Railway detection method, device and equipment for inspection trolley, medium and product

Abstract

The invention discloses a railway detection method, device and equipment for an inspection trolley, a medium and a product, and relates to the technical field of unmanned driving. The method comprises the following steps: acquiring radar data in a preset sensing area in front of an inspection trolley; determining a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in the advancing direction, and mapping the nearest reflection point and the other reflection points into a target image; and determining advancing attention information of the inspection trolley according to the mapped target image. According to the technical scheme provided by the invention, the radar data in the specific area is used and mapped to the plane image to analyze the advancing attention information of the inspection trolley, so that refined sensing of attention events in different areas is realized, and the detection and sensing capability of the inspection trolley to the railway state is greatly improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of unmanned driving technology, and in particular to a railway inspection method, device, equipment, medium and product for an inspection vehicle. Background Art

[0002] The safety inspection and maintenance of railway lines has always been one of the important tasks of the railway department. With the continuous maturity of autonomous driving technology, low-speed unmanned railway inspection vehicles are being put into use. During the movement of the inspection vehicle, it is necessary to timely perceive the status of the railway track to ensure driving safety. For example, obstacles on the track surface will directly affect the smooth driving of the inspection vehicle. When entering a curved track or a switch, the inspection vehicle is generally light and the vehicle body is prone to tilting at too fast a speed, which indirectly affects driving safety. Existing technologies usually detect the status of the railway based on vision. Detection itself is a difficult point in the safety perception of railway autonomous driving. At the same time, due to the influence of factors such as the outdoor working environment, the existing detection of the railway status still has problems of missed detection or low accuracy. Summary of the Invention

[0003] The embodiments of the present invention provide a railway detection method, device, equipment, medium and product for an inspection vehicle to enhance the inspection vehicle's ability to detect and perceive railway conditions.

[0004] In a first aspect, an embodiment of the present invention provides a railway inspection method for an inspection vehicle, the method comprising:

[0005] Obtain radar data within the preset sensing area in front of the inspection vehicle;

[0006] Determining a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in the direction of travel, and mapping the nearest reflection point and the other reflection points into a target image;

[0007] The traveling attention information of the inspection vehicle is determined according to the mapped target image.

[0008] Optionally, mapping the nearest reflection point and the other reflection points into a target image includes:

[0009] Mapping the nearest reflection point to a preset coordinate position in the target image, and setting the pixel points in the target area around the preset coordinate position to 255, wherein the target image is initially an image with all zero pixels;

[0010] Determine the mapping position of the corresponding other reflection points in the target image based on the relative position of the other reflection points and the nearest reflection point in the travel direction and the relative position in the vertical track direction, and set the pixel points in the target area around the mapping position to 255.

[0011] Optionally, the target area is an N×N rectangular area, where N is the product of the radar distance of the corresponding reflection point and the radar horizontal angular resolution.

[0012] Optionally, determining the traveling attention information of the inspection vehicle according to the mapped target image includes:

[0013] Performing region segmentation on the mapped target image;

[0014] The maximum segmented region area obtained by segmentation is compared with the preset area threshold to determine whether there is a target attention event.

[0015] Optionally, the preset sensing area includes a preset track surface obstacle sensing area and a preset curved track switch sensing area; correspondingly, the traveling attention information includes track surface obstacle information and curved track switch information.

[0016] Optionally, a multi-line laser radar is provided above the front of the inspection vehicle, and the track surface obstacle information includes first track surface obstacle information determined based on multi-line radar data collected by the multi-line laser radar; and / or, a single-line laser radar is provided below the front of the inspection vehicle, and the track surface obstacle information includes second track surface obstacle information determined based on single-line radar data collected by the single-line laser radar.

[0017] In a second aspect, an embodiment of the present invention further provides a railway inspection device for an inspection vehicle, the device comprising:

[0018] Radar data acquisition module, used to obtain radar data within the preset sensing area in front of the inspection vehicle;

[0019] a reflection point mapping module, configured to determine a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in a direction of travel, and map the nearest reflection point and the other reflection points to a target image;

[0020] The attention information determination module is used to determine the traveling attention information of the inspection vehicle according to the mapped target image.

[0021] In a third aspect, an embodiment of the present invention further provides a patrol vehicle device, the patrol vehicle device comprising:

[0022] one or more processors;

[0023] a memory for storing one or more programs;

[0024] When the one or more programs are executed by the one or more processors, the one or more processors implement the railway inspection method for inspection vehicles provided in any embodiment of the present invention.

[0025] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the railway inspection method for inspection vehicles provided by any embodiment of the present invention.

[0026] In a fifth aspect, an embodiment of the present invention further provides a computer program product, which includes a computer program, and when the program is executed by a processor, it implements the railway inspection method for inspection vehicles provided by any embodiment of the present invention.

[0027] An embodiment of the present invention provides a railway inspection method for inspection vehicles. The method first acquires radar data within a preset sensing area in front of the inspection vehicle. It then determines the closest reflection point in the radar data and other reflection points within a preset range from the closest reflection point in the direction of travel. The closest reflection point and all other reflection points are mapped into a target image. The inspection vehicle's travel attention information is then determined based on the mapped target image. The railway inspection method for inspection vehicles provided by an embodiment of the present invention uses radar data within a specific area and maps it onto a planar image to analyze the inspection vehicle's travel attention information. This method achieves refined regional perception of attention events, significantly improving the inspection vehicle's ability to detect and perceive railway conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of a railway inspection method for an inspection vehicle provided in the first embodiment of the present invention;

[0029] Figure 2 A schematic diagram of an exemplary distribution of preset sensing areas provided in the first embodiment of the present invention;

[0030] Figure 3 A side view of the inspection vehicle structure provided in the first embodiment of the present invention;

[0031] Figure 4 A front view of the inspection vehicle structure provided in the first embodiment of the present invention;

[0032] Figure 5 A schematic structural diagram of a railway inspection device for inspection vehicles provided in the second embodiment of the present invention;

[0033] Figure 6 This is a structural diagram of the inspection vehicle equipment provided in Example 3 of the present invention. DETAILED DESCRIPTION

[0034] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0035] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the steps as sequential processes, many of the steps can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the steps can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0036] Example 1

[0037] Figure 1 This is a flow chart of a railway inspection method for inspection vehicles provided in the first embodiment of the present invention. This embodiment is applicable to situations where inspection vehicles detect railway conditions for driving safety considerations when conducting railway inspections. This method can be executed by a railway inspection device for inspection vehicles provided in the embodiment of the present invention. The device can be implemented in hardware and / or software and can generally be integrated into the inspection vehicle equipment. Figure 1 As shown, the specific steps include:

[0038] S11. Acquire radar data within a preset sensing area in front of the inspection vehicle.

[0039] S12: Determine the nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in the direction of travel, and map the nearest reflection point and the other reflection points into a target image.

[0040] S13. Determine the traveling attention information of the inspection vehicle according to the mapped target image.

[0041] Specifically, a laser radar can be set at the front of the inspection vehicle to collect radar raw data in front of the inspection vehicle, and the collected radar raw data can be first converted to the vehicle center coordinate system for subsequent processing. Specifically, the rotation and translation matrix of the laser radar relative to the vehicle center can be calculated in advance based on the installation position of the laser radar, so that the coordinate conversion process can be achieved by multiplying the radar raw data with the rotation and translation matrix. It is also possible to extract valid data within a certain range from the center of the vehicle body for subsequent processing based on the effective perception distance of the laser radar. For example, for a single-line laser radar, valid data within 30 meters can be extracted, and for a multi-line laser radar, valid data within 50 meters can be extracted.

[0042] The radar raw data can then be filtered to obtain radar data within the required preset sensing area, thereby removing excessive background interference for the specified attention event, saving computing resources and improving the accuracy of the detection results. Optionally, the preset sensing area includes a preset track surface obstacle sensing area and a preset curved track switch sensing area. The preset track surface obstacle sensing area can further include a left track surface obstacle sensing area and a right track surface obstacle sensing area, so as to respectively detect obstacles on the track surfaces on both sides and the curved track switch in front. The preset sensing area can be set based on the empirical value obtained during the actual measurement process. The preset track surface obstacle sensing area can specifically be an area extending along the travel direction at a certain width and a certain height directly above each track surface at the current position of the inspection vehicle. The preset curved track switch sensing area can specifically be an area extending along the travel direction at a certain width and a certain height centered between the two tracks at the current position of the inspection vehicle. Based on the fixed spacing of railway rails (usually 1435mm), the width and height of the preset track surface obstacle sensing area are preferably 20-60cm, so as to cover the track surface area as much as possible and reduce the problem of false alarms of obstacle warnings caused by external obstacles without hidden dangers. The width and height of the preset curved track and switch sensing area are preferably 60-100cm and 10-30cm, so as to detect the presence of curved tracks or switches in time. The exemplary distribution of the preset sensing areas is as follows: Figure 2 As shown in the figure, the left and right track obstacle sensing areas A and B are 40 cm wide and 5 cm apart above the track surface. The preset curved switch sensing area C is 80 cm wide and 15 cm high, 5 cm apart below the track surface. Since the positions of the preset sensing areas are known relative to the vehicle center, the radar raw data can be filtered based on the vehicle center coordinate system.

[0043] After obtaining the required radar data, the reflection points can be mapped into a two-dimensional target image. The mapped target image can reflect the objects within the preset sensing area, so as to analyze the travel warning events in front of the inspection vehicle, such as whether there are obstacles, curved tracks or switches, etc. Specifically, when there are reflection points, they can first be sorted from near to far in the direction of travel to determine the closest reflection point and other reflection points within a preset range from the closest reflection point in the direction of travel. These reflection points can reflect the situation of the closest reflecting object. The preset range can be set according to the specific travel warning event being analyzed, preferably 1-2 meters, so as to ensure calculation speed and improve real-time performance based on accurate analysis.

[0044] The coordinates of the nearest reflection point and other reflection points can then be directly mapped to the target image. For example, in the vehicle center coordinate system, the travel direction can be used as the X-axis, the vertical track direction as the Y-axis, and the height direction as the Z-axis. The Z coordinate of the reflection point is not important, while the x and y coordinates of the reflection point can be used to calculate the distance and direction from the center of the vehicle. Specifically, the mapping can be based on the x and y coordinates of each reflection point. The x and y coordinates of each reflection point can be uniformly converted to a value in a specified unit (such as mm) to correspond to a single pixel in the target image. For example, if the reflection point coordinates are (3m, 3m), they can be converted to (3000mm, 3000mm), which can be mapped to the pixel position (3000, 3000) in the target image.

[0045] Alternatively, the nearest reflection point can be first mapped to a preset coordinate position in the target image. Then, using the nearest reflection point as a reference, all other reflection points can be mapped to the target image based on the positional relationship between each other reflection point and the nearest reflection point. The preset coordinate position can be (0, h / 2), where h represents the height of the target image. This allows the nearest reflection point to be mapped to the position in the target image with the smallest horizontal coordinate and the center vertical coordinate, so that the distribution of the other reflection points is symmetrical about the nearest reflection point along the vertical axis. For example, if the target image size is 1200×400, the nearest reflection point is mapped to pixel position (0, 200). If one of the other reflection points has the same y-coordinate as the nearest reflection point and an x-coordinate that is 0.6m greater than the nearest reflection point, which translates to 600mm, then the other reflection point can be mapped to pixel position (600, 200). This mapping method effectively reduces the required target image size while ensuring that the nearest reflection point and all other reflection points are mapped to the target image, thereby reducing memory usage and improving computational speed.

[0046] After the mapping is completed, the mapping points in the target image can be used to reflect the situation of the reflective objects in the preset perception area. For example, the contour of the reflective object can be estimated based on the mapping points. Specifically, the mapping points with a close distance can be used as the contour points of the same reflective object and connected, so that the contours of one or more reflective objects can be obtained based on the mapping points.

[0047] Optionally, mapping the nearest reflection point and the other reflection points into the target image includes: mapping the nearest reflection point to a preset coordinate position in the target image, and setting the pixel points in the target area around the preset coordinate position to 255, and the target image is initially a full 0-pixel image; determining the corresponding mapping position of the other reflection points in the target image based on the relative position of the other reflection points and the nearest reflection point in the direction of travel and the relative position in the vertical track direction, and setting the pixel points in the target area around the mapping position to 255.

[0048] Specifically, each time mapping begins, a target image with all 0 (black) pixels can be pre-initialized. After mapping the nearest reflection point to a preset coordinate position, the pixels in the target area surrounding the preset coordinate position in the target image can be set to 255 (white). The corresponding mapping positions of other reflection points in the target image are then determined based on their relative positions to the nearest reflection point in the travel direction (X-axis) and in the perpendicular track direction (Y-axis). Each other reflection point is then mapped to its corresponding mapping position, and the pixels in the target area surrounding the mapping position are similarly set to 255. After mapping is complete, the white points in the target image clearly represent the reflective objects within the preset sensing area, and the binarized result is more easily processed by various algorithms. The target area represents the range of possible reflective objects around the mapping position of each reflection point. Because LiDAR uses a sector scanning method with a certain angular resolution, scanning at fixed angular intervals, there is a certain scanning interval between the two closest scanning points. Therefore, reflective objects may exist within the scanning interval around a reflection point. By setting the pixel values ​​within the target area to 255, the areas of the same reflective object in the target image can be made as connected as possible, allowing for direct determination of the outline of the reflective object within the preset perception area, thus facilitating analysis of moving warning events. The scanning interval can be determined based on the product of the radar position of the reflection point and the radar's horizontal angular resolution. For example, if a lidar collects data every 0.1 radian, i.e., the radar's horizontal angular resolution is 0.1 radian, then at a distance of 1 meter from the radar, the scanning interval is 0.1 × 1m = 0.1m. Furthermore, the target area can be determined based on the mapped position of the corresponding reflection point and the corresponding scanning interval. Specifically, it can be a circular area centered at the mapped position of the corresponding reflection point and with the corresponding scanning interval as the radius.

[0049] Optionally, the target area is an N×N rectangular area centered at the mapped location of the corresponding reflection point, where N is the product of the radar range of the corresponding reflection point and the radar's horizontal angular resolution, i.e., the scanning interval, to facilitate algorithm processing. The scanning interval can be uniformly converted to a value in a specified unit (e.g., mm) to determine the corresponding pixel location in the target image.

[0050] After the mapping is completed, the traveling attention events of the inspection car can be analyzed based on the outline of the reflective object in the target image to determine the traveling attention information. Optionally, corresponding to the preset track surface obstacle perception area and the preset curved track switch perception area, the traveling attention information may include track surface obstacle information and curved track switch information, and further the track surface obstacle information may include left track surface obstacle information and right track surface obstacle information. Specifically, it can be determined based on the size of the outline whether there is an obstacle with hidden dangers, or whether there is a curved track or switch, etc., and then the distance and angle of the obstacle or curved track switch can be determined for further analysis, so as to adaptively adjust the traveling speed of the inspection car. By dividing different perception areas specifically for corresponding traveling attention event analysis, other useless and potentially interfering areas can be better filtered out, thereby improving the accuracy of the analysis results.

[0051] Optionally, determining the inspection vehicle's travel attention information based on the mapped target image includes: segmenting the mapped target image; and comparing the area of ​​the largest segmented region obtained by segmentation with a preset area threshold to determine whether a target attention event exists. Specifically, the target image obtained by the binarization scheme can be directly segmented using the findContours algorithm in OpenCV to obtain subregions of each reflective object in the target image, and the subregions can be sorted by area size. The areas of each subregion can then be compared with a preset area threshold, and if a subregion exists that has an area greater than the preset area threshold, a target attention event can be determined to exist. Specifically, the area of ​​the largest segmented region can be compared first. If the area of ​​the largest segmented region is not greater than the preset area threshold, it can be directly determined that a target attention event does not exist without comparing other subregions. If the area of ​​the largest segmented region is greater than the preset area threshold, the other subregions can be compared sequentially to obtain all travel attention information. Of course, it is also possible to determine only whether the area of ​​the largest segmented region is greater than the preset area threshold, such as to determine only the presence of an obstacle, a curved track, or a switch. Among them, target attention events may include the appearance of obstacles, the existence of curved tracks or switches, etc. The preset area thresholds for different target attention events can be set to be the same or different, preferably 2×N×N, where N is the product of the radar distance of the corresponding reflecting object and the radar horizontal angular resolution.

[0052] In an optional embodiment, a multi-line laser radar is provided above the front of the inspection vehicle, and the track surface obstacle information includes first track surface obstacle information determined based on multi-line radar data collected by the multi-line laser radar; and / or, a single-line laser radar is provided below the front of the inspection vehicle, and the track surface obstacle information includes second track surface obstacle information determined based on single-line radar data collected by the single-line laser radar.

[0053] Specifically, such as Figure 3 and Figure 4 As shown, a multi-line laser radar 101 can be installed in the middle position above the vehicle head, and a single-line laser radar 102 can be installed in the middle position below the vehicle head. The scanning direction of each laser radar is as follows: Figure 3 Specifically, the multi-line laser radar 101 can be set at a height of 60-200 cm from the rail surface, preferably 82 cm, and the single-line laser radar 102 can be set at a height of 5-10 cm from the rail surface, preferably 5 cm, and installed horizontally to achieve better detection.

[0054] For track surface obstacle detection, a driving warning event, single-line radar data can be used to detect small obstacles at close range, generating secondary track surface obstacle information. Multi-line radar data (multi-line lidar has a longer range) can be used to detect medium-to-large obstacles at long range, generating primary track surface obstacle information. By utilizing a combined radar sensing approach, the advantages of both single-line and multi-line radars can be fully utilized, improving the accuracy of the final obstacle analysis results.

[0055] For example, for close-range small-volume obstacle detection, the left track surface obstacle detection is taken as an example to illustrate. The single-line radar data in the left track surface obstacle sensing area is filtered. If there is a reflection point in the area, the reflection points are sorted in order from near to far in the direction of travel. A 1200×400 full-pixel target image is pre-constructed. The height of 400 corresponds to the 40cm width of the left track surface obstacle sensing area, and the width of 1200 is the above-mentioned preset range, that is, corresponding to the 1.2m travel direction range. The nearest reflection point is mapped to the preset coordinate position (0, 200) in the target image, and the nearby N×N target area pixels are mapped to the preset coordinate position (0, 200). All points are set to 255. All other reflection points within a 1.2m distance from the nearest reflection point in the direction of travel are then mapped into the target image. The mapping positions are determined as above. The pixel values ​​of the N×N target area near each mapping position are all set to 255. Region segmentation is then performed, and the resulting sub-regions are sorted by area. The area of ​​the largest segmented region is then compared with a preset area threshold of 2×N×N. If the area of ​​the largest segmented region is greater than 2×N×N, it is considered that there is a small, close-range obstacle on the left track surface. The distance and angle of the obstacle are calculated as the second track surface obstacle information. Otherwise, it is considered to be noise interference. The process for right track surface obstacle detection is similar to that for the left track surface, and only the single-line radar data within the right track surface obstacle sensing area is used.

[0056] For long-distance medium and large-volume obstacle detection, the left track surface obstacle detection is taken as an example to illustrate. The multi-line radar data in the left track surface obstacle sensing area is filtered. If there is a reflection point in the area, the reflection points are sorted in order from near to far in the direction of travel. A 2000×400 full-pixel target image is pre-constructed. The height 400 corresponds to the 40cm width of the left track surface obstacle sensing area, and the width 2000 is the above preset range, that is, corresponding to the 2m travel direction range. The nearest reflection point is mapped to the preset coordinate position (0, 200) in the target image, and all the pixel points in the nearby N×N target area are mapped to the preset coordinate position (0, 200). The pixel values ​​of the N×N target regions around the nearest reflection point are set to 255. All other reflection points within a 2m distance from the nearest reflection point in the direction of travel are then mapped into the target image. The mapping locations are determined as above. The pixel values ​​of the N×N target regions surrounding each mapping location are then set to 255. Region segmentation is then performed, and the resulting subregions are sorted by area. The area of ​​the largest segmented region is then compared with a preset area threshold of 2×N×N. If the area of ​​the largest segmented region is greater than 2×N×N, a large-volume obstacle is considered to exist on the left track surface at a distance. The distance and angle of the obstacle are calculated as the first obstacle information. Otherwise, it is considered noise interference. The process for right track obstacle detection is similar to that for the left track surface, requiring only multi-line radar data within the right track obstacle sensing area. Due to the limited vertical angular resolution between multi-line laser radars and the expansion effect of point clouds at longer distances, severe point jitter can occur, resulting in unstable point spacing. Therefore, the target region and preset area threshold in the above scheme can be appropriately increased.

[0057] Furthermore, the combined lidar system can achieve long-range detection and close-range verification of track surface obstacles, thereby improving obstacle detection accuracy. Specifically, the obstacle detection process can be performed in real time at preset intervals. A cache of historical obstacle perception information can be pre-established. When an obstacle is detected in real time, its distance and angle are calculated and stored in the cache. During the next detection, if the vehicle moves, the new distance and angle of each historical obstacle in the cache at the vehicle's new position can be estimated and updated based on the odometer information stored on the vehicle. Simultaneously, a new detection result is obtained and then matched against the updated obstacle perception information in the cache. If a historical obstacle with a matching distance and angle error within a preset error range (e.g., 10%) is found, the match is considered successful. Successfully matched historical obstacles are considered successfully verified, increasing their credibility and allowing them to be added to a higher-level warning queue. If the match fails, the new detection result is stored in the cache. If a piece of historical obstacle information in the historical obstacle perception information cache queue fails to be successfully matched multiple times (for example, three times) in a row, it is considered to be noise interference and is discarded.

[0058] Furthermore, multi-line radar data can be used to detect curved track switches, a travel warning event. Exemplarily, multi-line radar data within a preset curved track and switch sensing area is filtered and obtained. If a reflection point exists in the area, the reflection points are sorted in order from near to far in the direction of travel. A 2000×800 all-zero-pixel target image is pre-constructed. The height of 800 corresponds to the width of the preset curved track and switch sensing area of ​​80 cm, and the width of 2000 corresponds to the above-mentioned preset range, i.e., a range of 2 m in the direction of travel. The nearest reflection point is mapped to the preset coordinate position (0, 400) in the target image, and all pixels in the nearby N×N target area are set to 255. Then, all other reflection points within a 2 m distance from the nearest reflection point in the direction of travel are mapped to the target image. The mapping position is determined as described above, and all pixels in the N×N target area near each mapping position are set to 255. Then, region segmentation is performed, and the segmented sub-regions are sorted by area. The area of ​​the largest segmented region is then compared with a preset area threshold of 2×N×N. If the area of ​​the largest segmented region is greater than 2×N×N, it can be considered that a curved track or switch is present ahead. Otherwise, it can be considered as noise interference. Similarly, the target area and preset area threshold in the above solution can be appropriately increased. Since no other objects except curved rails and switches are generally allowed within the trackbed, the multi-line LiDAR can accurately detect curved rails and switches by sensing the preset curved rail and switch sensing area.

[0059] Furthermore, the accuracy of curved track and switch detection can be improved by analyzing the results of curved track and switch detection on consecutive frames. Specifically, the curved track and switch detection process can be performed in real time at preset time intervals. If a curved track or switch is detected multiple times (e.g., three times), the presence of a curved track or switch can be determined, and the inspection vehicle can be decelerated, thereby increasing the reliability of the detection results. If a curved track or switch is not detected for the specified number of consecutive times, the previous detection result can be discarded, and the most recent curved track and switch detection result can be retained for subsequent determination.

[0060] The technical solution provided by the embodiments of the present invention first acquires radar data within a preset sensing area in front of the inspection vehicle. It then determines the closest reflection point in the radar data and other reflection points within a preset range from the closest reflection point in the direction of travel. These closest reflection point and other reflection points are mapped into a target image. The resulting mapped target image is then used to determine the inspection vehicle's travel attention information. By using radar data within a specific area and mapping it onto a planar image to analyze the inspection vehicle's travel attention information, this system achieves refined regional perception of attention events, significantly improving the inspection vehicle's ability to detect and perceive railway conditions.

[0061] Example 2

[0062] Figure 5 This is a schematic diagram of the structure of a railway inspection device for inspection vehicles provided in the second embodiment of the present invention. The device can be implemented by hardware and / or software and can generally be integrated into an inspection vehicle device to execute the railway inspection method for inspection vehicles provided in any embodiment of the present invention. Figure 5 As shown, the device includes:

[0063] Radar data acquisition module 21, used to acquire radar data in a preset sensing area in front of the inspection vehicle;

[0064] a reflection point mapping module 22 for determining a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in the direction of travel, and mapping the nearest reflection point and the other reflection points to a target image;

[0065] The attention information determination module 23 is used to determine the traveling attention information of the inspection vehicle according to the mapped target image.

[0066] The technical solution provided by the embodiments of the present invention first acquires radar data within a preset sensing area in front of the inspection vehicle. It then determines the closest reflection point in the radar data and other reflection points within a preset range from the closest reflection point in the direction of travel. These closest reflection point and other reflection points are mapped into a target image. The resulting mapped target image is then used to determine the inspection vehicle's travel attention information. By using radar data within a specific area and mapping it onto a planar image to analyze the inspection vehicle's travel attention information, this system achieves refined regional perception of attention events, significantly improving the inspection vehicle's ability to detect and perceive railway conditions.

[0067] Based on the above technical solution, optionally, the reflection point mapping module 22 is specifically configured to:

[0068] Mapping the nearest reflection point to a preset coordinate position in the target image, and setting the pixel points in the target area around the preset coordinate position to 255, wherein the target image is initially an image with all zero pixels;

[0069] Determine the mapping position of the corresponding other reflection points in the target image based on the relative position of the other reflection points and the nearest reflection point in the travel direction and the relative position in the vertical track direction, and set the pixel points in the target area around the mapping position to 255.

[0070] On the basis of the above technical solution, optionally, the target area is an N×N rectangular area, where N is the product of the radar distance of the corresponding reflection point and the radar horizontal angular resolution.

[0071] Based on the above technical solution, optionally, the attention information determination module 23 is specifically configured to:

[0072] Performing region segmentation on the mapped target image;

[0073] The maximum segmented region area obtained by segmentation is compared with the preset area threshold to determine whether there is a target attention event.

[0074] On the basis of the above technical solution, optionally, the preset sensing area includes a preset track surface obstacle sensing area and a preset curved track switch sensing area; correspondingly, the traveling attention information includes track surface obstacle information and curved track switch information.

[0075] On the basis of the above technical solution, optionally, a multi-line laser radar is provided above the front of the inspection vehicle, and the track surface obstacle information includes first track surface obstacle information determined based on multi-line radar data collected by the multi-line laser radar; and / or, a single-line laser radar is provided below the front of the inspection vehicle, and the track surface obstacle information includes second track surface obstacle information determined based on single-line radar data collected by the single-line laser radar.

[0076] The railway inspection device for inspection vehicles provided in the embodiment of the present invention can execute the railway inspection method for inspection vehicles provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0077] It is worth noting that in the above-mentioned embodiment of the railway detection device for inspection vehicles, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0078] Example 3

[0079] Figure 6 This is a structural diagram of an inspection vehicle device provided in Example 3 of the present invention, showing a block diagram of an exemplary inspection vehicle device suitable for implementing an embodiment of the present invention. Figure 6 The inspection vehicle device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention. Figure 6 As shown, the inspection vehicle device includes a processor 31, a memory 32, an input device 33 and an output device 34; the number of processors 31 in the inspection vehicle device can be one or more. Figure 6 Taking a processor 31 as an example, the processor 31, memory 32, input device 33 and output device 34 in the inspection vehicle device can be connected through a bus or other means. Figure 6 The bus connection is taken as an example.

[0080] Memory 32, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the railway inspection method for inspection vehicles in the embodiments of the present invention (for example, the radar data acquisition module 21, the reflection point mapping module 22, and the attention information determination module 23 in the railway inspection device for inspection vehicles). By running the software programs, instructions, and modules stored in memory 32, the processor 31 executes various functional applications and data processing of the inspection vehicle device, thereby implementing the aforementioned railway inspection method for inspection vehicles.

[0081] The memory 32 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created based on the use of the inspection vehicle device, etc. In addition, the memory 32 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory 32 may further include a memory remotely located relative to the processor 31, and these remote memories may be connected to the inspection vehicle device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0082] The input device 33 can be used to obtain the radar raw data collected by the laser radar, and generate key signal input related to the user settings and function control of the inspection vehicle equipment, etc. The output device 34 can be used to control the movement of the inspection vehicle, etc.

[0083] Example 4

[0084] The fourth embodiment of the present invention further provides a storage medium containing computer-executable instructions. When the computer-executable instructions are executed by a computer processor, the computer-executable instructions are used to execute a railway inspection method for an inspection vehicle. The method includes:

[0085] Obtain radar data within the preset sensing area in front of the inspection vehicle;

[0086] Determining a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in the direction of travel, and mapping the nearest reflection point and the other reflection points into a target image;

[0087] The traveling attention information of the inspection vehicle is determined according to the mapped target image.

[0088] The storage medium can be any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROMs, floppy disks, or tape devices; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disks or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the computer system in which the program is executed, or may be located in a different second computer system that is connected to the computer system via a network (such as the Internet). The second computer system may provide program instructions to the computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations (e.g., in different computer systems connected via a network). The storage medium may store program instructions (e.g., embodied as a computer program) that may be executed by one or more processors.

[0089] Of course, the storage medium containing computer-executable instructions provided in an embodiment of the present invention is not limited to the method operations described above, and can also execute related operations in the railway inspection method for inspection vehicles provided in any embodiment of the present invention.

[0090] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0091] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0092] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware. Of course, it can also be implemented with hardware, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the existing technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, and includes a number of instructions for enabling an electronic device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments of the present invention.

[0093] Example 5

[0094] Embodiment 5 of the present invention also provides a computer program product, which includes a computer program (also referred to as code, instructions). The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it is used to execute the railway inspection method for inspection vehicles provided in any of the above embodiments, and has the corresponding beneficial effects of the execution method.

[0095] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A railway inspection method for inspection vehicles, characterized in that: include: Obtain radar data within the preset sensing area in front of the inspection vehicle; Determining a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in the direction of travel, and mapping the nearest reflection point and the other reflection points into a target image; The traveling attention information of the inspection vehicle is determined according to the mapped target image.

2. The railway inspection method for inspection vehicles according to claim 1, characterized in that: Mapping the nearest reflection point and the other reflection points into a target image includes: Mapping the nearest reflection point to a preset coordinate position in the target image, and setting the pixel points in the target area around the preset coordinate position to 255, wherein the target image is initially an image with all zero pixels; Determine the mapping position of the corresponding other reflection points in the target image based on the relative position of the other reflection points and the nearest reflection point in the travel direction and the relative position in the vertical track direction, and set the pixel points in the target area around the mapping position to 255.

3. The railway inspection method for inspection vehicles according to claim 2, characterized in that: The target area is an N×N rectangular area, where N is the product of the radar distance of the corresponding reflection point and the radar horizontal angular resolution.

4. The railway inspection method for inspection vehicles according to claim 2, characterized in that: The determining of the traveling attention information of the inspection vehicle according to the mapped target image includes: Performing region segmentation on the mapped target image; The maximum segmented region area obtained by segmentation is compared with the preset area threshold to determine whether there is a target attention event.

5. The railway inspection method for inspection vehicles according to any one of claims 1 to 4, characterized in that: The preset sensing area includes a preset track surface obstacle sensing area and a preset curved track switch sensing area; correspondingly, the traveling attention information includes track surface obstacle information and curved track switch information.

6. The railway inspection method for inspection vehicles according to claim 5, characterized in that: A multi-line laser radar is provided above the front of the inspection vehicle, and the rail surface obstacle information includes first rail surface obstacle information determined based on multi-line radar data collected by the multi-line laser radar; And / or, a single-line laser radar is provided under the front of the inspection vehicle, and the track surface obstacle information includes second track surface obstacle information determined based on single-line radar data collected by the single-line laser radar.

7. A railway inspection device for inspection vehicles, characterized in that: include: Radar data acquisition module, used to obtain radar data within the preset sensing area in front of the inspection vehicle; a reflection point mapping module, configured to determine a nearest reflection point in the radar data and other reflection points within a preset range from the nearest reflection point in a direction of travel, and map the nearest reflection point and the other reflection points to a target image; The attention information determination module is used to determine the traveling attention information of the inspection vehicle according to the mapped target image.

8. A patrol vehicle device, characterized in that: include: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the railway inspection method for inspection vehicles as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the railway inspection method for inspection vehicles as described in any one of claims 1 to 6 is implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the railway inspection method for inspection vehicles as described in any one of claims 1 to 6.

Images