Dynamic fast moving screen system based on unmanned aerial vehicle cluster and implementation method thereof

Abstract

The invention discloses a dynamic fast moving screen system based on an unmanned aerial vehicle cluster and an implementation method of the dynamic fast moving screen system, and relates to the technical field of unmanned aerial vehicle application. Each unmanned aerial vehicle hangs a lightweight high-reflection gain flexible micro display curtain comprising an ultra-lightweight rigid / semi-rigid frame and a micro tensioning mechanism through a quick-release connecting mechanism of an integrated attitude fine-tuning motor, and the whole cluster forms a continuous display plane which can be controlled to move and deform in the air through close cooperative arrangement; the continuous display plane is formed through cooperation of the multiple unmanned aerial vehicles, image projection is achieved in combination with the projection subsystem, the display resolution is determined by the projector performance and the arrangement density of the unmanned aerial vehicles, and the problem that the resolution of an existing air display technology is limited is effectively solved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) application technology, specifically to a dynamic, fast-moving screen system based on UAV swarms and its implementation method. Background Technology

[0002] With the rapid development of science and technology, drone application technology, swarm control technology, and projection display technology have all made significant progress in their respective fields. Drones, with their flight capabilities and maneuverability, have demonstrated enormous application potential in many areas. Swarm control technology enables multiple drones to work collaboratively to complete complex tasks. Projection display technology is constantly pursuing higher resolution, larger display size, and more flexible display methods to meet diverse visual display needs. Against this backdrop, combining drone swarms with projection display technology to build a new type of dynamic display system has become a highly innovative and forward-looking research direction.

[0003] Currently, aerial display technologies are mainly divided into two categories: drone swarm light show technology and traditional projection display technology. However, both have significant limitations. Drone swarm light show technology uses multiple drones equipped with LED lights to form a swarm, and their arrangement is programmed to form light spot patterns. However, the display resolution of this technology directly depends on the number of drones. To achieve high-resolution display, a large number of drones are required, which not only increases costs but also increases the complexity of control. In addition, LED lights require continuous power, further limiting the drones' endurance and application scenarios. On the other hand, traditional projection display technology uses fixed walls, large fixed screens, or limited movable screens as projection carriers, and projects images through projectors. The range of movement of the projection carrier in this technology is constrained by the physical structure, making it impossible to achieve complex shape deformations. Furthermore, it is inadequate in scenarios that require rapid movement or changes in display position. Therefore, existing technologies cannot meet the needs of constructing a movable, deformable, continuous display plane in three-dimensional space and achieving precise image projection. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a dynamic, fast-moving screen system based on a drone swarm and its implementation method. It utilizes multiple drones with collaborative control capabilities to form a display unit array. Each drone carries a passive micro-display screen, and a preset swarm collaboration algorithm is used to achieve spatial arrangement, forming a continuous display plane that can be moved, rotated, scaled, and deformed. The projection subsystem employs a high-brightness laser projector, supporting multiple expansions. A mechanical adjustment structure adjusts the projection angle to achieve precise image projection onto the display plane. The positioning and attitude stabilization system utilizes a multi-sensor fusion and data processing module to provide the drones with centimeter-level spatial position information and highly stable attitude angle data, ensuring system stability and display accuracy. The central control system is responsible for swarm motion planning, image processing and correction, synchronization control, and offline calibration and online calibration, ensuring the coordinated operation of the entire system. The high-speed data transmission network employs a combination of communication architectures and a built-in channel priority allocation mechanism to ensure efficient transmission of various types of data. This effectively solves the problems of resolution dependence on the number of drones and poor dynamics of the projection carrier in existing technologies, demonstrating broad application prospects and significant innovative value.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In one aspect, a dynamic, fast-moving screen system based on a drone swarm, the system comprising: Display unit array: It consists of multiple drones with collaborative control functions. Each drone is connected by a quick-release mechanism with an integrated attitude fine-tuning motor to hang a lightweight, high-reflectivity, high-gain flexible micro display screen containing an ultra-lightweight rigid / semi-rigid frame and a micro tensioning mechanism. The entire cluster is arranged in close collaboration to form a continuous display plane in the air that can be moved and deformed in a controlled manner. Projection subsystem: includes at least one high-brightness laser projector, configured to project its output image beam onto the display plane formed by the drone cluster; Positioning and attitude stabilization system: including a multi-sensor fusion module and a data processing module; using multi-sensor fusion technology, it provides each UAV with real-time centimeter-level position information and highly stable attitude angle data, and actively controls the UAV to maintain the preset shape and spatial orientation of the entire display plane; The central control system includes a swarm motion planning module, an image processing and correction module, a synchronization control module, and an offline calibration and online calibration module. The swarm motion planning module generates the flight path and formation change sequence of the UAV swarm and issues control commands. The image processing and correction module calculates the projection transformation matrix and remaps the video source signal frame by frame. The synchronization control module controls the timing of image frame output. The offline calibration and online calibration module establishes coordinate mapping relationships and fine-tunes the correction parameters. High-speed data transmission network: It adopts one or more combined architectures of cellular network, self-organizing mesh network, and point-to-point microwave communication, with a built-in channel priority allocation mechanism; the network is connected to the central control system, UAV swarm, projection subsystem, and positioning and attitude stabilization system respectively.

[0006] Furthermore, the display unit array includes multiple drones with collaborative control capabilities; each drone is connected to a passive micro-display screen via a connecting mechanism, which is a quick-release structure integrating an attitude fine-tuning motor. The output end of the attitude fine-tuning motor is connected to the edge frame of the screen to adjust the relative angle between the screen and the drone body; the micro-display screen is made of a lightweight, high-reflectivity, flexible material, and the edge of the screen is fixedly connected to an ultra-lightweight rigid or semi-rigid frame. A micro-tensioning mechanism is mounted on the frame, and the two ends of the micro-tensioning mechanism are respectively connected to the opposite sides of the frame; multiple drones are spatially arranged through a cluster collaborative algorithm to form a continuous display plane, which can be moved, rotated, scaled, and deformed through the collaborative movement of the drones.

[0007] Furthermore, the UAVs achieve spatial arrangement through a cluster collaboration algorithm. The specific steps are as follows: Each UAV first receives the target formation topology parameters issued by the central control system, including the relative coordinates of each UAV, the spacing threshold, and the arrangement sequence. It also acquires its own centimeter-level pose data output by the positioning and attitude stabilization system in real time, as well as the pose information of neighboring UAVs synchronized through a high-speed data transmission network. By adopting distributed control logic, combined with preset safety distance constraints and motion response models, the desired position and speed commands of each UAV are calculated through a consensus protocol. At the same time, neighbor collision avoidance technology is incorporated to dynamically adjust the trajectory of individual UAVs to avoid spatial conflicts. Based on the calculation results, the UAVs autonomously drive the flight control system to adjust their flight status. Through continuous pose feedback and command correction, the UAVs achieve precise positioning and relative position maintenance in three-dimensional space. Finally, they complete a compact and orderly spatial arrangement according to the preset target formation, forming a continuous display plane.

[0008] Furthermore, the projection subsystem includes at least one high-brightness laser projector. When it is necessary to expand the display area or increase the display brightness, the projection subsystem can be expanded into an array of multiple high-brightness laser projectors. The multiple projectors are arranged in a light field overlap in the physical space, and the projectors are signal connected to the central control system to receive the corrected image signal output by the central control system.

[0009] Furthermore, the projector in the projection subsystem can be deployed in at least one of the following ways: fixed ground deployment, vehicle-mounted mobile deployment, and aerial hovering deployment with the assistance of unmanned aerial vehicles.

[0010] Furthermore, the positioning and attitude stabilization system includes a multi-sensor fusion module and a data processing module. The multi-sensor fusion module integrates a real-time dynamic differential global navigation satellite system receiver and a miniature inertial measurement unit, and can additionally integrate an ultra-wideband wireless pulse positioning sensor or a computer vision positioning sensor as needed. The data processing module has a built-in data filtering algorithm, and its input end is connected to the output end of the multi-sensor fusion module through a signal line. It is used to process the raw data collected by the multi-sensor system, outputting centimeter-level spatial position information and high-stability attitude angle data for each UAV. It is also connected to the UAV flight control system and the central control system through signal interfaces to transmit the attitude data to the UAV flight control system and simultaneously feed the attitude data back to the central control system.

[0011] Furthermore, the data processing module incorporates a data filtering algorithm. It first performs timestamp synchronization and outlier removal preprocessing on the RTKGNSS and MIMU data input from the multi-sensor fusion module. It then constructs a multi-source data fusion model using extended Kalman filtering or complementary filtering algorithms. Based on the dynamic response characteristics and measurement accuracy of each sensor, it assigns adaptive weights and updates the filtering gain in real time through state equations and observation equations. This suppresses and compensates for random noise and measurement errors in the original data. Simultaneously, it combines the UAV flight kinematics model to perform state prediction and observation correction, continuously outputting continuous and stable spatial position information and highly stable attitude angle data for each UAV.

[0012] Furthermore, the swarm motion planning module incorporates an airspace path planning algorithm and a formation transformation algorithm. The airspace path planning algorithm presets airspace constraints and obstacle avoidance rules. By gridding the spatial coordinates of the preset flight area and combining the number of UAVs in the swarm with the motion performance parameters of individual UAVs, it generates an overall swarm flight path that meets spatial safety distance requirements. The path data includes continuous coordinate nodes, corresponding speed thresholds, and motion direction parameters. The formation transformation algorithm analyzes the topological structure parameters and transformation timing requirements of the preset target formation. Through interpolation, it generates a smooth transition trajectory for the swarm from the current formation to the target formation, while matching the motion response characteristics of each UAV to ensure the synchronization of the formation transformation. The module's output is connected to the input of a high-speed data transmission network via a signal line. The generated overall swarm flight path and formation transformation sequence are decomposed into independent control commands for each UAV according to a preset data format, including position control commands, attitude control commands, and speed control commands. These commands are then sent to the corresponding UAV flight control system through a high-priority channel of the high-speed data transmission network, achieving precise control of the UAV swarm's flight status and formation transformation.

[0013] Furthermore, the input of the image processing and correction module is connected to the positioning and attitude stabilization system and an external video source via signal lines. It has built-in projection transformation matrix calculation algorithm and pixel remapping algorithm. It receives the real-time 3D point cloud model of the UAV cluster in real time. The projection transformation matrix calculation algorithm first calls the preset projector intrinsic and extrinsic parameters, and then uses the preset calibration coordinates of each screen corner point in the image pixel coordinate system and the corresponding 3D world coordinates as constraints. It solves the perspective projection equation system by least squares to generate an initial projection transformation matrix. Then, combined with the real-time spatial attitude and position offset data of each screen fed back by the 3D point cloud model, the initial matrix is ​​dynamically corrected screen by screen to obtain the final projection transformation matrix that adapts to the overall shape of the current cluster display plane. At the same time, the built-in pixel remapping algorithm performs pixel-by-pixel coordinate mapping calculation on the original video frame input by external input based on the final projection transformation matrix, and establishes a precise correspondence between each pixel point in the pixel coordinate system and the 3D world coordinate system to realize image geometric correction and multi-projector screen fusion.

[0014] Furthermore, the synchronization control module incorporates a time synchronization algorithm. With a unified time reference as its core, it collects the position update timestamps of the UAV cluster, the data output cycle of the positioning and attitude stabilization system, and the refresh synchronization signal of the projector in real time through a high-speed data transmission network. Simultaneously, it records the single-frame processing time and data transmission delay within the central control system. Employing a timestamp alignment and dynamic compensation mechanism, it calculates the time deviation values ​​of each subsystem. Through linear interpolation, it corrects the timing difference between the UAV position update cycle and the projector refresh cycle, generating a precise image frame output trigger signal. This ensures that the transmission time of the corrected image frame is strictly aligned with the acquisition time of the current pose data of the UAV cluster and the projection time of the projector's light signal. Moreover, it monitors the timing fluctuations of each subsystem in real time, dynamically adjusts the synchronization trigger threshold, and controls the output timing of the corrected image frame.

[0015] Furthermore, the offline calibration and online calibration module incorporates a photogrammetry algorithm and a parameter fine-tuning algorithm. The photogrammetry algorithm controls the UAV to deploy a miniature display screen according to a preset spatial distribution pattern. It collects the standard calibration pattern projected by the projector through ground-based or UAV-mounted visual sensors, extracts the corresponding coordinates of each screen corner point in the image pixel coordinate system and the three-dimensional world coordinate system, and establishes an initial mapping relationship between the projector pixel coordinate system and the UAV screen corner point world coordinate system through multi-view geometric calculations. During the operation phase, the parameter fine-tuning algorithm receives the feature point data of the projected image collected by the visual sensors in real time, compares it with the preset standard feature data, calculates the projection distortion error and brightness uniformity deviation, and dynamically adjusts the projection transformation matrix parameters, brightness correction coefficient, and edge blending parameters through the gradient descent method. The adjusted parameters are fed back to the image processing and correction module of the central control system in real time to achieve closed-loop optimization of the projection correction parameters.

[0016] On the other hand, a method for implementing a dynamically and rapidly moving screen based on a drone swarm, characterized in that the method is applied to the system as described in any one of claims 1 to 11, the method comprising the following steps: System initialization and cluster deployment: Control the drone cluster to fly to the preset starting airspace, start the screen unfolding and tensioning mechanism to form the initial display plane; deploy the projection subsystem and adjust the light field coverage through the mechanical adjustment structure; set up the positioning base station and complete the coordinate system calibration; establish the coordinate mapping relationship between the projector pixels and the screen corner points; import the preset flight path, formation sequence and video source data; Continuous precise positioning and active stabilization: The positioning and attitude stabilization system collects the original pose data of the UAV at a preset frequency, and outputs centimeter-level pose data after data filtering algorithm processing; the pose data is transmitted to the UAV flight control system to adjust the flight state, and at the same time fed back to the central control system; Dynamic trajectory planning and real-time image rendering: The cluster motion planning module generates and issues UAV target trajectory and formation instructions at preset intervals; the image processing and correction module performs frame-by-frame correction of the video source signal based on the cluster's real-time 3D model to generate adapted image frames. Frame-synchronized projection display: The synchronization control module sends image frames to the projection subsystem at preset times, and the projector projects the image onto the display plane; if the drone's position deviation exceeds the limit or communication is interrupted, the formation is re-planned and the correction parameters are adjusted.

[0017] Compared with existing technologies, this dynamic, fast-moving screen system based on drone swarms and its implementation method have the following advantages: This invention utilizes multiple drones to collaboratively form a continuous display plane, combined with a projection subsystem to project images. The display resolution is determined by the projector performance and the drone density, effectively solving the resolution limitations of existing aerial display technologies. Secondly, in terms of display flexibility, the drones in the display unit array can move, rotate, scale, and deform the display plane through coordinated movement, breaking through the limitations of the static structure of traditional projection carriers and providing more possibilities for dynamic displays. Furthermore, the display unit array uses a passive micro-display screen, significantly reducing drone energy consumption and extending flight time. In addition, the projection subsystem supports multi-projector expansion, the high-speed data transmission network supports multi-priority channel allocation, and the central control system has offline calibration and online calibration functions, enabling the system to adapt to application scenarios of different scales. Moreover, in terms of system response speed, when the central control system detects excessive drone position deviation or communication interruption, it can quickly re-plan the formation and adjust correction parameters, ensuring the stability and continuity of the display effect and providing reliable technical support for dynamic and fast-moving aerial displays.

[0018] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0020] Figure 1 This is a structural block diagram of a dynamic, fast-moving screen system based on a drone swarm; Figure 2 This is a flowchart of a method for implementing a dynamically and rapidly moving screen based on a drone swarm. Detailed Implementation

[0021] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0022] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a dynamic, fast-moving screen system based on unmanned aerial vehicle (UAV) swarms and its implementation method, such as... Figure 1 As shown, the system comprises a display unit array, a projection subsystem, a positioning and attitude stabilization system, a central control system, and a high-speed data transmission network. The display unit array consists of multiple drones with collaborative control capabilities. Each drone mounts a passive miniature display screen, and a pre-defined swarm collaboration algorithm enables spatial arrangement, forming a continuous display plane that can move, rotate, scale, and deform. The projection subsystem uses a high-brightness laser projector, supporting multiple expansions. A mechanical adjustment structure adjusts the projection angle to achieve precise image projection onto the display plane. The positioning and attitude stabilization system utilizes a multi-sensor fusion and data processing module to provide the drones with centimeter-level spatial position information and highly stable attitude angle data, ensuring system stability and display accuracy. The central control system is responsible for swarm motion planning, image processing and correction, synchronization control, and offline and online calibration, ensuring the coordinated operation of the entire system. The high-speed data transmission network employs a combination of communication architectures and a built-in channel priority allocation mechanism to ensure efficient transmission of various data types. This effectively solves the problems of resolution dependence on the number of drones and poor dynamics of the projection carrier in existing technologies, demonstrating broad application prospects and significant innovative value. This embodiment applies the above-mentioned dynamic fast-moving screen system and its implementation method based on drone swarms, and is suitable for medium-sized outdoor performance scenarios.

[0023] like Figure 2 As shown, the system uses multiple multi-rotor UAVs with collaborative control capabilities to form a display unit array. Each UAV is equipped with a quick-release connection mechanism with an integrated attitude fine-tuning motor, and a lightweight, high-reflectivity, high-gain flexible micro-display screen is mounted on it. An ultra-lightweight semi-rigid frame is fixed to the edge of the screen, and a micro-tensioning mechanism is installed on the frame. The projection subsystem uses an array of multiple high-brightness laser projectors. The projectors are equipped with mechanical adjustment structures and are deployed using fixed ground supports around the perimeter of the outdoor venue. They are connected to the central control system via signal lines. The multi-sensor fusion module of the positioning and attitude stabilization system integrates an RTK GNSS receiver, a MIMU, and a UWB wireless pulse positioning sensor. The data processing module has a built-in extended Kalman filter algorithm. The central control system uses an industrial control host as the hardware carrier. The swarm motion planning module has built-in path planning and formation transformation algorithms. The image processing and correction module has built-in projection transformation matrix calculation and pixel remapping algorithms. The synchronization control module has built-in time synchronization algorithms. The offline calibration and online calibration module has built-in SFM-based photogrammetry algorithms and gradient descent parameter fine-tuning algorithms. The high-speed data transmission network adopts a multi-architecture combination mode, allocating priority channels according to data type to ensure reliable transmission of various types of data.

[0024] The system controls a swarm of drones to take off synchronously from a pre-defined take-off and landing area on the site, flying along a pre-defined route to the initial airspace suitable for the scenario. During takeoff, the curtain unfolding mechanism and the micro-tensioning mechanism are activated simultaneously to rapidly unfold the curtain and keep it flat. Each drone receives initial formation topology parameters from the central control system and achieves spatial arrangement through a pre-defined swarm collaboration algorithm. Using distributed control logic, combined with safety distance constraints and motion response models, the system calculates the desired position and speed commands for each drone through a consensus protocol. At the same time, it incorporates a neighboring drone collision avoidance algorithm to dynamically adjust the trajectory of individual drones. The drones combine their own pose data with the synchronization information of neighboring drones to accurately form an initial rectangular display plane, and multiple high-brightness laser projectors are deployed, with the projection angle adjusted by a mechanical adjustment structure. The projector's light field is adjusted to fully cover the pre-defined activity airspace of the drone swarm, ensuring that the overlapping areas of the light fields meet the fusion requirements. Simultaneously, positioning base stations are set up at suitable locations around the site, and the base stations are connected to the positioning and attitude stabilization system to complete coordinate system calibration. The offline calibration module of the central control system is then activated, and the drones used for calibration are deployed in the airspace according to a pre-defined distribution pattern. The standard calibration pattern projected by the projectors is collected by the visual sensor, and the corresponding coordinates of each screen corner point in the image pixel coordinate system and the three-dimensional world coordinate system are extracted using photogrammetry algorithms. The coordinate mapping relationship between the projector pixels and the screen corner points is established through multi-view geometric calculations. Subsequently, the pre-defined flight path data, formation change sequence, and high-definition performance video source data are imported into the central control system.

[0025] The multi-sensor fusion module of the positioning and attitude stabilization system synchronously collects RTKGNSS data, MIMU data, and UWB wireless pulse positioning data from each UAV at a preset frequency. The data processing module's built-in data filtering algorithm first performs timestamp synchronization and outlier removal preprocessing on the input data. Then, it uses a Kalman filter algorithm to construct a multi-source data fusion model, assigning adaptive weights based on the dynamic response characteristics and measurement accuracy of each sensor. The filtering gain is updated in real time through state equations and observation equations to suppress and compensate for random noise and measurement errors in the raw data. Simultaneously, it combines the UAV flight kinematics model for state prediction and observation correction, continuously outputting stable UAV attitude data. This data is transmitted to the UAV flight control system and the central control system via a high-speed data transmission network. The UAV flight control system uses a closed-loop control algorithm to adjust the flight state in real time based on the attitude data, resisting external environmental interference and maintaining the overall stability of the display plane.

[0026] The central control system's swarm motion planning module reads preset data at preset intervals. The path planning algorithm presets airspace constraints and obstacle avoidance rules, and generates an overall swarm flight path that meets safety distance requirements by meshing the spatial coordinates of the preset flight area. The formation transformation algorithm analyzes the topological parameters and transformation timing requirements of the preset target formation, and generates a smooth trajectory for the swarm to transition from the initial rectangle to the target wavy shape through interpolation calculation. Simultaneously, it matches the motion response characteristics of each UAV to ensure synchronization. The generated trajectory and formation parameters are then decomposed into independent control commands for each UAV and sent through a high-priority channel. The image processing and correction module receives real-time feedback from the positioning and attitude stabilization system. A real-time 3D point cloud model of a human-machine cluster is generated. This model contains the 3D world coordinates of the corner points of the effective display area of ​​each drone screen. The module first calls the pre-stored intrinsic and extrinsic parameters of the projector. With the preset calibration coordinates and corresponding 3D world coordinates of each screen corner point as constraints, the perspective projection equations are solved by the least squares method to generate an initial projection transformation matrix. Then, the initial matrix is ​​corrected by combining the real-time pose offset data of the screen to obtain the final projection transformation matrix. Subsequently, a pixel remapping algorithm is used to perform pixel-by-pixel coordinate mapping operations on the high-definition video source signal based on this matrix to establish a precise correspondence between the pixel coordinate system and the 3D world coordinate system. Color and brightness adjustments are performed simultaneously to generate a corrected image frame adapted to the current dynamic display plane.

[0027] The built-in time synchronization algorithm of the synchronization control module takes the unified time base of the system as its core. It collects the position update timestamps of the UAV cluster, the data output cycle of the positioning and attitude stabilization system, and the refresh synchronization signal of the projector in real time through a high-speed data transmission network. At the same time, it records the image processing time and data transmission delay within the central control system. It calculates the time deviation value of each subsystem using timestamp alignment and dynamic compensation mechanisms. It corrects the timing difference between the UAV position update cycle and the projector refresh cycle through linear interpolation, generates a precise image frame output trigger signal, and sends the corrected image frame to the projection subsystem according to the projector refresh cycle. After receiving the signal, the multiple projectors convert the electrical signal into an optical signal. Through digital edge blending technology, the projected images from multiple projectors are merged into a single continuous image and projected onto the dynamic display. In the planar mode, the central control system monitors the pose data of each drone at preset intervals. When a drone is detected to have an excessive positional deviation or communication abnormality, the swarm motion planning module quickly re-plans the formation of the remaining drones. Based on distributed control logic and consistency protocols, and combined with neighbor collision avoidance rules, the system adjusts the target position of each drone. Simultaneously, the image processing and correction module uses parameter fine-tuning algorithms to compare the feature point data of the projected image with preset standard data, calculates the distortion error, and dynamically adjusts the projection transformation matrix and brightness correction coefficient to ensure that the projected image remains continuous. During the performance, the system completes multiple formation changes and video switching according to a preset sequence. Each change completes trajectory planning, image correction, and synchronous projection through the corresponding algorithm. After the performance, the drone swarm disperses in an orderly manner and returns to the take-off and landing area for recovery.

[0028] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A dynamic, fast-moving screen system based on a drone swarm, characterized in that, The system includes: Display unit array: It consists of multiple drones with collaborative control functions. Each drone is connected by a quick-release mechanism with an integrated attitude fine-tuning motor to hang a lightweight, high-reflectivity, high-gain flexible micro display screen containing an ultra-lightweight rigid / semi-rigid frame and a micro tensioning mechanism. The entire cluster is arranged in close collaboration to form a continuous display plane in the air that can be moved and deformed in a controlled manner. Projection subsystem: includes at least one high-brightness laser projector, configured to project its output image beam onto the display plane formed by the drone cluster; Positioning and attitude stabilization system: including a multi-sensor fusion module and a data processing module; using multi-sensor fusion technology, it provides each UAV with real-time centimeter-level position information and highly stable attitude angle data, and actively controls the UAV to maintain the preset shape and spatial orientation of the entire display plane; The central control system includes a swarm motion planning module, an image processing and correction module, a synchronization control module, and an offline calibration and online calibration module. The swarm motion planning module generates the flight path and formation change sequence of the UAV swarm and issues control commands. The image processing and correction module calculates the projection transformation matrix and remaps the video source signal frame by frame. The synchronization control module controls the timing of image frame output. The offline calibration and online calibration module establishes coordinate mapping relationships and fine-tunes the correction parameters. High-speed data transmission network: It adopts one or more combined architectures of cellular network, self-organizing mesh network, and point-to-point microwave communication, with a built-in channel priority allocation mechanism; the network is connected to the central control system, UAV swarm, projection subsystem, and positioning and attitude stabilization system respectively.

2. The dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The display unit array includes multiple drones with collaborative control capabilities. Each drone is connected to a passive micro-display screen via a connecting mechanism. The connecting mechanism is a quick-release structure with an integrated attitude fine-tuning motor. The output end of the attitude fine-tuning motor is connected to the edge frame of the screen to adjust the relative angle between the screen and the drone's body. The micro-display screen is made of a lightweight, high-reflectivity, flexible material. The edge of the screen is fixedly connected to an ultra-lightweight rigid or semi-rigid frame, and a micro-tensioning mechanism is mounted on the frame. The two ends of the micro-tensioning mechanism are connected to the opposite sides of the frame, respectively. Multiple drones are spatially arranged through a cluster collaborative algorithm to form a continuous display plane. This display plane can move, rotate, scale, and deform through the collaborative movement of the drones.

3. The dynamic, fast-moving screen system based on a drone swarm according to claim 2, characterized in that, The UAVs achieve spatial arrangement through a swarm collaboration algorithm. The specific steps are as follows: Each UAV first receives the target formation topology parameters issued by the central control system, including the relative coordinates of each UAV, the spacing threshold, and the arrangement sequence. It also acquires its own centimeter-level pose data output by the positioning and attitude stabilization system in real time, as well as the pose information of neighboring UAVs synchronized through a high-speed data transmission network. By adopting distributed control logic, combined with preset safety distance constraints and motion response models, the desired position and velocity commands of each UAV are calculated through a consensus protocol. At the same time, neighbor collision avoidance technology is incorporated to dynamically adjust the trajectory of individual UAVs to avoid spatial conflicts. Based on the calculation results, the UAVs autonomously drive the flight control system to adjust their flight status. Through continuous pose feedback and command correction, the UAVs achieve precise positioning and relative position maintenance in three-dimensional space. Finally, they complete a compact and orderly spatial arrangement according to the preset target formation, forming a continuous display plane.

4. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The projection subsystem includes at least one high-brightness laser projector. When it is necessary to expand the display area or increase the display brightness, the projection subsystem can be expanded into an array of multiple high-brightness laser projectors. The multiple projectors are arranged in a light field overlap in the physical space, and the projectors are signal connected to the central control system to receive the corrected image signal output by the central control system.

5. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The projection subsystem can be deployed in at least one of the following ways: fixed ground deployment, vehicle-mounted mobile deployment, and aerial hovering deployment with the assistance of unmanned aerial vehicles.

6. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The positioning and attitude stabilization system includes a multi-sensor fusion module and a data processing module. The multi-sensor fusion module integrates a real-time dynamic differential global navigation satellite system receiver and a miniature inertial measurement unit, and can additionally integrate an ultra-wideband wireless pulse positioning sensor or a computer vision positioning sensor as needed. The data processing module has a built-in data filtering algorithm, and its input end is connected to the output end of the multi-sensor fusion module through a signal line. It is used to process the raw data collected by the multi-sensor system, outputting centimeter-level spatial position information and high-stability attitude angle data for each UAV. It is also connected to the UAV flight control system and the central control system through signal interfaces to transmit attitude data to the UAV flight control system and simultaneously feed the attitude data back to the central control system.

7. A dynamic, fast-moving screen system based on a drone swarm according to claim 6, characterized in that, The data processing module incorporates a data filtering algorithm. It first performs timestamp synchronization and outlier removal preprocessing on the RTKGNSS and MIMU data input from the multi-sensor fusion module. It then constructs a multi-source data fusion model using extended Kalman filtering or complementary filtering algorithms. Based on the dynamic response characteristics and measurement accuracy of each sensor, it assigns adaptive weights and updates the filtering gain in real time through state equations and observation equations. This suppresses and compensates for random noise and measurement errors in the original data. Simultaneously, it combines the UAV flight kinematics model to perform state prediction and observation correction, continuously outputting continuous and stable spatial position information and highly stable attitude angle data for each UAV.

8. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The swarm motion planning module incorporates an airspace path planning algorithm and a formation transformation algorithm. The airspace path planning algorithm pre-sets airspace constraints and obstacle avoidance rules. By gridding the spatial coordinates of the preset flight area and combining the number of UAVs in the swarm with the motion performance parameters of individual UAVs, it generates an overall swarm flight path that meets spatial safety distance requirements. The path data includes continuous coordinate nodes, corresponding speed thresholds, and motion direction parameters. The formation transformation algorithm analyzes the topological structure parameters and transformation timing requirements of the preset target formation. Through interpolation, it generates a smooth transition trajectory for the swarm from the current formation to the target formation, while matching the motion response characteristics of each UAV to ensure the synchronization of the formation transformation. The module's output is connected to the input of a high-speed data transmission network via a signal line. The generated overall swarm flight path and formation transformation sequence are decomposed into independent control commands for each UAV according to a preset data format, including position control commands, attitude control commands, and speed control commands. These commands are then sent to the corresponding UAV flight control system through a high-priority channel of the high-speed data transmission network, achieving precise control of the UAV swarm's flight status and formation transformation.

9. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The input terminals of the image processing and correction module are connected to the positioning and attitude stabilization system and an external video source via signal lines. It incorporates a projection transformation matrix calculation algorithm and a pixel remapping algorithm. It receives real-time 3D point cloud models of the UAV cluster. The projection transformation matrix calculation algorithm first calls preset projector intrinsic and extrinsic parameters, then uses the preset calibration coordinates of each screen corner point in the image pixel coordinate system and the corresponding 3D world coordinates as constraints. It solves the perspective projection equations using the least squares method to generate an initial projection transformation matrix. Subsequently, it combines the real-time spatial attitude and position offset data of each screen fed back from the 3D point cloud model to dynamically correct the initial matrix screen by screen, obtaining a final projection transformation matrix that adapts to the overall shape of the current cluster display plane. Simultaneously, the built-in pixel remapping algorithm, based on this final projection transformation matrix, performs pixel-by-pixel coordinate mapping calculations on the externally input original video frames, establishing a precise correspondence between each pixel point in the pixel coordinate system and the 3D world coordinate system, thus achieving image geometric correction and multi-projector image fusion.

10. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The synchronization control module incorporates a time synchronization algorithm. With a unified time reference as its core, it collects the position update timestamps of the UAV cluster, the data output cycle of the positioning and attitude stabilization system, and the refresh synchronization signal of the projector in real time through a high-speed data transmission network. Simultaneously, it records the single-frame consumption time of image processing and data transmission delay within the central control system. Employing a timestamp alignment and dynamic compensation mechanism, it calculates the time deviation values ​​of each subsystem. Through linear interpolation, it corrects the timing difference between the UAV position update cycle and the projector refresh cycle, generating a precise image frame output trigger signal. This ensures that the transmission time of the corrected image frame is strictly aligned with the acquisition time of the current pose data of the UAV cluster and the projection time of the projector's light signal. Furthermore, it monitors the timing fluctuations of each subsystem in real time, dynamically adjusts the synchronization trigger threshold, and controls the output timing of the corrected image frame.

11. A dynamic, fast-moving screen system based on a drone swarm according to claim 1, characterized in that, The offline calibration and online calibration module incorporates a photogrammetry algorithm and a parameter fine-tuning algorithm. The photogrammetry algorithm controls the UAV to unfold a miniature display screen according to a preset spatial distribution pattern. It collects the standard calibration pattern projected by the projector through the visual sensors on the ground or on the UAV, extracts the corresponding coordinates of each corner point of the screen in the image pixel coordinate system and the three-dimensional world coordinate system, and establishes the initial mapping relationship between the projector pixel coordinate system and the UAV screen corner point world coordinate system through multi-view geometric calculation. During the operation phase, the parameter fine-tuning algorithm receives feature point data of the projected image collected by the visual sensor in real time, compares it with preset standard feature data, calculates the projection distortion error and brightness uniformity deviation, and dynamically adjusts the projection transformation matrix parameters, brightness correction coefficient and edge blending parameters through the gradient descent method. The adjusted parameters are fed back to the image processing and correction module of the central control system in real time to realize closed-loop optimization of projection correction parameters.

12. A method for implementing a dynamically and rapidly moving screen based on a drone swarm, characterized in that, This method is applied to the system as described in any one of claims 1 to 11, the method comprising the following steps: System initialization and cluster deployment: Control the drone cluster to fly to the preset starting airspace, start the screen unfolding and tensioning mechanism to form the initial display plane; deploy the projection subsystem and adjust the light field coverage through the mechanical adjustment structure; set up the positioning base station and complete the coordinate system calibration; establish the coordinate mapping relationship between the projector pixels and the screen corner points; import the preset flight path, formation sequence and video source data; Continuous precise positioning and active stabilization: The positioning and attitude stabilization system collects the original pose data of the UAV at a preset frequency, and outputs centimeter-level pose data after data filtering algorithm processing; the pose data is transmitted to the UAV flight control system to adjust the flight state, and at the same time fed back to the central control system; Dynamic trajectory planning and real-time image rendering: The cluster motion planning module generates and issues UAV target trajectory and formation instructions at preset intervals; the image processing and correction module performs frame-by-frame correction of the video source signal based on the cluster's real-time 3D model to generate adapted image frames. Frame-synchronized projection display: The synchronization control module sends image frames to the projection subsystem at preset times, and the projector projects the image onto the display plane; if the drone's position deviation exceeds the limit or communication is interrupted, the formation is re-planned and the correction parameters are adjusted.

Images