Low-Error Surveying Method and System Based on UAV Surveying and Mapping

Abstract

The present invention relates to the technical field of unmanned aerial vehicles, and discloses a low-error mapping method and system based on unmanned aerial vehicle mapping. The method includes: using an unmanned aerial vehicle to collect ground point cloud images, and obtaining all overlapping regions of consecutive frame point cloud images; obtaining a set of three-dimensional coordinates of each position point in the overlapping region in all relevant frames, calculating the degree of dispersion of each set of three-dimensional coordinates, and optimizing each set of three-dimensional coordinates of the overlapping region based on a preset dispersion threshold to obtain an optimal set of three-dimensional coordinates for each position point; the relevant frames are all frame images containing a position point for any position point; calculating the mean value of each dimension in the optimal set of three-dimensional coordinates to obtain the optimal three-dimensional coordinates of the corresponding position point, and then using the optimal three-dimensional coordinates of all position points to draw a topographic map. The present invention can improve the accuracy of mapping and make the drawn topographic map more accurately reflect the actual terrain and landform.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and particularly to a low-error mapping method and system based on UAV mapping. Background Art

[0002] UAV mapping is an efficient geographic information acquisition technology based on the UAV platform. This technology integrates professional mapping sensors such as multispectral cameras and Light Detection and Ranging (LiDAR) on UAVs to achieve real-time acquisition of high-resolution geospatial data during low-altitude flight operations. After professional data processing, diverse mapping results such as high-precision Digital Line Graphic (DLG), Digital Elevation Model (DEM), Digital Orthophoto Map (DOM), and three-dimensional real-scene models can be generated. This technology significantly improves the acquisition efficiency and measurement accuracy of geographic information data, providing important technical support for fields such as national land surveying and mapping, smart city construction, disaster monitoring, and agricultural census.

[0003] During the actual operation process, environmental noise generated under complex environments or adverse weather conditions will significantly reduce the three-dimensional geometric accuracy of point cloud data, and thus have an adverse impact on the accuracy of subsequent three-dimensional modeling. More critically, due to the pose estimation error in the registration process of consecutive frame point clouds, the occurrence of registration misalignment will be caused. This registration error will not only result in inconsistent three-dimensional coordinates in the overlapping area, but also significantly reduce the spatial consistency of the overall point cloud data, thereby affecting the overall quality of three-dimensional reconstruction. Summary of the Invention

[0004] To solve the above technical problem of how to achieve high-precision and low-error mapping, the present invention provides solutions in the following aspects.

[0005] In the first aspect, a low-error mapping method based on UAV mapping includes:

[0006] A low-error mapping method based on UAV mapping, characterized by including:

[0007] Using a UAV to collect ground point cloud images and obtain all overlapping areas of consecutive frame point cloud images;

[0008] Obtaining the three-dimensional coordinate sets of each position point in the overlapping area in all relevant frames, calculating the dispersion degree of each three-dimensional coordinate set, and optimizing each three-dimensional coordinate set in the overlapping area based on a preset dispersion threshold to obtain the optimal three-dimensional coordinate set of each position point; the relevant frames are all frame images containing a position point for any one position point;

[0009] Calculating the mean value of each dimension in the optimal three-dimensional coordinate set to obtain the optimal three-dimensional coordinates of the corresponding position point, and then using the optimal three-dimensional coordinates of all position points to draw a topographic map.

[0010] In the present invention, by obtaining the three-dimensional coordinate sets of each position point in the overlapping region in all relevant frames, calculating the degree of dispersion, and optimizing based on a preset dispersion threshold, it is possible to effectively eliminate coordinate data with large errors and obtain the optimal three-dimensional coordinate sets of each position point. Then, the mean value of each dimension in the optimal three-dimensional coordinate set is calculated as the final optimal three-dimensional coordinate. This series of operations can significantly reduce coordinate deviations caused by single measurement errors, UAV attitude changes, environmental factor interferences, etc., thereby improving the accuracy of surveying and mapping and enabling the topographic map drawn to more accurately reflect the actual terrain and landform.

[0011] Preferably, the process of obtaining the overlapping region includes:

[0012] Select one frame of image as the reference frame and another frame of image as the target frame. Move the reference frame point by point on the target frame. Each time a movement is made, an overlapping region is obtained, and the overlapping coefficient of the overlapping region is calculated. When the overlapping coefficient reaches the minimum value, stop the movement to obtain all the overlapping regions.

[0013] By calculating the overlapping regions at different positions and their overlapping coefficients, the overlapping region corresponding to the minimum overlapping coefficient indicates the best matching position of the reference frame on the target frame. By obtaining these overlapping regions, the images can be stitched into a complete panoramic image.

[0014] Preferably, the overlapping coefficient is the average value of the absolute values of the differences in the three-dimensional coordinates of each position point in the overlapping region between the reference frame and the target frame.

[0015] The overlapping coefficient can intuitively reflect the accuracy of point cloud matching between the reference frame and the target frame. If the overlapping coefficient is small, it indicates that the difference in the three-dimensional coordinates of the corresponding position points in the two frames of images is small and the matching accuracy is high; on the contrary, if the overlapping coefficient is large, it indicates that there are large errors in the matching. For example, during the UAV surveying and mapping process, due to factors such as changes in flight attitude and different shooting angles, there may be deviations in point cloud matching between different frames. By calculating the overlapping coefficient, these problems can be discovered in a timely manner.

[0016] Preferably, the process of obtaining the degree of dispersion includes:

[0017] Obtain the position of the center of the minimum circumscribed sphere of the three-dimensional coordinate set, calculate the Euclidean distance from each three-dimensional coordinate in the three-dimensional coordinate set to the center of the sphere, use the cosine value of the angle between each three-dimensional coordinate and the center of the sphere in the three dimensions as the angular deviation in the corresponding dimension, apply the hyperbolic tangent function transformation to both the standard deviation of all the Euclidean distances and the standard deviation of all the angular deviations in each dimension, and then multiply them and take the maximum value to obtain the degree of dispersion of the three-dimensional coordinate set.

[0018] By obtaining the position of the center of the minimum circumscribed sphere of the three-dimensional coordinate set and calculating the Euclidean distance from each three-dimensional coordinate to the center of the sphere and the angular deviation in each dimension, the distribution of the three-dimensional coordinates in space is comprehensively considered from two dimensions of distance and angle. The Euclidean distance reflects the difference in the distance between the coordinate point and the center of the sphere, and the angular deviation reflects the directional distribution of the coordinate point around the center of the sphere. The combination of the two can more accurately describe the degree of dispersion of the three-dimensional coordinate set.

[0019] Preferably, the optimization of each three-dimensional coordinate set in the overlapping region based on a preset dispersion threshold includes:

[0020] When the degree of dispersion of the three-dimensional coordinate set is greater than the preset dispersion threshold, calculate the deviation coefficient of each three-dimensional coordinate in the three-dimensional coordinate set, and take the three-dimensional coordinate with the largest deviation coefficient as the high-error three-dimensional coordinate and remove it to obtain a new three-dimensional coordinate set, and repeat the calculation of the degree of dispersion and the removal of the high-error three-dimensional coordinate until the degree of dispersion is less than or equal to the preset dispersion threshold. At this time, the new three-dimensional coordinate set composed of the remaining three-dimensional coordinates is the optimal three-dimensional coordinate set; when the degree of dispersion of the three-dimensional coordinate set is less than or equal to the preset dispersion threshold, the three-dimensional coordinate set at this time is the optimal three-dimensional coordinate set.

[0021] When the degree of dispersion of the three-dimensional coordinate set is greater than the preset dispersion threshold, by calculating the deviation coefficient and removing the high-error three-dimensional coordinate with the largest deviation coefficient, the outliers and points with large errors in the data can be effectively reduced. For example, in the process of UAV mapping, due to reasons such as sensor failure, environmental factor interference, or unstable flight attitude, some three-dimensional coordinates may have large deviations. These high-error data will seriously affect the accuracy of the mapping results. Through this optimization process, these high-error data can be eliminated, thereby improving the quality of the coordinate data; as the high-error three-dimensional coordinates are continuously removed, the degree of dispersion of the new three-dimensional coordinate set gradually decreases, and the consistency of the data is enhanced. This means that each three-dimensional coordinate in the set is closer to the real topographic and geomorphic information and can better reflect the actual situation.

[0022] Preferably, the process of obtaining the deviation coefficient includes:

[0023] Calculate the degree of distance deviation of each three-dimensional coordinate from the center of the sphere; calculate the degree of angular deviation of each three-dimensional coordinate from the center of the sphere coordinates;

[0024] Take the product of the degree of distance deviation and the degree of angular deviation as the deviation coefficient of each three-dimensional coordinate.

[0025] Calculate the distance deviation degree and the angle deviation degree of each three-dimensional coordinate from the center of the sphere respectively, and comprehensively evaluate the deviation of the three-dimensional coordinates from the position of the center of the sphere from two dimensions of distance and angle. The distance deviation degree reflects the difference in the distance between the coordinate point and the center of the sphere in the spatial position, while the angle deviation degree reflects the directional distribution of the coordinate point around the center of the sphere. Combining the two can more accurately describe the comprehensive deviation characteristics of the three-dimensional coordinates.

[0026] Preferably, the process of obtaining the distance deviation degree includes:

[0027] Take the ratio of the difference between the distance from the three-dimensional coordinate to the center of the sphere and the average value of the distances from all three-dimensional coordinates to the center of the sphere to the average value of the distances from all three-dimensional coordinates to the center of the sphere as the distance deviation degree.

[0028] Directly comparing the numerical values of the distances from the three-dimensional coordinates to the center of the sphere may be affected by the data magnitude and it is difficult to accurately reflect the relative size of the distance deviation. By dividing the difference between the distance and the average value of the distances from all three-dimensional coordinates to the center of the sphere by the average value, the distance deviation degree is obtained. This normalization process eliminates the influence of the dimension, enabling the distance deviation degrees of different coordinate points to be compared on the same scale.

[0029] Preferably, the process of obtaining the angle deviation degree includes:

[0030] Obtain the mean value of the sum of the cosine values of the angles between the three-dimensional vector from the three-dimensional coordinate to the center of the sphere and the three-dimensional vectors from all other three-dimensional coordinates to the center of the sphere, and normalize this mean value as the angle deviation degree.

[0031] By calculating the mean value of the sum of the cosine values of the angles between the three-dimensional vector from the three-dimensional coordinate to the center of the sphere and the three-dimensional vectors from all other three-dimensional coordinates to the center of the sphere, the relationship between this three-dimensional coordinate and all other coordinates in terms of direction is comprehensively considered. The cosine value of the angle reflects the size of the angle between two vectors, and the mean value reflects the average angular distribution of this coordinate in the entire coordinate set. For example, in a complex terrain mapping, the three-dimensional vectors of each point are directionally related to other points, and this calculation method can comprehensively capture these correlation information.

[0032] Preferably, the process of obtaining the position of the center of the sphere includes:

[0033] Calculate the Euclidean distance between any two three-dimensional coordinates, obtain the two three-dimensional coordinates corresponding to the maximum Euclidean distance, and calculate the center coordinates of these two three-dimensional coordinates. The center coordinates are the center coordinates of the minimum circumscribed sphere.

[0034] Obtaining the center coordinates of the minimum circumscribed sphere provides a unified spatial reference benchmark for the three-dimensional coordinate set.

[0035] Second aspect, an on-line monitoring system for a pressure transmitter, comprising: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, any one of the low-error mapping systems based on unmanned aerial vehicle mapping is implemented.

[0036] Compared with the prior art, the beneficial effects of the low-error mapping method based on unmanned aerial vehicle mapping in the embodiments of the present invention are as follows:

[0037] First, the present invention calculates the dispersion degree of the three-dimensional coordinate set and optimizes it based on a preset dispersion threshold, which can remove high-error three-dimensional coordinate points, and then calculates the mean value of each dimension in the optimized three-dimensional coordinate set to obtain the optimal three-dimensional coordinates of each position point, further improving the mapping accuracy of the topographic map. Description of the Drawings

[0038] Figure 1 is a flowchart of the method from step S1 to step S3 in the low-error mapping method based on unmanned aerial vehicle mapping in the embodiments of the present invention.

[0039] Figure 2 is a three-dimensional point cloud map composed of the optimal three-dimensional coordinates corresponding to all position points in the embodiments of the present invention, wherein the X-axis is the extension of the terrain in the east-west direction, the Y-axis is the extension of the terrain in the north-south direction, and the Z-axis is the height information of the terrain.

[0040] Figure 3 is the terrain surface formed by mapping based on the optimal three-dimensional coordinates in the embodiments of the present invention, wherein the X-axis is the extension of the terrain in the east-west direction, the Y-axis is the extension of the terrain in the north-south direction, and the Z-axis is the height information of the terrain. Detailed Embodiments

[0041] The following further describes in detail the specific embodiments of the present invention in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0042] As Figure 1 shown, the preferred embodiment of the low-error mapping method based on unmanned aerial vehicle mapping in the embodiments of the present invention is as follows:

[0043] S1: Use an unmanned aerial vehicle to collect ground point cloud images and obtain all overlapping regions of consecutive frame point cloud images.

[0044] In one embodiment, according to factors such as the operation range and terrain complexity, a suitable unmanned aerial vehicle is selected, and a suitable sensor and other collection devices are installed on the unmanned aerial vehicle, and it is ensured that the collection device is normally connected to the flight control system of the unmanned aerial vehicle. After installation, preliminary communication testing and calibration are carried out.

[0045] Start the drone and let it fly stably according to the requirements of the operation manual and the planned flight route.

[0046] During the flight, collect ground point cloud images in real time. Each position point on each frame of the collected point cloud images contains a three-dimensional coordinate.

[0047] It should be noted that during the flight of the drone, when its carried sensors (such as lidar, depth camera, etc.) continuously collect three-dimensional point cloud data of the ground or target objects, due to motion, sensor characteristics or environmental factors, there may be deviations or inconsistent overlaps in the spatial positions of the point cloud data of adjacent frames.

[0048] In one embodiment, select one frame of image as the reference frame and another frame of image as the target frame. Move the reference frame point by point on the target frame. Each time a movement is made, an overlapping area is obtained. Calculate the overlapping coefficient of the overlapping area. When the overlapping coefficient reaches the minimum value, stop the movement to obtain all the overlapping areas.

[0049] Among them, calculate the mean value of the absolute values of the differences in the three-dimensional coordinates (i.e., the mean value of the differences in the three dimensions) between the target frame and the reference frame for all the position points in the overlapping area, which is the overlapping coefficient of the above-mentioned overlapping area.

[0050] Furthermore, all the overlapping areas of all the frames of point cloud images collected can be obtained.

[0051] S2: Obtain the set of three-dimensional coordinates of each position point in the overlapping area in all relevant frames, calculate the degree of dispersion of each set of three-dimensional coordinates, and optimize each set of three-dimensional coordinates of the overlapping area based on a preset dispersion threshold to obtain the optimal set of three-dimensional coordinates for each position point; the relevant frames are all the frame images containing a position point for any one position point.

[0052] In one embodiment, for each position point in the overlapping area, it is necessary to collect and sort out the set of coordinates of this point in three-dimensional space from all the frame images containing this point, that is, obtain the set of three-dimensional coordinates of each position point in the overlapping area in all relevant frames. For example, the three-dimensional coordinate set of a position point is: , where the three-dimensional coordinates of .

[0053] When the drone collects three-dimensional information, due to the influence of rainy days or interference from complex environments, there may be differences in the three-dimensional information on different frame images. Therefore, it is necessary to identify which three-dimensional information units have high errors, so as to denoise this information and improve the accuracy of the subsequent obtained three-dimensional information points.

[0054] In one embodiment, first, find a minimum circumscribed sphere for the obtained three-dimensional coordinate set. Specifically, calculate the Euclidean distance between any two three-dimensional coordinates in the three-dimensional coordinate set, obtain the two three-dimensional coordinates corresponding to the maximum Euclidean distance and use them as the sphere diameter group, and calculate the central coordinates of these two three-dimensional coordinates. This central coordinate is the center coordinate of the minimum circumscribed sphere.

[0055] The center of the above-mentioned minimum circumscribed sphere provides a common reference point for the three-dimensional coordinate set, which helps to uniformly describe and compare the relative positional relationships of each three-dimensional coordinate with the overall set.

[0056] Furthermore, for each three-dimensional coordinate in the three-dimensional coordinate set, calculate its Euclidean distance from the center coordinate and obtain the standard deviation of all Euclidean distances; then, use the cosine function to calculate the angular deviation degree of each three-dimensional coordinate from the center coordinate in the x-axis direction and obtain the standard deviation of all angular deviation degrees. Use the hyperbolic tangent function to combine these two standard deviation sets to obtain the dispersion degree of the three-dimensional coordinate set in the x-axis direction. Similarly, the dispersion degrees of the three-dimensional coordinate set in the y-axis direction and the z-axis direction can be obtained. Take the maximum value of these three dispersion degrees as the dispersion degree of the entire three-dimensional coordinate set.

[0057] According to the above operations, the dispersion degrees of all three-dimensional coordinate sets can be obtained in the same way. Take the quartiles of all dispersion degrees as their dispersion thresholds. When the dispersion degree of each three-dimensional coordinate set is greater than this dispersion threshold, calculate the deviation coefficient of each three-dimensional coordinate in the three-dimensional coordinate set, and take the three-dimensional coordinate with the largest deviation coefficient as the high-error three-dimensional coordinate and remove it to obtain a new three-dimensional coordinate set. Repeat the calculation of the dispersion degree and the removal of high-error three-dimensional coordinates until the dispersion degree is less than or equal to the preset dispersion threshold. At this time, the new three-dimensional coordinate set composed of the remaining three-dimensional coordinates is the optimal three-dimensional coordinate set of the corresponding position points, and then the optimal three-dimensional coordinate sets of all position points in all overlapping regions can be obtained.

[0058] Conversely, when the dispersion degree of the three-dimensional coordinate set is less than or equal to the preset dispersion threshold, the three-dimensional coordinate set at this time is the optimal three-dimensional coordinate set.

[0059] Among them, the calculation process of the above-mentioned deviation coefficient is as follows:

[0060] First, take the ratio of the difference between the distance from each three-dimensional coordinate in the three-dimensional coordinate set to the center of the sphere and the average value of the distances from all three-dimensional coordinates to the center of the sphere to the average value of the distances from all three-dimensional coordinates to the center of the sphere as the distance deviation degree of the three-dimensional coordinate.

[0061] Then, obtain the mean value of the sum of the cosine values of the angles between the three-dimensional vector from the three-dimensional coordinate to the center of the sphere and the three-dimensional vectors from all other three-dimensional coordinates to the center of the sphere, and standardize this mean value as the angular deviation degree of the three-dimensional coordinate from the center coordinate.

[0062] Finally, the product of the above distance deviation degree and the angle deviation degree is used as the deviation coefficient of the corresponding three-dimensional coordinate.

[0063] Among them, the above angle deviation degree satisfies the relational expression as:

[0064]

[0065] In the formula, is the angle deviation degree from the th three-dimensional coordinate in the three-dimensional coordinate set to the center-of-sphere coordinate, is the total number of all other three-dimensional coordinates except the th three-dimensional coordinate in the three-dimensional coordinate set, is the three-dimensional vector of the th three-dimensional coordinate, is the three-dimensional vector of the th three-dimensional coordinate.

[0066] S3: Calculate the mean value of each dimension in the optimal three-dimensional coordinate set to obtain the optimal three-dimensional coordinate of the corresponding position point, and then use the optimal three-dimensional coordinates of all position points to draw a topographic map.

[0067] According to the optimal three-dimensional coordinate set of all position points obtained in the above S2, calculate the mean values of the coordinate values on the x-axis, y-axis, and z-axis in the optimal three-dimensional coordinate set, and then obtain the optimal three-dimensional coordinate of this optimal three-dimensional coordinate set (that is, obtain a three-dimensional coordinate point representing the "center" or "average" position of this group of three-dimensional coordinate data). Furthermore, after all position points undergo the above operations, their respective corresponding optimal three-dimensional coordinates can be obtained.

[0068] Then, using the optimal three-dimensional coordinates of all position points, these optimal three-dimensional coordinates can be connected through a drawing software to form a continuous terrain surface.

[0069] Refer to Figure 2 and Figure 3 , Figure 2 is the three-dimensional point cloud map composed of the optimal three-dimensional coordinates corresponding to all position points in the embodiment of the present invention. Among them, the X-axis is the extension of the terrain in the east-west direction, the Y-axis is the extension of the terrain in the north-south direction, and the Z-axis is the height information of the terrain.

[0070] Figure 3 is the terrain surface formed based on the optimal three-dimensional coordinates in the embodiment of the present invention. Among them, the X-axis is the extension of the terrain in the east-west direction, the Y-axis is the extension of the terrain in the north-south direction, and the Z-axis is the height information of the terrain.

[0071] The system includes a processor and a memory. The memory stores computer program instructions which, when executed by the processor, implement the low-error mapping method based on UAV mapping according to the first aspect of the present invention.

[0072] The system also includes other components well-known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art and thus will not be elaborated herein.

[0073] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art of the present technology, several improvements and replacements can be made without departing from the counting principle of the present invention, and these improvements and replacements should also be regarded as the protection scope of the present invention.

Claims

1. A low-error surveying method based on UAV surveying and mapping, characterized in that, Including: Using a drone to collect ground point cloud images and obtaining all overlapping regions of consecutive frame point cloud images; Obtaining the three-dimensional coordinate sets of each position point in the overlapping region in all relevant frames, calculating the dispersion degree of each three-dimensional coordinate set, and optimizing each three-dimensional coordinate set in the overlapping region based on a preset dispersion threshold to obtain the optimal three-dimensional coordinate set for each position point; the relevant frames are all frame images containing a position point for any position point; the optimizing each three-dimensional coordinate set in the overlapping region based on a preset dispersion threshold includes: When the dispersion degree of the three-dimensional coordinate set is greater than the preset dispersion threshold, calculating the deviation coefficient of each three-dimensional coordinate in the three-dimensional coordinate set, taking the three-dimensional coordinate with the largest deviation coefficient as the high-error three-dimensional coordinate and removing it to obtain a new three-dimensional coordinate set, and repeating the calculation of the dispersion degree and the removal of the high-error three-dimensional coordinate until the dispersion degree is less than or equal to the preset dispersion threshold. At this time, the new three-dimensional coordinate set composed of the remaining three-dimensional coordinates is the optimal three-dimensional coordinate set; when the dispersion degree of the three-dimensional coordinate set is less than or equal to the preset dispersion threshold, the three-dimensional coordinate set at this time is the optimal three-dimensional coordinate set; Calculating the mean value of each dimension in the optimal three-dimensional coordinate set to obtain the optimal three-dimensional coordinates of the corresponding position point, and then using the optimal three-dimensional coordinates of all position points to draw a topographic map.

2. The low-error mapping method based on UAV mapping according to claim 1, characterized in that, The obtaining process of the overlapping region includes: Selecting one frame image as the reference frame and another frame image as the target frame, moving the reference frame point by point on the target frame, obtaining an overlapping region each time, calculating the overlapping coefficient of the overlapping region, and stopping moving when the overlapping coefficient reaches the minimum value to obtain all overlapping regions.

3. The low-error mapping method based on UAV mapping according to claim 2, wherein, The overlapping coefficient is the average value of the absolute values of the differences in the three-dimensional coordinates of each position point in the overlapping region between the reference frame and the target frame.

4. The low-error mapping method based on drone mapping according to claim 3, characterized in that, The obtaining process of the dispersion degree includes: Obtaining the center position of the minimum circumscribed sphere in the three-dimensional coordinate set, calculating the Euclidean distance from each three-dimensional coordinate in the three-dimensional coordinate set to the center of the sphere, taking the cosine value of the angle between each three-dimensional coordinate and the center of the sphere in three dimensions as the angular deviation in the corresponding dimension, applying the hyperbolic tangent function to transform both the standard deviation of all Euclidean distances and the standard deviation of all angular deviations in each dimension, and then multiplying and taking the maximum value to obtain the dispersion degree of the three-dimensional coordinate set.

5. The low-error mapping method based on UAV mapping according to claim 4, wherein The obtaining process of the deviation coefficient includes: Calculating the distance deviation degree from each three-dimensional coordinate to the center of the sphere; calculating the angular deviation degree from each three-dimensional coordinate to the center coordinate of the sphere; Taking the product of the distance deviation degree and the angular deviation degree as the deviation coefficient of each three-dimensional coordinate.

6. The low-error mapping method based on UAV mapping according to claim 5, wherein, The obtaining process of the distance deviation degree includes: Taking the ratio of the difference between the distance from the three-dimensional coordinate to the center of the sphere and the average value of the distances from all three-dimensional coordinates to the center of the sphere to the average value of the distances from all three-dimensional coordinates to the center of the sphere as the distance deviation degree.

7. The low-error mapping method based on UAV mapping according to claim 6, characterized in that, The obtaining process of the angular deviation degree includes: Obtaining the mean value of the sum of the cosine values of the angles between the three-dimensional vector from the three-dimensional coordinate to the center of the sphere and the three-dimensional vectors from all other three-dimensional coordinates to the center of the sphere, and normalizing this mean value as the angular deviation degree.

8. The low-error mapping method based on UAV mapping according to claim 7, wherein, The obtaining process of the center position of the sphere includes: Calculate the Euclidean distance between any two three-dimensional coordinates, obtain the two three-dimensional coordinates corresponding to the maximum Euclidean distance, and calculate the central coordinates of these two three-dimensional coordinates. The central coordinates are the center coordinates of the minimum circumscribed sphere.

9. A low-error surveying and mapping system based on UAV surveying and mapping, characterized in that, Including: A processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the low-error mapping method based on unmanned aerial vehicle mapping according to any one of claims 1-8 is implemented.

Images