A method for planning a patrol route

Abstract

The application discloses a kind of inspection route planning method, the application is fitted with the fitting multi-fold line of crack with geographic information, automatically carries out the design of collection route, the route designed only for specific crack collection, original data is accurate, data volume is less and has operation personnel safety risk, collection efficiency is high, route can be automatically repeated execution etc. Characteristic, very suitable for subsequent aerial triangulation modeling operation, can very real and accurate restore crack and surrounding various subtle material texture and geometric structure, provide data basis for later calculation simulation and prediction warning.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) photography technology, and in particular to a method for planning inspection routes. Background Technology

[0002] With the increasing prevalence of UAVs in aerial photogrammetry in recent years, the requirements for accuracy have become increasingly stringent, and the scope of application has expanded. Conventional UAV aerial surveying flight path planning software and methods primarily target surface terrain and man-made structures for data acquisition, but current solutions rarely include flight path planning methods for high-precision close-range photogrammetry of individual target points. However, there is a real-world need for this. As is well known, high-precision close-range photogrammetry requires the acquisition equipment to be directly facing the target point, and a constant relative distance and overlap must be maintained throughout the acquisition process to obtain data suitable for subsequent modeling and analysis.

[0003] Existing measurement data acquisition methods often rely heavily on the raw data of the site to be measured, which increases the amount of additional measurement work and results in poor measurement accuracy. This problem urgently needs to be solved. Summary of the Invention

[0004] This invention provides a method for planning inspection routes, which can at least solve one of the problems pointed out in the background art.

[0005] A method for planning inspection routes includes the following steps:

[0006] Step 1: Based on the parameters of the acquisition device, the expected resolution (GSD), and the expected sampling overlap, calculate the relative distance L between the acquisition device and the measured point. relative And the horizontal acquisition interval L of the acquisition device H and the acquisition interval L in the vertical direction V ;

[0007] Step 2: Resample the existing crack location data;

[0008] Step 3: Extract the corresponding image coverage area for each point in the resampled crack location data to obtain the point cloud model for each point;

[0009] Step 4: Calculate the normal vector for each extracted point cloud model to obtain the average normal vector of the image coverage area;

[0010] Step 5: Use the angle between the average normal vector and the vertical direction as the gimbal's depression angle when acquiring waypoints;

[0011] Step Six: Using the normal vector and the relative distance L between the acquisition device and the measured point. relative Calculate the location of the waypoints;

[0012] Step 7: Perform bidirectional buffering on multiple waypoints in a direction perpendicular to the direction of interpolation sampling points during resampling, resulting in an even number of lateral acquisition routes;

[0013] Step 8: Connect the lateral flow path and the main crack measurement flow path in a bow-shaped sequence to generate the crack measurement flow path.

[0014] In step one, the parameter of the acquisition device is the pixel P on the X-axis. x Pixel P on the Y-axis y Lens focal length f and pixel size a; expected sampling overlap is horizontal overlap H% and vertical overlap V%; where;

[0015] L relative The calculation formula is:

[0016] L H The calculation formula is: L H =P x ×(1-%)×GSD;

[0017] L V The calculation formula is: L V =P y ×(1-%)×GSD.

[0018] The resampling method in step two is as follows:

[0019] The crack location data was converted from a geographic coordinate system using latitude and longitude to a projected coordinate system using meters.

[0020] Set a threshold; if the slope from point n to point n+1 in the crack location data is greater than the threshold, then the horizontal acquisition interval L is set. H Calculate the location of the interpolation sampling points and perform interpolation; otherwise, use the vertical sampling interval L. V Calculate the positions of the interpolation sampling points and perform interpolation;

[0021] After each interpolation, calculate the distance from the interpolated sampling point to point n+1 until the distance is less than the distance from the previous interpolated sampling point to point n+1. This completes the interpolation from point n to point n+1. Then, replace the position of point n+1 with the position of the last interpolated point and repeat the above steps for the interpolation from point n+1 to point n+2.

[0022] In step two, the threshold is set as follows:

[0023] Calculate the aspect ratio of the acquisition device and use the obtained aspect ratio value as the threshold;

[0024] The formula for calculating the threshold is:

[0025] The formula for calculating the slope from point n to point n+1 in the crack location data is as follows:

[0026]

[0027] Calculate the yaw angle from point n to point n+1 in the crack location data. n The calculation formula is as follows:

[0028]

[0029] If the slope is greater than the threshold, then the horizontal sampling interval L is used. H Calculate the location of the interpolation sampling point North i East i Alititude i And perform interpolation;

[0030] Among them, if Altitude n+1 -Altitude n >0 i Then Alititude i =Altitude n +L V ;

[0031] If Altitude n+1 -Altitude n ≤0 i Then Alititude i =Altitude n -L V ;

[0032] North i =North n +cos(Heading n )*Dist V ;

[0033] East i =East n +sin(Heading n )*Dist V ;

[0034] If the slope is not greater than the threshold, then the vertical sampling interval L is used. V Calculate the location of the interpolation sampling point North i East i Alititude i And perform interpolation;

[0035] Among them, if Altitude n+1 -Altitude n >0 i Then Alititude i =Altitude n +L V ;

[0036] If Altitude n+1 -Altitude n ≤0 i Then Alititude i =Altitude n -L V ;

[0037] North i =North n +cos(Heading n )*Dist H ;

[0038] East i =East n +sin(Heading n )*Dist H .

[0039] The extraction method in step three is as follows:

[0040] Using the resampling point as the center of the coordinate system, the yaw angle is calculated using the arctangent function of two adjacent points, as shown in the following formula:

[0041]

[0042] Extract the point cloud data within the area enclosed by the four vertices where the center point intersects the yaw angle direction and the perpendicular yaw angle direction. The yaw angle direction of the center point is: The direction of the vertical yaw angle is:

[0043] Extract point cloud data within the coverage area defined by the elevation of the center point from the point cloud data enclosed by the four vertices mentioned above, where the elevation is [value missing].

[0044] The method for calculating the average normal vector of the image coverage area in step four is as follows:

[0045] After removing the normal vectors with the maximum and minimum values, the average normal vector of the remaining normal vectors is calculated using a function that calculates the average value.

[0046] In step seven, if the interpolation sampling points in the resampling process are at horizontal sampling intervals L... H For interpolation, the sampling interval L is used in the vertical direction. V Perform integer-multiple buffering if the interpolation sampling points during the resampling process are at acquisition intervals L in the vertical horizontal direction. V For interpolation, the sampling interval L is used in the horizontal direction. H Perform buffering at integer multiples.

[0047] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention automatically designs the data collection route by fitting multi-polylines with geographic information to cracks. The designed route only collects data from specific cracks, resulting in accurate raw data, small data volume, no personnel safety risks during operation, high data collection efficiency, and the ability to automatically repeat the route execution. It is very suitable for subsequent aerial triangulation modeling operations and can very realistically and accurately reproduce the cracks and various subtle material textures and geometric structures around them, providing a data foundation for later calculation simulation and prediction and early warning. Attached Figure Description

[0048] Figure 1 This is based on existing crack location data;

[0049] Figure 2 The points resampled on the crack and the enclosing quadrilateral of those points;

[0050] Figure 3 for Figure 2 Point cloud data clipped by enclosing quadrilaterals in the middle;

[0051] Figure 4 for Figure 3 A schematic diagram of the cropped point cloud data from a 3D perspective;

[0052] Figure 5 Points resampled on the crack are aligned with elevation within the visible range. Figure 3 A diagram showing the cropped point cloud data.

[0053] Figure 6 A schematic diagram of point cloud data cropped according to elevation in a 3D perspective;

[0054] Figure 7 This is a schematic diagram for calculating the normal vector of a point cloud model.

[0055] Figure 8 A schematic diagram of several waypoints along the main route for crack measurement;

[0056] Figure 9 A schematic diagram of the main data acquisition path for crack measurement and the side acquisition path after bidirectional buffering;

[0057] Figure 10 The crack measurement route is generated by connecting the main and lateral acquisition routes in a bow-shaped sequence. Detailed Implementation

[0058] The following detailed description of a specific embodiment of the present invention is provided in conjunction with the accompanying drawings. However, it should be understood that the scope of protection of the present invention is not limited to the specific embodiment.

[0059] like Figures 1 to 10 As shown in the figure, an inspection route planning method provided by this invention utilizes pre-prepared crack location data and a digital model of the structural surface. By inputting relevant parameters of the acquisition device and considering requirements such as acquisition resolution and overlap, the crack location data is resampled. Combined with the digital model of the structural surface, a safe, efficient, and repeatable inspection route for the cracks is calculated. The method includes the following steps:

[0060] Step 1: Based on the parameters of the acquisition device, the expected resolution (GSD), and the expected sampling overlap, calculate the relative distance L between the acquisition device and the measured point. relative And the horizontal acquisition interval L of the acquisition device H and the acquisition interval L in the vertical direction V ;

[0061] In step one, the parameter of the acquisition device is the pixel P on the X-axis. x Pixel P on the Y-axis y Lens focal length f and pixel size a; expected sampling overlap is horizontal overlap H% and vertical overlap V%; where;

[0062] L relative The calculation formula is:

[0063] L H The calculation formula is: L H =P x ×(1-H%)×GSD;

[0064] L V The calculation formula is: L V =P y ×(1-V%)×GSD;

[0065] Step 2: Resample the existing crack location data;

[0066] The resampling method in step two is as follows:

[0067] The crack location data (which consists of continuous points of latitude, longitude, and elevation, and fitted polylines formed by connecting the points in sequence) was transformed from a geographic coordinate system using latitude and longitude units to a projected coordinate system using meters.

[0068] Set a threshold; if the slope from point n to point n+1 in the crack location data is greater than the threshold, then the horizontal acquisition interval L is set. H Calculate the location of the interpolation sampling points and perform interpolation; otherwise, use the vertical sampling interval L. V Calculate the location of the interpolation sampling points and perform interpolation; the threshold is set as follows:

[0069] Calculate the aspect ratio of the acquisition device and use the obtained aspect ratio value as the threshold;

[0070] The formula for calculating the threshold is:

[0071] For example, for a data acquisition device with a resolution of 8192×5460, the threshold is...

[0072] The formula for calculating the slope from point n to point n+1 in the crack location data is as follows:

[0073]

[0074] By comparing the calculated slope with the slope threshold, the appropriate method for using L is determined. V Or L H To perform reference calculations for interpolation sampling points, if the calculated N... slope >T slope Then use L V This serves as the basis for calculating interpolation sampling points, and the horizontal interval distance is calculated according to this slope; otherwise, L is used. H As the basis for calculating interpolation sampling points;

[0075]

[0076]

[0077] Calculate the yaw angle from point n to point n+1 in the crack location data. n The calculation formula is as follows:

[0078]

[0079] If the slope is greater than the threshold, then the horizontal sampling interval L is used. H Calculate the location of the interpolation sampling point North i East iAlititude i And perform interpolation;

[0080] Among them, if Altitude n+1 -Altitude n >0 i Then Alititude i =Altitude n +L V ;

[0081] If Altitude n+1 -Altitude n ≤0 i Then Alititude i =Altitude n -L V ;

[0082] North i =North n +cos(Headingn)*Dist V ;

[0083] East i =East n +sin(Heading n )*Dist V ;

[0084] If the slope is not greater than the threshold, then the vertical sampling interval L is used. V Calculate the location of the interpolation sampling point North i East i Alititude i And perform interpolation;

[0085] Among them, if Altitude n+1 -Altitude n >0 i Then Alititude i =Altitude n +L V ;

[0086] If Altitude n+1 -Altitude n ≤0 i Then Alititude i =Altitude n -L V ;

[0087] Northi =North n +cos(Heading n )*Dist H ;

[0088] East i =East n +sin(Heading n )*Dist H ;

[0089] Interpolation points, i.e., interpolation sampling points, are marked with attribute identifiers for horizontal or vertical interpolation. Subsequent waypoint lateral buffering will be calculated based on this identifier. Waypoints interpolated in the horizontal direction need to be buffered according to the vertical interval, and waypoints interpolated in the vertical direction need to be buffered according to the horizontal interval.

[0090] After each interpolation, calculate the distance from the interpolated sampling point to point n+1 until the distance is less than the distance from the previous interpolated sampling point to point n+1. This completes the interpolation from point n to point n+1. Then, replace the position of point n+1 with the position of the last interpolated point and repeat the above steps for the interpolation from point n+1 to point n+2.

[0091] Step 3: Extract the corresponding image coverage area for each point in the resampled crack location data to obtain the point cloud model for each point;

[0092] The extraction method in step three above is as follows:

[0093] Using the resampling point as the center of the coordinate system, the yaw angle is calculated using the arctangent function of two adjacent points, as shown in the following formula:

[0094]

[0095] Extract the area enclosed by the four vertices that intersect the center point along the yaw angle direction and the perpendicular yaw angle direction. Figure 2 Point cloud data within a quadrilateral formed by four vertices ( Figure 3 To crop the point cloud using a quadrilateral, retaining only the points within the quadrilateral, where the yaw angle of the center point is: The direction of the vertical yaw angle is: (CoverageRatio is 1 by default. If the structure is particularly complex or the tilt angle is large, CoverageRatio can be increased to obtain more point cloud data for calculation.)

[0096] like Figure 5 and Figure 6As shown, point cloud data within a coverage area with the elevation of the center point as the vertical direction is extracted from the point cloud data enclosed by the four vertices, where the elevation is...

[0097] Step 4: Calculate the normal vector for each extracted point cloud model (e.g., Figure 7 As shown in the figure, the average normal vector of the image coverage area is obtained;

[0098] The average normal vector of the image coverage area is calculated as follows: after removing the normal vectors with the maximum and minimum values, the average normal vector of the remaining normal vectors is calculated using a function that calculates the average value.

[0099] Step 5: Use the angle between the average normal vector and the vertical direction as the gimbal's depression angle when acquiring waypoints;

[0100] Step Six: Using the normal vector and the relative distance between the acquisition device and the measured point. Calculate the location of the waypoints;

[0101] Upon reaching step six, multiple consecutive waypoints directly facing the crack are obtained. The route connecting these multiple consecutive waypoints (interpolated sampling points) and the resampled points is the main route for crack measurement.

[0102] Step 7: Perform bidirectional buffering on multiple waypoints in a direction perpendicular to the direction of interpolation sampling points during resampling, using integer multiples (positive integers), to obtain an even number of lateral acquisition lines; if the interpolation sampling points in the resampling process are acquired at horizontal intervals L... H For interpolation, the sampling interval L is used in the vertical direction. V Perform integer-multiple buffering if the interpolation sampling points during the resampling process are at acquisition intervals L in the vertical horizontal direction. V For interpolation, the sampling interval L is used in the horizontal direction. H Perform integer multiple buffering ( Figure 9 The diagram shown is a schematic of obtaining two lateral data collection routes.

[0103] Step 8: Connect the lateral route (composed of multiple buffered points) and the main crack measurement route in a bow-shaped sequence to generate the crack measurement route (e.g., ...). Figure 10 (As shown).

[0104] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit and essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0105] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for planning inspection routes, characterized in that, Includes the following steps: Step 1: Based on the parameters of the acquisition device, the expected resolution (GSD), and the expected sampling overlap, calculate the relative distance between the acquisition device and the measured point. and the horizontal acquisition interval of the acquisition device. and the acquisition interval in the vertical direction ; Step 2: Resample the existing crack location data; Step 3: Extract the corresponding image coverage area for each point in the resampled crack location data to obtain the point cloud model for each point; Step 4: Calculate the normal vector for each extracted point cloud model to obtain the average normal vector of the image coverage area; Step 5: Use the angle between the average normal vector and the vertical direction as the gimbal's depression angle when acquiring waypoints; Step Six: Using the normal vector and the relative distance between the acquisition device and the measured point. Calculate the location of the waypoints; Step 7: Perform bidirectional buffering on multiple waypoints in a direction perpendicular to the direction of interpolation sampling points during resampling, resulting in an even number of lateral acquisition routes; Step 8: Connect the lateral flow path and the main crack measurement flow path in a bow-shaped sequence to generate the crack measurement flow path; The resampling method in step two is as follows: The crack location data was converted from a geographic coordinate system using latitude and longitude to a projected coordinate system using meters. A threshold is set; if the slope from point n to point n+1 in the crack location data is greater than the threshold, then the horizontal acquisition interval is set. Calculate the location of the interpolation sampling points and perform interpolation; otherwise, use the vertical sampling interval. Calculate the positions of the interpolation sampling points and perform interpolation; After each interpolation, calculate the distance from the interpolated sampling point to point n+1 until the distance is less than the distance from the previous interpolated sampling point to point n+1. This completes the interpolation from point n to point n+1. Then, replace the position of point n+1 with the position of the last interpolated point and perform the interpolation from point n+1 to point n+2 according to the above steps. The extraction method in step three is as follows: Using the resampling point as the center of the coordinate system, the yaw angle is calculated using the arctangent function of two adjacent points, as shown in the following formula: ； Extract the point cloud data within the area enclosed by the four vertices where the center point intersects the yaw angle direction and the perpendicular yaw angle direction. The yaw angle direction of the center point is: The direction of the vertical yaw angle is: ; Extract point cloud data within the coverage area defined by the elevation of the center point from the point cloud data enclosed by the four vertices mentioned above, where the elevation is [value missing]. .

2. The inspection route planning method as described in claim 1, characterized in that, In step one, the parameters of the acquisition device are pixels along the X-axis. Pixels along the Y-axis Lens focal length and pixel size The expected sampling overlap is the horizontal overlap. Vertical overlap ;in: The calculation formula is: ; The calculation formula is: ; The calculation formula is: .

3. The inspection route planning method as described in claim 1, characterized in that, In step two, the threshold is set as follows: Calculate the aspect ratio of the acquisition device and use the obtained aspect ratio value as the threshold; The formula for calculating the threshold is: ; The formula for calculating the slope from point n to point n+1 in the crack location data is as follows: ； Calculate the yaw angle from point n to point n+1 in the crack location data. The calculation formula is as follows: ； If the slope is greater than the threshold, then the horizontal sampling interval will be used. Calculate the position of the interpolation sampling point And perform interpolation; Among them, if ,but ; like ,but ; ； ； If the slope is not greater than the threshold, then the sampling interval in the vertical direction is used. Calculate the position of the interpolation sampling point And perform interpolation; Among them, if ,but ; like ,but ; ； 。 4. The inspection route planning method as described in claim 1, characterized in that, The method for calculating the average normal vector of the image coverage area in step four is as follows: After removing the normal vectors with the maximum and minimum values, the average normal vector of the remaining normal vectors is calculated using a function that calculates the average value.

5. The inspection route planning method as described in claim 1, characterized in that, In step seven, if the interpolation sampling points in the resampling process are at horizontal sampling intervals... For interpolation, the sampling interval is used in the vertical direction. Perform integer-multiple buffering if the interpolation sampling points during the resampling process are at acquisition intervals perpendicular to the horizontal direction. For interpolation, the data is collected at horizontal intervals. Perform buffering at integer multiples.

Images