Method for correcting high-precision surveying and mapping error based on flight state of unmanned aerial vehicle

Abstract

The invention discloses a method for correcting a high-precision surveying and mapping error based on the flight state of an unmanned aerial vehicle, and relates to the technical field of data error correction, and the method comprises the steps: obtaining task flight path surveying and mapping attitude data of a to-be-surveyed area of the unmanned aerial vehicle, and carrying out the dynamic weight distribution according to federated filtering; generating a task flight path space-time synchronization matrix of the surveying and mapping area of the unmanned aerial vehicle; establishing an exposure control-distortion compensation function according to the task flight path space-time synchronization matrix of the surveying and mapping area of the unmanned aerial vehicle, and correcting task initialization image parameters of the surveying and mapping area of the unmanned aerial vehicle to obtain task correction image parameters of the surveying and mapping area of the unmanned aerial vehicle; and updating the optimal estimation of the surveying and mapping attitude data of the task flight path of the surveying and mapping area of the unmanned aerial vehicle, and performing iterative matching with the map point cloud of the surveying and mapping area to generate a high-precision image of the surveying and mapping area of the unmanned aerial vehicle. The method has the advantages that the surveying and mapping precision of the unmanned aerial vehicle is improved, error accumulation is reduced, and image quality is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of data error correction, and in particular to a method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle. Background Art

[0002] In existing UAV high-precision mapping, the accuracy of flight state estimation is insufficient, which leads to the continuous accumulation of mapping errors during long-term flight, affecting the accuracy of the final image and point cloud data. For complex terrain or environmental conditions, sensor errors and inaccurate attitude estimation will aggravate the propagation of errors, resulting in an increase in the deviation of the overall mapping results, making it difficult to meet high-precision requirements. Summary of the invention

[0003] In order to solve the above technical problems, a method for correcting high-precision mapping errors based on the flight status of an unmanned aerial vehicle is provided. This technical solution solves the problem that in the above-mentioned existing high-precision mapping of unmanned aerial vehicles, the flight status estimation accuracy is insufficient, resulting in the continuous accumulation of mapping errors during long-term flight, affecting the accuracy of the final image and point cloud data. For complex terrain or environmental conditions, sensor errors and inaccurate attitude estimation will aggravate the propagation of errors, resulting in an increase in the deviation of the overall mapping results, making it difficult to meet high-precision requirements.

[0004] In order to achieve the above purpose, the technical solution adopted by the present invention is: A method for correcting high-precision mapping errors based on the flight status of an unmanned aerial vehicle, comprising: Based on the UAV mapping management background, determine the mission flight path of the UAV in the area to be mapped; Based on heterogeneous sensors, the mission flight path mapping attitude data of the UAV in the area to be mapped is obtained, the multi-source errors in the mission flight path mapping attitude data are marked, and dynamic weight allocation is performed according to the federated filter to generate the mission flight path time-space synchronization matrix of the UAV's mapping area; Based on the mission flight path mapping attitude data of the UAV's mapping area, establish the mission initialization image parameters of the UAV's mapping area; An exposure control-distortion compensation function is established according to the space-time synchronization matrix of the mission flight path of the UAV's surveying and mapping area, and the mission initialization image parameters of the UAV's surveying and mapping area are corrected to obtain the mission correction image parameters of the UAV's surveying and mapping area; According to the mission of the UAV's mapping area, the image parameters are corrected, the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area is updated, and iterative matching is performed with the map point cloud of the mapping area to generate a high-precision image of the UAV's mapping area.

[0005] Preferably, based on heterogeneous sensors, obtaining the mission flight path mapping attitude data of the UAV in the area to be mapped, marking the multi-source errors in the mission flight path mapping attitude data, performing dynamic weight allocation according to federated filtering, and generating the mission flight path spatiotemporal synchronization matrix of the UAV in the mapping area specifically includes: Based on heterogeneous sensors, the mission flight path mapping attitude data of the UAV's mapping area is obtained for preprocessing; the mapping attitude data includes: IMU data, RTK-GNSS data, and point cloud data; According to the federated filtering architecture, a child individual Kalman filter is established based on each type of mapping attitude data in the mission flight path mapping attitude data of the mapping area of ​​the UAV, and a local optimal estimate is generated and substituted into the parent Kalman filter to obtain the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area. According to the optimal estimate of the attitude data of the mission flight path of the UAV's mapping area, the time synchronization matrix is ​​formed by aligning the timestamps of each data; According to the optimal estimate of the attitude data of the mission flight path of the UAV's mapping area, the carrier coordinates of each data are converted to the navigation coordinate system to establish a space conversion matrix; Based on the time synchronization matrix and the space conversion matrix, a coordinate rotation correction-time delay compensation function is established to generate the space-time synchronization matrix of the mission flight path of the UAV's mapping area; , in, The measured position coordinates of the mapping attitude data of the i-th mission flight path of the UAV’s mapping area, is the acquisition timestamp of the mapping attitude data of the flight path of the i-th mission in the mapping area of ​​the UAV, is the space transformation matrix, is the time synchronization matrix, is the measured velocity vector of the UAV inertial measurement unit, is the time synchronization error of the UAV mapping attitude data, is the global coordinate position of the i-th mission flight path mapping attitude data of the UAV's mapping area after time-space alignment, It is the timestamp after synchronization of the mapping attitude data of the flight path of the i-th mission in the mapping area of ​​the UAV.

[0006] Preferably, an exposure control-distortion compensation function is established according to the spatiotemporal synchronization matrix of the mission flight path of the surveying and mapping area of ​​the UAV, and the mission initialization image parameters of the surveying and mapping area of ​​the UAV are corrected. The mission correction image parameters of the surveying and mapping area of ​​the UAV are obtained, which specifically include: Based on the spatiotemporal synchronization matrix of the mission flight path of the surveying and mapping area of ​​the UAV, the illumination data and image data of the mission flight path of the surveying and mapping area of ​​the UAV are marked according to the unit time; Based on the illumination data of the mission flight path of the UAV's surveying and mapping area, the AE illumination intensity distribution function is constructed by associating the point cloud reflectivity with the visual RGB value, and the image illumination intensity distribution value of the mission flight path of the UAV's surveying and mapping area is generated; Substituting the image illumination intensity distribution value of the mission flight path of the surveying and mapping area of ​​the UAV into the illumination field model, the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV is generated; Based on the CNN convolutional neural network, the original image data and IMU data of the mission flight path of the UAV's surveying area are used as input, and the distortion-compensated image data of the mission flight path of the UAV's surveying area is used as output; Using the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV and the distortion compensation image data of the mission flight path of the surveying and mapping area of ​​the UAV, the mission initialization image parameters of the surveying and mapping area of ​​the UAV are corrected to obtain the mission correction image parameters of the surveying and mapping area of ​​the UAV; Among them, the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV is specifically: , In the formula, is the dynamic image brightness value of the i-th mission flight path in the surveying area of ​​the UAV, is the illumination intensity fusion value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the visual RGB value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the point cloud reflectivity value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the visual RGB sensor weight coefficient, is the cloud reflectivity weight coefficient (dynamically updated through federated filtering), is the regional weight of the i-th mission flight path in the surveying and mapping area of ​​the UAV, and N is the total number of mission flight paths in the surveying and mapping area of ​​the UAV; The distortion-compensated image data of the mission flight path of the surveying area of ​​the UAV is specifically: , In the formula, is the distortion-compensated image data of the flight path of the ith mission in the mapping area of ​​the UAV, is the original image data of the flight path of the ith mission in the surveying area of ​​the UAV, is the IMU data of the flight path of the i-th mission in the mapping area of ​​the UAV, is the attitude angle, It is the CNN convolutional neural network interface function.

[0007] Preferably, according to the mission correction image parameters of the surveying and mapping area of ​​the UAV, the optimal estimate of the mission flight path surveying and mapping attitude data of the surveying and mapping area of ​​the UAV is updated, and iterative matching is performed with the map point cloud of the surveying and mapping area to generate a high-precision image of the surveying and mapping area of ​​the UAV; Based on the mission correction image parameters of the UAV's mapping area, the Kalman filter update formula is used to update the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area, and the optimal state estimate of the mission flight path mapping attitude data of the UAV's mapping area is generated; Based on the optimal state estimation of the mission flight path mapping posture data of the UAV's mapping area and the map point cloud of the mapping area, iterative matching is performed according to the ICP iterative closest point algorithm to generate a high-precision image of the UAV's mapping area; Among them, the update formula of the Mann filter is specifically: , In the formula, is the optimal state estimate of the flight path of the ith mission in the mapping area of ​​the UAV, is the optimal estimate of the flight path of the ith mission in the mapping area of ​​the UAV, The state transfer matrix of the attitude data for the mission flight path mapping of the UAV's mapping area, is the Kalman gain, The original value of the mission flight path mapping attitude data for the UAV's mapping area, An observation matrix of attitude data for the mission flight path of the UAV mapping area.

[0008] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes a high-precision mapping error correction scheme based on the flight status of a UAV. The mapping error is corrected by using the flight status of the UAV, and the multi-source errors generated by the UAV during flight are accurately corrected by combining heterogeneous sensor data and federal filtering technology. By dynamically adjusting the weights of the flight path data, a spatiotemporal synchronization matrix is ​​generated, and then the initialization image parameters are established and corrected. The exposure control and distortion compensation models are used to optimize the image quality, and high-precision complete image parameters are generated through iterative matching with the map point cloud. The beneficial effects of this scheme are: improving the mapping accuracy of UAVs, reducing error accumulation, and optimizing image quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1This is a flow chart of a method for correcting high-precision mapping errors based on the flight status of a UAV; Figure 2 Flowchart of the method for generating a spatiotemporal synchronization matrix of the mission flight path of the UAV's mapping area; Figure 3 Flowchart of the method for correcting image parameters for the task of obtaining the mapping area of ​​the UAV; Figure 4 Flowchart of a method for generating high-precision images of an area being mapped by a drone. DETAILED DESCRIPTION

[0010] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0011] Reference Figure 1 As shown, a method for correcting high-precision mapping errors based on the flight status of a UAV includes: Based on the UAV mapping management background, determine the mission flight path of the UAV in the area to be mapped; Based on heterogeneous sensors, the mission flight path mapping attitude data of the UAV in the area to be mapped is obtained, the multi-source errors in the mission flight path mapping attitude data are marked, and dynamic weight allocation is performed according to the federated filter to generate the mission flight path time-space synchronization matrix of the UAV's mapping area; Based on the mission flight path mapping attitude data of the UAV's mapping area, establish the mission initialization image parameters of the UAV's mapping area; An exposure control-distortion compensation function is established according to the space-time synchronization matrix of the mission flight path of the UAV's surveying and mapping area, and the mission initialization image parameters of the UAV's surveying and mapping area are corrected to obtain the mission correction image parameters of the UAV's surveying and mapping area; According to the mission of the UAV's mapping area, the image parameters are corrected, the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area is updated, and iterative matching is performed with the map point cloud of the mapping area to generate a high-precision image of the UAV's mapping area.

[0012] This solution uses the UAV flight status to correct the mapping error, combines heterogeneous sensor data and federated filtering technology to accurately correct the multi-source errors generated by the UAV during flight. By dynamically adjusting the weight of the flight path data, a spatiotemporal synchronization matrix is ​​generated, and then the initialization image parameters are established and corrected. The exposure control and distortion compensation model is used to optimize the image quality, and high-precision complete image parameters are generated through iterative matching with the map point cloud. The beneficial effects of this solution are: improving the accuracy of UAV mapping, reducing error accumulation, and optimizing image quality.

[0013] Reference Figure 2 As shown, based on heterogeneous sensors, the task flight path mapping attitude data of the UAV in the area to be mapped is obtained, the multi-source errors in the task flight path mapping attitude data are marked, and dynamic weight allocation is performed according to the federated filter to generate the task flight path spatiotemporal synchronization matrix of the UAV's mapping area. Specifically, it includes: Based on heterogeneous sensors, the mission flight path mapping attitude data of the UAV's mapping area is obtained for preprocessing; the mapping attitude data includes: IMU data, RTK-GNSS data, and point cloud data; According to the federated filtering architecture, a child individual Kalman filter is established based on each type of mapping attitude data in the mission flight path mapping attitude data of the mapping area of ​​the UAV, and a local optimal estimate is generated and substituted into the parent Kalman filter to obtain the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area. According to the optimal estimate of the attitude data of the mission flight path of the UAV's mapping area, the time synchronization matrix is ​​formed by aligning the timestamps of each data; According to the optimal estimate of the attitude data of the mission flight path of the UAV's mapping area, the carrier coordinates of each data are converted to the navigation coordinate system to establish a space conversion matrix; Based on the time synchronization matrix and the space conversion matrix, a coordinate rotation correction-time delay compensation function is established to generate the space-time synchronization matrix of the mission flight path of the UAV's mapping area; , in, The measured position coordinates of the mapping attitude data of the i-th mission flight path of the UAV’s mapping area, is the acquisition timestamp of the mapping attitude data of the flight path of the i-th mission in the mapping area of ​​the UAV, is the space transformation matrix, is the time synchronization matrix, is the measured velocity vector of the UAV inertial measurement unit, is the time synchronization error of the UAV mapping attitude data, is the global coordinate position of the i-th mission flight path mapping attitude data of the UAV's mapping area after time-space alignment, It is the timestamp after synchronization of the mapping attitude data of the flight path of the i-th mission in the mapping area of ​​the UAV.

[0014] This solution is based on the data fusion of heterogeneous sensors, combined with technologies such as federated filtering and Kalman filtering, to optimize the attitude data of drones in the flight path of mapping area missions. Through efficient preprocessing and error marking of multi-source data (IMU, RTK-GNSS, point cloud, etc.), the weight of the data is dynamically adjusted, and the optimal flight path mapping attitude estimation value is generated. Through the establishment of time synchronization and spatial transformation matrix, this solution effectively eliminates the temporal and spatial differences between different sensors, realizes high-precision fusion and synchronization of mapping data, and provides effective data support for high-precision mapping tasks.

[0015] Reference Figure 3 As shown, according to the spatiotemporal synchronization matrix of the mission flight path of the UAV's surveying and mapping area, an exposure control-distortion compensation function is established to correct the mission initialization image parameters of the UAV's surveying and mapping area, and the mission correction image parameters of the UAV's surveying and mapping area are obtained, which specifically include: Based on the spatiotemporal synchronization matrix of the mission flight path of the surveying and mapping area of ​​the UAV, the illumination data and image data of the mission flight path of the surveying and mapping area of ​​the UAV are marked according to the unit time; Based on the illumination data of the mission flight path of the UAV's surveying and mapping area, the AE illumination intensity distribution function is constructed by associating the point cloud reflectivity with the visual RGB value, and the image illumination intensity distribution value of the mission flight path of the UAV's surveying and mapping area is generated; Substituting the image illumination intensity distribution value of the mission flight path of the surveying and mapping area of ​​the UAV into the illumination field model, the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV is generated; Based on the CNN convolutional neural network, the original image data and IMU data of the mission flight path of the UAV's surveying area are used as input, and the distortion-compensated image data of the mission flight path of the UAV's surveying area is used as output; Using the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV and the distortion compensation image data of the mission flight path of the surveying and mapping area of ​​the UAV, the mission initialization image parameters of the surveying and mapping area of ​​the UAV are corrected to obtain the mission correction image parameters of the surveying and mapping area of ​​the UAV; Among them, the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV is specifically: , In the formula, is the dynamic image brightness value of the i-th mission flight path in the surveying area of ​​the UAV, is the illumination intensity fusion value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the visual RGB value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the point cloud reflectivity value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the visual RGB sensor weight coefficient, is the cloud reflectivity weight coefficient (dynamically updated through federated filtering), is the regional weight of the i-th mission flight path in the surveying and mapping area of ​​the UAV, and N is the total number of mission flight paths in the surveying and mapping area of ​​the UAV; The distortion-compensated image data of the mission flight path of the surveying area of ​​the UAV is specifically: , In the formula, is the distortion-compensated image data of the flight path of the ith mission in the mapping area of ​​the UAV, is the original image data of the flight path of the ith mission in the surveying area of ​​the UAV, is the IMU data of the flight path of the i-th mission in the mapping area of ​​the UAV, is the attitude angle, It is the CNN convolutional neural network interface function.

[0016] This solution combines the spatiotemporal synchronization matrix of the flight path of the UAV's mapping area mission, uses the association between illumination data and image data, and constructs an exposure control and distortion compensation model, which can effectively correct the image parameters of the UAV in the mapping mission. Through the generation of illumination intensity distribution function and dynamic image brightness value, combined with the distortion compensation training of CNN convolutional neural network, this solution can eliminate the impact of illumination change and image distortion on image quality, thereby improving the accuracy and precision of the image, enhancing the reliability of mapping data, and providing more accurate visual information support for subsequent mapping analysis.

[0017] Reference Figure 4 As shown, according to the mission correction image parameters of the UAV's mapping area, the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area is updated, and iterative matching is performed with the map point cloud of the mapping area to generate a high-precision image of the UAV's mapping area. Specifically; Based on the mission correction image parameters of the UAV's mapping area, the Kalman filter update formula is used to update the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area, and the optimal state estimate of the mission flight path mapping attitude data of the UAV's mapping area is generated; Based on the optimal state estimation of the mission flight path mapping posture data of the UAV's mapping area and the map point cloud of the mapping area, iterative matching is performed according to the ICP iterative closest point algorithm to generate a high-precision image of the UAV's mapping area; Among them, the update formula of the Mann filter is specifically: , In the formula, is the optimal state estimate of the flight path of the ith mission in the mapping area of ​​the UAV, is the optimal estimate of the flight path of the ith mission in the mapping area of ​​the UAV, The state transfer matrix of the attitude data for the mission flight path mapping of the UAV's mapping area, is the Kalman gain, The original value of the mission flight path mapping attitude data for the UAV's mapping area, An observation matrix of attitude data for the mission flight path of the UAV mapping area.

[0018] The solution corrects image parameters by combining the UAV's mapping area mission, uses Kalman filtering to update the optimal estimate of the flight path mapping attitude data, and uses the ICP iterative closest point algorithm to accurately match the map point cloud of the mapping area to generate a high-precision image. This method can effectively improve the accuracy of attitude estimation, accurately align map point clouds with image data, eliminate flight path errors and image distortion, improve data accuracy and reliability in UAV mapping missions, and support higher-quality mapping and map reconstruction.

[0019] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for correcting high-precision mapping errors based on the flight status of an unmanned aerial vehicle, characterized in that: include: Based on the UAV mapping management background, determine the mission flight path of the UAV in the area to be mapped; Based on heterogeneous sensors, the mission flight path mapping attitude data of the UAV in the area to be mapped is obtained, the multi-source errors in the mission flight path mapping attitude data are marked, and dynamic weight allocation is performed according to the federated filter to generate the mission flight path time-space synchronization matrix of the UAV's mapping area; Based on the mission flight path mapping attitude data of the UAV's mapping area, establish the mission initialization image parameters of the UAV's mapping area; An exposure control-distortion compensation function is established according to the space-time synchronization matrix of the mission flight path of the UAV's surveying and mapping area, and the mission initialization image parameters of the UAV's surveying and mapping area are corrected to obtain the mission correction image parameters of the UAV's surveying and mapping area; According to the mission of the UAV's mapping area, the image parameters are corrected, the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area is updated, and iterative matching is performed with the map point cloud of the mapping area to generate a high-precision image of the UAV's mapping area.

2. According to claim 1, a method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle is characterized in that: Based on heterogeneous sensors, the task flight path mapping attitude data of the UAV in the area to be mapped is obtained, the multi-source errors in the task flight path mapping attitude data are marked, and dynamic weight allocation is performed according to the federated filter to generate the task flight path time-space synchronization matrix of the UAV mapping area. Specifically, it includes: Based on heterogeneous sensors, the mission flight path mapping attitude data of the UAV's mapping area is obtained for preprocessing; the mapping attitude data includes: IMU data, RTK-GNSS data, and point cloud data; According to the federated filtering architecture, a child individual Kalman filter is established based on each type of mapping attitude data in the mission flight path mapping attitude data of the mapping area of ​​the UAV, and a local optimal estimate is generated and substituted into the parent Kalman filter to obtain the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area. According to the optimal estimate of the attitude data of the mission flight path of the UAV's mapping area, the time synchronization matrix is ​​formed by aligning the timestamps of each data; According to the optimal estimate of the attitude data of the mission flight path of the UAV's mapping area, the carrier coordinates of each data are converted to the navigation coordinate system to establish a space conversion matrix; Based on the time synchronization matrix and the space conversion matrix, a coordinate rotation correction-time delay compensation function is established to generate the space-time synchronization matrix of the mission flight path of the UAV's mapping area; ， in, The measured position coordinates of the mapping attitude data of the i-th mission flight path of the UAV’s mapping area, is the acquisition timestamp of the mapping attitude data of the flight path of the i-th mission in the mapping area of ​​the UAV, is the space transformation matrix, is the time synchronization matrix, is the measured velocity vector of the UAV inertial measurement unit, is the time synchronization error of the UAV mapping attitude data, is the global coordinate position of the i-th mission flight path mapping attitude data of the UAV's mapping area after time-space alignment, It is the timestamp after synchronization of the mapping attitude data of the flight path of the i-th mission in the mapping area of ​​the UAV.

3. The method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle according to claim 2 is characterized in that: An exposure control-distortion compensation function is established according to the spatiotemporal synchronization matrix of the mission flight path of the surveying and mapping area of ​​the UAV, and the mission initialization image parameters of the surveying and mapping area of ​​the UAV are corrected to obtain the mission correction image parameters of the surveying and mapping area of ​​the UAV. Specifically, the illumination data and image data of the mission flight path of the surveying and mapping area of ​​the UAV are marked according to the unit time based on the spatiotemporal synchronization matrix of the mission flight path of the surveying and mapping area of ​​the UAV; Based on the illumination data of the mission flight path of the UAV's surveying and mapping area, the AE illumination intensity distribution function is constructed by associating the point cloud reflectivity with the visual RGB value, and the image illumination intensity distribution value of the mission flight path of the UAV's surveying and mapping area is generated; Substituting the image illumination intensity distribution value of the mission flight path of the surveying and mapping area of ​​the UAV into the illumination field model, the dynamic image brightness value of the mission flight path of the surveying and mapping area of ​​the UAV is generated; Based on the CNN convolutional neural network, the original image data and IMU data of the mission flight path of the UAV's surveying area are used as input, and the distortion-compensated image data of the mission flight path of the UAV's surveying area is used as output; The dynamic image brightness value of the mission flight path of the UAV's surveying and mapping area and the distortion compensation image data of the mission flight path of the UAV's surveying and mapping area are used to correct the mission initialization image parameters of the UAV's surveying and mapping area to obtain the mission correction image parameters of the UAV's surveying and mapping area.

4. The method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle according to claim 3 is characterized in that: The dynamic image brightness value of the mission flight path of the surveying area of ​​the UAV is specifically: ， In the formula, is the dynamic image brightness value of the i-th mission flight path in the surveying area of ​​the UAV, is the illumination intensity fusion value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the visual RGB value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the point cloud reflectivity value of the image coordinate point of the i-th mission flight path in the surveying area of ​​the UAV, is the visual RGB sensor weight coefficient, is the cloud reflectivity weight coefficient (dynamically updated through federated filtering), is the regional weight of the i-th mission flight path in the surveying area of ​​the UAV, and N is the total number of mission flight paths in the surveying area of ​​the UAV.

5. The method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle according to claim 4 is characterized in that: The distortion-compensated image data of the mission flight path of the surveying area of ​​the UAV is specifically: , In the formula, The distortion-compensated image data of the i-th mission flight path of the UAV’s mapping area, is the original image data of the flight path of the ith mission in the surveying area of ​​the UAV, is the IMU data of the flight path of the ith mission in the mapping area of ​​the UAV, is the attitude angle, It is the CNN convolutional neural network interface function.

6. The method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle according to claim 5 is characterized in that: According to the mission of the UAV's mapping area, the image parameters are corrected, the optimal estimate of the mapping attitude data of the mission flight path of the UAV's mapping area is updated, and iterative matching is performed with the map point cloud of the mapping area to generate a high-precision image of the UAV's mapping area. Specifically, it includes: Based on the mission correction image parameters of the UAV's mapping area, the Kalman filter update formula is used to update the optimal estimate of the mission flight path mapping attitude data of the UAV's mapping area, and the optimal state estimate of the mission flight path mapping attitude data of the UAV's mapping area is generated; Based on the optimal state estimation of the mapping posture data of the mission flight path of the UAV's mapping area and the map point cloud of the mapping area, iterative matching is performed according to the ICP iterative closest point algorithm to generate a high-precision image of the UAV's mapping area.

7. The method for correcting high-precision surveying and mapping errors based on the flight status of an unmanned aerial vehicle according to claim 6 is characterized in that: The update formula of the Mann filter is specifically: , In the formula, is the optimal state estimate of the flight path of the ith mission in the mapping area of ​​the UAV, is the optimal estimate of the flight path of the ith mission in the mapping area of ​​the UAV, The state transfer matrix of the attitude data for the mission flight path mapping of the UAV's mapping area, is the Kalman gain, The original value of the mission flight path mapping attitude data for the UAV's mapping area, An observation matrix of attitude data for the mission flight path of the UAV mapping area.

Images