Railway gauge recognition method and device based on millimeter wave radar point cloud

Abstract

A track gauge recognition method and device based on millimeter wave radar point cloud, the method comprising: acquiring millimeter wave radar point cloud collected by a millimeter wave radar at a front end of a railway vehicle, wherein the millimeter wave radar point cloud comprises a plurality of first point cloud points; extracting a plurality of second point cloud points located in a preset position interval from the plurality of first point cloud points; extracting a plurality of third point cloud points within a preset speed range from the plurality of second point cloud points; clustering the plurality of third point cloud points according to the distribution of the plurality of third point cloud points to obtain a plurality of skeleton points corresponding to a track center position; determining the direction of the skeleton points according to the positions of the third point cloud points within a preset range around the skeleton points; determining track gauge boundary points located on both sides of each skeleton point according to the position and direction of each skeleton point and the track width; and determining the track gauge according to the plurality of track gauge boundary points. The present application can accurately recognize the track gauge based on millimeter wave radar point cloud, and can overcome the influence of complex weather and light interference.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, and more specifically, to a method and apparatus for track clearance identification based on millimeter-wave radar point clouds. Background Technology

[0002] In intelligent driving systems for rail transit (or fully autonomous operation systems), the identification of the drivable track area ahead, i.e., the track clearance, is a necessary part of environmental perception. Currently, mainstream technologies are based on LiDAR and cameras, or a fusion of both, to delineate the track clearance. However, LiDAR and camera sensor equipment have difficulty solving problems related to complex weather and light interference. Summary of the Invention

[0003] The summary section introduces a series of simplified concepts, which will be further explained in detail in the detailed description section. The summary section of this invention is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0004] To address the shortcomings of existing technologies, the first aspect of this invention proposes a method for track boundary identification based on millimeter-wave radar point clouds, comprising:

[0005] Acquire millimeter-wave radar point cloud data collected by millimeter-wave radar, wherein the millimeter-wave radar point cloud includes multiple first point cloud points, and the millimeter-wave radar is installed at the front end of the rail vehicle;

[0006] Based on the location information of the point cloud points, extract a plurality of second point cloud points located in a preset location range from the plurality of first point cloud points;

[0007] Based on the velocity information of the second point cloud, extract multiple third point cloud points within a preset velocity range from the plurality of second point cloud points;

[0008] Based on the distribution of the multiple third point cloud points, the multiple third point cloud points are clustered to obtain multiple skeleton points corresponding to the center position of the orbit;

[0009] The orientation of the skeleton point is determined based on the position of the third point cloud point within a preset range around the skeleton point;

[0010] Based on the position and orientation of each skeleton point and the track width, determine the track boundary points located on both sides of each skeleton point;

[0011] The track clearance is determined based on multiple track clearance boundary points.

[0012] In some embodiments, clustering the plurality of third point cloud points according to their distribution to obtain a plurality of skeleton points corresponding to the orbital center position includes:

[0013] Determine a bounding box of a preset shape and size centered on each of the aforementioned third point cloud points;

[0014] For each of the third point cloud points, determine the total number of all third point cloud points within the bounding box;

[0015] The third point cloud point at the center of the bounding box whose total number of all third point cloud points is greater than a preset threshold is identified as a key point.

[0016] The skeleton points are determined based on the key points.

[0017] In some embodiments, determining the skeleton points based on the key points includes:

[0018] The intersection-union ratio of at least two bounding boxes is determined based on the number of all third point cloud points within at least two of the bounding boxes;

[0019] Merge at least two bounding boxes whose intersection-union ratio is greater than a preset threshold, and use the center point of the merged bounding box as the skeleton point.

[0020] In some embodiments, the merged bounding box includes all the third point cloud points within at least two bounding boxes where the intersection-union ratio is greater than a preset threshold.

[0021] In some embodiments, the bounding box is rectangular in shape, and the two sets of opposite sides of the bounding box are parallel to the horizontal and vertical axes of the millimeter-wave radar coordinate system, respectively.

[0022] In some embodiments, determining the orientation of the skeleton point based on the position of the third point cloud points within a preset range around the skeleton point includes:

[0023] The coordinates of multiple third point cloud points within a preset range around each skeleton point are fitted with a straight line, and the direction of the fitted straight line is taken as the direction of the skeleton point.

[0024] In some embodiments, determining the track boundary points located on both sides of each skeleton point based on the position and orientation of each skeleton point and the track width includes:

[0025] Determine a straight line centered on the skeleton point and perpendicular to the direction of the skeleton point;

[0026] Two track boundary points symmetrical to the skeleton point are determined on the straight line, such that the distance between the two track boundary points is equal to the track width.

[0027] In some embodiments, before extracting a plurality of third point cloud points within a preset speed range from the plurality of second point cloud points based on the speed information of the second point cloud points, the method further includes:

[0028] Determine the mean and mode of the velocities of the plurality of second point cloud points;

[0029] The preset speed range is determined based on the mean and mode of the speeds of the plurality of second point cloud points.

[0030] A second aspect of the present invention provides a track boundary identification device based on millimeter-wave radar point clouds. The device includes a memory and a processor. The memory stores a computer program that is run by the processor. When the computer program is run by the processor, it executes the track boundary identification method based on millimeter-wave radar point clouds as described above.

[0031] A third aspect of the present invention provides a computer storage medium storing a computer program thereon, characterized in that the computer program, when executed, implements the orbital boundary identification method based on millimeter-wave radar point clouds as described above.

[0032] A fourth aspect of the present invention provides a rail vehicle, the rail vehicle including a car body, a millimeter-wave radar disposed at the front of the car body, and a track boundary identification device based on millimeter-wave radar point cloud as described above.

[0033] The track clearance identification method based on millimeter-wave radar point clouds proposed in this invention achieves automatic identification of the track clearance area in front of rail transit vehicles through steps such as coarse extraction of point cloud position, point cloud velocity filtering, track skeleton point extraction, and track area clearance boundary point identification. This provides basic information for the detection of rail transit obstacles. Compared with LiDAR and camera sensor solutions, this method is more accurate in identifying track clearances and can also overcome the influence of complex weather and light interference. Attached Figure Description

[0034] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of embodiments thereof in conjunction with the accompanying drawings. The drawings are provided to further illustrate embodiments of the invention and form part of the specification. They are used together with the embodiments to explain the invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same parts or steps.

[0035] Figure 1This is a schematic flowchart of a track boundary identification method based on millimeter-wave radar point cloud according to an embodiment of the present invention;

[0036] Figure 2 This is a schematic diagram illustrating the relationship between a millimeter-wave radar and its orbit according to an embodiment of the present invention.

[0037] Figure 3 A schematic diagram of an orbital skeleton vector according to an embodiment of the present invention;

[0038] Figure 4 This is a schematic diagram illustrating the merging of bounding boxes according to an embodiment of the present invention;

[0039] Figure 5 This is a schematic diagram illustrating the relationship between track clearance boundary points and skeleton points according to an embodiment of the present invention;

[0040] Figure 6 This is a schematic block diagram of a track boundary identification device based on millimeter-wave radar point clouds according to an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application more apparent, exemplary embodiments according to this application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein. Based on the embodiments of this application described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of this application.

[0042] The following description provides numerous specific details to offer a more thorough understanding of this application. However, it will be apparent to those skilled in the art that this application can be practiced without one or more of these details. In other instances, certain technical features well-known in the art have not been described to avoid confusion with this application.

[0043] It should be understood that this application can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. Rather, providing these embodiments will make the disclosure thorough and complete, and will fully convey the scope of this application to those skilled in the art.

[0044] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising” and / or “including,” when used in this specification, identify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.

[0045] To fully understand this application, a detailed structure will be presented in the following description to illustrate the technical solution proposed in this application. Optional embodiments of this application are described in detail below; however, in addition to these detailed descriptions, this application may have other implementation methods.

[0046] The method, apparatus, and computer storage medium for track boundary identification based on millimeter-wave radar point clouds proposed in embodiments of the present invention will now be described with reference to the accompanying drawings. First, see... Figure 1 , Figure 1 A schematic flowchart of a trajectory boundary identification method 100 based on millimeter-wave radar point clouds according to an embodiment of the present invention is shown. Figure 1 As shown, the orbit boundary identification method 100 based on millimeter-wave radar point clouds according to an embodiment of the present invention includes the following steps:

[0047] In step S110, a millimeter-wave radar point cloud is acquired by the millimeter-wave radar. The millimeter-wave radar point cloud includes multiple first point cloud points. The millimeter-wave radar is installed at the front end of the rail vehicle.

[0048] In step S120, based on the location information of the first point cloud points, a plurality of second point cloud points located in a preset location range are extracted from the plurality of first point cloud points;

[0049] In step S130, based on the velocity information of the second point cloud point, a plurality of third point cloud points within a preset velocity range are extracted from the plurality of second point cloud points;

[0050] In step S140, the multiple third point cloud points are clustered according to their distribution to obtain multiple skeleton points corresponding to the center position of the orbit.

[0051] In step S150, the direction of the skeleton point is determined based on the position of the third point cloud point within a preset range around the skeleton point;

[0052] In step S160, the track boundary points located on both sides of each skeleton point are determined based on the position and orientation of each skeleton point and the track width.

[0053] In step S170, the track clearance is determined based on the plurality of track clearance boundary points.

[0054] Based on the advantages of millimeter-wave radar—its greater sensitivity to metallic targets, its ability to adapt to complex weather conditions, and its ability to overcome light interference—this invention utilizes millimeter-wave radar point clouds to identify track areas, thereby delineating track clearances. Compared to existing technologies that use lidar and cameras, this method provides more accurate track clearance identification and overcomes the effects of complex weather and light interference. The track clearance refers to the drivable area of ​​the track in front of the vehicle.

[0055] In step S110, the millimeter-wave radar point cloud acquired by the millimeter-wave radar located at the front of the rail vehicle is obtained. The onboard millimeter-wave radar is positioned at the center of the front of the rail vehicle, serving as the main detection radar. The millimeter-wave radar point cloud is structured data comprising multiple point cloud points. Each point cloud point contains information such as the target's planar coordinates, velocity, and radar cross-section (RCS) in the millimeter-wave radar coordinate system. Typically, a single frame of the millimeter-wave radar point cloud can contain 30-255 or more point cloud points. The millimeter-wave radar can transmit the millimeter-wave radar point cloud to a track clearance identification device based on the millimeter-wave radar point cloud via a communication protocol for track clearance identification. For ease of description, the point cloud points in the millimeter-wave radar point cloud acquired by the millimeter-wave radar are defined as the first point cloud point.

[0056] In step S120, based on the location information of the first point cloud points, multiple second point cloud points located within a preset location range are extracted from the millimeter-wave radar point cloud. This step is used to perform coarse extraction of the millimeter-wave radar point cloud. Based on the characteristics of a fixed orbit width and infinite forward extension, point cloud points that cannot belong to the orbit are initially excluded, while multiple second point cloud points that may belong to the orbit are retained.

[0057] like Figure 2 As shown, in the XOY coordinate system of the millimeter-wave radar, the distance D between the outer boundaries L1 and L2 of the two orbits is... L It is always greater than the distance D from the center of the millimeter-wave radar to the outer boundary of any orbit. R Additionally, the width of the vehicle's outer contour boundary is D. T In general, D T >D L Let ξ be an empirical parameter for track clearance width, the value of which varies depending on the track and vehicle type. Let P be the millimeter-wave radar point cloud obtained in step S110, then P is the set of second point cloud points coarsely extracted from the first point cloud based on the position information of the first point cloud points. B It can be represented as:

[0058] P B ={(x,y)|x∈[0,X]}limit ],y∈[-D R -ξ,D R +ξ],D R +ξ>D T D T >0,X limit >0}, and

[0059] In the above formula, X iimit This represents the maximum value in the positive x-axis direction of the millimeter-wave radar's coordinate system. Its specific value can be preset, for example, based on factors such as the millimeter-wave radar's parameters and orbital parameters.

[0060] In step S130, velocity filtering is performed on the coarsely extracted millimeter-wave radar point cloud. Specifically, based on the velocity information of each second point cloud point, multiple third point cloud points within a preset velocity range are extracted from the multiple second point cloud points.

[0061] The velocity filtering is based on the assumption that the relative velocity of the track point cloud points measured by millimeter-wave radar is consistent with the vehicle speed. This step eliminates noise data that cannot be track point cloud points in the most direct way. It is fast to calculate, the algorithm is relatively intuitive, and it can effectively eliminate interference from the point cloud of moving vehicles ahead.

[0062] Millimeter-wave radar inherently possesses the capability to measure the velocity of targets. In each frame of the millimeter-wave radar point cloud P, any point p(x,y,v)∈P always contains its relative velocity v with respect to the millimeter-wave radar itself (i.e., the rail vehicle). The preset velocity range can be obtained by statistically analyzing the velocities of all the second point cloud points; specifically, the average velocity of the multiple second point cloud points is determined. mode V mod Based on the average velocity of multiple second-point cloud points mode V mod Determine the preset speed range, and calculate the upper and lower boundary thresholds T of the preset speed range for speed filtering using the following formula. max and T min :

[0063]

[0064]

[0065] For all P B The points in the middle are filtered using the following formula to obtain the velocity in T. max and T min The set P of the third point cloud points between v :

[0066]

[0067] P v The third point cloud in the diagram represents all possible track point cloud points within a general track area, and their relative speeds are related to the vehicle's speed. Speed ​​filtering can filter out a large number of track point cloud points while eliminating noise points with abnormal speeds or interference from vehicle point cloud points with relative motion speeds.

[0068] In step S140, the multiple third point cloud points are clustered according to their distribution to obtain multiple skeleton points corresponding to the orbit center. The set P of the third point cloud points extracted in the previous step... v The dataset contains potential orbital point cloud points, as well as other background points. Feature recognition and analysis are still required to extract the orbital skeleton points. Orbital skeleton extraction refers to the process of extracting orbital skeleton points from the set P of third-party point cloud points. v The process further extracts track skeleton points, which correspond to the track center. Next, the orientation of each track skeleton point needs to be determined to obtain the skeleton description vector used to describe the track skeleton. For example... Figure 3 As shown, the orbital skeleton consists of multiple skeleton description vectors. It includes its starting coordinates (x, y) in the millimeter-wave radar coordinate system and the angle θ with the x-axis. The starting point of the skeleton description vector is the orbit skeleton point.

[0069] For example, the extraction of track skeleton points includes point cloud clustering and merging. Point cloud clustering is based on the characteristic that the track extends infinitely along the running direction while its width is fixed. It clusters the third point cloud points according to the track shape characteristics, extracting point cloud clusters that are denser along the length direction. This removes remaining noise points within a preset track boundary, effectively eliminating track edge noise. Merging the clustered point clouds can remove many duplicate skeleton points while preserving the local spatial features of the millimeter-wave radar point cloud to the greatest extent possible.

[0070] Specifically, a bounding box of a preset shape and size is determined centered on each third point cloud point. For each third point cloud point, the total number of third point cloud points within the bounding box is determined. The third point cloud point at the center of the bounding box where the total number of third point cloud points exceeds a preset threshold is identified as a keypoint. Skeleton points can then be determined based on these keypoints. If the number of third point cloud points within a bounding box exceeds a preset threshold, it indicates a relatively dense distribution of third point cloud points within that bounding box, making the third point cloud point at the center of that bounding box more likely to be a track skeleton point.

[0071] For example, considering the shape of the orbit, the bounding box is rectangular, with its two pairs of opposite sides parallel to the horizontal and vertical axes of the millimeter-wave radar coordinate system, respectively. Figure 3 The x-axis and y-axis in the diagram. For example... Figure 4As shown, for the set P of the third point cloud points v Any third point P in the cloud i (x i ,y i The rectangle centered on it can be represented as R. i (x i ,y i Let R be the width of the rectangle along the x-axis and the width along the y-axis, respectively. i When the number of third point cloud points within the rectangle exceeds a preset threshold μ, then the center point P of the rectangle is considered to be... i As the key point, the rectangle R i The third point in the cloud is called its center point P. i neighborhood point P j Furthermore, if P... j The rectangular frame R centered on j If the number of points in the third inner cloud is also greater than μ, then P j Also a key point, and P j neighborhood points and P i All neighborhood points belong to the same category; this step can be called clustering. After clustering, we can obtain a set P of all key points whose number of neighborhood points is greater than μ. c .

[0072] Obtain the set of key points P c Afterwards, in P c Based on this, skeleton points are extracted. As mentioned above, the keypoint set P c Each keypoint in the diagram represents the center point of the corresponding rectangle, while the track skeleton points are the centroids of all points within the rectangle, i.e., their average positions. Additionally, there may be... Figure 4 As shown, multiple rectangles overlap in height. Therefore, this embodiment of the invention uses a rectangle merging method based on intersection-union ratio to remove duplicate key points in the overlapping bounding boxes.

[0073] Specifically, let P c The bounding box centered at any key point in the bounding box is R. i Then it is connected to any other bounding box R j The intersection ratio of I ij for:

[0074]

[0075] With R i Merged bounding box R j It will no longer merge with other bounding boxes. Let there be N bounding boxes with R... i The intersection ratio of I ij R > 0.9 jThen, these N+1 bounding boxes can be merged to obtain the merged bounding box R′. i The center point of the merged bounding box is the skeleton point.

[0076] The merged bounding box includes all third point cloud points within at least two bounding boxes with an intersection-union ratio greater than a preset threshold. Therefore, the merged bounding box R′ i Width on the x-axis i and the width h on the y-axis i They are respectively:

[0077] w i =abs(max(x) n )-min(x n ))

[0078] h i =abs(max(y n )-min(y n ))

[0079] In the above formula, x n and y n Let x and y be the x-coordinates and y-coordinates of all third-point cloud points within these N+1 bounding boxes, respectively. max(x n ) and min(x n ) are the maximum and minimum x-axis coordinates of all third-point cloud points in the N+1 bounding boxes, respectively. n ) and min(y n Let be the maximum and minimum y-axis coordinates of all third-point cloud points, and let abs represent the absolute value. Then the merged bounding box R′ i center point p′ i The coordinates are:

[0080]

[0081] Where, p′ i These are the key points corresponding to the merged bounding boxes, i.e., the skeleton points. The x-axis coordinate of a skeleton point is the average of the maximum and minimum x-axis coordinates among all third-point cloud points in at least two bounding boxes, and the y-axis coordinate of a skeleton point is the average of the maximum and minimum y-axis coordinates among all third-point cloud points in at least two bounding boxes.

[0082] To describe the orientation of the track skeleton, the orientation of the skeleton points also needs to be determined. Specifically, in step S150, the orientation of the skeleton point is determined based on the position of a third point cloud point within a preset range around the skeleton point. For skeleton point p... i For bounding box R iAll third-point cloud points within the system are fitted with straight lines using a fitting algorithm such as least squares. The acute angle θ between the fitted line and the x-axis of the millimeter-wave radar coordinate system represents the direction of the skeleton points. Repeating the above calculations yields P. c The orientation of all skeleton points. The coordinates and orientations of the skeleton points constitute the skeleton vector. Let B be the set of skeleton vectors, then:

[0083]

[0084] The orbital skeleton can be described by combining multiple skeleton vectors, and the orbital boundary can be determined based on the orbital skeleton and the prior orbital boundary width. This method can not only highly fit the orbital boundary for straight segments, but also generate the orbital boundary for curved segments, greatly improving the usability of millimeter-wave radar point clouds.

[0085] In step S160, based on the position and orientation of each skeleton point in the track skeleton vector set B, and the track width D... L Determine the track boundary points located on both sides of each skeleton point. Specifically, such as... Figure 5 As shown, first, a straight line is determined with the skeleton point as the center and perpendicular to the direction of the skeleton point; then, two track boundary points symmetrical about the skeleton point are determined on this line, such that the distance between the two track boundary points is equal to the track width. That is, for each skeleton vector... The two corresponding orbital boundary points q can be generated according to the following formula. i ,q′ i :

[0086]

[0087]

[0088] Where, x i ,y i They are skeleton vectors The x-axis and y-axis coordinates.

[0089] In addition to the track boundary points generated based on the above formula, it also includes, for example, Figure 5 The starting vertex pair (q0, q′0) of the orbital boundary shown in the figure, wherein, Therefore, after all the above calculation steps, multiple track clearance boundary points in the millimeter-wave radar coordinate system can be obtained, that is, the track traffic clearance in front of the vehicle is identified through the millimeter-wave radar point cloud.

[0090] In summary, the track clearance identification method 100 based on millimeter-wave radar point cloud data of this invention achieves automatic identification of the track clearance area in front of rail transit vehicles through steps such as coarse extraction of point cloud position, point cloud velocity filtering, track skeleton point extraction, and track area clearance boundary point identification. This provides basic information for the detection of rail transit obstacles. Compared with the existing technology that uses lidar and cameras, the identification of track clearance is more accurate and can also overcome the influence of complex weather and light interference.

[0091] This invention also provides a track clearance identification device based on millimeter-wave radar point clouds, which can be used to implement the track clearance identification method 100 based on millimeter-wave radar point clouds described above. See also Figure 6 , Figure 6 A schematic block diagram of a track clearance identification device 600 based on millimeter-wave radar point clouds according to an embodiment of the present invention is shown. The track clearance identification device 600 based on millimeter-wave radar point clouds according to this embodiment of the present invention can be implemented as a controller for a rail vehicle, a cloud processor, or any electronic device.

[0092] like Figure 6 As shown, the track boundary identification device 600 based on millimeter-wave radar point cloud includes a memory 610, a processor 620, and a computer program stored in the memory 610 and running on the processor 620. When the processor 620 executes the computer program, it can implement the track boundary identification method 100 based on millimeter-wave radar point cloud as described above.

[0093] The memory 610 is a memory for storing processor-executable instructions, such as processor-executable program instructions for implementing corresponding steps in the orbital boundary identification method 100 based on millimeter-wave radar point clouds according to an embodiment of the present invention. The memory 610 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0094] The processor 620 can execute the program instructions stored in the memory 610 to implement the functions (implemented by the processor) in the embodiments of the present invention described herein, and / or other desired functions, such as performing corresponding steps of the orbital boundary identification method 100 based on millimeter-wave radar point clouds according to an embodiment of the present invention. Various application programs and various data, such as various data used and / or generated by the application programs, can also be stored in the computer-readable storage medium.

[0095] Processor 620 may be a central processing unit (CPU), graphics processing unit (GPU), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other form of processing unit with data processing and / or instruction execution capabilities, and may control other components in the millimeter-wave radar point cloud-based track boundary identification device 600 to perform desired functions. Processor 620 is capable of executing the instructions stored in memory 610 to perform the path planning method described herein. For example, processor 620 may include one or more embedded processors, processor cores, microprocessors, logic circuits, hardware finite state machines (FSMs), digital signal processors (DSPs), or combinations thereof.

[0096] This invention also proposes a computer storage medium storing a computer program. When executed, this computer program can implement the orbital boundary identification method 100 based on millimeter-wave radar point clouds according to this invention. The computer storage medium may include, for example, a memory card of a smartphone, a storage component of a tablet computer, a hard disk of a personal computer, a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a portable compact disc read-only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

[0097] This invention also proposes a rail vehicle, which includes a car body, a millimeter-wave radar disposed at the front of the car body, and a track clearance identification device 600 based on millimeter-wave radar point clouds as described above. The millimeter-wave radar is used to collect millimeter-wave radar point clouds at the front of the rail vehicle and transmits these point clouds to the track clearance identification device 600. The track clearance identification device 600 executes a track clearance identification method 100 based on millimeter-wave radar point clouds to determine the track clearance based on the millimeter-wave radar point clouds.

[0098] The track clearance identification device, computer storage medium, and track vehicle based on millimeter-wave radar point cloud in this embodiment of the invention are used to implement the track clearance identification method 100 based on millimeter-wave radar point cloud described above, and therefore also have similar advantages.

[0099] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of this application. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of this application. All such changes and modifications are intended to be included within the scope of this application as claimed in the appended claims.

[0100] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0102] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0103] Similarly, it should be understood that, in order to streamline this application and aid in understanding one or more of the various inventive aspects, features of this application may sometimes be grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of this application. However, this approach should not be construed as reflecting an intention that the claimed application requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with features fewer than all features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.

[0104] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus so disclosed can be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0105] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of this application and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0106] The various component embodiments of this application can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules according to the embodiments of this application. This application can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such an implementation of this application can be stored on a computer-readable medium, or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0107] It should be noted that the above embodiments are illustrative of this application and not limiting of it, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. This application can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0108] The above description is merely a specific embodiment or illustration of the embodiments of this application. The scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. The scope of protection of this application shall be determined by the scope of the claims.

Claims

1. A method for track boundary identification based on millimeter-wave radar point clouds, characterized in that, The method includes: Acquire millimeter-wave radar point cloud data collected by millimeter-wave radar, wherein the millimeter-wave radar point cloud includes multiple first point cloud points, and the millimeter-wave radar is installed at the front end of the rail vehicle. Based on the location information of the first point cloud point, extract a plurality of second point cloud points located in a preset location range from the plurality of first point cloud points; Based on the velocity information of the second point cloud, extract multiple third point cloud points within a preset velocity range from the plurality of second point cloud points; Based on the distribution of the multiple third point cloud points, the multiple third point cloud points are clustered to obtain multiple skeleton points corresponding to the orbital center position, specifically including: Determine a bounding box of a preset shape and size centered on each of the aforementioned third point cloud points; For each of the third point cloud points, determine the total number of all third point cloud points within the bounding box; The third point cloud point at the center of the bounding box whose total number of all third point cloud points is greater than a preset threshold is identified as a key point. The intersection-union ratio of at least two bounding boxes is determined based on the number of all third point cloud points within at least two of the bounding boxes; Merge at least two bounding boxes whose intersection-union ratio is greater than a preset threshold, and use the center point of the merged bounding box as the skeleton point. The orientation of the skeleton point is determined based on the position of the third point cloud point within a preset range around the skeleton point; Based on the position and orientation of each skeleton point and the track width, determine the track boundary points located on both sides of each skeleton point; The track clearance is determined based on multiple track clearance boundary points.

2. The orbital boundary identification method based on millimeter-wave radar point clouds according to claim 1, characterized in that, The merged bounding box includes all the third point cloud points within at least two bounding boxes where the intersection-union ratio is greater than a preset threshold.

3. The orbital boundary identification method based on millimeter-wave radar point clouds according to any one of claims 1 or 2, characterized in that, The bounding box is rectangular in shape, and the two sets of opposite sides of the bounding box are parallel to the horizontal and vertical axes of the millimeter-wave radar coordinate system, respectively.

4. The orbital boundary identification method based on millimeter-wave radar point clouds according to claim 1, characterized in that, The step of determining the orientation of the skeleton point based on the position of the third point cloud points within a preset range around the skeleton point includes: The coordinates of multiple third point cloud points within a preset range around each skeleton point are fitted with a straight line, and the direction of the fitted straight line is taken as the direction of the skeleton point.

5. The orbital boundary identification method based on millimeter-wave radar point clouds according to claim 1, characterized in that, The step of determining the track boundary points located on both sides of each skeleton point based on the position and orientation of each skeleton point and the track width includes: Determine a straight line centered on the skeleton point and perpendicular to the direction of the skeleton point; Two track boundary points symmetrical to the skeleton point are determined on the straight line, such that the distance between the two track boundary points is equal to the track width.

6. The orbital boundary identification method based on millimeter-wave radar point clouds according to claim 1, characterized in that, Before extracting multiple third point cloud points within a preset speed range from the plurality of second point cloud points based on the speed information of the second point cloud points, the method further includes: Determine the mean and mode of the velocities of the plurality of second point cloud points; The preset speed range is determined based on the mean and mode of the speeds of the plurality of second point cloud points.

7. A track boundary identification device based on millimeter-wave radar point clouds, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program, when executed by the processor, performs the orbital boundary identification method based on millimeter-wave radar point clouds as described in any one of claims 1-6.

8. A computer storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the orbital boundary identification method based on millimeter-wave radar point clouds as described in any one of claims 1-6.

9. A rail vehicle, characterized in that, The rail vehicle includes a car body, a millimeter-wave radar disposed at the front of the car body, and a track boundary identification device based on millimeter-wave radar point cloud as described in claim 7.

Images