Near-natural unmanned aerial vehicle intelligent bullet-seeding afforestation device and method

Abstract

The invention discloses a near-natural unmanned aerial vehicle intelligent missile-seeding afforestation device and method, and relates to the technical field of forestry intelligence and ecological restoration intellectualization, the near-natural unmanned aerial vehicle intelligent missile-seeding afforestation device comprises an unmanned aerial vehicle platform as a flight bearing foundation of a system; the ejection seeding device is used for storing and ejecting the plant growing body with controllable ejection force; the plant growing body is used for carrying out coating treatment on target plant seeds; the multi-source information acquisition module is used for acquiring image information of the afforestation land; the intelligent cloud platform is used for processing the image information of the multi-source information acquisition module, generating a flight and ejection strategy for simulating natural seeding and controlling the unmanned aerial vehicle platform and the ejection seeding device to execute tasks; and the edge computing device is used for performing automatic microhabitat scene identification, unmanned aerial vehicle flight path auxiliary decision making and data signal transmission. By integrating the unmanned aerial vehicle, artificial intelligence, multi-source remote sensing and ecology principles, precise, efficient and nature-imitating throwing of the vegetation is achieved, and the afforestation quality and ecological benefits are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of forestry intelligence and ecological restoration intelligence, in particular to a near-natural unmanned aerial vehicle wisdom bomb-sowing afforestation device and method. BACKGROUND

[0002] Traditional afforestation methods mainly rely on artificial seeding or seedling planting, which has problems such as high labor intensity, low operation efficiency, high cost, and difficulty in implementation in steep mountainous areas and remote areas. In recent years, unmanned aerial vehicle seeding technology has been applied. This method mainly uses the "air sowing" method, that is, the unmanned aerial vehicle uniformly sows seeds along the preset grid path at a fixed height. This method has high efficiency, but has obvious defects: poor seeding accuracy, large-area sowing leading to uneven distribution of seeds, low matching degree with site conditions (such as soil, light, and slope), serious seed waste, and low seedling emergence rate; strong randomness of seeding position, seeds may fall on rocks, water bodies, or dense shrubs and cannot germinate; ignoring microhabitats, uniform sowing cannot take advantage of the microenvironment (such as under bare slopes and shady places under stones) that can provide shelter for seedlings, resulting in low survival rate of seeds. Intense competition between species, dense and uniform seeding will lead to intense competition for water and nutrients among seedlings, which is not conducive to the formation of dominant trees; seeds are easily damaged: exposed seeds are easily eaten by birds and rodents, or cannot germinate and survive due to poor contact with soil; lack of intelligent decision-making: flight path and seeding decisions rely on preset programs and cannot be dynamically optimized and adjusted according to real-time geographic information and forest conditions. Therefore, there is an urgent need for a near-natural unmanned aerial vehicle wisdom bomb-sowing afforestation device and method. SUMMARY

[0003] Therefore, the present application provides a near-natural unmanned aerial vehicle wisdom bomb-sowing afforestation device and method, which integrates unmanned aerial vehicles, artificial intelligence, multi-source remote sensing, and ecological principles, effectively solves the problems of high cost, low efficiency, and poor survival rate of traditional afforestation methods, realizes the precise, efficient, and natural-like release of plant bodies, significantly improves afforestation quality and ecological benefits, and is particularly suitable for ecological restoration and afforestation engineering in difficult sites.

[0004] In order to achieve the above purpose, the present application adopts the following technical solutions: A near-natural unmanned aerial vehicle wisdom bomb-sowing afforestation device, comprising: An unmanned aerial vehicle platform as a flight carrying basis of the system; A bomb-sowing device installed on the unmanned aerial vehicle platform for storing and bomb-sowing plant bodies with controllable bomb-sowing force; A plant body, which is subjected to coating treatment to form a pellet containing native plant seeds, a nutrient layer, a substrate layer, and a structure layer from the inside to the outside; A multi-source information acquisition module installed on the unmanned aerial vehicle platform for acquiring image information of the afforestation site; The intelligent cloud platform is used for processing image information of the multi-source information acquisition module, continuously optimizing an image recognition model by means of an image recognition algorithm, generating flight and ejection strategies simulating natural seeding, and controlling the unmanned aerial vehicle platform and the ejection seeding device to perform a task. The edge computing device is used in cooperation with the unmanned aerial vehicle platform to automatically identify a microhabitat scene, assist in decision-making of a flight path of the unmanned aerial vehicle, and transmit data signals.

[0005] Further, the ejection seeding device comprises a seed storage bin, a vibration assembly, a quantitative seed distributor and an ejection device, the seed storage bin is fixedly installed at a lower end of the unmanned aerial vehicle platform, the outlet end of the seed storage bin is respectively provided with the vibration assembly, the quantitative seed distributor and the ejection device, and the vibration assembly, the quantitative seed distributor and the ejection device are electrically connected with the intelligent cloud platform.

[0006] Further, the ejection device is an electromagnetic ejection, a rotating flywheel, a mechanical ejection or a pneumatic ejection mechanism, which can control the ejection force and direction to ensure that the plant body can be embedded in the surface layer of the soil or stably stay in the gap between the bare slope, and the ejection device further comprises a real-time calibration module based on a YOLO machine vision model to make a final confirmation on the seeding point below before ejection, and if the seeding point is covered by an obstacle, the ejection target point is automatically adjusted to the nearest suitable position.

[0007] Further, the composition of the nutrient layer comprises organic fertilizer 10-15%, planting soil 80-85%, plant glue 1-3% and rooting powder 1-3%, the substrate layer comprises planting soil 85-90%, microbial inoculant 5-10%, plant glue 1-3% and water retaining agent 1-3%, and the structure layer comprises planting soil 80-85%, basalt fiber 10-15%, plant glue 1-3% and repellent 1-3%.

[0008] Further, the plant body has a polyhedral shape.

[0009] Further, the multi-source information acquisition module comprises a high-precision positioning unit, a spectral camera and a laser radar.

[0010] Further, the intelligent cloud platform comprises: A microhabitat identification unit analyzes collected images and terrain data based on a lightweight deep learning model to identify suitable seeding points meeting preset ecological conditions; A seeding path planning unit generates a non-uniform and nonlinear flight path according to the identified suitable seeding points.

[0011] Further, the preset ecological conditions of the microhabitat identification unit include one or a combination of vegetation coverage, ground roughness, slope, slope direction and indirect indicators of soil moisture to find a local environment capable of providing shade, wind protection and moisture retention.

[0012] A near-natural unmanned aerial vehicle wisdom bomb afforestation method, comprising: The native plant seed is coated and pelleted, The native plant seed is coated and pelleted, and the plant body comprises the native plant seed, a nutrient layer, a substrate layer and a structure layer from inside to outside, the nutrient layer promotes plant germination and the required nutrients at the beginning, the substrate layer guarantees the soil conditions required by the plant at the beginning of growth, and the structure layer maintains the coated seed from being damaged during the ejection process; Intelligent sensing of site conditions and planning of the seeding area of the afforested land: An unmanned aerial vehicle platform equipped with a spectral camera, a high-precision positioning unit and a laser radar is used to carry out aerial survey on the afforestation area, obtain high-precision topographic data, vegetation coverage data and soil information, and add afforestation requirements; Based on the obtained data, intelligent division of site types is carried out through image recognition and scene classification artificial intelligence algorithms, suitable micro-sites for seeding are identified, and a three-dimensional seeding planning map containing the best seeding point position, recommended seeding tree species and seeding amount is generated; Unmanned aerial vehicle wisdom bomb seeding operation: The unmanned aerial vehicle platform equipped with a special ejection seeding device carries out flight navigation according to the three-dimensional seeding planning map; the ejection seeding device precisely ejects single or a small amount of plant bodies to the planned seeding point with a certain initial speed and angle, simulating the natural seeding process.

[0013] Further, the artificial intelligence algorithm is an image recognition and scene classification model based on a convolutional neural network combined with an expert system, which is used to analyze multi-source remote sensing data and realize intelligent division of site types to determine the most suitable seeding point.

[0014] The beneficial effects of the present application are: 1. Precise and efficient: through the mode of surveying first, planning second and seeding last, the seeds are precisely placed in the most suitable micro-site for growth, greatly improving the seed utilization rate and operation efficiency, and reducing the labor cost.

[0015] 2. High degree of intelligence: AI decision-making, automatic navigation, real-time feedback and adaptive adjustment are integrated in the whole process, realizing digitalization, intelligence and fine management of afforestation engineering and ecological restoration.

[0016] 3. Strong adaptability: especially suitable for ecological restoration and afforestation engineering in difficult site conditions such as high mountains, steep slopes and degraded forest lands where manpower is difficult to reach. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only constitute a part of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of the provided drawings.

[0018] Figure 1 For the overall flowchart of the present application; Figure 2 For the structural diagram of the plant body; Figure 3 For the structural diagram of the unmanned aerial vehicle platform; Figure 4 For the enlarged plan view of A; Figure 5 For the architectural design diagram of the present application; In the figure: 1-native plant seed; 2-nutrition layer; 3-substrate layer; 4-structure layer; 5-intelligent cloud platform; 6-unmanned aerial vehicle platform; 7-seed storage bin; 8-multi-source information acquisition module; 9-vibration assembly; 10-quantitative seed dispenser; 11-ejection device. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments only constitute a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] Please refer to the accompanying Figures 1-5 The present application provides a near-natural unmanned aerial vehicle intelligent ejection afforestation device, comprising: The unmanned aerial vehicle platform 6 serves as the flight carrying basis of the system; The ejection seeding device is installed on the unmanned aerial vehicle platform 6 and is used for storing and ejecting the plant body with controllable ejection force; The plant body is subjected to coating treatment on the target plant seed to form a pellet containing the native plant seed 1, the nutrition layer 2, the substrate layer 3 and the structure layer 4 from inside to outside; The multi-source information acquisition module 8 is installed on the unmanned aerial vehicle platform 6 and is used for acquiring image information of the afforestation land, which contains geographic, vegetation and topographic data; The intelligent cloud platform 5 is used for processing the image information of the multi-source information acquisition module 8, continuously optimizing the image recognition model by means of the image recognition algorithm, generating the flight and ejection strategy simulating natural seeding, and controlling the unmanned aerial vehicle platform 6 and the ejection seeding device to perform the task. Edge computing device, used with UAV platform 6, automatically identifies microhabitat scene, assists in decision-making of UAV flight path, and transmits data signals.

[0021] Preferably, the ejection seeding device includes a seed storage bin 7, a vibration assembly 9, a quantitative seed dispenser 10, and an ejection device 11. The seed storage bin 7 is fixedly installed at the lower end of the UAV platform 6. The outlet end of the seed storage bin 7 is respectively provided with the vibration assembly 9, the quantitative seed dispenser 10, and the ejection device 11. The vibration assembly 9, the quantitative seed dispenser 10, and the ejection device 11 are all electrically connected with the intelligent cloud platform 5. The ejection device 11 can adjust the initial speed and angle of ejection to achieve accurate delivery of different weight plant bodies and different ejection distances.

[0022] Preferably, the ejection device 11 is an electromagnetic ejection, a rotating flywheel, a mechanical ejection, or a pneumatic ejection mechanism, which can control the ejection force and direction to ensure that the plant body can be embedded in the soil surface layer or stably stay in the gap between the exposed slope. The ejection device 11 also includes a real-time calibration module based on a YOLO machine vision model to make the last confirmation of the seeding point below before ejection. If the seeding point is covered by obstacles, the ejection target point will be automatically adjusted to the nearest suitable position.

[0023] Before the device is operated, the prepared plant body is placed in the seed storage bin 7. During the process of accurate ejection, the vibration assembly 9 enables the plant body in the seed storage bin 7 to move downward smoothly without stacking in the seed storage bin 7. The plant body moving downward through the vibration assembly 9 determines the number of ejection plant bodies through the quantitative seed dispenser 10 under the design of the intelligent program. The ejection device 11 provides power conditions and ejection angles for the ejection plant body under the setting of the intelligent program.

[0024] Preferably, the composition of the nutrient layer 2 includes organic fertilizer 10-15%, planting soil 80-85%, plant glue 1-3%, and rooting powder 1-3%, which promotes plant germination and provides the nutrients needed at the beginning. The substrate layer 3 includes planting soil 85-90%, microbial inoculant 5-10%, plant glue 1-3%, and water-retaining agent 1-3%, which ensures the soil conditions needed for plant growth at the beginning. The structure layer (4) includes planting soil 80-85%, basalt fiber 10-15%, plant glue 1-3%, and repellent 1-3% (mainly camphor, which ensures that the ejection plant body is not disturbed by rodents and pests), which maintains the seed after coating from being damaged during the ejection process.

[0025] Preferably, the shape of the plant body is a polyhedral shape, which prevents the plant body from rolling after ejection and landing, and makes the plant body land on the predetermined ejection position.

[0026] Preferably, the multi-source information acquisition module 8 comprises a high-precision positioning unit, a spectral camera and a laser radar.

[0027] Preferably, the intelligent cloud platform 5 comprises: A microhabitat identification unit analyzes the collected images and terrain data based on a lightweight deep learning model, and identifies suitable seeding points that meet the preset ecological conditions. A seeding path planning unit generates a non-uniform and nonlinear flight path according to the identified suitable seeding points.

[0028] Preferably, the preset ecological conditions of the microhabitat identification unit include one or a combination of vegetation coverage, ground roughness, slope, slope direction and soil moisture indirect indicators, to find a local environment that can provide shade, wind protection and moisture retention.

[0029] A near-natural unmanned aerial vehicle intelligent bomb-seeding afforestation method comprises: A native plant seed 1 is coated, and the phytobiotic includes, from the inside out, the native plant seed 1, a nutrient layer 2, a substrate layer 3 and a structure layer 4. The nutrient layer 2 promotes plant germination and provides the necessary nutrients at the beginning, the substrate layer 3 ensures the soil conditions required for plant growth at the beginning, and the structure layer 4 maintains the integrity of the coated seed during the ejection process. Intelligent sensing of afforestation site conditions and planning of seeding area: An unmanned aerial vehicle platform 6 equipped with a spectral camera, a high-precision positioning unit and a laser radar is used to conduct aerial survey of the afforestation area, obtain high-precision terrain data, vegetation coverage data and soil information, and add afforestation requirements. Based on the obtained data, an image recognition and scene classification artificial intelligence algorithm is used to intelligently divide the site types, identify suitable micro-sites for seeding, and generate a three-dimensional seeding plan that includes the best seeding point location, recommended tree species and seeding amount. Unmanned aerial vehicle intelligent bomb-seeding operation: The unmanned aerial vehicle platform 6 equipped with a special bomb-seeding device performs flight navigation according to the three-dimensional seeding plan; the bomb-seeding device precisely ejects single or a small amount of phytobiotics at a certain initial speed and angle to the planned seeding point, simulating the natural seed dropping process.

[0030] Preferably, the artificial intelligence algorithm is an image recognition and scene classification model based on a convolutional neural network combined with an expert system, which is used to analyze multi-source remote sensing data and realize intelligent division of site types to determine the most suitable seeding point.

[0031] The specific operation process of the present application is as follows: First, deploy the job vehicle equipped with edge computing equipment near the target afforestation area. The edge computing equipment serves as the on-site command and processing center, and its hardware configuration is as follows: Vehicle-mounted computer: equipped with Intel Core i9 high-performance processor, NVIDIA RTX series independent graphics card, 64GB DDR4 memory and 1TB or more high-speed SSD, used for processing complex AI inference and path planning calculation.

[0032] Communication and power supply: equipped with high-speed Wi-Fi or Ethernet interface, multi-modal communication module, and stable power supply provided by vehicle-mounted power system or independent battery pack.

[0033] Referring to Figure 1 , the present application includes three processes of intelligent perception of afforestation site conditions and seeding area planning, pelletization of native plant seeds 1, and ejection seeding of phytobodies.

[0034] Referring to Figure 3 , use the UAV platform 6 equipped with RTK, P1 aerial survey camera and L1 laser radar to conduct aerial survey of the afforestation area, generate centimeter-level precision digital elevation model and orthographic image map. Through image recognition and scene classification and other artificial intelligence algorithms, the microsite suitable for seeding is identified, and a three-dimensional seeding planning map containing the best seeding point position, recommended seeding species and seeding amount is generated.

[0035] When the UAV platform 6 completes the survey flight, it returns to the edge computing equipment. Through a high-bandwidth dedicated data link or a high-speed wired interface, the raw aerial survey data stored on board is quickly unloaded to the edge computing equipment. The edge computing equipment fuses and preprocesses the received data.

[0036] Position: combined with the centimeter-level positioning data provided by the RTK of the UAV platform 6, and the internal and external parameters of the P1 aerial survey camera, the pixel coordinates of the identified microhabitat in the image are calculated to the accurate geographic coordinates (such as WGS84 coordinate system) in the real world.

[0037] Size: use the three-dimensional point cloud data generated by the laser radar to directly measure the actual physical size (length, width, depth, etc.) of the microhabitat.

[0038] Secondly, through the AI image recognition analysis module on the intelligent cloud platform 5, different site types such as sunny slope, shady slope, semi-sunny slope and valley are identified, and further on the slope surface, rock exposed area, thick soil layer area and herbaceous cover area are identified, and their accurate coordinates and sizes are recorded. The system finally plans the optimal three-dimensional operation map.

[0039] A path planning unit in the edge computing device plans a seeding operation path according to the microhabitat point distribution map generated in step AI.

[0040] Planning algorithm: an improved traveling salesman problem algorithm is used to cluster the microhabitat points and sort the paths, and a non-uniform, non-linear three-dimensional flight path that can avoid obstacles and minimize flight distance is generated.

[0041] Instruction issuing: the planned path and seeding parameters are sent to the unmanned aerial vehicle platform 6 through a high-speed data link.

[0042] Referring to Figure 2 The native plant seeds 1 are subjected to pelletization treatment, and the particle size of the native plant seeds 1 after pelletization treatment is 20-30 mm, and the mass is about 10-20 g.

[0043] Finally, the ejection seeding device added to the unmanned aerial vehicle platform 6 is used for operation.

[0044] Referring to Figure 4 The native plant seeds 1 are placed in the seed storage bin 7, and the unmanned aerial vehicle platform 6 autonomously flies according to the planned route, reaches the predetermined seeding point, and then the quantitative seed dispenser 10 takes 1-3 pellets each time, which are ejected downward by the electromagnetic ejector, the wheeled ejector or the pneumatic ejector 11 at a speed of about 5-10 m / s and an angle of 45°, so that the plant bodies can be gently “planted” into the soil, litter layer, rock crevices and low-lying areas. Before ejection, the unmanned aerial vehicle platform 6 can use the on-board visual sensor to make a final confirmation of the point below, and if the original point is covered by new obstacles, the target ejection point can be fine-tuned.

[0045] The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the protection scope of the application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on these embodiments, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of the present application. Although the present application has been described in detail with reference to the above embodiments, a person of ordinary skill in the art can still combine, add or delete the features of the embodiments of the present application according to the circumstances without creative labor, so as to obtain different other technical solutions which do not deviate from the concept of the present application in essence, and these technical solutions also belong to the scope of the present application.

Claims

1. A near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device, characterized in that, include: The unmanned aerial vehicle platform (6) serves as the flight support basis for the system; A catapult seeding device is installed on an unmanned aerial vehicle platform (6) for storing and launching vegetation with a controllable catapult force; Vegetation body, the target plant seeds are coated to form pellets containing native plant seeds (1), nutrient layer (2), matrix layer (3) and structural layer (4) from the inside out; A multi-source information acquisition module (8) is installed on the UAV platform (6) to acquire image information of the afforestation area; The intelligent cloud platform (5) is used to process the image information of the multi-source information acquisition module (8), continuously optimize the image recognition model with the help of image recognition algorithm, generate flight and catapult strategies to simulate natural seeding, and control the UAV platform (6) and catapult seeding device to perform tasks. Edge computing devices are used in conjunction with drone platforms (6) to automatically identify micro-habitat scenes, assist in drone flight path decision-making, and transmit data signals.

2. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 1, characterized in that, The ejector seeding device includes a seed storage bin (7), a vibration component (9), a quantitative seed dispenser (10), and an ejector device (11). The seed storage bin (7) is fixedly installed at the lower end of the drone platform (6). The outlet end of the seed storage bin (7) is respectively provided with a vibration component (9), a quantitative seed dispenser (10), and an ejector device (11). The vibration component (9), the quantitative seed dispenser (10), and the ejector device (11) are all electrically connected to the intelligent cloud platform (5).

3. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 2, characterized in that, The ejection device (11) is an electromagnetic ejection, a rotating flywheel, a mechanical ejection, or a pneumatic ejection mechanism, which can control the ejection force and direction to ensure that the vegetation can be embedded in the soil surface or remain stably in the gaps of the bare slope. The ejection device (11) also includes a real-time calibration module based on the YOLO machine vision model to make a final confirmation of the seeding point below before ejection. If the seeding point is covered by an obstacle, the ejection target point is automatically adjusted to the nearest suitable position.

4. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 1, characterized in that, The nutrient layer (2) consists of 10-15% organic fertilizer, 80-85% planting soil, 1-3% plant gum and 1-3% rooting powder. The substrate layer (3) consists of 85-90% planting soil, 5-10% microbial agent, 1-3% plant gum and 1-3% water-retaining agent. The structural layer (4) consists of 80-85% planting soil, 10-15% basalt fiber, 1-3% plant gum and 1-3% repellent.

5. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 4, characterized in that, The plant has a polyhedral shape.

6. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 1, characterized in that, The multi-source information acquisition module (8) includes a high-precision positioning unit, a spectral camera, and a lidar.

7. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 1, characterized in that, The intelligent cloud platform (5) includes: The microhabitat identification unit analyzes the collected image and terrain data based on a lightweight deep learning model to identify suitable sowing sites that meet the preset ecological conditions. The seeding path planning unit generates a non-uniform, non-linear flight path based on the identified suitable seeding points.

8. The near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation device according to claim 7, characterized in that, The preset ecological conditions of the microhabitat identification unit include one or a combination of indirect indicators such as vegetation cover, surface roughness, slope, aspect, and soil moisture, in order to find local environments that can provide shade, wind protection, and moisture retention.

9. A near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation method, characterized in that, include: Seeds of native plants (1) Coating and pelleting treatment: The native plant seeds (1) are coated. The plant body includes, from the inside out: native plant seed (1), nutrient layer (2), substrate layer (3), and structural layer (4). The nutrient layer (2) promotes the germination of the plant and provides the nutrients needed at the beginning. The substrate layer (3) ensures the soil conditions needed when the plant begins to grow. The structural layer (4) ensures that the coated seeds are not damaged during the ejection process. Intelligent sensing of afforestation site conditions and planning of planting areas: Using a drone platform (6) equipped with a spectral camera, a high-precision positioning unit and a lidar, aerial surveys were conducted in the area to be afforested to obtain high-precision terrain data, vegetation cover data and soil information, and afforestation requirements were added. Based on the acquired data, site types are intelligently classified using image recognition and scene classification artificial intelligence algorithms, suitable micro-sites for sowing are identified, and a three-dimensional sowing planning map containing the optimal sowing point location, suggested tree species, and sowing quantity is generated. Intelligent drone seeding operation: The drone platform (6) equipped with a special catapult seeding device navigates according to the three-dimensional seeding plan; the catapult seeding device accurately launches single or a small number of plants at a certain initial velocity and angle to the planned seeding point, simulating the natural seeding process.

10. A near-natural unmanned aerial vehicle (UAV) intelligent seeding afforestation method according to claim 9, characterized in that, The artificial intelligence algorithm is an image recognition and scene classification model based on the combination of convolutional neural networks and expert systems. It is used to analyze multi-source remote sensing data to achieve intelligent classification of site types in order to determine the most suitable planting location.

Images