line is as Figure 5 , as shown by line segment L in the figure).

[0059] In steps S105 and S106, the point cloud points corresponding to the tunnel wall are divided into point cloud groups, and then straight line fitting is performed on each point cloud group obtained by the division. This processing method can avoid the point cloud point clustering processing performed in the prior art, and can effectively reduce the amount of data and the amount of calculation; in addition, through the straight line fitting method, it is convenient to control the generation of parallel fitting straight lines (setting the straight line slope during straight line fitting), which is conducive to the generation of the final tunnel limit.

[0060] Step S107: Smoothing the plurality of fitted straight lines to obtain tunnel limits.

[0061] In this step, a plurality of fitted straight lines obtained by straight line fitting are smoothed. Specifically, the fitted straight lines corresponding to the left and right tunnel walls are smoothed to obtain the tunnel limits corresponding to the tunnel walls on both sides. The smoothing process can adopt the least square method, the quadratic exponential smoothing method, etc.

[0062] In some embodiments, before smoothing the plurality of fitted straight lines, the method further includes: for each of the fitted straight lines, translating the fitted straight line along its normal line in a direction away from the tunnel wall (toward the track) by a predetermined distance. The translation distance needs to be determined according to parameters such as the specifications of the rail vehicle and the specifications of the tunnel. Considering that the tunnel limit is located between the side wall of the rail vehicle and the tunnel wall, the translation processing in this embodiment can more accurately describe the range of the tunnel limit area, thereby reducing the amount of calculation for subsequent judgment of whether an obstacle intrudes the limit.

[0063] refer to Figure 5 and Figure 6 , the fitted straight line after translation is Figure 5 The dotted line segment L' in the figure is shown; the tunnel limit obtained after smoothing is as follows Figure 6 As shown by the curves L1 and L2 in (L1 and L2 correspond to the left and right tunnel walls respectively).

[0064] It can be seen from the above embodiments that the tunnel limit determination method disclosed in the present invention effectively removes useless parts in the point cloud data through filtering processing, and then after rasterization processing, determines the point cloud points corresponding to the tunnel wall through simple and efficient segmentation processing, and then through the straight line fitting method, the tunnel limit is determined with a small amount of calculation. It can be seen that the tunnel limit determination scheme disclosed in the present invention has the advantages of high accuracy and data processing efficiency and small amount of data calculation, and can better realize the safe driving assistance of rail vehicles.

[0065] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0066] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the tunnel boundary determination method described in any of the above embodiments is implemented.

[0068] Figure 7 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0069] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0070] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0071] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0072] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0073] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0074] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0075] The electronic device of the above embodiment is used to implement the corresponding tunnel limit determination method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0076] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the tunnel limit determination method described in any of the above embodiments.

[0077] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0078] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the tunnel limit determination method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0079] Those 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 the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, 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 the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0080] The embodiments of the present disclosure are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A method for determining a tunnel limit, comprising: Acquire first point cloud data corresponding to the area ahead of the rail vehicle; Performing filtering processing on the first point cloud data to obtain second point cloud data; wherein the filtering processing is used to remove the parts corresponding to the top and bottom of the tunnel in the first point cloud data; Performing rasterization processing on the second point cloud data to obtain a two-dimensional point cloud image corresponding to the second point cloud data; the two-dimensional point cloud image includes a plurality of point cloud points; Segmenting the two-dimensional point cloud image to obtain point cloud points corresponding to the tunnel wall; In the running direction of the rail vehicle, the point cloud points corresponding to the tunnel wall are divided into several point cloud point groups; For each of the point cloud point groups, performing straight line fitting on the point cloud points included therein to obtain a number of fitting straight lines corresponding one-to-one to the point cloud point groups; Smoothing a number of the fitted straight lines to obtain tunnel limits; The first point cloud data and the second point cloud data are both based on a predefined three-dimensional rectangular coordinate system; the three-dimensional rectangular coordinate system has the running direction of the rail vehicle as the X-axis, the tunnel width direction as the Y-axis, and the tunnel height direction as the Z-axis; The filtering process of the first point cloud data specifically includes: Obtaining a predetermined coordinate value interval; Removing a portion of the first point cloud data whose z-axis coordinate value does not fall within the coordinate value interval; The two-dimensional point cloud image includes a plurality of grid units arranged in an array, and at least some of the grid units include the point cloud points; The segmentation process of the two-dimensional point cloud image to obtain point cloud points corresponding to the tunnel wall specifically includes: Determining the center coordinates of the grid cell; Selecting a plurality of target grid cells from the plurality of grid cells; the target grid cells include point cloud points, and the absolute value of the y-axis coordinate value of the center coordinate is the largest; The point cloud points included in the target grid unit are used as the point cloud points corresponding to the tunnel wall.

2. The method according to claim 1, wherein: In the running direction of the rail vehicle, the point cloud points corresponding to the tunnel wall are divided into several point cloud point groups, including: Determining a radar detection distance applied when collecting the first point cloud data; Determining a segment distance according to the radar detection distance; In the x-axis direction, a plurality of point cloud point groups are divided in sequence according to the segment distances.

3. The method according to claim 1, wherein: The linear fitting of the point cloud points includes: The point cloud points are fitted with a straight line using the Ransac algorithm.

4. The method according to claim 3, wherein: Based on the running direction of the rail vehicle, the point cloud point group includes: a plurality of left point cloud point groups corresponding to the left tunnel wall, and a plurality of right point cloud point groups corresponding to the right tunnel wall; the left point cloud point group corresponds to the right point cloud point group one by one; The linear fitting of the point cloud points by the Ransac algorithm specifically includes: For each pair of left point cloud point group and right point cloud point group, the following processing is performed: Determine a first straight line equation of the left point cloud point group, and determine a second straight line equation of the right point cloud point group; the first straight line equation and the second straight line equation have the same slope; Randomly select three point cloud points from the point cloud point group corresponding to the left tunnel wall and the point cloud point group corresponding to the right tunnel wall; wherein two of the three point cloud points come from the point cloud point group corresponding to the left tunnel wall and the point cloud point group corresponding to the right tunnel wall, respectively; Determine the candidate fitting straight lines corresponding to the left point cloud point group and the right point cloud point group respectively through the selected three point cloud points; According to a predetermined inlier determination threshold, determine the number of inliers from the left point cloud point group to the corresponding candidate fitting straight line, and determine the number of inliers from the right point cloud point group to the corresponding candidate fitting straight line; The above four steps are repeated until a predetermined number of iterations is reached, and a pair of candidate fitting lines with the largest number of internal points is determined as the fitting lines corresponding to the left point cloud point group and the right point cloud point group respectively.

5. The method according to claim 1, wherein: The smoothing of the plurality of fitting straight lines specifically includes: The plurality of fitted straight lines are smoothed by a least square smoothing algorithm.

6. The method according to claim 1, wherein: The smoothing of the plurality of fitted straight lines may also include: For each of the fitting straight lines, the fitting straight line is translated along its normal line by a predetermined distance in a direction away from the tunnel wall.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the program.

8. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to any one of claims 1 to 6.

Images