Dynamic alignment device for antenna of remote controller of unmanned aerial vehicle

Abstract

The embodiment of the invention relates to an unmanned aerial vehicle remote controller antenna dynamic alignment device. The device comprises an unmanned aerial vehicle positioning module which is used for obtaining real-time position data of an unmanned aerial vehicle, determining an azimuth angle and a pitch angle of the unmanned aerial vehicle according to an unmanned aerial vehicle image shot by a preset camera and camera calibration parameters, and determining a relative position of the unmanned aerial vehicle according to the intensity of an image transmission signal transmitted by the unmanned aerial vehicle and a multi-antenna signal receiving condition; the dynamic adjustment module is used for determining the relative direction of the unmanned aerial vehicle and the antenna according to the unmanned aerial vehicle position data, the azimuth angle and the pitch angle of the unmanned aerial vehicle and the relative position of the unmanned aerial vehicle, and controlling the antenna steering mechanism to adjust the antenna direction according to the relative direction of the unmanned aerial vehicle and the antenna; and the antenna angle adjusting mechanism is used for adjusting the antenna direction under the control of the dynamic adjusting module. According to the technical scheme, the antenna direction of the remote controller can be dynamically adjusted, and it is ensured that the signal transmitting direction is aligned with the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) control, and in particular to a dynamic alignment device for an antenna of an UAV remote controller. Background Art

[0002] Current mainstream drone remote controls use whip-type omnidirectional antennas, which have an apple-shaped or "swimming ring" signal transmission pattern. Horizontal gain is high (0.1-dBi), while vertical gain drops sharply (-12-15dBi). Users must manually adjust the antenna's orientation, ensuring the line connecting the antenna to the drone is perpendicular to the antenna's axis (i.e., facing the drone horizontally). Otherwise, the signal attenuates significantly. However, the drone's position changes dynamically during flight, and frequent antenna adjustments can be distracting and increase operational complexity.

[0003] Signal attenuation is exacerbated in complex scenarios (such as building obstructions and areas with dense electromagnetic interference sources). The omnidirectional nature of the whip antenna causes signal energy to disperse, making it difficult to transmit it in the direction of the drone. This can easily lead to image transmission lag or even loss of connection due to multipath effects or co-channel interference. Some users have attempted to enhance the signal by modifying third-party directional antennas (such as high-gain parabolic antennas), but such modifications may violate radio management regulations (such as exceeding the radiated power limit in a local direction) and sacrifice omnidirectional coverage, requiring frequent manual adjustments to the antenna direction. Summary of the Invention

[0004] The present invention provides a dynamic alignment device for an antenna of a remote controller of an unmanned aerial vehicle (UAV), the purpose of which is to keep the signal transmission direction of the remote controller aligned with the UAV.

[0005] An embodiment of the present invention provides a dynamic alignment device for an antenna of a drone remote controller, comprising:

[0006] The drone positioning module is used to obtain the real-time location data of the drone, determine the azimuth and pitch angle of the drone based on the drone image captured by the preset camera and the camera calibration parameters, and determine the relative position of the drone based on the image transmission signal strength transmitted by the drone and the multi-antenna signal reception status;

[0007] A dynamic adjustment module is used to determine the relative direction of the drone and the antenna based on the drone position data, the azimuth and pitch angle of the drone and the relative position of the drone, and to control the antenna steering mechanism to adjust the antenna direction according to the relative direction of the drone and the antenna;

[0008] The antenna angle adjustment mechanism is used to adjust the antenna direction under the control of the dynamic adjustment module.

[0009] Optionally, determining the azimuth and pitch angle of the drone based on the drone image includes:

[0010] The drone image is input into a neural network model with an improved YOLOv8 architecture. Multi-scale feature fusion is performed through the SSFF scale sequence feature fusion module to output the target bounding box and predict the coordinates of the center points of the drone's four blades. The improved YOLOv8 architecture is configured to introduce a Ghost module to replace the original convolutional layer or bottleneck structure of YOLOv8.

[0011] Perform perspective transformation using camera calibration parameters to convert the pixel coordinates of the drone image into normalized plane coordinates;

[0012] Select the direction vector of the drone's symmetry axis and calculate the angle between it and the north direction through plane projection as the drone's azimuth;

[0013] The pitch angle of the drone is determined based on the perspective projection principle using the blade size of the drone and the detection width in the drone image.

[0014] Optionally, the relative position of the drone is determined based on the strength of the image transmission signal transmitted by the drone and the signal reception status of multiple antennas, including:

[0015] Four orthogonal ring antenna arrays are used to receive the drone's image transmission signal and simultaneously record the original IQ data and RSSI value of each channel. The antenna spacing of the antenna array is at least twice the signal wavelength.

[0016] The phase difference of multi-antenna reception is used to calculate the signal incident direction and determine the direction of the drone relative to the antenna.

[0017] Optionally, the antenna angle adjustment mechanism includes a dual-degree-of-freedom pan-tilt platform; wherein the dual-degree-of-freedom pan-tilt platform is configured to have horizontal rotation and pitch adjustment; and the remote control is arranged on the dual-degree-of-freedom pan-tilt platform.

[0018] Optionally, the dynamic alignment device for the drone remote control antenna also includes a remote control connection mechanism, including a clamping structure and / or a magnetic interface for fixed connection with the drone remote control.

[0019] Optionally, the UAV remote control antenna dynamic alignment device also includes:

[0020] When acquiring the image transmission signal transmitted by the drone, the signal is adaptively filtered. The receiving end signal is the superposition of the useful signal and the interference d(n) = s(n) + v(n), where s(n) is the received signal and v(n) is the co-channel interference. The reference signal is the interference-related signal x(n) obtained through multiple antennas or spectrum sensing. n represents a continuous time representation.

[0021] The filter output is set to generate an estimate of the interfering signal where w k(n) is the time-varying filter coefficient, L is the filter order;

[0022] The error signal is the difference between the mixed signal and the interference signal estimation value e(n)=d(n)-y(n), and the error signal is the purified useful signal s(n);

[0023] Based on the weight update formula: w(n+1)=w(n)+μ·e(n)·x(n)

[0024] Where μ is the step size factor, which controls the convergence speed and steady-state error; the step size satisfies the stability condition: λ max is the maximum eigenvalue of the autocorrelation matrix of the input signal;

[0025] Based on normalized LMS, the convergence is optimized for input signal power variations where ∈ is a very small constant used to prevent the denominator from being zero.

[0026] Optionally, if the relative direction of the drone and the antenna is determined based on the drone position data, it includes converting the drone position into the azimuth and elevation angles required by the antenna.

[0027] Optionally, controlling the antenna steering mechanism to adjust the antenna direction according to the relative direction between the drone and the antenna includes:

[0028] Based on a dual closed-loop control architecture, the antenna steering mechanism is controlled to adjust the antenna direction according to the relative direction between the drone and the antenna. The outer loop control of the dual closed-loop control architecture is set to calculate the theoretical pointing angle of the antenna based on the drone's position, and the inner loop control is set to search for the signal peak direction based on the real-time intensity gradient change of the image transmission signal using a particle swarm optimization algorithm.

[0029] Optionally, the UAV remote control antenna dynamic alignment device and dynamic adjustment module are also used to:

[0030] Based on the motion prediction model, the drone's future position is predicted through its speed and acceleration data, and the antenna direction is adjusted in advance.

[0031] Optionally, the communication data of the dynamic alignment device for the drone remote controller antenna is exchanged with the drone remote controller via Bluetooth and / or USB HID protocol.

[0032] An embodiment of the present invention provides a dynamic alignment device for a drone remote control antenna. By tracking the drone's position and dynamically adjusting the remote control antenna's direction, this device addresses existing issues such as inefficient manual adjustment, susceptibility to signal interference, and the risk of illegal modification, thereby ensuring that the remote control signal transmission direction remains aligned with the drone. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic structural diagram of a dynamic alignment device for a drone remote control antenna provided by an embodiment of the present invention;

[0034] Figure 2 This is a structural schematic diagram of an antenna angle adjustment mechanism in a dynamic alignment device for an unmanned aerial vehicle remote control antenna provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0036] Example

[0037] Figure 1 This is a schematic diagram of the structure of a dynamic alignment device for a drone remote control antenna provided in Example 1 of the present invention. This embodiment is applicable to situations where a drone is controlled by a remote control, especially when the drone is operating in complex scenarios (such as building obstructions or areas with dense electromagnetic interference sources) where signal attenuation occurs. The dynamic alignment device for a drone remote control antenna includes:

[0038] The drone positioning module 110 is used to obtain the real-time location data of the drone, determine the azimuth and pitch angle of the drone based on the drone image captured by the preset camera and the camera calibration parameters, and determine the relative position of the drone based on the image transmission signal strength transmitted by the drone and the multi-antenna signal reception status;

[0039] A dynamic adjustment module 120 is configured to determine the relative direction of the drone and the antenna based on the drone position data, the azimuth and pitch angles of the drone, and the relative position of the drone, and to control the antenna steering mechanism to adjust the antenna direction based on the relative direction of the drone and the antenna;

[0040] The antenna angle adjustment mechanism 130 is used to adjust the antenna direction under the control of the dynamic adjustment module.

[0041] The drone positioning module 110 can receive the drone's location data, which can be the drone's latitude and longitude coordinates. For example, the drone's latitude and longitude coordinates can be obtained through the remote control's built-in GPS module or an external RTK (Real-Time Kinematic Positioning) receiver. For example, the latitude and longitude coordinates can be accurate to the centimeter level. The drone positioning module 110 can also use visual recognition, such as using a remote control camera or an independent visual sensor to capture drone images and calculate the drone's azimuth and pitch angles using image processing algorithms (such as YOLO target detection). The drone positioning module 110 can also use multiple orthogonal ring antenna arrays with an antenna spacing of twice the signal wavelength of the drone's image transmission signal to form a spatial sampling baseline. The drone image transmission signal is received and the raw IQ data and RSSI values ​​of each channel are synchronously recorded. The recorded signal values ​​are then analyzed to determine the drone's relative position. The dynamic adjustment module 120 integrates multi-source positioning data with signal strength feedback to form a closed-loop control system for perception, decision-making, and execution, avoiding the risk of single sensor failure. The relative direction between the drone and the antenna is determined, and the antenna angle adjustment mechanism 130 can then be controlled to adjust the antenna direction. The antenna angle adjustment mechanism 130 may include a dual-degree-of-freedom gimbal for horizontal rotation and pitch adjustment. The remote control is placed on the platform of the dual-degree-of-freedom gimbal. Under the control of the dynamic adjustment module 120, the servo motor of the dual-degree-of-freedom gimbal controls the rotation and pitch of the adjustment platform, driving the antenna of the remote control to change direction and aim at the working drone.

[0042] The solution in this embodiment can solve the problem of novice users losing control of their drone due to incorrect antenna pointing. Although modifying a directional antenna can increase gain, it violates the restrictions on transmit power and antenna pattern in the "Radio Management Regulations of the People's Republic of China" and requires destroying the original structure, which will affect the warranty. However, the solution in this embodiment can be configured to be fixed to the original remote control via a clamping structure or magnetic interface, without the need for hardware modification, and meets the requirements of Radio Equipment Type Approval (SRRC certification).

[0043] Optionally, determining the azimuth and pitch angle of the drone based on the drone image includes:

[0044] Drone images are fed into a neural network model based on the improved YOLOv8 architecture. The SSFF (Scale Sequence Feature Fusion) module performs multi-scale feature fusion, outputs a bounding box, and predicts the coordinates of the center points of the drone's four propeller blades. The improved YOLOv8 architecture introduces the Ghost module, which replaces the original convolutional layers or bottleneck structure of YOLOv8. The Ghost module replaces the convolutional layers in the Backbone architecture. In the YOLOv8 backbone network, the Ghost module replaces the traditional convolutional layers. The Ghost module separates the primary and secondary paths and generates residual features using inexpensive linear operations, reducing the number of parameters while preserving feature representation. The captured drone images are fed into the neural network model, which is first scaled to a preset size and normalized. The backbone network containing the Ghost module then extracts multi-level feature maps of the image, with each level corresponding to a different resolution (e.g., 80x80, 40x40, 20x20). The SSFF module sequentially fuses upsampled high-level features with downsampled low-level features to generate an enhanced multi-scale feature map. The detection head predicts the target's center point, width, height, and class probability, filters overlapping boxes using non-maximum suppression (NMS), and outputs the final detection box and class label. In one implementation, an output layer is added to predict the coordinates of the drone's four propeller centers, represented as (x1, y1), (x2, y2), (x3, y3), and (x4, y4), respectively. The output layer that predicts the coordinates of the drone's four propeller centers is integrated into the detection head, which employs a decoupled design consisting of three branches: bounding box regression, classification probability, and keypoint prediction. The newly added propeller center prediction layer serves as an extension of the keypoint prediction branch. The YOLOv8-pose model natively supports human pose keypoint detection. Propeller center prediction adjusts the loss function by adjusting the number of keypoints (e.g., COCO has four keypoints). In the keypoint branch of the detection head, the number of output channels is adjusted from the original COCO keypoints to 4x3, representing the coordinates of the center point of the blade and the confidence. The blade coordinates are decoupled from the target detection box and predicted by each branch.

[0045] The pixel coordinates of the drone image are converted to normalized plane coordinates by performing a perspective transformation using the camera calibration parameters. The pixel coordinates are converted to image physical coordinates by translating the principal point and dividing by the focal length. This is achieved by taking the inverse matrix K-1 of the camera's intrinsic parameter matrix K. The normalized plane is the plane with Z = 1 in the camera coordinate system. The normalized coordinates corresponding to the image physical coordinates (x', y') on this plane are:

[0046] Select the direction vector of the drone's symmetry axis and calculate the angle between it and the north direction through plane projection as the drone's azimuth;

[0047] The azimuth angle can be calculated using the following formula:

[0048]

[0049] Where Δx, Δy are the coordinate differences of the endpoints of the symmetry axis in the image. The coordinate difference refers to the coordinates of the endpoints of the symmetry axis in the two images. y , V x The meaning is the component vector of the drone's symmetry axis direction vector V in the x and y directions, and the azimuth refers to the azimuth of the drone relative to the remote control antenna.

[0050] The pitch angle of the drone is determined based on the perspective projection principle using the blade size of the drone and the detection width in the drone image.

[0051] Among them, the pitch angle can be calculated using the following formula

[0052]

[0053] Where f is the focal length, Z is the estimated depth (via binocular parallax or laser ranging fusion). The pitch angle refers to the pitch angle of the drone relative to the antenna.

[0054] Optionally, the relative position of the drone is determined based on the strength of the image transmission signal transmitted by the drone and the signal reception status of multiple antennas, including:

[0055] Four groups of orthogonal circular antenna arrays are used to receive the UAV image transmission signal and simultaneously record the original IQ data and RSSI value of each channel; the antenna spacing of the antenna array is at least 2 times the signal wavelength; the received signal power calculation Where V rms is the effective value of the receiving end voltage, R in is the input impedance.

[0056] The signal incident direction is calculated using the phase difference of multiple antenna reception to determine the direction of the drone relative to the antenna.

[0057] Phase difference formula (distance d between two antennas):

[0058] Phase difference formula (distance d between two antennas):

[0059] Δφ represents the phase difference.

[0060] Based on the nonlinear least squares solution to the direction of the UAV relative to the antenna, the joint multi-antenna observation equation is constructed

[0061]

[0062] Where (x i ,yi ,z i ) is the antenna coordinate, d i To estimate the distance.

[0063] Optionally, the antenna angle adjustment mechanism includes a dual-degree-of-freedom platform; wherein the dual-degree-of-freedom platform is configured to have horizontal rotation and pitch adjustment; the remote control is configured on the dual-degree-of-freedom platform, such as Figure 2 As shown, the remote controller 200 is set on a dual-degree-of-freedom gimbal 300.

[0064] The coverage area can be 0-360° horizontally and 0-90° in pitch. Because the remote control itself also has an antenna, the remote control can be placed on the platform to realize the corresponding functions. The gimbal can use servo motors or stepper motors to drive the rotation in two degrees of freedom.

[0065] Optionally, the dynamic alignment device for the drone remote control antenna also includes a remote control connection mechanism, including a clamping structure and / or a magnetic interface for fixed connection with the drone remote control.

[0066] Optionally, the UAV remote control antenna dynamic alignment device also includes:

[0067] When acquiring the image transmission signal transmitted by the drone, the signal is adaptively filtered. The receiving end signal is the superposition of the useful signal and the interference d(n) = s(n) + v(n), where s(n) is the received signal and v(n) is the co-frequency interference. The reference signal is the interference-related signal x(n) obtained through multiple antennas or spectrum sensing.

[0068] The filter output is set to generate an estimate of the interfering signal where w k (n) is the time-varying filter coefficient, L is the filter order;

[0069] The error signal is the difference between the mixed signal and the interference signal estimation value e(n)=d(n)-y(n), and the error signal is the purified useful signal s(n);

[0070] Based on the weight update formula: w(n+1)=w(n)+μ·e(n)·x(n)

[0071] Where μ is the step size factor, which controls the convergence speed and steady-state error; the step size satisfies the stability condition: λ max is the maximum eigenvalue of the autocorrelation matrix of the input signal;

[0072] Based on normalized LMS, the convergence is optimized for input signal power variations where ∈ is a very small constant used to prevent the denominator from being zero.

[0073] Optionally, if the relative direction of the drone and the antenna is determined based on the drone position data, it includes converting the drone position into the azimuth and elevation angles required by the antenna.

[0074] Optionally, controlling the antenna steering mechanism to adjust the antenna direction according to the relative direction between the drone and the antenna includes:

[0075] Based on a dual closed-loop control architecture, the antenna steering mechanism is controlled to adjust the antenna direction according to the relative direction between the drone and the antenna. The outer loop control of the dual closed-loop control architecture is set to calculate the theoretical pointing angle of the antenna based on the drone's position, and the inner loop control is set to search for the signal peak direction based on the real-time intensity gradient change of the image transmission signal using a particle swarm optimization algorithm.

[0076] Each particle represents a possible signal receiving direction (azimuth angle θ and pitch angle ), the position in the three-dimensional search space is expressed as: X i =[θ i ,φ i ] T . Randomly generate the initial velocity vector Where V max is the maximum allowed speed. The RSSI / SNR value S(θ,φ) in the current direction is measured by the multi-antenna array. The signal strength gradient is calculated Used to guide the particle search direction, Approximate calculation by difference method: The objective function is to maximize the consistency between signal strength and gradient direction. Among them, λ is the gradient weight coefficient (0.2-0.5), α is the angle between the particle velocity direction and the gradient direction The particle motion is corrected by the gradient direction to accelerate convergence. The particle velocity update formula introduces the improved PSO with gradient information: Where W is the inertia weight, W = W max -(W max -W min )·t / T max , W max is the maximum value of the inertia weight, W max is the minimum value of inertia weight, t is time, T max is the time corresponding to the maximum allowed speed, c1, c2, c3 are gradient learning factors (0.5-1.0), r1, r2, r3 are random numbers in the range [0, 1]. Particle position update

[0077] Optionally, the UAV remote control antenna dynamic alignment device and dynamic adjustment module are also used to:

[0078] Based on the motion prediction model, the drone's future position is predicted through its speed and acceleration data, and the antenna direction is adjusted in advance.

[0079] The input data includes IMU three-axis accelerometer, gyroscope, and GPS speed information. The state vector contains position, velocity, acceleration X k =[x,y,z,v x ,v y ,v z ,a x ,a y ,a z ] T

[0080] The observation equation fuses GPS velocity and IMU acceleration.

[0081] The kinematic equations of the trajectory prediction model are discretized using a third-order uniform acceleration model: in:

[0082]

[0083] B=0 means no external control input.

[0084] The Runge-Kutta method is used for numerical integration to predict the future T p Second Track X pred =RK4(x k ,x=f(x),T p ).

[0085] Optionally, the communication data of the dynamic alignment device for the drone remote controller antenna is exchanged with the drone remote controller via Bluetooth and / or USB HID protocol.

[0086] Although the present invention has been described in detail above using general explanations, specific embodiments, and experiments, it will be apparent to those skilled in the art that modifications and improvements may be made based on the present invention. Therefore, such modifications and improvements, which do not depart from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A dynamic alignment device for an aerial vehicle remote control, characterized in that: include: The drone positioning module is used to obtain the real-time location data of the drone, determine the azimuth and pitch angle of the drone based on the drone image captured by the preset camera and the camera calibration parameters, and determine the relative position of the drone based on the image transmission signal strength transmitted by the drone and the multi-antenna signal reception status; A dynamic adjustment module is used to determine the relative direction of the drone and the antenna based on the drone position data, the azimuth and pitch angle of the drone and the relative position of the drone, and to control the antenna steering mechanism to adjust the antenna direction according to the relative direction of the drone and the antenna; The antenna angle adjustment mechanism is used to adjust the antenna direction under the control of the dynamic adjustment module.

2. The dynamic alignment device for the remote control antenna of a UAV according to claim 1, characterized in that: Determining the azimuth and pitch angle of the drone based on the drone image includes: The drone image is input into a neural network model with an improved YOLOv8 architecture. Multi-scale feature fusion is performed through the SSFF scale sequence feature fusion module to output the target bounding box and predict the coordinates of the center points of the drone's four blades. The improved YOLOv8 architecture is configured to introduce a Ghost module to replace the original convolutional layer or bottleneck structure of YOLOv8. Perform perspective transformation using camera calibration parameters to convert the pixel coordinates of the drone image into normalized plane coordinates; Select the direction vector of the drone's symmetry axis and calculate the angle between it and the north direction through plane projection as the drone's azimuth; The pitch angle of the drone is determined based on the perspective projection principle using the blade size of the drone and the detection width in the drone image.

3. The dynamic alignment device for the remote control antenna of a UAV according to claim 1 or 2, characterized in that: Determining the relative position of the drone based on the image transmission signal strength transmitted by the drone and the multi-antenna signal reception status includes: Four orthogonal ring antenna arrays are used to receive the drone's image transmission signal and simultaneously record the original IQ data and RSSI value of each channel. The antenna spacing of the antenna array is at least twice the signal wavelength. The phase difference of multi-antenna reception is used to calculate the signal incident direction and determine the direction of the drone relative to the antenna.

4. The dynamic alignment device for the remote control antenna of a UAV according to claim 1, characterized in that: The antenna angle adjustment mechanism includes a dual-degree-of-freedom platform, wherein the dual-degree-of-freedom platform is configured to have horizontal rotation and pitch adjustment; and the remote control is arranged on the dual-degree-of-freedom platform.

5. The dynamic alignment device for the remote control antenna of a UAV according to claim 4, characterized in that: It also includes a remote control connection mechanism, including a clamping structure and / or a magnetic interface for fixed connection with the remote control of the drone.

6. The dynamic alignment device for the remote control antenna of a UAV according to claim 3, characterized in that: Also includes: When acquiring the image transmission signal transmitted by the drone, the signal is adaptively filtered. The receiving end signal is the superposition of the useful signal and the interference d(n) = s(n) + v(n), where s(n) is the received signal and v(n) is the co-frequency interference. The reference signal is the interference-related signal x(n) obtained through multiple antennas or spectrum sensing. The filter output is set to generate an estimate of the interfering signal where w k (n) is the time-varying filter coefficient, L is the filter order; The error signal is the difference between the mixed signal and the interference signal estimation value e(n)=d(n)-y(n), and the error signal is the purified useful signal s(n); Based on the weight update formula: w(n+1)=w(n)+μ·e(n)·x(n) Where μ is the step size factor, which controls the convergence speed and steady-state error; the step size satisfies the stability condition: λ max is the maximum eigenvalue of the autocorrelation matrix of the input signal; Based on normalized LMS, the convergence is optimized for input signal power variations where ∈ is a very small constant used to prevent the denominator from being zero.

7. The dynamic alignment device for the remote control antenna of a UAV according to claim 1, characterized in that: If the relative direction of the drone and the antenna is determined based on the drone position data, it includes converting the drone position into the azimuth and elevation angles required by the antenna.

8. The dynamic alignment device for the remote control antenna of a UAV according to claim 1, characterized in that: Control the antenna steering mechanism to adjust the antenna direction according to the relative direction between the drone and the antenna, including: Based on a dual closed-loop control architecture, the antenna steering mechanism is controlled to adjust the antenna direction according to the relative direction between the drone and the antenna. The outer loop control of the dual closed-loop control architecture is set to calculate the theoretical pointing angle of the antenna based on the drone's position, and the inner loop control is set to search for the signal peak direction based on the real-time intensity gradient change of the image transmission signal using a particle swarm optimization algorithm.

9. The dynamic alignment device for the remote control antenna of a UAV according to claim 1, characterized in that: The dynamic adjustment module is also used to: Based on the motion prediction model, the drone's future position is predicted through its speed and acceleration data, and the antenna direction is adjusted in advance.

10. The dynamic alignment device for the remote control antenna of a UAV according to claim 1, characterized in that: The communication data of the UAV remote control antenna dynamic alignment device is exchanged with the UAV remote control via Bluetooth and / or USB HID protocol.

Images