Method for collecting grassland spatial data based on UAV mapping

Abstract

The present invention provides a method for collecting grassland spatial data based on unmanned aerial vehicle (UAV) mapping, which relates to the field of mapping technology. When mapping grasslands, the rapidly changing external environment can lead to inaccurate grassland mapping data. For example, strong winds can cause the UAV to deviate from its flight path. At this time, it is necessary to re-plan the mapping path and let the UAV perform the mapping task again. In the present invention, the UAV is made to sequentially perform the flight mapping tasks corresponding to each mapping sub-path. According to the data volume of the initial actual mapping data and the predicted mapping data corresponding to the mapping sub-path corresponding to the flight mapping task, it is determined whether it is necessary to re-perform the flight mapping task to ensure data accuracy. Even if the external environment of the mapping sub-geographical area changes during the process of the UAV performing the flight mapping task, only the sub-path corresponding to the flight mapping task needs to be re-flown, reducing unnecessary repetitive work and being conducive to improving the overall efficiency of mapping.

Description

Technical Field

[0001] The present invention relates to the field of surveying and mapping technology, and particularly to a method for collecting grassland spatial data based on unmanned aerial vehicle (UAV) surveying and mapping. Background Art

[0002] Through surveying and mapping, maps, terrain models, and geographic information system data can be generated, which are crucial for multiple fields such as urban planning, land management, environmental monitoring, disaster prevention, and agricultural development. With the development of UAV technology, UAV-based surveying and mapping has gradually become an emerging method. A UAV can carry various sensors, such as lidar and camera acquisition devices, to collect high-resolution spatial data during flight. UAV-based surveying and mapping requires planning a survey path that can cover the entire survey area, enabling the UAV to execute the survey task so that the UAV can travel from the starting point to the end point of the survey path along the survey path. After the UAV reaches the end point of the survey path, the survey is completed based on the data collected by the sensors in the UAV.

[0003] However, the above method also has the following technical problems:

[0004] When surveying areas with rapidly changing external environments such as grasslands, environmental factors such as wind speed, light, and temperature will have a significant impact on the flight stability of the UAV and the working effect of the sensors. Due to the flat and open terrain of the grassland, the wind speed is usually large, and strong wind weather is likely to form, which may cause the actual flight path of the UAV to deviate from the survey path. In addition, the temperature difference between day and night in grassland areas is large. Along with the sharp change in light conditions, suddenly increasing or decreasing light will affect the imaging quality. At the same time, the sensors on the UAV are sensitive to temperature, and temperature changes may cause an increase in sensor measurement errors. Therefore, when surveying a grassland survey area based on a UAV, the rapidly changing external environment may result in inaccurate grassland survey data. For example, strong winds may cause the UAV to deviate from its course, extreme light conditions will reduce the image quality, and temperature fluctuations will affect the accuracy of the sensors. When it is found that the data collected by the sensors in the UAV is inaccurate, in order to ensure data quality and survey accuracy, it is often necessary to re-plan the survey path and let the UAV execute the survey task again, which increases the additional workload and reduces the overall efficiency of the survey. Summary of the Invention

[0005] In view of the above technical problems, the technical solution adopted by the present invention is as follows:

[0006] A method for collecting grassland spatial data based on UAV surveying and mapping, the method comprising the following steps:

[0007] S1. When the UAV starts to execute A for the first time jExecute steps S2 - S4 of the corresponding flight mapping task simultaneously, A j is the j-th mapping sub-path, where j ranges from 2 to m, and m is the number of mapping sub-paths.

[0008] S2. If the UAV has only completed the corresponding flight mapping task once, then obtain j-1 the data volume E of the initial actual mapping data corresponding to A j-1 where A 1 j-1 is the (j - 1)-th mapping sub-path, and the initial actual mapping data corresponding to A j-1 is the three-dimensional point cloud data collected by the lidar device in the UAV during the first execution of the flight mapping task corresponding to A j-1 j-1 j-1 j-1

[0009] S3. Input a number of initial mapping images corresponding to A j-1 into the target CNN model to obtain the data volume E of the predicted mapping data output by the target CNN model corresponding to A j-1 where the initial mapping images corresponding to A 2 j-1 are the visible light images collected by the camera acquisition device in the UAV during the first execution of the flight mapping task corresponding to A j-1 j-1 j-1 j-1

[0010] S4. If |E 2 j-1 - E 1 j-1 | / E 2 j-1 ≥ C 0 then control the UAV to re-execute the flight mapping task corresponding to A j after completing the flight mapping task corresponding to A j-1 so as to re-map the mapping sub-geographical area corresponding to A j-1 where C 0 is a preset data volume deviation degree value.

[0011] The present invention has at least the following beneficial effects:

[0012] The present invention provides a method for collecting grassland spatial data based on UAV mapping. In this method, a number of mapping sub-paths are determined, and the UAV is made to sequentially execute the flight mapping tasks corresponding to each mapping sub-path. When the UAV starts to execute the current flight mapping task, the number of completions of the previous flight mapping task of the current flight mapping task is determined. If it has been completed only once, the data volume of the initial actual mapping data and the predicted mapping data corresponding to the mapping sub-path of the previous flight mapping task are obtained. According to the data volume of the initial actual mapping data, the data volume of the predicted mapping data, and the preset data volume deviation degree value, the data collected by the sensor in the UAV during the execution of the previous flight mapping task is dynamically evaluated for accuracy, and based on the evaluation result, it is judged whether it is necessary to re-execute the previous flight mapping task to ensure data accuracy when the current flight mapping task is completed. Even if the external environment of the mapping sub-geographical area changes during the execution of the UAV flight mapping task, resulting in inaccurate data collected by the sensor in the UAV during the execution of the flight mapping task, it is only necessary to re-fly the mapping sub-path corresponding to the flight mapping task, without the need to re-plan the entire mapping path or repeat the execution of all flight mapping tasks, reducing unnecessary repetitive work and being beneficial to improving the overall efficiency of mapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0014] Figure 1 It is a flowchart of the method for collecting grassland spatial data based on UAV mapping provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0016] It should be noted that the terms "first", "second", etc. in the specification, claims and the above-mentioned drawings of the present invention are used to distinguish similar tasks and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0017] An embodiment of the present invention provides a method for collecting grassland spatial data based on unmanned aerial vehicle (UAV) mapping. The method includes the following steps, as Figure 1 shown:

[0018] S1. While the UAV starts to execute the A j corresponding flight mapping task for the first time, execute steps S2 - S4. A j is the j-th mapping sub-path, where j ranges from 2 to m, and m is the number of mapping sub-paths.

[0019] Specifically, the mapping sub-path is a path obtained by dividing the target mapping path corresponding to the UAV according to a preset path distance. Among them, the target mapping path corresponding to the UAV can cover the entire grassland mapping area. The preset path distance is a path distance preset by those skilled in the art according to actual needs and will not be elaborated here.

[0020] Specifically, the mapping sub-path corresponds one-to-one with the mapping sub-geographical area. The mapping geographical sub-area is a sub-area in the grassland mapping area. The mapping sub-geographical areas corresponding to all the mapping sub-paths can cover the entire grassland mapping area.

[0021] Specifically, the starting point of the first mapping sub-path is the starting point of the target mapping path, the ending point of the m-th mapping sub-path is the ending point of the target mapping path, and the ending point of the (j - 1)-th mapping sub-path is the starting point of the j-th mapping sub-path.

[0022] Specifically, the flight mapping task corresponding to the mapping sub-path is to make the UAV fly along the mapping sub-path from the starting point of the mapping sub-path to the ending point of the mapping sub-path.

[0023] S2. If the UAV has only completed the A j-1 corresponding flight mapping task once, then obtain the data volume E j-1 of the initial actual mapping data corresponding to A 1 j-1, A j-1 is the (j - 1)-th surveying sub-path, A j-1 The corresponding initial actual surveying data is the three-dimensional point cloud data collected by the lidar device in the unmanned aerial vehicle during the first execution of A j-1 by the unmanned aerial vehicle during the corresponding flight surveying mission.

[0024] Specifically, in step S2, if the unmanned aerial vehicle has completed two A j-1 corresponding flight surveying missions and j ≠ m, then control the unmanned aerial vehicle to execute A j corresponding flight surveying mission after completing the A j+1 corresponding flight surveying mission so as to survey the A j+1 corresponding surveying sub-geographical area, A j+1 is the (j + 1)-th surveying sub-path; if the unmanned aerial vehicle has completed two A j-1 corresponding flight surveying missions and j = m, then end the surveying after the unmanned aerial vehicle completes the A j corresponding flight surveying mission.

[0025] In the above steps, for each flight surveying mission, if the unmanned aerial vehicle has only completed one flight surveying mission currently, it means that the unmanned aerial vehicle has not re-flown the surveying sub-path corresponding to the flight surveying mission. It is necessary to determine whether to re-fly the surveying sub-path based on the data collected by the sensors (camera acquisition device and lidar device) in the unmanned aerial vehicle during the execution of the flight surveying mission to ensure the accuracy of the data collected by the sensors in the unmanned aerial vehicle. If the unmanned aerial vehicle has completed two flight surveying missions currently, it means that the unmanned aerial vehicle has re-flown the surveying sub-path corresponding to the flight surveying mission, and the accuracy of the data collected by the sensors in the unmanned aerial vehicle can already be guaranteed. There is no need to re-fly the surveying sub-path again. Even if the external environment of the surveying sub-geographical area changes during the execution of the flight surveying mission by the unmanned aerial vehicle, resulting in inaccurate data collected by the sensors in the unmanned aerial vehicle during the execution of the flight surveying mission, only the surveying sub-path corresponding to the flight surveying mission needs to be re-flown, without the need to re-plan the entire surveying path or repeat all the flight surveying missions, reducing unnecessary repetitive work and being beneficial to improving the overall efficiency of surveying.

[0026] S3. Input a number of initial surveying images corresponding to A j-1 into the target CNN model to obtain the data volume E of the predicted surveying data output by the target CNN model corresponding to A j-1 2 j-1 , A j-1 The corresponding initial surveying image is the image collected by the camera acquisition device in the unmanned aerial vehicle during the first execution of A j-1 ​Visible light images collected during the corresponding flight mapping mission.

[0027] Specifically, the visible light image acquisition frequency of the camera acquisition device in the drone is set by those skilled in the art according to actual needs, for example: collecting a visible light image once per second, collecting a visible light image once per 0.5 seconds, which will not be repeated here.

[0028] Specifically, the target CNN model is a trained CNN model used to extract image features, and the amount of predicted mapping data can be understood as the amount of mapping data collected by the predicted UAV during the performance of the flight mapping mission.

[0029] S4. If |E 2 j-1 -E 1 j-1 | / E 2 j-1 ≥C 0 , then control the drone to complete A j After the corresponding flight mapping mission, re-execute A j-1 The corresponding flight mapping mission is to make A j-1 The corresponding sub-geographical area is re-surveyed, C 0 It is a preset data volume deviation degree value. Those skilled in the art know that the preset data volume deviation degree value is a value pre-set by those skilled in the art according to actual needs, for example: 0.1, 0.15, 0.2, and will not be repeated here.

[0030] Specifically, step S4 also includes: if |E 2 j-1 -E 1 j-1 | / E 2 j-1 <C 0 , then control the drone to complete A j After the corresponding flight mapping mission, execute A j+1 The corresponding flight mapping mission is to make A j+1 The corresponding sub-geographical area is surveyed.

[0031] Specifically, step S4 also includes: j-1 The corresponding C j-1 Stored in the first surveying and mapping database, C j-1 A j-1 The corresponding data volume difference ratio, C j-1 =|E 2 j-1 -E 1 j-1 | / E 2j-1 。

[0032] Specifically, when C j-1 ≥C 0 , generate the corresponding abnormal label for A j-1 and store the abnormal label in the first mapping database.

[0033] Specifically, the abnormal label indicates that there are abnormalities in the initial actual mapping data and the initial mapping image corresponding to the mapping sub-path, and they may not be accurate enough.

[0034] In the above steps, according to the data volume of the initial actual mapping data, the data volume of the predicted mapping data, and the preset data volume deviation degree value, it is evaluated whether the data collected by the sensor in the UAV during the flight mapping task is accurate. If the data volume difference ratio obtained based on the data volume of the initial actual mapping data and the data volume of the predicted mapping data is not less than the preset data volume deviation degree value, it means that the difference between the data volume of the initial actual mapping data and the data volume of the predicted mapping data is relatively large, and the data collected by the sensor in the UAV during the flight mapping task corresponding to the initial actual mapping data is not accurate enough. During the process of the UAV performing the flight mapping task, the external environment of the mapping sub-geographical area may have changed. Therefore, it is necessary to re-execute the flight mapping task to ensure the accuracy of the data collected by the sensor in the UAV. If the data volume difference ratio obtained based on the data volume of the initial actual mapping data and the data volume of the predicted mapping data is less than the preset data volume deviation degree value, it means that the difference between the data volume of the initial actual mapping data and the data volume of the predicted mapping data is relatively small, and the data collected by the sensor in the UAV during the flight mapping task corresponding to the initial actual mapping data is relatively accurate. There is no need to re-execute the flight mapping task. Even if the external environment of the mapping sub-geographical area changes during the process of the UAV performing the flight mapping task, resulting in inaccurate data collected by the sensor in the UAV during the process of the UAV performing the flight mapping task, it only needs to re-fly the mapping sub-path corresponding to the flight mapping task, without re-planning the entire mapping path or repeating all the flight mapping tasks, reducing unnecessary repetitive work and being beneficial to improving the overall efficiency of mapping.

[0035] Specifically, before step S1, control the UAV to start performing the flight mapping task corresponding to the first mapping sub-path to map the mapping sub-geographical area corresponding to the first mapping sub-path, and after controlling the UAV to complete the flight mapping task corresponding to the first mapping sub-path, continue to perform the flight mapping task corresponding to the second mapping sub-path to map the mapping sub-geographical area corresponding to the second mapping sub-path.

[0036] Specifically, when the drone first completes i the corresponding flight mapping task, the i corresponding initial actual mapping data and the i corresponding initial mapping images are stored in the first mapping database. For the i-th mapping sub-path, where i ranges from 1 to m, i the corresponding initial actual mapping data is the 3D point cloud data collected by the lidar device in the drone during the first execution of i the corresponding flight mapping task by the drone, and i the corresponding initial mapping images are the visible light images collected by the camera acquisition device in the drone during the first execution of i the corresponding flight mapping task by the drone. i Specifically, when the drone secondarily completes

[0037] the corresponding flight mapping task, the i corresponding key actual mapping data and the i corresponding key mapping images are stored in the second mapping database. The i corresponding key actual mapping data is the 3D point cloud data collected by the lidar device in the drone during the second execution of i the corresponding flight mapping task by the drone, and i the corresponding key mapping images are the visible light images collected by the camera acquisition device in the drone during the second execution of i the corresponding flight mapping task by the drone. i In the above steps, when the drone first completes the flight mapping task, the initial actual mapping data and the initial mapping images corresponding to the mapping sub-path of the flight mapping task are stored in the first mapping database. When the drone secondarily completes the flight mapping task, the key actual mapping data and the key mapping images corresponding to the mapping sub-path of the flight mapping task are stored in the second mapping database. Separated storage can ensure that the data will not be confused, facilitating subsequent data analysis and processing.

[0038] In a specific embodiment, in step S2, if the drone has currently completed the

[0039] corresponding flight mapping task twice and j = m, then obtain C j-1 , where C j is the data volume difference ratio corresponding to j A. If C j ≥C j , then control the drone to re-execute 0 A after completing the j corresponding flight mapping task. jThe corresponding flight mapping task so that for A j The corresponding surveyed sub-geographical area is resurveyed. If C j < C 0 , then after the UAV completes the flight mapping task corresponding to A j , the surveying ends. C j Meets the following conditions:

[0040] C j = |E 2 j - E 1 j | / E 2 j , E 2 j is the data volume of the initial actual surveyed data corresponding to A j , E 1 j is the data volume of the predicted surveyed data output by the target CNN model obtained by inputting several initial surveyed images corresponding to A j into the target CNN model. The initial actual surveyed data corresponding to A j is the three-dimensional point cloud data collected by the lidar device in the UAV during the first execution of the flight mapping task corresponding to A j by the UAV. The initial surveyed images corresponding to A j are the visible light images collected by the camera acquisition device in the UAV during the first execution of the flight mapping task corresponding to A j by the UAV. j

[0041] ​In the above steps, if the current flight mapping task is the last flight mapping task and the previous flight mapping task of the current flight mapping task has been completed twice, it means that after completing the current flight mapping task, there is no need to re-execute the previous flight mapping task, nor is there a next flight mapping task to execute. At this time, the data volume difference ratio corresponding to the current flight mapping task can be obtained. If the data volume difference ratio is not less than the preset data volume deviation degree value, it indicates that the data volume of the initial actual mapping data collected by the sensor in the UAV during the execution of the current flight mapping task differs greatly from the data volume of the predicted mapping data, and the data collected by the sensor in the UAV during the execution of the current flight mapping task is not accurate enough. During the execution of the current flight mapping task by the UAV, the external environment of the mapped sub-geographical area may have changed. Therefore, it is necessary to re-execute the current flight mapping task to ensure the accuracy of the data collected by the sensor in the UAV. If the data volume difference ratio is less than the preset data volume deviation degree value, it indicates that the data volume of the initial actual mapping data collected by the sensor in the UAV during the execution of the current flight mapping task differs little from the data volume of the predicted mapping data, and the data collected by the sensor in the UAV during the execution of the current flight mapping task is relatively accurate. There is no need to re-execute the current flight mapping task, and the mapping can be ended after the UAV completes the current flight mapping task.

[0042] Specifically, after step S4, the following steps S5 - S9 are further included:

[0043] S5. While the UAV starts to execute the e corresponding flight mapping task for the second time, execute steps S6 - S9. e A is the e-th mapping sub-path, and the value of e ranges from 1 to m - 1.

[0044] S6. If the current UAV has completed the e+1 corresponding flight mapping task once, then enter step S7. e+1 A is the (e + 1)-th mapping sub-path; if the current UAV has never executed the e+1 corresponding flight mapping task, then control the UAV to execute the e corresponding flight prediction task after completing the e+1 corresponding flight mapping task for the second time, so as to map the e+1 corresponding mapped sub-geographical area.

[0045] S7. Obtain the data volume E e+1 of the initial actual mapping data corresponding to 1 e+1 A e+1The corresponding initial actual surveying and mapping data is the three-dimensional point cloud data collected by the lidar device in the unmanned aerial vehicle (UAV) during the first execution of mission A by the UAV. e+1 The corresponding flight surveying and mapping mission is the visible light image collected by the camera acquisition device in the UAV during the first execution of mission A by the UAV.

[0046] S8. Input a plurality of initial surveying and mapping images corresponding to A into the target CNN model to obtain the data volume E of the predicted surveying and mapping data output by the target CNN model. e+1 The corresponding initial surveying and mapping images are the visible light images collected by the camera acquisition device in the UAV during the first execution of mission A by the UAV. e+1 The corresponding predicted surveying and mapping data data volume E 2 e+1 A e+1 The corresponding initial surveying and mapping images are the visible light images collected by the camera acquisition device in the UAV during the first execution of mission A by the UAV. e+1 The corresponding flight surveying and mapping mission is the visible light image collected by the camera acquisition device in the UAV during the first execution of mission A by the UAV.

[0047] S9. Obtain C e+1 And control the UAV to re-execute the flight prediction mission corresponding to A after the UAV completes the flight surveying and mapping mission corresponding to A for the second time, so as to re-survey the surveyed sub-geographical area corresponding to A. C e The corresponding flight prediction mission is to re-survey the surveyed sub-geographical area corresponding to A. C e+1 The corresponding flight prediction mission is to re-survey the surveyed sub-geographical area corresponding to A. e+1 The corresponding surveyed sub-geographical area is re-surveyed. C e+1 A e+1 The corresponding data volume difference ratio C e+1 Meets the following conditions:

[0048] C e+1 =|E 2 e+1 -E 1 e+1 | / E 2 e+1 .

[0049] Specifically, in step S9, it further includes storing C e+1 In the first surveying and mapping database.

[0050] Specifically, when C e+1 ≥C 0 Generate the corresponding anomaly label for C e+1 And store the anomaly label in the first surveying and mapping database.

[0051] Through the above steps, while the UAV starts to execute the current flight mapping task for the second time, it is determined whether the UAV has completed the subsequent flight mapping task of the current flight mapping task. If the UAV has never executed the subsequent flight mapping task, it means that after the UAV executed the current flight mapping task for the first time, it re-executed the previous flight mapping task of the current flight mapping task, and after re-completing the previous flight mapping task, it re-executed the current flight mapping task. At this time, it is only necessary to control the UAV to continue to execute the subsequent flight mapping task of the current flight mapping task after completing the current flight mapping task; if the UAV has completed the subsequent flight mapping task once, it means that after the UAV executed the subsequent flight mapping task for the first time, it re-executed the current flight mapping task. At this time, the data volume difference ratio corresponding to the subsequent flight mapping task can be obtained and stored in the first mapping database to improve the data processing efficiency.

[0052] In a specific embodiment, after step S4, steps S5 - S10 are further included, where step S9 includes:

[0053] When C e+1 ≥C 0 , control the UAV to re-execute the flight prediction task corresponding to A e after the UAV completes the flight mapping task corresponding to A e+1 for the second time, so as to re-map the mapping sub-geographical area corresponding to A e+1 ; when C e+1 <C 0 , set the preset task execution duration corresponding to A e+1 and enter step S10.

[0054] S10. Control the UAV to re-execute the flight prediction task corresponding to A e after the UAV completes the flight mapping task corresponding to A e+1 for the second time, and complete the flight mapping task corresponding to A e+1 within the preset task execution duration corresponding to A e+1 ; it can be understood as accelerating the flight speed of the UAV when executing the flight mapping task corresponding to A e+1 .

[0055] In the above steps, if the difference in data volume corresponding to the next flight mapping task of the current flight mapping task is not less than the preset data volume deviation degree value, it indicates that during the execution of the next flight mapping task of the current flight mapping task, the difference in the data volume between the initially actual mapping data collected by the sensors in the UAV and the predicted mapping data is relatively large. The sensors in the UAV collect inaccurate data during the execution of the next flight mapping task of the current flight mapping task. During the execution of the next flight mapping task of the current flight mapping task, the external environment of the mapped sub-geographical area may have changed. Therefore, it is necessary to re-execute the next flight mapping task of the current flight mapping task to ensure the accuracy of the data collected by the sensors in the UAV. Otherwise, it indicates that the difference in the data volume between the initially actual mapping data collected by the sensors in the UAV and the predicted mapping data is relatively small during the execution of the next flight mapping task of the current flight mapping task. The sensors in the UAV collect relatively accurate data during the execution of the next flight mapping task of the current flight mapping task, and there is no need to collect data again. At this time, set the preset task execution duration, and let the UAV execute the next flight mapping task of the current flight mapping task and complete the flight mapping task within the preset task execution duration, which helps to reduce the overall mapping duration and improve the overall mapping efficiency.

[0056] Specifically, when the UAV needs to complete the flight mapping task within the preset task execution duration, when executing the flight mapping task, there is no need to turn on the data collection functions of the camera acquisition device and the lidar device, which helps to reduce resource consumption.

[0057] In a specific embodiment, when the UAV needs to complete the flight mapping task within the preset task execution duration, if the camera acquisition device and the lidar device in the UAV collect data during the execution of the flight mapping task, there is no need to store the collected data in the second mapping database, which helps to save storage space.

[0058] In a specific embodiment, when there are abnormal labels corresponding to the mapped sub-path in the first mapping database, those skilled in the art can process the initially actual mapping data, initial mapping images corresponding to the mapped sub-path in the first mapping database and the key actual mapping data, key mapping images corresponding to the mapped sub-path in the second mapping database to obtain the target mapping data corresponding to the mapped sub-path, so as to complete the mapping based on the target mapping data, which helps to improve the accuracy of the mapping. Among them, the processing method in the above steps is determined by those skilled in the art according to actual needs. For example: fusion processing, which will not be elaborated here.

[0059] The present invention provides a method for collecting grassland spatial data based on UAV mapping. In this method, a number of mapping sub-paths are determined, and the UAV is made to sequentially execute the flight mapping tasks corresponding to each mapping sub-path. When the UAV starts to execute the current flight mapping task, the number of completions of the previous flight mapping task of the current flight mapping task is determined. If it has been completed only once, the data volume of the initial actual mapping data and the predicted mapping data corresponding to the mapping sub-path of the previous flight mapping task are obtained. According to the data volume of the initial actual mapping data, the data volume of the predicted mapping data, and the preset data volume deviation degree value, the data collected by the sensor in the UAV during the execution of the previous flight mapping task is dynamically evaluated for accuracy, and based on the evaluation result, it is judged whether it is necessary to re-execute the previous flight mapping task to ensure data accuracy when the current flight mapping task is completed. Even if the external environment of the mapping sub-geographical area changes during the execution of the UAV flight mapping task, resulting in inaccurate data collected by the sensor in the UAV during the execution of the flight mapping task, only the mapping sub-path corresponding to the flight mapping task needs to be re-flown, without the need to re-plan the entire mapping path or repeat the execution of all flight mapping tasks, reducing unnecessary repetitive work and being conducive to improving the overall efficiency of mapping.

[0060] An embodiment of the present invention further provides a non-transitory computer-readable storage medium, which can be set in an electronic device to store a computer program related to a method for implementing a method in a method embodiment. The computer program is loaded and executed by the processor to implement the method provided in the above embodiment.

[0061] An embodiment of the present invention further provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method provided in the above embodiment is implemented.

[0062] An embodiment of the present invention further provides a computer program product, which includes program code. When the program product runs on an electronic device, the program code is used to cause the electronic device to execute the steps in the method according to various exemplary embodiments of the present invention described above in this specification.

[0063] Although some specific embodiments of the present invention have been described in detail by way of examples, those skilled in the art should understand that the above examples are only for illustration and not for limiting the scope of the present invention. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present invention.

Claims

1. A grassland spatial data collection method based on drone mapping, characterized in that: The method comprises the following steps: S1. When the drone starts executing A for the first time j The corresponding flight mapping task is performed at the same time as steps S2-S4, A j is the jth mapping subpath, the value of j ranges from 2 to m, and m is the number of mapping subpaths; S2. If the drone has only completed A once j-1 Corresponding flight mapping mission, then obtain A j-1 The corresponding initial actual surveying and mapping data volume E 1 j-1 , A j-1 is the j-1th mapping subpath, A j-1 The corresponding initial actual surveying data is the laser radar device in the UAV when the UAV performs A for the first time. j-1 The three-dimensional point cloud data collected during the corresponding flight mapping mission; S3, A j-1 The corresponding initial mapping images are input into the target CNN model to obtain the output of the target CNN model A j-1 The corresponding amount of predicted surveying and mapping data E 2 j-1 , A j-1 The corresponding initial mapping image is the image acquisition device in the UAV when the UAV performs A for the first time. j-1 The visible light images collected during the corresponding flight mapping mission; S4. If |E 2 j-1 -E 1 j-1 | / E 2 j-1 ≥C 0 , then control the drone to complete A j After the corresponding flight mapping mission, re-execute A j-1 The corresponding flight mapping mission is to make A j-1 The corresponding sub-geographical area is re-surveyed, C 0 is a preset data volume deviation value; step S4 also includes: When C j-1 ≥C 0 When A is generated j-1 The corresponding abnormal label and the abnormal label and C j-1 Stored in the first surveying and mapping database, where C j-1 A j-1 The corresponding data volume difference ratio, C j-1 =|E 2 j-1 -E 1 j-1 | / E 2 j-1 ,The abnormal label represents that the initial actual surveying data corresponding to the surveying sub-path is abnormal.

2. The grassland spatial data collection method based on drone mapping according to claim 1 is characterized in that: The mapping sub-path is a path obtained by dividing the target mapping path corresponding to the UAV according to the preset path distance, wherein the target mapping path corresponding to the UAV can cover the entire grassland mapping area.

3. The grassland spatial data collection method based on drone mapping according to claim 2 is characterized in that: The mapping sub-paths correspond one-to-one to the mapping sub-geographical areas, and the mapping geographic sub-areas are sub-areas in the grassland mapping area.

4. The grassland spatial data collection method based on drone mapping according to claim 2 is characterized in that: The flight mapping task corresponding to the mapping sub-path is a task of causing the UAV to follow the mapping sub-path from the starting point of the mapping sub-path to the end point of the mapping sub-path.

5. The grassland spatial data collection method based on drone mapping according to claim 1 is characterized in that: In step S2, if the drone has completed A twice j-1 If the corresponding flight mapping task is j≠m, the drone is controlled to complete A j After the corresponding flight mapping mission, execute A j+1 The corresponding flight mapping mission is to make A j+1 The corresponding sub-geographical area is surveyed and mapped. j+1 is the j+1th mapping subpath; if the drone has completed A twice j-1 The corresponding flight mapping task and j = m, then when the UAV completes A j After completing the corresponding flight mapping mission, the mapping is ended.

6. The grassland spatial data collection method based on drone mapping according to claim 5 is characterized in that: Step S4 also includes: If |E 2 j-1 -E 1 j-1 | / E 2 j-1 <C 0 , then control the drone to complete A j After the corresponding flight mapping mission, execute A j+1 The corresponding flight mapping mission is to make A j+1 The corresponding sub-geographical area is surveyed.

7. The grassland spatial data collection method based on drone mapping according to claim 1 is characterized in that: When the drone completes A for the first time i When the corresponding flight surveying mission is in progress, A i The corresponding initial actual surveying data and A i The corresponding initial mapping image is stored in the first mapping database. i is the i-th mapping subpath, i ranges from 1 to m, A i The corresponding initial actual surveying data is the laser radar device in the UAV when the UAV performs A for the first time. i The 3D point cloud data collected during the corresponding flight mapping mission, A i The corresponding initial mapping image is the image acquisition device in the UAV when the UAV performs A for the first time. i Visible light images collected during the corresponding flight mapping mission.

8. The grassland spatial data collection method based on drone mapping according to claim 7 is characterized in that: When the drone completes A for the second time i When the corresponding flight surveying mission is in progress, A i The corresponding key actual surveying and mapping data and A i The corresponding key surveying and mapping images are stored in the second surveying and mapping database. i The corresponding key actual surveying data is the laser radar device in the UAV when the UAV performs A for the second time. i The 3D point cloud data collected during the corresponding flight mapping mission, A i The corresponding key mapping image is the image acquisition device in the UAV when the UAV performs A for the second time. i Visible light images collected during the corresponding flight mapping mission.

Images