Railway high-precision surveying and mapping method

Abstract

The invention relates to a railway high-precision surveying and mapping method, which is characterized by comprising the following steps: S1, surveying and mapping by an unmanned aerial vehicle, S2, measuring ground engineering, and S3, integrating data and generating a map, the S1 step comprises a preparation stage, a flight and data acquisition step and a control point selection step, the S2 step comprises a ground data acquisition step, and the S3 step comprises a data integration step and a map generation step. The step S3 comprises a data fusion step and a map generation step, comprehensive coverage of a railway line is realized through cooperative work of the unmanned aerial vehicle and the inspection trolley, and detailed surveying and mapping data can be obtained whether in the air or on the ground. The high-definition camera and the RTK technology of the unmanned aerial vehicle and various sensors of the inspection trolley jointly guarantee high precision and reliability of surveying and mapping data. The data fusion technology enables data from different sources to be seamlessly connected, and accuracy and integrity of surveying and mapping are further improved. And finally, the generated railway electronic map is visual and easy to understand, and convenient tools and references are provided for railway planning, design, construction and operation.

Description

Technical Field

[0001] The invention relates to a railway surveying and mapping method, in particular to a railway high-precision surveying and mapping method. Background Art

[0002] As an important infrastructure of the national economy, the construction and maintenance of railways place extremely high demands on surveying and mapping technology. Traditional railway surveying and mapping methods mainly rely on manual operation and ground measurement equipment, but these methods have many limitations in practical applications.

[0003] First, the efficiency of traditional surveying and mapping methods is low. Since manual field measurement is required, it is not only time-consuming and labor-intensive, but also difficult to carry out under complex terrain and bad weather conditions. Secondly, the accuracy of traditional surveying and mapping methods is difficult to guarantee. Limited by human factors and the accuracy of measuring equipment, the accuracy of measurement data is often questioned, especially when dealing with subtle changes and complex structures of railway lines, the measurement error may be large. In addition, with the continuous development of railway engineering construction, the requirements for the comprehensiveness and real-time nature of surveying and mapping data are becoming higher and higher. Traditional surveying and mapping methods often find it difficult to provide sufficient data support and meet the needs of modern railway engineering for high-precision and high-efficiency surveying and mapping. Therefore, it has become an urgent problem to develop a new railway surveying and mapping method to improve surveying and mapping efficiency, ensure measurement accuracy, and meet the comprehensiveness and real-time nature of railway engineering construction data. Summary of the invention

[0004] In view of the above situation, it is necessary to provide a railway high-precision surveying and mapping method that solves at least one of the above problems, characterized in that it includes the following steps:

[0005] S1: UAV mapping;

[0006] S2: Ground engineering survey;

[0007] S3: Data integration and map generation;

[0008] The S1 step includes the following steps:

[0009] Preparation stage: clarify the scope of the survey area, plan the flight route, and determine the landing point;

[0010] Flight and data collection: Use drones equipped with high-definition cameras for aerial photography to obtain centimeter-level accurate positioning data and high-resolution images and graphics;

[0011] Control point selection: The control points selected in the aerial image are measured with precise coordinates using RTK technology;

[0012] The S2 step includes the following steps:

[0013] Ground data collection: A comprehensive intelligent railway inspection vehicle with integrated Beidou navigation, inertial navigation system, and laser scanner sensor is used for ground data collection. Point cloud data, high-definition images, and spatial position and posture information of the railway line are obtained through the inspection vehicle.

[0014] The S3 step includes the following steps:

[0015] Data fusion: Accurately align and fuse the drone aerial survey images, point cloud data and ground engineering survey data;

[0016] Map generation: Use GIS software to vectorize the integrated data and generate an electronic railway map containing railway lines, stations, bridges, and tunnel elements.

[0017] Preferably, the preparation stage steps also include the following steps: setting up the drone before aerial photography, the image overlap rate is greater than 35%, the route overlap rate is greater than 65%, and the polygon composed of the edge routes contains at least one or more areas of railway lines, stations, bridges, and tunnels.

[0018] Preferably, in the ground data collection step, the comprehensive intelligent railway inspection vehicle also includes a track image acquisition module, which is used to collect high-definition images of the railway track in real time, and combines the data of the inertial navigation system and the laser scanner to achieve three-dimensional reconstruction and feature extraction of the railway track.

[0019] Preferably, when the drone measures a railway curve, the following formula is used to calculate the curvature radius R of the curve:

[0020] R=(h2+l2) / (2*h*Δθ)

[0021] Among them, h is the shooting height of the drone, l is the arc length of the curve in the aerial image, and Δθ is the arc of the central angle corresponding to the curve in the aerial image.

[0022] Preferably, when measuring a straight railway track, the comprehensive intelligent railway inspection vehicle uses an inertial navigation system to obtain the real-time speed v and acceleration a of the vehicle, and calculates the length L of the straight track by the following formula:

[0023] L=(v2-u2) / (2*a)

[0024] Among them, u is the initial velocity of the car when it starts measuring.

[0025] Preferably, when measuring a straight railway track, the drone uses the high-definition camera on board to obtain the coordinates of the two endpoints of the straight line segment (x1, y1) and (x2, y2), and calculates the length L of the straight line by the following formula:

[0026] L=√[(x2-x1)2+(y2-y1)2].

[0027] Preferably, when measuring a railway curve, the comprehensive intelligent railway inspection vehicle uses a laser scanner to obtain multiple point cloud data on the curve, and calculates the curvature radius R of the curve by the following formula:

[0028] R=(180*ΔS) / (π*Δθ)

[0029] Among them, ΔS is the length of the arc that the inspection car travels on the curve, and Δθ is the change in the center angle of the curve measured by the laser scanner during the car's driving.

[0030] Preferably, when measuring railway curves, the railway comprehensive intelligent inspection vehicle can also combine the data of the inertial measurement unit (IMU) to estimate the curvature of the curve in real time through the following formula:

[0031] κ=(a_y+g*tan(θ)) / v2

[0032] Among them, κ is the curvature of the curve, a_y is the acceleration of the car perpendicular to the forward direction, g is the acceleration of gravity, θ is the inclination angle of the car in the vertical plane, and v is the real-time speed of the car.

[0033] Preferably, in the data fusion step, a deep learning-based algorithm is used to automatically align and fuse the drone aerial survey images and the ground engineering measurement data, including the following steps: extracting the features of the drone images and the ground measurement data through a deep learning network; using these features to perform similarity matching and determine the optimal alignment parameters; and fusing the drone images and the ground measurement data according to the alignment parameters to generate a data set.

[0034] Preferably, in the map generation step, a three-dimensional modeling technology is used to construct a three-dimensional model of the railway line based on the integrated data.

[0035] The working principle of the drone mapping part is to use the high-definition camera and advanced navigation and positioning system carried by the drone to take aerial photos and obtain centimeter-level precision positioning data and high-resolution images. The drone flies according to the pre-planned flight route and captures detailed information of the railway line and its surrounding environment through the high-definition camera. At the same time, using RTK technology, the drone can receive the differential signal sent by the ground base station in real time, correct its own positioning error, and thus obtain accurate control point coordinates. At the same time, the working principle of the ground engineering survey part is to collect ground data through the railway comprehensive intelligent inspection car that integrates Beidou navigation, inertial navigation system, and laser scanner sensor. The inspection car can drive autonomously or remotely on the railway line, obtain the point cloud data of the line through the laser scanner, shoot high-definition images through the high-definition camera, and use the Beidou navigation and inertial navigation system to determine its spatial position and attitude information. The acquisition of these data provides an important basis for the detailed mapping of the railway line. Next, the working principle of the data integration and map generation part is to accurately align and fuse the images, point cloud data and ground engineering survey data of the drone aerial survey. This process requires the use of algorithms to perform spatial registration, coordinate conversion and error correction on the data to ensure seamless connection between various data sources. The fused data is vectorized through GIS software to extract the geometric and attribute information of elements such as railway lines, stations, bridges, tunnels, etc., and finally generate an electronic railway map containing these elements.

[0036] Compared with the prior art, the beneficial effect of the present invention is that the working principle of the high-precision railway mapping method is to combine UAV mapping and ground engineering measurement, and realize comprehensive and high-precision mapping of railway lines through data fusion and map generation technology. The coordinated work of UAVs and inspection vehicles can give full play to their respective advantages and improve the efficiency and accuracy of mapping. At the same time, the application of data processing technology and GIS software enables the mapping data to be converted into intuitive and easy-to-read electronic maps, providing strong support for railway engineering construction and management. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention more clear, the railway high-precision surveying and mapping method of the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0038] In the description of the present invention, unless otherwise specified, "multiple" means two or more; the terms "center", "longitudinal", "lateral", "up", "down", "left", "right", "inside", "outside", "front end", "rear end", "head", "tail", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the shown positions or positional relationships, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0039] In the description of this utility, it should be noted that, unless otherwise clearly specified and limited, the terms "installation", "connection" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this utility can be understood by specific circumstances.

[0040] The high-precision railway surveying and mapping method of the embodiment of the present invention is characterized in that it includes the following steps: S1: UAV surveying and mapping; S2: ground engineering surveying; S3: data integration and map generation; wherein, step S1 includes the following steps: preparation stage: clarify the scope of the surveying and mapping area, plan the flight route, and determine the landing point; flight and data acquisition: use a UAV equipped with a high-definition camera to take aerial photos to obtain centimeter-level precision positioning data and high-resolution images and images; control point selection: the control points selected in the aerial images are measured using RTK technology to measure their precise coordinates; step S2 includes the following steps: ground data acquisition: use a comprehensive intelligent railway inspection car that integrates Beidou navigation, inertial navigation system, and laser scanner sensors to collect ground data, and obtain point cloud data, high-definition images, and spatial position and posture information of the railway line through the inspection car; step S3 includes the following steps: data fusion: accurately align and fuse the images and point cloud data of the UAV aerial survey and the data of the ground engineering survey; map generation: use GIS software to vectorize the integrated data to generate a railway electronic map containing railway lines, stations, bridges, and tunnel elements.

[0041] In the above embodiment, first, the working principle of the UAV surveying and mapping part is to use the high-definition camera and advanced navigation and positioning system carried by the UAV to take aerial photos and obtain centimeter-level precision positioning data and high-resolution images. The UAV flies according to the pre-planned flight route and captures detailed information of the railway line and its surrounding environment through the high-definition camera. At the same time, using RTK technology, the UAV can receive the differential signal sent by the ground base station in real time, correct its own positioning error, and thus obtain accurate control point coordinates. At the same time, the working principle of the ground engineering measurement part is to collect ground data through the railway comprehensive intelligent inspection car that integrates Beidou navigation, inertial navigation system, and laser scanner sensor. The inspection car can travel autonomously or remotely on the railway line, obtain the point cloud data of the line through the laser scanner, shoot high-definition images through the high-definition camera, and use the Beidou navigation and inertial navigation system to determine its spatial position and attitude information. The acquisition of these data provides an important basis for the detailed surveying and mapping of the railway line. Next, the working principle of the data integration and map generation part is to accurately align and fuse the images, point cloud data and ground engineering measurement data of the UAV aerial survey. This process requires the use of algorithms to spatially align data, perform coordinate conversion, and perform error correction to ensure seamless integration between various data sources. The fused data is vectorized using GIS software to extract the geometric and attribute information of railway lines, stations, bridges, tunnels, and other elements, and ultimately generate an electronic railway map containing these elements.

[0042] More specifically, data integration involves the fusion of data obtained from drone mapping and ground engineering surveying. This includes the calibration, alignment and integration of information from different data sources to ensure data accuracy and consistency. Drones obtain high-resolution images, point cloud data, and spatial position and attitude information through high-definition cameras and other sensors on board, while ground intelligent inspection vehicles obtain ground data through integrated Beidou navigation, inertial navigation systems, laser scanners and other sensors. The data obtained by these two methods may differ in format, accuracy and coordinate system, so they need to be effectively integrated through data processing technology. GIS technology can perform spatial analysis and visualization of the fused data to generate electronic maps containing elements such as railway lines, stations, bridges, and tunnels. By vectorizing these data, the characteristic information of the line, such as length, curvature, slope, etc., can be further extracted to provide accurate data support for the design and construction of railway projects. In addition, GIS technology can also use spatial analysis functions to optimize railway lines and improve the safety and economy of the lines. In addition to GIS technology, data processing technology also includes some other methods, such as artificial intelligence (AI) technology. AI technology can improve the accuracy of prediction and decision support by learning and processing large amounts of surveying and mapping data. For example, AI can be used to analyze measurement data to detect potential structural risks and provide corresponding treatment recommendations.

[0043] In summary, the working principle of the railway high-precision mapping method is to combine UAV mapping and ground engineering measurement, and realize comprehensive and high-precision mapping of railway lines through data fusion and map generation technology. The collaborative work of UAVs and inspection vehicles can give full play to their respective advantages and improve the efficiency and accuracy of mapping. At the same time, the application of data processing technology and GIS software enables the mapping data to be converted into intuitive and easy-to-read electronic maps, providing strong support for railway engineering construction and management.

[0044] The advantages of this method are: first, through the collaborative work of drones and inspection vehicles, comprehensive coverage of railway lines can be achieved, and detailed surveying and mapping data can be obtained both in the air and on the ground. Secondly, the high-definition camera and RTK technology of the drone, as well as the various sensors of the inspection vehicle, jointly ensure the high accuracy and reliability of the surveying and mapping data. Furthermore, the application of data fusion technology enables data from different sources to be seamlessly connected, further improving the accuracy and completeness of surveying and mapping. Finally, the generated electronic railway map is intuitive and easy to understand, providing a convenient tool and reference for railway planning, design, construction and operation.

[0045] Preferably, the preparation stage steps also include the following steps: setting up the drone before aerial photography, the image overlap rate is greater than 35%, the route overlap rate is greater than 65%, and the polygon composed of the edge routes contains at least one or more areas of railway lines, stations, bridges, and tunnels.

[0046] In the above embodiment, the image overlap rate is set to ensure that there is a certain overlap between two adjacent photos during the drone aerial photography process. An overlap rate greater than 35% means that at least 35% of the areas of two adjacent photos are overlapping. The purpose of this is to accurately match and splice the information of the overlapping area in the subsequent image processing and three-dimensional modeling process, thereby improving the accuracy and consistency of the entire surveying and mapping results. At the same time, the overlapping area can also be used to check the accuracy and consistency of the data. By comparing the same area in different photos, possible errors can be detected and corrected. The setting between routes involves the flight path planning of drone aerial photography. The setting of more than 65% between routes means that the intervals between different routes will not be too large, which can ensure that the entire surveying and mapping area is fully covered and avoid missing or repeated measurements. At the same time, reasonable route planning can also improve the efficiency of aerial photography and reduce the flight time and energy consumption of drones. When planning routes, the polygons composed of edge routes need to include key areas such as railway lines, stations, bridges, tunnels, etc. This is because these areas are the most important and complex parts of railway engineering and require more refined and accurate surveying and mapping data. By ensuring that these areas are included in the aerial photography range, more comprehensive and detailed surveying and mapping information can be obtained, providing strong support for subsequent design and construction.

[0047] To summarize, the working principles of image overlap rate and route planning in the pre-setup of UAV aerial photography are designed to ensure that UAV aerial photography can obtain high-precision, comprehensive coverage surveying and mapping data through reasonable parameter settings and route planning, providing a reliable data basis for high-precision railway surveying and mapping.

[0048] Preferably, in the ground data collection step, the comprehensive intelligent railway inspection vehicle also includes a track image acquisition module, which is used to collect high-definition images of the railway track in real time, and combines the data of the inertial navigation system and the laser scanner to achieve three-dimensional reconstruction and feature extraction of the railway track.

[0049] In the above embodiment, the track image acquisition module is equipped with a high-definition camera that can capture the subtle texture and structure of the track surface. As the inspection car travels, the camera continuously captures continuous images of the track, and these images are transmitted to the processing unit in real time through a high-speed data transmission system. The inertial navigation system provides accurate navigation and positioning information for the inspection car to ensure that the car travels stably and accurately on the railway line. The laser scanner is used to obtain three-dimensional point cloud data of the track, which can accurately reflect the geometric shape and spatial position of the track. When the track image acquisition module acquires high-definition images, the three-dimensional reconstruction of the railway track can be achieved by combining the data of the inertial navigation system and the laser scanner. Through image processing and computer vision technology, key information such as the contour, size, slope, etc. of the track can be extracted, and an accurate three-dimensional model can be constructed. This process not only improves the accuracy and efficiency of surveying and mapping, but also provides strong data support for subsequent railway engineering design and construction. In addition, the track image acquisition module can also realize feature extraction. Through deep learning and pattern recognition of track images, defects, damage or abnormal areas on the track can be automatically identified and extracted. This characteristic information is crucial for railway safety monitoring and maintenance, and can help staff detect and deal with potential safety hazards in a timely manner.

[0050] In summary, the track image acquisition module of the railway comprehensive intelligent inspection vehicle realizes the three-dimensional reconstruction and feature extraction of railway tracks by real-time acquisition of high-definition images and combining the data of the inertial navigation system and laser scanner. The application of this technology not only improves the accuracy and efficiency of railway surveying and mapping, but also provides strong support for the safety monitoring and maintenance of railway projects.

[0051] Preferably, when the drone measures a railway curve, the following formula is used to calculate the curvature radius R of the curve:

[0052] R=(h2+l2) / (2*h*Δθ)

[0053] Among them, h is the shooting height of the drone, l is the arc length of the curve in the aerial image, and Δθ is the arc of the central angle corresponding to the curve in the aerial image.

[0054] In the above embodiment, the drone takes aerial photos of the railway curve at a specific shooting height h. During the aerial photography process, the drone captures the image of the curve and measures the arc length l of the curve in the image. This arc length represents the actual length of the curve from the perspective of the drone. Next, the drone also needs to determine the central angle arc Δθ corresponding to the curve in the aerial image. This central angle arc represents the angular span of the curve in the image. The drone identifies and calculates this angle through an internal image processing system. Finally, the data obtained from the above measurements are substituted into the formula R=(h2+l2) / (2*h*Δθ) for calculation. This formula calculates the curvature radius R of the curve through the drone shooting height, the arc length of the curve in the aerial image, and the central angle arc corresponding to the curve. The curvature radius is an important parameter for measuring the curvature degree of a curve.

[0055] Preferably, when measuring a straight railway track, the comprehensive intelligent railway inspection vehicle uses an inertial navigation system to obtain the real-time speed v and acceleration a of the vehicle, and calculates the length L of the straight track by the following formula:

[0056] L=(v2-u2) / (2*a)

[0057] Among them, u is the initial velocity of the car when it starts measuring.

[0058] In the above embodiment, the inspection car travels on a straight railway track, and the inertial navigation system records the speed v and acceleration a of the car in real time. The inertial navigation system can accurately measure and calculate these parameters through built-in sensors and algorithms. Secondly, in order to calculate the length L of the straight track, it is necessary to know the initial velocity u when the car starts measuring. The initial velocity u can be obtained and recorded by the inertial navigation system before the car starts measuring. Then, the real-time velocity v, acceleration a and initial velocity u are substituted into the formula L = (v2-u2) / (2*a) for calculation. This formula is based on the kinematic principles in physics, and the length of the straight track is deduced through the relationship between velocity, acceleration and initial velocity.

[0059] Preferably, when measuring a straight railway track, the drone uses the high-definition camera on board to obtain the coordinates of the two endpoints of the straight line segment (x1, y1) and (x2, y2), and calculates the length L of the straight line by the following formula:

[0060] L=√[(x2-x1)2+(y2-y1)2].

[0061] In the above embodiment, the drone takes aerial photos over the railway straight track and captures the image of the railway straight track through a high-definition camera. The high resolution and accuracy of the high-definition camera ensure that the two endpoints of the straight line segment can be clearly identified. Then, the drone automatically or manually identifies the aerial image through image processing technology to extract the coordinate information of the two endpoints of the straight line segment. These coordinate information are expressed in the form of two-dimensional plane coordinates, namely (x1, y1) and (x2, y2), which represent the positions of the two endpoints in the image respectively. Then, the drone uses the built-in computing system to substitute the extracted endpoint coordinates into the formula L=√[(x2-x1)2+(y2-y1)2] for calculation. This formula is based on the definition of Euclidean distance and calculates the length of the railway straight track by calculating the straight-line distance between the two endpoints on a two-dimensional plane.

[0062] Preferably, when measuring a railway curve, the comprehensive intelligent railway inspection vehicle uses a laser scanner to obtain multiple point cloud data on the curve, and calculates the curvature radius R of the curve by the following formula:

[0063] R=(180*ΔS) / (π*Δθ)

[0064] Among them, ΔS is the length of the arc that the inspection car travels on the curve, and Δθ is the change in the center angle of the curve measured by the laser scanner during the car's driving.

[0065] Preferably, when measuring railway curves, the railway comprehensive intelligent inspection vehicle can also combine the data of the inertial measurement unit (IMU) to estimate the curvature of the curve in real time through the following formula:

[0066] κ=(a_y+g*tan(θ)) / v2

[0067] Among them, κ is the curvature of the curve, a_y is the acceleration of the car perpendicular to the forward direction, g is the acceleration of gravity, θ is the inclination angle of the car in the vertical plane, and v is the real-time speed of the car.

[0068] In the above embodiment, the inspection car travels on the railway curve, and the laser scanner continuously emits laser beams and receives reflected signals, thereby obtaining multiple point cloud data on the curve. These point cloud data accurately reflect the three-dimensional shape and spatial position of the curve. Then, the inspection car processes and analyzes the acquired point cloud data through the internal processing system. The system can identify the arc length ΔS that the car has traveled on the curve, which is usually obtained by calculating the distance between adjacent points in the point cloud data and accumulating them. At the same time, the system can also measure the change in the center angle of the curve Δθ measured by the laser scanner during the car's travel, which represents the change in direction of the car when it travels on the curve. Then, the calculated arc length ΔS and center angle change Δθ are substituted into the formula R=(180*ΔS) / (π*Δθ) for calculation. This formula is based on the principles of geometry and trigonometry, and the curvature radius R of the curve is calculated through the relationship between the arc length and the center angle. The curvature radius is an important parameter for measuring the curvature degree of the curve, which is of great significance to the design and construction of railway projects.

[0069] Preferably, in the data fusion step, a deep learning-based algorithm is used to automatically align and fuse the drone aerial survey images and the ground engineering measurement data, including the following steps: extracting the features of the drone images and the ground measurement data through a deep learning network; using these features to perform similarity matching and determine the optimal alignment parameters; and fusing the drone images and the ground measurement data according to the alignment parameters to generate a data set.

[0070] In the above embodiment, the deep learning network can automatically extract information such as texture, shape, edge, etc. in the image, as well as geometric and position features in the ground measurement data by learning abstract features in the data layer by layer. Then, the extracted features are used for similarity matching. The deep learning algorithm can learn how to compare the features of different data sources and find the correlation and similarity between them. By calculating the similarity between the features, the correspondence between the drone image and the ground measurement data can be determined, and then the optimal registration parameters can be determined. Finally, according to the determined registration parameters, the drone image and the ground measurement data are fused. This step usually involves geometric transformation of the image to align it with the ground measurement data, and fuse the two into a comprehensive data set in some way (such as weighted averaging, maximum value selection, etc.). The fused data set contains both the rich texture and detail information of the drone image and the accuracy and reliability of the ground measurement data. Through this process, the algorithm based on deep learning can automatically and efficiently realize the registration and fusion of drone aerial survey images and ground engineering measurement data, generating a comprehensive, accurate and comprehensive data set, providing strong data support for subsequent railway engineering construction and management.

[0071] Preferably, in the map generation step, a three-dimensional modeling technology is used to construct a three-dimensional model of the railway line based on the integrated data.

[0072] In the above embodiments, the 3D modeling technology allows the creation of highly realistic 3D models using integrated data (including drone aerial survey images, ground engineering survey data, etc.). These models not only show the topography of the railway line, but also accurately restore the structural features of key facilities such as bridges, tunnels, and stations.

[0073] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A railway high-precision surveying and mapping method, characterized in that: The following steps are involved: S1: UAV mapping; S2: Ground engineering survey; S3: Data integration and map generation; The S1 step includes the following steps: Preparation stage: clarify the scope of the survey area, plan the flight route, and determine the landing point; Flight and data collection: Use drones equipped with high-definition cameras for aerial photography to obtain centimeter-level accurate positioning data and high-resolution images and graphics; Control point selection: The control points selected in the aerial image are measured with precise coordinates using RTK technology; The S2 step includes the following steps: Ground data collection: A comprehensive intelligent railway inspection vehicle with integrated Beidou navigation, inertial navigation system, and laser scanner sensor is used for ground data collection. Point cloud data, high-definition images, and spatial position and posture information of the railway line are obtained through the inspection vehicle. The S3 step includes the following steps: Data fusion: Accurately align and fuse the drone aerial survey images, point cloud data and ground engineering survey data; Map generation: Use GIS software to vectorize the integrated data and generate an electronic railway map containing railway lines, stations, bridges, and tunnel elements.

2. The high-precision railway surveying and mapping method according to claim 1, characterized in that: The preparation stage steps also include the following steps: setting up the drone before aerial photography, the image overlap rate is greater than 35%, the route overlap rate is greater than 65%, and the polygon composed of the edge routes contains at least one or more areas of railway lines, stations, bridges, and tunnels.

3. The high-precision railway surveying and mapping method according to claim 1, characterized in that: In the ground data collection step, the railway comprehensive intelligent inspection vehicle also includes a track image acquisition module, which is used to collect high-definition images of the railway track in real time, and combines the data of the inertial navigation system and the laser scanner to achieve three-dimensional reconstruction and feature extraction of the railway track.

4. The high-precision railway surveying and mapping method according to claim 1, characterized in that: When the UAV measures the railway curve, the following formula is used to calculate the curvature radius R of the curve: R=(h2+l2) / (2*h*Δθ) Among them, h is the shooting height of the drone, l is the arc length of the curve in the aerial image, and Δθ is the arc of the central angle corresponding to the curve in the aerial image.

5. The high-precision railway surveying and mapping method according to claim 1, characterized in that: When measuring a straight railway track, the railway comprehensive intelligent inspection vehicle uses the inertial navigation system to obtain the vehicle's real-time speed v and acceleration a, and calculates the length L of the straight track using the following formula: L=(v2-u2) / (2*a) Among them, u is the initial velocity of the car when it starts measuring.

6. The high-precision railway surveying and mapping method according to claim 1, characterized in that: When measuring a straight railway track, the drone uses the high-definition camera on board to obtain the coordinates of the two endpoints of the straight line segment (x1, y1) and (x2, y2), and calculates the length L of the straight line using the following formula: L=√[(x2-x1)2+(y2-y1)2].

7. The high-precision railway surveying and mapping method according to claim 1, characterized in that: When measuring railway curves, the railway comprehensive intelligent inspection vehicle uses a laser scanner to obtain multiple point cloud data on the curve and calculates the curvature radius R of the curve using the following formula: R=(180*ΔS) / (π*Δθ) Among them, ΔS is the length of the arc that the inspection car travels on the curve, and Δθ is the change in the center angle of the curve measured by the laser scanner during the car's driving.

8. The high-precision railway surveying and mapping method according to claim 1, characterized in that: When measuring railway curves, the railway comprehensive intelligent inspection vehicle can also combine the data of the inertial measurement unit (IMU) to estimate the curvature of the curve in real time through the following formula: κ=(a_y+g*tan(θ)) / v2 Among them, κ is the curvature of the curve, a_y is the acceleration of the car perpendicular to the forward direction, g is the acceleration of gravity, θ is the inclination angle of the car in the vertical plane, and v is the real-time speed of the car.

9. The high-precision railway surveying and mapping method according to claim 1, characterized in that: In the data fusion step, a deep learning-based algorithm is used to automatically register and fuse the UAV aerial survey images and ground engineering measurement data, including the following steps: extracting the features of the UAV images and ground measurement data through a deep learning network; using these features for similarity matching to determine the optimal registration parameters; and fusing the UAV images and ground measurement data according to the registration parameters to generate a data set.

10. The high-precision railway surveying and mapping method according to claim 1, characterized in that: In the map generation step, 3D modeling technology is used to construct a 3D model of the railway line based on the integrated data.