Unmanned aerial vehicle artificial wind field precision construction method and system based on rotor airflow regulation

Abstract

The application discloses a kind of unmanned plane artificial wind field precision construction method and system based on rotor airflow regulation, belong to unmanned plane automatic control technical field.For solving existing unmanned plane rotor airflow random diffusion, it is difficult to accurately construct specific direction and form artificial wind field in target area, the method of the present application includes: setting target parameter of demand artificial wind field;Rotor airflow regulation strategy is generated based on target parameter, through the collaborative control of each rotor speed, phase and lift ratio, the downwash airflow is superimposed or interfered in the target area, while generating the attitude regulation strategy of three degrees of freedom control of the flow guide device below the unmanned plane, the downwash airflow is guided and shaped;Dynamic correction is carried out on the regulation strategy based on feedback information during operation to maintain the stability of the wind field.The present application realizes the active control of wind field form and action intensity, and can be used for crop pollination, fruit dew removal, precision spraying and crown layer microclimate regulation and other operations.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) automatic control technology. More specifically, this invention relates to a method and system for accurately constructing artificial wind fields for UAVs based on rotor airflow regulation. Background Technology

[0002] In recent years, with the rapid development of drone technology, its applications in agricultural plant protection, environmental monitoring, engineering operations, and even special transportation have been continuously deepened and expanded. In this process, the downwash airflow generated by the rotor of a drone, as a significant physical phenomenon, has been widely observed and recognized. In scenarios such as pesticide spraying, this airflow has been noted to assist droplets in penetrating the crop canopy, and thus it has been passively utilized in some applications. However, from the perspective of mainstream technology research and engineering, rotor airflow has long been regarded as an auxiliary factor or accompanying effect closely related to the flight stability, maneuverability, and safety of drones. The research focus of academia and industry is highly concentrated on optimizing the overall aerodynamic layout, robust stability control of flight attitude, and reducing flight time and energy consumption, with the core objective of improving the performance of the drone as a flight platform itself. In other words, the current technological paradigm does not recognize the airflow generated by the drone rotor itself as an independently designable, precisely controlled, and actively utilized operational carrier or working medium. This understanding limits the value of airflow to an auxiliary role, failing to explore its enormous potential as a core operational method.

[0003] In the few applications that attempt to actively utilize this airflow, existing technologies have revealed significant limitations. For example, in agricultural spraying operations, existing solutions attempt to enhance the deposition and penetration of pesticide droplets in dense canopies by using the downwash airflow from hovering or flying drones. However, these technologies generally rely on fixed-structure rotors and uniform speed control strategies. All rotors typically operate at the same or simple differential speeds, resulting in airflows with strong randomness and uncontrollable diffusion characteristics in spatial distribution. As the airflow propagates downwards from the rotor disk, it mixes violently with the surrounding air, rapidly dissipating energy and preventing the formation of an artificial wind field with a clear direction, preset intensity gradient, or clear boundary within the target area. Furthermore, existing technologies mostly influence airflow by adjusting the overall flight altitude, horizontal speed, or flight attitude of the drone—an indirect and coarse control method. It alters the relative position of the airflow source (drone) and the target, rather than finely refining the structure and morphology of the airflow itself, lacking a sophisticated, multi-parameter-based control mechanism for the rotor aerodynamics subsystem itself.

[0004] When operational demands expand to a wider range of non-spraying precision intervention scenarios, the shortcomings of existing technologies become even more apparent. For example, in hybrid crop seed production, it is necessary to efficiently and directionally transport pollen from the male parent to the stigma of the female parent; before harvesting high-value fruits and vegetables (such as cherries), it is necessary to remove dew from the fruit surface to prevent diseases; in facility agriculture, it is necessary to regulate the canopy microclimate to promote photosynthesis or reduce humidity; and in industrial settings, it is necessary to perform targeted dust removal. For these tasks, traditional ground fans or fixed ventilation equipment have inherent drawbacks such as poor mobility, limited range of action, and difficulty in adapting to complex terrain. While conventional multi-rotor drones possess excellent maneuverability and flexibility, their airflow, as mentioned above, is uncontrollable and randomly diffused, making it difficult to achieve precise and customized wind field effects on specific areas or targets on the ground or in space. This not only results in a large amount of valuable propulsion energy being wasted on ineffective air disturbances but may also cause unnecessary interference or even damage to non-target areas (such as adjacent crops or sensitive equipment), limiting the commercial application of drones in such high-value operations.

[0005] In summary, the existing technological system suffers from a fundamental technological gap in both concept and method: it has not yet established a method for the proactive construction and precise delivery of artificial wind fields with UAV rotor airflow as the independent and core operational object. Specifically, existing technologies cannot achieve on-demand design and real-time control of the three-dimensional spatial morphology (e.g., cross-sectional shape, depth of action), intensity distribution (e.g., wind speed gradient, core area location), and dynamic characteristics (e.g., directional stability) of the artificial wind field synthesized from rotor downwash airflow according to complex operational needs. Airflow remains a byproduct of the primary task of flight; its state is a passive result determined by flight parameters, rather than an active variable that can be directly targeted and precisely controlled. This deficiency severely restricts the evolution of UAVs from flight and spraying platforms to mobile, intelligent, and multifunctional physical field generation and intervention platforms, hindering their greater role in fields requiring precise airflow intervention, such as precision agriculture and environmental engineering. Therefore, the industry urgently needs an innovative methodology and system to achieve the controllable generation, dynamic shaping, and precise spatial delivery of UAV-derived artificial wind fields, which is precisely the core technical challenge that this invention aims to solve. Summary of the Invention

[0006] One object of the present invention is to solve at least the above-mentioned problems and to provide at least the advantages that will be described later.

[0007] Another objective of this invention is to provide a method for precisely constructing artificial wind fields using UAVs based on rotor airflow regulation. This method transforms uncontrollable airflow into artificial wind fields that can be precisely constructed on demand through coordinated regulation of the rotor and the airflow guiding device, thereby achieving active control over the shape, direction, and intensity of the wind field.

[0008] To achieve these objectives and other advantages according to the present invention, a method for accurately constructing an artificial wind field using a drone based on rotor airflow regulation is provided, comprising: S1. Set the target parameters for the required artificial wind field. The target parameters include at least the wind speed range, the main direction of action, the depth of action, and the spatial distribution pattern. S2. Based on the target parameters set in step S1, a rotor airflow control strategy for the UAV is generated. The rotor airflow control strategy includes a set of instructions for independent or coordinated control of the rotation speed, phase, and lift ratio of each rotor of the UAV, so that the downwash airflow generated by each rotor forms superposition or interference in the target working area. At the same time, an attitude control strategy for the airflow guiding device below the UAV is generated. The attitude control strategy includes instructions for controlling the three degrees of freedom of the airflow guiding device, so as to guide and shape the downwash airflow, thereby constructing an artificial wind field with target parameters in the target working area. S3. During the operation, based on the feedback of the UAV's attitude, position and environmental perception information, the rotor airflow control strategy and the attitude control strategy of the flow guiding device are dynamically corrected to maintain the stability and delivery accuracy of the required artificial wind field.

[0009] In the above technical solution, step S1 sets the target parameters of the required artificial wind field, specifically including one or a combination of the following methods: First, setting based on the geometric features of the target operation area and the physical characteristics of the object being operated on, specifically: obtaining the spatial location information of the target operation area, the spatial location information including at least the boundary coordinates of the operation area, the horizontal distance and vertical height difference relative to the drone, and the canopy height or surface curvature of the object being operated on; determining the physical action threshold that the artificial wind field needs to achieve according to the operation requirements, the physical action threshold including at least the minimum wind speed required to carry pollen or fog droplets, the wind pressure range required to overcome the water film tension on the fruit surface, or the airflow coverage depth required to regulate the canopy microclimate; coupling the spatial location information and the physical action threshold to generate target parameters adapted to the current operation scenario, so that the effect range of the artificial wind field covers the boundary of the operation area, and the wind speed and wind pressure meet the physical action threshold. Secondly, settings are made based on manual input through the human-computer interaction interface: through the human-computer interaction interface of the ground station software or handheld terminal, the parameter values ​​directly input or selected by the operator are received. The parameter values ​​include at least the desired wind speed value, the angle value of the main direction of action, the specific value of the depth of action, and the shape and size of the spatial distribution; the manually input parameter values ​​are directly used as the target parameters of step S1.

[0010] In the above technical solution, the attitude change rate of the flow guiding device is preset to be below a threshold to ensure smooth and continuous movement of the flow guiding body and avoid disturbance to the UAV's flight attitude caused by additional inertial forces or aerodynamic changes generated by the rapid movement of the flow guiding device. When the method is executed in an environment with natural wind interference, the dynamic correction in step S3 further includes: based on the real-time acquired natural wind vector information, introducing a feedforward compensation term into the rotor airflow control strategy and the attitude control strategy of the flow guiding device to counteract the disturbance of natural wind on the delivery direction and wind speed of the artificial wind field.

[0011] Preferably, in step S2, at least one flow guiding device is provided below the drone, the flow guiding device comprising: A ball joint assembly, comprising a mating ball head and a ball socket, the ball socket being connected to the underside of the drone body via a mounting bracket; The guide body is connected to the ball head, enabling it to move in three degrees of freedom relative to the UAV body. With the vertical axis of the UAV body as the Z-axis, the three rotational degrees of freedom include rotational freedom around the Z-axis and tilting freedom around mutually orthogonal X-axis and Y-axis, respectively. By controlling the rotation of the guide body around the Z-axis, the horizontal delivery direction of the artificial wind field can be changed. By controlling the tilt of the guide body around the X-axis and / or Y-axis, the depth of action and / or lateral delivery position of the artificial wind field can be changed.

[0012] In the above technical solution, the flow guiding device further includes a three-degree-of-freedom drive mechanism. The three-degree-of-freedom drive mechanism is built into the ball joint assembly or at the connection between the flow guiding body and the ball joint assembly. The three-degree-of-freedom drive mechanism includes a first motor, a second motor, and a third motor. The rotation axes of the three motors are orthogonal to each other in space and correspond to the three rotational degrees of freedom of the flow guiding body, respectively. The first motor is used to drive the flow guiding body to rotate relative to the ball joint about the vertical axis of the UAV body, with a rotation angle range of 0~360° continuously and a control accuracy of ±0.5°. The second motor is used to drive the flow guiding body to tilt relative to the ball joint about the X-axis, with a tilt angle range of -30°~+30° and a control accuracy of ±0.3°. The third motor is used to drive the flow guiding body to tilt relative to the ball joint about the Y-axis, with a tilt angle range of -30°~+30° and a control accuracy of ±0.3°. The stator portions of the first, second, and third motors are connected to the fixed portion of the ball joint assembly, and the rotor portions are connected to the corresponding moving portions of the guide body, so as to realize independent driving and precise control of the guide body in three rotational degrees of freedom.

[0013] Preferably, the depth of action in step S1 is the absolute depth value D of the artificial wind field in the direction perpendicular to the ground plane, 0.2m≤D≤10m; the spatial distribution form includes the spatial gradient variation characteristics of the wind speed of the artificial wind field in the target working area, and / or the boundary shape characteristics of the artificial wind field in the horizontal direction, wherein the boundary shape characteristics are circular, elliptical, rectangular or fan-shaped.

[0014] Preferably, the specific steps in step S2 for generating the UAV rotor airflow control strategy include: S21. Based on the target parameters, determine the basic rotational speeds of each rotor required to maintain the basic flight state of the UAV using the UAV dynamics model. S22. Based on the main action direction and spatial distribution form in the target parameters, the rotor airflow superposition model is used to calculate, under the premise of meeting the flight state constraints, the adjustment amount of the basic rotation speed of each rotor, the phase relationship between each rotor and the lift ratio required to form the superposition or interference airflow in the target operating area through the optimization algorithm. S23. The basic rotational speeds of each rotor obtained in step S21, adjusted by the adjustment amount obtained in step S22, the phase relationship and lift ratio obtained in step S22, are mapped into independent or cooperative control commands for each rotor, under the premise of satisfying the flight state constraints, so as to form the command set. The flight state constraints include at least the following: the balance constraint between the total lift generated by all rotors and the weight of the UAV and the mission payload; the balance constraint between the torque generated by each rotor and the control torque required to maintain the attitude of the UAV; and the upper limit constraint of the power of each rotor.

[0015] Among the above technical solutions, the UAV dynamics models that can be selected include rigid body six-degree-of-freedom dynamics models, dynamics models based on Lagrange equations, multibody system dynamics models, and aerodynamic and rotor aerodynamic models. The multibody system dynamics model employs the principle of virtual work and the second kind of Lagrange equations, incorporating the forces and moments of each rigid body into a unified dynamic framework. By characterizing the constraints between rigid bodies using Lagrange multipliers, it can accurately describe the impact of the airflow guidance device's attitude adjustment on the overall dynamic characteristics of the UAV, providing theoretical support for the control of the airflow guidance device.

[0016] In the above technical solution, the rotor airflow superposition model is pre-constructed in the following way: through wind tunnel tests and / or actual flight tests, using airflow field data under different control parameters, a mathematical expression or neural network model of the rotor airflow superposition model is generated based on the test data. A specific pre-construction method for the rotor airflow superposition model includes: Wind tunnel test data acquisition: The UAV was fixed on a six-degree-of-freedom wind tunnel test rig. Under zero incoming wind speed conditions, the rotors were controlled to operate at different combinations of rotational speeds (speed range: 2000~6000 rpm), different phase differences (0~2π), and different lift ratios (0.1~0.5). A three-dimensional particle image velocimeter (PIV) was used to measure the three-dimensional wind speed vector field of multiple horizontal sections (0.5m interval) within a range of 0.5m~10m below the UAV to construct an initial dataset.

[0017] Actual flight test calibration: In the absence of natural wind, the UAV hovered at heights of 3m, 5m, and 8m above the ground, and the above parameter combinations were repeated. The wind speed vector reaching the ground was measured using a ground-based ultrasonic wind speed sensor array (5×5 grid, grid spacing 0.5m) to correct and supplement the wind tunnel data.

[0018] When using a neural network model for fitting, a backpropagation (BP) neural network was employed. The input layer consisted of the rotor speed, phase, lift ratio, and target point coordinates (x, y, z). There were two hidden layers (20 neurons each), and the output layer was the wind speed vector (u, v, w) of the target point. 80% of the experimental data was used as the training set, and 20% as the validation set, with the training error controlled within 5%.

[0019] When using an analytical model for fitting, a superposition formula for the induced velocity fields of each rotor is established based on the rotor free vortex theory: ; where Q i f(n) is the circulation of the i-th rotor, which is related to the lift generated by the rotor and is determined by fitting experimental data; i ,p i (n) represents the rotational speed. i Lift ratio p i The correction function; r i Let e ​​be the position vector pointing from the i-th rotor reference point to the target point, where the reference point is taken as the rotor center; i Let N be the unit vector of the i-th rotor's position direction, where the position direction is taken from the direction of the rotor axis, and N is the number of rotors.

[0020] Flight tests were conducted under untrained conditions, and the average relative error between model predictions and measured values ​​was ≤8%, meeting the requirements for engineering applications.

[0021] In the above technical solution, the flight state constraints include: Overall lift balance: ;in, k T The thrust coefficient of the motor-blade combination is taken as 1.0 × 10⁻⁶. -5~ 1.5×10 -5 N / rpm2 m represents the mass of the UAV, and F represents the mission payload. load For mission load, during real-time monitoring, the allowable deviation is ≤2%.

[0022] Attitude torque balance ;R i M is the lever arm vector from the rotor to the center of gravity. reaction,i τ is the rotor counter-torque, N·m; max The maximum torque to be compensated by the flight control system is 0~5 N·m. During real-time monitoring, the torque deviation of each axis (roll, pitch, yaw) must be kept within the expected torque range calculated by the flight control system; if it exceeds this range, the amplitude limit will be triggered.

[0023] Power limit constraint: Where kp = 1.5 × 10 -6 W / rpm 3 , where P is the power coefficient. max This represents the upper limit of the motor's rated power, ranging from 300 to 800W. During real-time monitoring, an early warning is triggered when the power of any rotor exceeds 90% of the rated value, and power reduction protection is implemented when it exceeds 100%.

[0024] Rotational speed change rate constraint (to prevent sudden airflow changes from affecting flight safety): ;n max The maximum permissible rate of change of rotational speed is 400~800 rpm / s.

[0025] The optimization algorithm mentioned in step S22 is one of gradient descent, sequential quadratic programming, genetic algorithm or particle swarm optimization. The objective function of the optimization algorithm is the weighted sum of squares of the deviations between the simulated wind field parameters calculated by the rotor airflow superposition model and the target parameters set in step S1 within the target operating area.

[0026] Preferably, the attitude control strategy for generating the flow guiding device in step S2 specifically includes: S2a. Based on the main action direction in the target parameters and combined with the UAV heading, determine the target rotation angle of the flow guiding device around the Z-axis so that the main action direction of the flow guiding device is aligned with the main action direction; S2b. Based on the target parameters, including the depth of action and spatial distribution, and combined with the height of the UAV relative to the target operating area, determine the target tilt angle of the flow guiding device around the X-axis and / or Y-axis; the target tilt angle around the X-axis and / or Y-axis is used to change the landing point of the downwash airflow and the airflow component in the vertical direction to meet the requirements of the depth of action and the lateral delivery position. S2c: Based on the target rotation angle and target tilt angle, generate drive commands for the flow guiding device to control the flow guiding device to move to the target posture.

[0027] Preferably, the attitude control strategy and the rotor airflow control strategy are generated in concert, and when the target parameters change, the artificial wind field is changed by adjusting the attitude control strategy first. If adjusting the attitude control strategy alone cannot meet the requirements, the rotor airflow control strategy is adjusted within the range allowed by the flight state constraints.

[0028] In the above technical solution, the flight state constraint conditions have a higher priority than the wind field construction requirements. When the adjustment of the rotor airflow control strategy touches the boundary of the flight state constraint conditions, further adjustment of the rotor airflow is stopped, and the maximum adjustable wind field parameter corresponding to the current boundary value is output.

[0029] Preferably, step S3, which dynamically corrects the rotor airflow control strategy and the attitude control strategy of the airflow guiding device, specifically includes: S31. During the operation, real-time feedback information including UAV attitude, position, environmental perception information and the actual effect of the artificial wind field is obtained; and based on the feedback information, the deviation between the currently constructed artificial wind field and the target parameters is evaluated. S32. Prioritize correction by adjusting the attitude control strategy; if the adjustment still does not meet the requirements, adjust the rotor airflow control strategy within the range allowed by the flight state constraints. S33. If the deviation still fails to meet the requirements after adjustment in step S32, then based on real-time feedback information, with the optimization objective of minimizing the deviation between the current artificial wind field and the target parameters, and under the constraints of the flight state, at least one control parameter in the rotor airflow control strategy and at least one control parameter in the attitude control strategy are jointly optimized; the set of control parameters obtained from the optimization calculation is simultaneously applied to the correction instructions of the rotor airflow control strategy and the attitude control strategy to collaboratively reshape the artificial wind field.

[0030] In the above technical solution, the environmental perception information in step S31 includes at least the wind speed and direction information of natural wind, and the obstacle distribution information of the target operation area; the feedback information of the actual effect of the artificial wind field is obtained through an airborne or ground-based sensor network, which includes at least a wind speed sensor, a direction sensor, or a high-speed camera system; the obstacle information is generated in real time by a lidar mounted on a UAV. An extended card two-shift filter is used to fuse the airborne sensor, ground sensor, and visual data. The frequency of the joint optimization calculation in step S33 is lower than the frequency of the priority adjustment in step S32; or the joint optimization calculation is performed offline before each operation task begins, and its calculation results serve as the basis for generating the initial control strategy in step S2.

[0031] In the above technical solution, the specific implementation of the joint optimization calculation in step S33 is as follows: Optimize objective function Among them, V target For the target average wind speed, φ sim To simulate the main direction of the wind field (by V) sim Calculation); D sim To simulate the attenuation depth of the wind field in the vertical direction (defined as the depth at which the wind speed drops to 1 / e); θ prev Let wv be the actual attitude angle vector of the guiding device at the previous optimization time; wφ, wD, and wθ are weighting coefficients determined by trial and error. A hybrid particle swarm optimization-sequential quadratic programming (PSO-SQP) algorithm is used: Phase 1 (Global Search): Particle swarm optimization (30 particles, 20 iterations) performs a coarse global search in the variable space to obtain an initial solution; Phase 2 (Local Precise Search): Using the result of Phase 1 as the initial value, sequential quadratic programming is used for precise optimization, with a convergence accuracy set to 10. -4 Calculation frequency: During online execution, it is triggered once every 2 seconds; if the deviation is small (<0.3 m / s), the frequency is reduced to once every 5 seconds. In each iteration, the flight state constraints (lift, torque, power) are added to the objective function as a penalty function. When the constraints are violated, the objective function value is multiplied by the penalty factor (value 10). 3 ).

[0032] Optimize variable boundaries: the speed adjustment should not exceed 5% of the base speed, the attitude angle adjustment should be less than or equal to 10° in the Z-axis direction and less than or equal to 5° in the X-axis and Y-axis directions; 0.1 ≤ lift ratio ≤ 0.4, and the sum of the lift ratios should be 1; 0 ≤ phase ≤ 2π.

[0033] Preferably, the operations described are crop pollination, fruit surface dehydration, enhanced spray deposition, or canopy microclimate regulation. Specifically, the fruit dehydration operation includes: setting target parameters for an artificial wind field based on the distribution of water film or dew on the fruit surface, ensuring the artificial wind field has a wind speed range sufficient to overcome the surface tension of the water film, and a spatial distribution form conforming to the shape of the fruit tree canopy; using a drone flying between rows of fruit trees, and coordinating the control of the rotor airflow and the attitude of the guide device, precisely delivering the artificial wind field to the fruit surface to remove or disperse the water film or dew on the fruit surface.

[0034] This invention further claims a UAV-based artificial wind field precision delivery system for implementing the aforementioned UAV-based artificial wind field precision construction method based on rotor airflow regulation, comprising: The target parameter setting module is used to set the target parameters of the artificial wind field based on the spatial location information of the target operation area and the operation requirements. A rotor airflow control module, connected to the target parameter setting module, is used to generate a UAV rotor airflow control strategy based on the target parameters; A flow guiding attitude control module is connected to the target parameter setting module and is used to generate an attitude control strategy for the flow guiding device based on the target parameters. The dynamic correction module is connected to the rotor airflow control module and the airflow attitude control module, and receives the attitude, position and environmental perception feedback information of the UAV. It is used to dynamically correct the rotor airflow control strategy and the attitude control strategy during operation. A collaborative control unit, connected to the rotor airflow control module, the guide air attitude control module, and the UAV's flight controller, coordinates rotor control commands, guide air attitude control commands, and the UAV's basic flight control commands, ensuring that the rotor airflow control strategy meets flight state constraints. The collaborative control unit is configured to: divide the UAV control system into a flight control layer and a wind field control layer; the flight control layer maintains the UAV's attitude stability, position control, and safe flight, with its control commands having higher priority than the wind field control layer; the wind field control layer generates the rotor airflow control strategy and the guide air device's attitude control strategy; when the flight control layer detects an attitude deviation or position error exceeding a preset threshold, the collaborative control unit restricts, corrects, or temporarily suppresses the output commands of the wind field control layer.

[0035] This invention further claims a drone for precise delivery of artificial wind fields, comprising: A multi-rotor unmanned aerial vehicle platform, which includes multiple independently driven rotors; At least one flow guiding device is disposed below the UAV platform. The flow guiding device includes a flow guiding body, a ball joint assembly, and a mounting bracket. The ball joint assembly includes a ball head and a ball socket that cooperate with each other. The ball head is connected to the flow guiding body, and the ball socket is connected to the UAV platform through the mounting bracket, so that the flow guiding body has three rotational degrees of freedom relative to the UAV platform: rotation about the Z-axis, tilting about the X-axis, and tilting about the Y-axis. The controller is communicatively connected to the drive units of the plurality of rotors and the drive unit of the air guide device; The controller is configured to execute the method for accurately constructing an artificial wind field for UAVs based on rotor airflow control as described in any one of claims 1 to 8, in order to control the rotational speed, phase and lift ratio of each rotor, and to control the attitude of the airflow guiding device, thereby accurately constructing and delivering an artificial wind field that conforms to the target parameters in the target operating area.

[0036] The present invention has at least the following beneficial effects: Firstly, this invention integrates rotor airflow regulation with the attitude regulation of the airflow guiding device, transforming the originally randomly diffused downwash airflow of the UAV into an artificial wind field that can be precisely constructed on demand. This enables the active design of wind speed, direction, depth, and spatial morphology, fundamentally breaking through the technical limitations of uncontrollable airflow in traditional UAVs and providing a brand-new technical means for airflow intervention operations. Secondly, this invention constructs a hierarchical constraint optimization algorithm, which, under the premise of strictly satisfying flight state constraints such as lift, torque and power, realizes the joint calculation of rotor speed fine adjustment and guide device attitude angle; at the same time, by establishing the mapping relationship between the guide body rotation angle and the target wind direction through coordinate system transformation, the wind field control can achieve centimeter-level delivery accuracy without affecting flight safety, which significantly improves the reliability and accuracy of wind field construction in complex environments. Thirdly, this invention establishes a flight-priority hierarchical collaborative control logic and dynamic correction mechanism. When the wind field target changes, the attitude of the guide device is adjusted first, and the rotor parameters are adjusted only when the attitude adjustment is insufficient. When the deviation exceeds the limit, multi-parameter joint optimization is initiated. This mechanism not only minimizes the interference to the flight state, but also ensures the stable output of the artificial wind field under external disturbances through real-time feedback of inner and outer loops. Fourth, this invention integrates target parameter setting, rotor control, airflow attitude control and dynamic correction functions into a modular system, and realizes three-degree-of-freedom motion through a ball joint airflow guide device, thus constructing a complete method-system-equipment technology system; enabling ordinary multi-rotor drones to be quickly upgraded into operating platforms with the ability to accurately deliver artificial wind fields, and widely applicable to various agricultural scenarios such as pollination, dew removal, spraying and microclimate regulation.

[0037] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0038] Figure 1 This is a flowchart illustrating the method for accurately constructing an artificial wind field using a drone based on rotor airflow control, as described in an embodiment of the present invention. Figure 2 This is a schematic diagram of the flow guiding device described in this invention; Among them, 1. mounting bracket; 2. ball socket; 3. ball head. Detailed Implementation

[0039] The present invention will now be described in further detail with reference to specific embodiments, so that those skilled in the art can implement it based on the description.

[0040] It should be understood that terms such as “having,” “comprising,” and “including” as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0041] Example: Directed pollination of maize from the male row to the female row based on artificial wind fields. This embodiment is applied to the directional pollination operation in the maize hybrid seed production process, aiming to accurately deliver the pollen produced by the male parent row to the adjacent female parent row area, thereby improving pollination efficiency and seed setting rate.

[0042] The work area is a standardized maize seed production field, with the male and female rows arranged in parallel intervals, and a row spacing of d. r =2.0 m, average height H of male parent plants f =2.5 m, average height H of the mother plant m =2.0m. The UAV uses a hexacoach platform with symmetrical airflow guides located beneath the fuselage. These guides have three rotational degrees of freedom: rotation around the Z-axis and tilting around the X and Y axes, independently driven by a built-in three-degree-of-freedom direct-drive mechanism. The UAV is equipped with a controller integrating a target parameter setting module, a rotor airflow control module, an airflow attitude control module, a dynamic correction module, and a collaborative control unit. It is also connected to an attitude sensor (IMU), RTK-GPS, an onboard anemometer, and a ground-based wireless wind speed sensor network.

[0043] like Figure 2 The flow guiding device includes: a ball joint assembly, which includes a mating ball head 3 and a ball socket 2, the ball socket 2 being connected to the underside of the UAV body via a mounting bracket 1; a flow guiding body (not shown in the attached drawings, its overall shape being plate-shaped or curved plate-shaped), which is connected to the ball head and allows the flow guiding body to have three degrees of freedom of movement relative to the UAV body, with the vertical axis of the UAV body as the Z-axis, the three rotational degrees of freedom including rotational degrees of freedom around the Z-axis and tilting degrees of freedom around mutually orthogonal X-axis and Y-axis respectively; by controlling the rotation of the flow guiding body around the Z-axis, the horizontal delivery direction of the artificial wind field can be changed; by controlling the tilt of the flow guiding body around the X-axis and / or Y-axis, the depth of action and / or lateral delivery position of the artificial wind field can be changed.

[0044] like Figure 1 Based on the relative positions of the parent and parent rows, canopy height, and pollen transport requirements, the artificial wind field parameters are set as follows: Main direction of action φ target Horizontal direction pointing towards the parent tree row (perpendicular to the row direction), at a 90° angle. Depth of influence: D = 2.0m (covering the parent tree canopy). Wind speed range: v min =3m / s, v max=5m / s, to ensure effective pollen transport without damaging the filaments. Spatial distribution: rectangular, width W=1.2m (covering the width of the parent plant row), length L matches the industry flight segment (L=50m), and wind speed is evenly distributed within the rectangular area.

[0045] Rotor airflow control strategy generation: A neural network model based on CFD simulation data is used as the rotor airflow superposition model to describe the mapping relationship between the rotor speed, phase, lift ratio, and wind speed vector field of the target area. For ease of optimization, it is simplified to the following analytical form (taking the wind speed at the center point of the target area as an example): ;where V i Let be the velocity vector generated by the i-th rotor at the target point, and its magnitude is approximately: ; where n i Let n be the rotational speed of the i-th rotor, in rpm; ref =4000rpm, which is the reference speed; pi is the lift proportional coefficient, dimensionless, which satisfies... ;m i Let be the phase angle in rad, which affects the multirotor airflow interference effect. This effect is not explicitly shown in the simplified model, but it has been factored into the CFD simulation; k v =0.02m / s, which is the velocity coefficient (calibrated experimentally); β=0.15m -1 , is the airflow attenuation coefficient; r i Let m be the horizontal distance from the i-th rotor to the target point.

[0046] To ensure stable flight of the drone, rotor control must meet the following constraints: Overall lift balance: ;in, k T =1.2×10 -5 N / rpm 2 (Motor-propeller combined thrust coefficient); UAV mass m = 15kg, mission payload F load =0 (no-load pollination), gravitational acceleration g = 9.8 m / s² 2 .

[0047] Attitude torque balance: ;R i M is the lever arm vector from the rotor to the center of gravity. reaction,i For rotor counter-torque, N·m.

[0048] Power limit constraint: ;where k p =1.5×10 -6 W / rpm 3 , where P is the power coefficient. max=Maximum rated power of the motor.

[0049] To optimize the objective function, the weighted least squares method is used to construct the objective function: Among them, V target =4m / s, which is the target average wind speed (taking the median value), φ sim To simulate the main direction of the wind field (by V) sim Calculation); D sim To simulate the attenuation depth of the wind field in the vertical direction (defined as the depth at which the wind speed drops to 1 / e); the weighting coefficient w v =1, w φ =10, w D =5, w θ =0.1, determined by trial and error.

[0050] A genetic algorithm is used to find the rotor parameters that minimize J within the feasible region. The population size is set to 100, the number of generations to 50, and the variable range is the rotor speed adjustment Δn. i ∈[-200.200]rpm (base speed n) i,0 =4000rpm), phase m i ∈[0,2π], lift ratio p i ∈[0.1,0.3] (the six rotors must satisfy...) After the algorithm converges, the optimal parameter combination is obtained: Δn1=+120rpm, Δn2=-80rpm, Δn3=+50rpm, Δn4=-30rpm, Δn5=+100rpm, Δn6=-60rpm, with all phases set to 0 (to simplify interference effects); lift ratio p i =[0.18,0.15,0.17,0.14,0.19,0.17].

[0051] Generation of attitude control strategy for flow guiding device: based on the target's main direction of action φ target With the drone's current heading φ UAV =0° (flying along the parent line), determine the target rotation angle θ of the guide device around the Z-axis. z =90°, so that the dominant direction of the guide body points towards the parent body row. Based on the target's depth of action, the UAV altitude H = 4.0m (above the ground) and the lateral delivery distance (row spacing) d r =2.0 m, calculate the target tilt angle θ of the guide device around the X-axis (or Y-axis) using geometric relationships. tilt The angle is 26.6°, which is the angle between the normal of the guide body and the vertical direction. This angle ensures that the center of the downwash airflow landing point is exactly located in the parent canopy. All guide devices are driven synchronously according to the above angle, generating guide device attitude control commands.

[0052] Dynamic correction during the process: Real-time feedback information is acquired, including drone attitude (Euler angles), position (RTK-GPS), and natural wind vector V. wind (Airborne anemometer), actual wind speed V of the parent canopy meas (Ground sensor network). Define the bias: e v =V meas -V target ;e φ =φ meas -φ target Correction strategy: Prioritize adjusting the attitude Δθ of the flow guiding device. tilt =Ae v +Bde v / dt; where A = 0.5° / (m / s), B = 0.1°s / m, and the adjustment range is limited to ±5°. If after adjustment |e v If |>0.5m / s, then the rotor speed fine-tuning amount Δn is adjusted under flight constraints. i (Not exceeding 5% of the base speed): Among them, K p =100rpm / (m / s), K i =10rpm / (m·s), K d =20rpm·s / m, weighted according to the optimized proportional coefficient when distributing to each rotor. If the deviation after joint adjustment is still greater than the threshold (|e v (>1m / s for 2s), initiate joint optimization: using the current state as the initial value, resolve the optimization problem under flight state constraints (objective function is the same as before, but V) target (Replace with real-time requirements), step size limited to Δn i ≤50rpm, Δθ tilt ≤2°, calculate the next cycle instruction.

[0053] Anti-natural wind feedforward compensation: When natural wind V is detected wind At this time, a feedforward term is introduced into the attitude control command of the flow guiding device: The orientation of the guide body is corrected according to the wind direction, and differential compensation is superimposed in the rotor control: Δn i,ff =k ff ·(V wind ·e i );e i Let k be the unit vector of the i-th rotor position direction. ff =20rpm / (m / s) to counteract the yaw moment caused by natural wind.

[0054] The collaborative control unit ensures that the flight control layer has higher priority than the wind field control layer. When the flight controller detects an attitude angle deviation greater than 5° or a position error greater than 0.2m, it temporarily suppresses wind field control commands until the flight status returns to stability.

[0055] Through the aforementioned coordinated regulation, pollen from the male parent row was effectively carried and directionally transported to the female parent row area by artificial wind. Field trials showed that, compared to the drone-assisted pollination method (conventional drone operation without rotor airflow regulation and guidance device coordination), the method in this embodiment increased the female parent seed setting rate by approximately 25% and reduced operational energy consumption by 15%. The drone-assisted pollination method involves the drone flying above the male parent row, relying solely on the downwash airflow naturally generated by the rotor to disturb the pollen, without targeted airflow shaping or directional delivery. The seed setting rate was calculated by randomly selecting 10 points in each area at harvest time, continuously surveying 50 ears of fruit at each point, and calculating the average seed setting rate. Energy consumption was based on the electrical energy consumed (battery energy consumption data) to complete the pollination operation over the same area (10 mu). The control area, due to low airflow diffusion efficiency, required multiple round trips (usually 2-3 times) to achieve acceptable pollination results; the experimental area, through precise airflow delivery, achieved better results in a single operation.

[0056] For other machine models or operating scenarios, the parameter calibration method is as follows: Thrust coefficient kT and power coefficient kp: The UAV is fixed on a thrust test bench, and the thrust and power are measured at different speeds. The quadratic coefficient (kT) and cubic coefficient (kp) are obtained by fitting using the least squares method. The calibrated speed range should cover ±20% of the actual operating speed.

[0057] Airflow attenuation coefficient β: In a windless environment, the wind speed at the center point of the ground is measured by hovering the drone at different altitudes, and the β value is obtained through nonlinear fitting. The β value is related to the rotor diameter and the number of blades, and the general formula is: .

[0058] w v Based on the sensitivity of the operation to wind speed, a value of 1.0~2.0 is used for highly sensitive operations such as pollination and decondensation, while a value of 0.5~1.0 is used for general ventilation operations; w φ For operations requiring precise orientation, such as inter-row pollination, use 5-10; for operations requiring only area coverage, use 1-3. D For operations within the canopy, such as spray penetration, use a rate of 3-5; for surface operations, such as dew removal, use a rate of 1-2. θ A value of 0.1 is used to suppress frequent vibrations of the flow guiding device.

[0059] The number of devices and processing scale described herein are for simplification of the invention. Applications, modifications, and variations of the present invention's method and system for precisely constructing artificial wind fields using UAVs based on rotor airflow regulation will be readily apparent to those skilled in the art.

[0060] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and the specific embodiments shown and described herein.

Claims

1. A method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation, characterized in that, include: S1. Set the target parameters for the required artificial wind field. The target parameters include at least the wind speed range, the main direction of action, the depth of action, and the spatial distribution pattern. S2. Based on the target parameters set in step S1, a rotor airflow control strategy for the UAV is generated. The rotor airflow control strategy includes a set of instructions for independent or coordinated control of the rotation speed, phase, and lift ratio of each rotor of the UAV, so that the downwash airflow generated by each rotor forms superposition or interference in the target working area. At the same time, an attitude control strategy for the airflow guiding device below the UAV is generated. The attitude control strategy includes instructions for controlling the three degrees of freedom of the airflow guiding device, so as to guide and shape the downwash airflow, thereby constructing an artificial wind field with target parameters in the target working area. S3. During the operation, based on the feedback of the UAV's attitude, position and environmental perception information, the rotor airflow control strategy and the attitude control strategy of the flow guiding device are dynamically corrected to maintain the stability and delivery accuracy of the required artificial wind field.

2. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 1, characterized in that, In step S2, at least one flow guiding device is installed below the drone, the flow guiding device comprising: A ball joint assembly, comprising a mating ball head and a ball socket, the ball socket being connected to the underside of the drone body via a mounting bracket; The guide body is connected to the ball head, enabling it to move in three degrees of freedom relative to the UAV body. With the vertical axis of the UAV body as the Z-axis, the three rotational degrees of freedom include rotational freedom around the Z-axis and tilting freedom around mutually orthogonal X-axis and Y-axis, respectively. By controlling the rotation of the guide body around the Z-axis, the horizontal delivery direction of the artificial wind field can be changed. By controlling the tilt of the guide body around the X-axis and / or Y-axis, the depth of action and / or lateral delivery position of the artificial wind field can be changed.

3. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 2, characterized in that, The working depth mentioned in step S1 is the absolute depth value D of the artificial wind field in the direction perpendicular to the ground plane, 0.2m≤D≤10m; the spatial distribution form includes the spatial gradient variation characteristics of the wind speed of the artificial wind field in the target working area, and / or the boundary shape characteristics of the artificial wind field in the horizontal direction, wherein the boundary shape characteristics are circular, elliptical, rectangular or fan-shaped.

4. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 3, characterized in that, The specific steps in step S2 to generate the UAV rotor airflow control strategy include: S21. Based on the target parameters, determine the basic rotational speeds of each rotor required to maintain the basic flight state of the UAV using the UAV dynamics model. S22. Based on the main action direction and spatial distribution form in the target parameters, the rotor airflow superposition model is used to calculate, under the premise of meeting the flight state constraints, the adjustment amount of the basic rotation speed of each rotor, the phase relationship between each rotor and the lift ratio required to form the superposition or interference airflow in the target operating area through the optimization algorithm. S23. The basic rotational speeds of each rotor obtained in step S21, adjusted by the adjustment amount obtained in step S22, the phase relationship and lift ratio obtained in step S22, are mapped into independent or cooperative control commands for each rotor, under the premise of satisfying the flight state constraints, so as to form the command set. The flight state constraints include at least the following: the balance constraint between the total lift generated by all rotors and the weight of the UAV and the mission payload; the balance constraint between the torque generated by each rotor and the control torque required to maintain the attitude of the UAV; and the upper limit constraint of the power of each rotor.

5. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 4, characterized in that, The attitude control strategy for generating the flow guiding device in step S2 specifically includes: S2a. Based on the main action direction in the target parameters and combined with the UAV heading, determine the target rotation angle of the flow guiding device around the Z-axis so that the main action direction of the flow guiding device is aligned with the main action direction; S2b. Based on the target parameters, including the depth of action and spatial distribution, and combined with the height of the UAV relative to the target operating area, determine the target tilt angle of the flow guiding device around the X-axis and / or Y-axis; the target tilt angle around the X-axis and / or Y-axis is used to change the landing point of the downwash airflow and the airflow component in the vertical direction to meet the requirements of the depth of action and the lateral delivery position. S2c: Based on the target rotation angle and target tilt angle, generate drive commands for the flow guiding device to control the flow guiding device to move to the target posture.

6. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 5, characterized in that, The attitude control strategy is generated in conjunction with the rotor airflow control strategy. When the target parameters change, the artificial wind field is changed by adjusting the attitude control strategy first. If the requirements cannot be met by adjusting the attitude control strategy alone, the rotor airflow control strategy is adjusted within the range allowed by the flight state constraints.

7. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 6, characterized in that, Step S3, which dynamically corrects the rotor airflow control strategy and the attitude control strategy of the airflow guiding device, specifically includes: S31. During the operation, real-time feedback information including UAV attitude, position, environmental perception information and the actual effect of the artificial wind field is obtained; and based on the feedback information, the deviation between the currently constructed artificial wind field and the target parameters is evaluated. S32. Prioritize correction by adjusting the attitude control strategy; if the adjustment still does not meet the requirements, adjust the rotor airflow control strategy within the range allowed by the flight state constraints. S33. If the deviation still fails to meet the requirements after adjustment in step S32, then based on real-time feedback information, with the optimization objective of minimizing the deviation between the current artificial wind field and the target parameters, and under the constraints of the flight state, at least one control parameter in the rotor airflow control strategy and at least one control parameter in the attitude control strategy are jointly optimized; the set of control parameters obtained from the optimization calculation is simultaneously applied to the correction instructions of the rotor airflow control strategy and the attitude control strategy to collaboratively reshape the artificial wind field.

8. The method for accurately constructing artificial wind fields using UAVs based on rotor airflow regulation as described in claim 1, characterized in that, The operations described include crop pollination, fruit surface dehydration, enhanced spray deposition, or regulation of canopy microclimate.

9. A UAV-based artificial wind field precision delivery system, used to implement the UAV-based artificial wind field precision construction method based on rotor airflow control as described in any one of claims 1 to 8, characterized in that, include: The target parameter setting module is used to set the target parameters of the artificial wind field based on the spatial location information of the target operation area and the operation requirements. A rotor airflow control module, connected to the target parameter setting module, is used to generate a UAV rotor airflow control strategy based on the target parameters; A flow guiding attitude control module is connected to the target parameter setting module and is used to generate an attitude control strategy for the flow guiding device based on the target parameters. The dynamic correction module is connected to the rotor airflow control module and the airflow attitude control module, and receives the attitude, position and environmental perception feedback information of the UAV. It is used to dynamically correct the rotor airflow control strategy and the attitude control strategy during operation. The collaborative control unit is connected to the rotor airflow control module, the guide attitude control module, and the UAV flight controller, respectively. It is used to coordinate rotor control commands, guide attitude control commands, and UAV basic flight control commands, and to ensure that the rotor airflow control strategy meets the flight state constraints.

10. A drone for precise delivery of artificial wind farms, characterized in that, include: A multi-rotor unmanned aerial vehicle platform, which includes multiple independently driven rotors; At least one flow guiding device is disposed below the UAV platform. The flow guiding device includes a flow guiding body, a ball joint assembly, and a mounting bracket. The ball joint assembly includes a ball head and a ball socket that cooperate with each other. The ball head is connected to the flow guiding body, and the ball socket is connected to the UAV platform through the mounting bracket, so that the flow guiding body has three rotational degrees of freedom relative to the UAV platform: rotation about the Z-axis, tilting about the X-axis, and tilting about the Y-axis. The controller is communicatively connected to the drive units of the plurality of rotors and the drive unit of the air guide device; The controller is configured to execute the method for accurately constructing an artificial wind field for UAVs based on rotor airflow control as described in any one of claims 1 to 8, in order to control the rotational speed, phase and lift ratio of each rotor, and to control the attitude of the airflow guiding device, thereby accurately constructing and delivering an artificial wind field that conforms to the target parameters in the target operating area.

Images