Integrated and adaptive general unmanned aerial vehicle cluster flight innovative development platform and implementation method thereof

Abstract

The invention discloses an integrated and self-adaptive general unmanned aerial vehicle cluster flight innovative development platform and an implementation method thereof. The invention discloses an integrated and adaptive general unmanned aerial vehicle cluster flight virtual-real combination platform, and aims to provide an efficient and comprehensive verification tool for an unmanned aerial vehicle cluster behavior algorithm. According to the platform, through combination of a prototype simulation system and an unmanned aerial vehicle cluster verification system, multi-level and multi-means algorithm verification from rapid prototype simulation to advanced simulation, physical experiment verification, virtual-real combination verification and the like is supported. The platform adopts a modular design, can be flexibly expanded according to research requirements, and supports cluster behavior research of various scales and complexities. The method comprises the following steps of: performing uniform MATLABamp; in a Simulink programming environment, a user can deploy and debug an algorithm in the same environment, and the complexity of cross-platform deployment is simplified. The platform also performs efficient communication with external systems through a standardized data interface, and supports parallel cooperation of a plurality of external systems. The design of the platform not only improves the verification efficiency and reduces the experiment cost, but also promotes the rapid development of the unmanned aerial vehicle cluster technology in practical application.

Description

Technical Field

[0001] This patent belongs to the field of UAV swarm control and simulation verification platforms, and relates to an integrated and adaptive virtual-real combined platform for verifying bionic swarm behavior algorithms and improving the R & D efficiency and verification accuracy of UAV swarm systems. Background Art

[0002] Swarm behavior is a common phenomenon in nature, manifested as multiple individuals forming coordinated collective behavior through interaction. Many organisms such as bird flocks, fish schools, insects, and mammals achieve efficient cooperation through self-organization, showing complex swarm behavior patterns. These phenomena reveal that simple local interactions can generate highly complex overall behaviors, which have important reference significance in UAV swarms and robotics. In multi-robot systems, swarm robots have become a research hotspot due to their efficient cooperation, strong robustness, and flexibility. UAV swarm technology, as part of swarm robot technology, shows broad application potential in fields such as disaster rescue, logistics transportation, and agricultural plant protection due to its aerial manipulation ability. UAV swarms can work together through self-organizing networks and complete complex tasks without central control.

[0003] However, in practical applications, UAV swarms face many challenges, especially in task execution in complex environments. For example, in unknown or dynamically changing environments such as canyons and forests, how to ensure the efficient crossing of the swarm becomes a major problem. Although swarm theory has the potential for large-scale task execution, limited by the perception ability of individual UAVs and limited interaction ability, how to achieve efficient swarm crossing under these conditions remains an unsolved problem.

[0004] In addition, it is also difficult to verify the effectiveness of swarm behavior algorithms. The current verification platforms have relatively single functions and are difficult to support the multi-level requirements of simulation, physical experiments, and virtual-real combined experiments. To promote the development of swarm robot technology, an integrated verification platform needs to be developed to support multiple verification methods, reduce experimental costs and operation complexity, and promote the transformation of theoretical research into practical applications. Summary of the Invention

[0005] The object of the present invention is to design and construct an integrated and adaptive general UAV swarm flight virtual-real combined platform for verifying bionic swarm behavior algorithms. With the rapid development of UAV swarms and swarm behavior algorithms, the existing verification platforms have relatively single functions and cannot meet various verification needs. This platform aims to provide multi-level verification means through a unified programming environment, such as rapid prototyping simulation, advanced simulation verification, physical experiment verification, and virtual-real combined verification, to help researchers more efficiently verify and optimize swarm behavior algorithms.

[0006] The technical solution of the present invention aims to solve the problems of single function and lack of flexibility in existing verification tools by designing an integrated and adaptive general virtual-real combined platform for the flight of UAV clusters. The platform adopts a modular design and can provide multi-level and comprehensive verification means in a unified programming environment to meet the verification requirements at different research stages and needs. The following is the detailed technical solution of the platform: Platform architecture design. This platform consists of two core components: a prototype simulation system and a UAV cluster verification system. These two systems work together to provide comprehensive support from theoretical simulation to physical verification.

[0007] Prototype simulation system: The core functions of this system include rapid prototype simulation, automatic parameter tuning, and parameter batch processing. By inputting the prototype of the cluster behavior algorithm, users can quickly verify the algorithm, adjust parameters, and evaluate performance in a virtual environment. The prototype simulation system not only supports manual adjustment of simulation parameters but also can automatically optimize parameters and output the optimized results, significantly improving the verification efficiency.

[0008] UAV cluster verification system: This system includes basic simulation, advanced simulation verification, physical experiment verification, and virtual-real combined verification. It combines the UAV control system in the real environment and the virtual simulation environment to comprehensively verify the cluster algorithm. Through the coordination of the ground control station and the communication relay module, the platform can effectively connect the virtual simulation with the actual hardware system to achieve a highly consistent verification process.

[0009] Function module design. Rapid prototype simulation: This function allows users to conduct experiments by adjusting input parameters, view the visual results of cluster behavior in real time, and observe the performance of the algorithm under different conditions. Users can intuitively understand the motion state and behavior evolution of the cluster.

[0010] Automatic parameter tuning function: This function uses optimization algorithms (such as genetic algorithms, particle swarm algorithms) to automatically adjust the parameters of the cluster algorithm. The system quickly searches for the best parameter combination according to the user-defined goals and constraints, thereby improving the performance of the cluster algorithm.

[0011] Parameter batch processing function: It supports users to batch input multiple parameter combinations. The system will automatically conduct multiple simulations, generate the cluster behavior performance under each parameter combination, and finally output the performance evaluation results. This function is particularly suitable for large-scale experiments and can significantly reduce manual intervention and optimization time.

[0012] Advanced simulation verification: Conduct simulations in a more complex three-dimensional environment, combine with the dynamic model of UAVs to verify the effectiveness of the algorithm. This verification can consider various physical factors, such as aerodynamics, inertial effects, wind speed changes, etc., to ensure the feasibility of the algorithm in the real environment.

[0013] Physical experiment verification: By interacting with real drones and environmental sensors, the performance of the algorithm is directly verified in the physical world, providing the most realistic performance feedback. This verification function supports connection with various hardware platforms (such as Tello EDU drones) for actual flight tests.

[0014] Verification combining virtual and real: By combining the virtual environment with the actual system for joint verification. During this process, users can simulate the behavior of the drone swarm in the virtual environment while testing the execution effect of the algorithm in the physical environment to verify the accuracy and reliability of the swarm behavior.

[0015] Programming environment and integration. To simplify user operations and lower the usage threshold, this platform is developed based on the MATLAB & Simulink programming environment, unifying the programming language and development tools. Users only need to deploy the swarm behavior algorithm or model in the MATLAB environment without switching between different programming languages and platforms. This platform utilizes the powerful numerical calculation and visualization capabilities of MATLAB to efficiently execute swarm behavior simulations and provides an intuitive operation interface for users.

[0016] User interface design: The platform designs an intuitive graphical user interface (GUI), enabling users to complete algorithm configuration, simulation settings, and verification tasks through simple click and drag operations. The graphical interface provides rich function options, allowing users to flexibly adjust simulation parameters, observe simulation results, and quickly obtain performance evaluation data.

[0017] Modular design: Each functional module of the platform adopts a hierarchical and modular design, ensuring the independence and scalability between different modules. For example, users can add new simulation modules or verification functions as needed without affecting other parts of the platform. This design method enables the platform to quickly adapt to research tasks of different scales and complexities.

[0018] System communication and verification extension. Ground control station and communication forwarding module: The ground control station communicates with external systems (such as drones, simulators, sensors, etc.) through the communication forwarding module. The communication protocol uses UDP, ensuring low latency and high real-time performance, which is suitable for real-time applications such as drone swarm control. The communication forwarding module can convert data protocols according to the requirements of external systems to achieve compatibility with various hardware platforms.

[0019] Interface and data management: The platform exchanges data with external systems through standardized data interfaces to ensure the efficient transmission and processing of data. In addition, all simulation and verification results are uniformly managed, enabling users to conveniently view, analyze, and export relevant data.

[0020] The seamless integration of virtual and physical environments. Through the seamless connection between virtual simulation and physical testing, the platform enhances the depth and breadth of swarm behavior research. Users can conduct preliminary verification in a virtual environment within a single platform and then further verify the actual effects of algorithms through physical experiments. This method of combining virtual and physical verification not only improves the efficiency of algorithm verification but also enhances the reliability of research results.

[0021] In summary, the present invention provides a comprehensive UAV swarm flight behavior verification tool through an integrated and adaptive platform architecture, which can efficiently support the development, optimization, and application promotion of swarm algorithms. Brief Description of the Drawings

[0022] Figure 1 is the overall framework diagram of the general integrated verification platform of the present invention; Figure 2 is the overall framework of the prototype simulation system of the present invention; Figure 3 is the internal framework of the upper module of the present invention; Figure 4 is the overall framework of the UAV swarm verification system of the present invention; Figure 5 is the communication block diagram of the UAV swarm verification system of the present invention. Detailed Description of the Invention

[0023] The following further elaborates on the present invention in detail with reference to the drawings and specific embodiments. For the step numbers in the following embodiments, they are only set for the convenience of elaboration and explanation, and no limitation is imposed on the order between steps. The execution order of each step in the embodiments can be adaptively adjusted according to the understanding of those skilled in the art.

[0024] S1. Overall design and architecture of the platform. The integrated and adaptive general UAV swarm flight virtual-real combination platform of the present invention consists of two core parts: a prototype simulation system and a UAV swarm verification system. These two parts work together to provide a complete UAV swarm behavior algorithm verification environment. The overall framework diagram of the general integrated verification platform is as Figure 1 shown.

[0025] The purpose of this platform is to provide a comprehensive and efficient verification environment for the research of swarm behavior algorithms, promoting the optimization and practical application of algorithms. Its core design concepts include comprehensiveness, modularity, scalability, user-friendliness, and efficiency, ensuring that the platform can meet the needs of different research stages and accelerate the development of UAV swarm technology.

[0026] First, comprehensiveness is a key feature of the platform, covering various functions such as rapid prototyping simulation, automatic parameter tuning, parameter batch processing, advanced simulation, physical experiment verification, and virtual-real combination verification. Researchers can complete various experiments and verifications on the same platform without frequently switching tools and environments, saving time and effort. The modular design enables independent development and maintenance of each component of the platform, facilitating quick response to technological progress and new functional requirements. Scalability means that the platform can adapt to cluster research of different scales and complexities, supporting the adjustment and expansion of small to large systems, ensuring good application adaptability. User-friendliness is reflected in simplifying the operation process and providing an intuitive interface, enabling even novices to easily get started and lowering the technical threshold. Efficiency is achieved by optimizing the verification process and quickly providing feedback results to help researchers timely adjust algorithms.

[0027] In the platform design, the functions of rapid prototyping simulation, automatic parameter tuning, and parameter batch processing are all centered around the simulation process, helping users observe the changes in cluster behavior by adjusting parameters, automatically optimizing, or batch simulating. They share a common underlying framework to improve numerical calculation efficiency and enhance overall performance. Advanced simulation, physical verification, and virtual-real combination verification focus on real-time communication and control with actual UAV systems to ensure the effectiveness of algorithms in practical applications. These functions are achieved by integrating existing advanced simulators and hardware devices and ensuring good scalability through standardized interfaces.

[0028] The overall architecture of the platform consists of a prototype simulation system and a UAV cluster verification system. Users can quickly conduct algorithm verification and optimization in the prototype simulator and perform intuitive operations through the graphical interface. The UAV cluster verification system includes basic simulation verification, advanced simulation verification, physical experiment verification, and virtual-real combination verification to ensure the reliability and performance of algorithms in different environments. The platform enables effective interaction between the ground control station and different external systems through a standardized communication protocol, further enhancing the breadth and depth of verification.

[0029] To improve the user experience, the platform is developed based on MATLAB&Simulink, using the M language as a unified programming environment, enabling efficient deployment of models and algorithms for cluster behavior. The M language not only supports powerful matrix operations and optimization algorithms but also allows graphical modeling through Simulink, simplifying the design and control of complex systems. The App Designer component provides a user-friendly interface, further enhancing the intuitiveness and convenience of operation.

[0030] S2. Implementation of the prototype simulation system. The prototype simulation system is a core tool in the development of swarm robotics technology, aiming to verify the effectiveness of swarm algorithms and model prototypes. Through modular design, the system has good scalability and flexibility, enabling users to adjust parameters and expand functions according to their needs. The main goal of the system is to achieve the dynamic display of swarm movement and provide optimized swarm parameters and performance evaluation. The overall framework diagram of the prototype simulation system is as shown in Figure 2 Figure

[0031] The prototype simulation system adopts a hierarchical modular design to ensure that each functional module is independent and highly collaborative. The core modules of the system include: Simulation module: Responsible for executing swarm algorithms in real time and generating simulation results.

[0032] Automatic parameter tuning module: Optimizes the calculation of swarm parameters according to user needs.

[0033] Parameter batch processing module: Processes the performance evaluation of different parameter combinations and supports large-scale simulations.

[0034] Each module has an independent function to ensure efficient code reuse and maintenance. For example, the swarm simulation module and the evaluation module are responsible for simulating swarm behavior and performance evaluation respectively, while the motion module simulates specific motion behaviors through kinematic modeling of individuals. To enhance the user experience, the system is also equipped with an intuitive graphical user interface (GUI), enabling users to conveniently set simulation parameters, view real-time results, and conduct data analysis.

[0035] User interface design. To improve the convenience of operation, the system designs a graphical user interface (GUI). Users can intuitively view and modify simulation parameters through the GUI and observe swarm behavior in real time. For example, the real-time visualization window allows users to track the swarm movement trajectory during the simulation and make adjustments as needed. In addition, the GUI supports importing and editing simulation parameter configuration files, facilitating users to preset parameters and quickly start simulations.

[0036] System underlying modules. The underlying modules of the system include multiple sub-modules, which are responsible for motion simulation, noise interference, map construction, sensor simulation, swarm control, and evaluation respectively. The following is the detailed design of each module: Motion module: Describes the motion behavior of individuals through the point mass model and the quadrotor UAV model. The motion module calculates the states of individuals such as speed and position, and adjusts the motion direction and speed through control signals.

[0037] Noise module: This module simulates uncertain factors in the environment by introducing noise to ensure more realistic simulation results. The noise model allows users to customize the noise type and standard deviation, increasing the complexity of the simulation environment.

[0038] Map Module: Responsible for the construction and simulation of a three-dimensional obstacle environment, supporting the use of geometric maps and elevation maps. The map module defines obstacles in STL format and allows users to modify them dynamically.

[0039] Sensor Module: Simulates sensors such as lidar to provide perception information. This module helps achieve environmental perception and navigation by calculating the distance between obstacles and individuals.

[0040] Cluster Module: Users can deploy different clustering algorithms to control the collaborative behavior of multiple agents. The module supports both centralized and distributed algorithms to ensure coordinated actions among multiple agents.

[0041] Evaluation Module: Conducts real-time evaluation of cluster behavior, supporting multiple performance metrics such as average speed and position error. The evaluation results can be used as a basis for optimization decisions.

[0042] Visualization Module: Implements the display of cluster behavior, obstacle environment, and performance changes through the visualization toolbox of the M language, supporting the drawing of dynamic trajectories and the display of performance metric charts.

[0043] Upper Layer Module. Based on the underlying modules, the prototype simulation system designs an upper layer module for easy user operation and analysis. The internal framework of the upper layer module is as Figure 3 shown. It mainly includes: Quick Prototype Module: Provides users with a simple function to adjust parameters and displays cluster behavior through visualization results. The simulation data will be saved for subsequent analysis.

[0044] Automatic Parameter Tuning Module: Automatically optimizes the parameters in cluster simulation. Users can define multiple parameters, and the optimization algorithm (such as genetic algorithm, particle swarm algorithm) automatically searches for the optimal parameters.

[0045] Batch Processing Module: Allows users to input multiple parameter combinations and perform batch simulations. The module supports multi-core parallel computing, significantly improving the computing efficiency.

[0046] Graphical User Interface Design and Implementation. The graphical user interface design plays an important role in the cluster simulation system, aiming to provide an intuitive and efficient way of interactive operation. The design includes two key parts: Obstacle Environment Editing Interface: Allows users to interactively add and modify obstacles in three-dimensional space. Users can set attributes such as the shape, size, and position of obstacles and view the modified effects in real time. This interface combines editing tools and visual display, simplifying the creation and editing of complex environments.

[0047] Cluster behavior simulation interface: Provides real-time display of cluster behavior, supports parameter configuration, result display, and data analysis. Users can control the simulation process through this interface to obtain the dynamic changes and performance evaluation of cluster movement.

[0048] Parameter configuration and management. The system conducts unified management through parameter configuration files in XML format. The parameters of each module are defined in the form of name-value pairs, supporting hierarchical structures and references, enabling different modules to share parameters without destroying the modular structure. The configuration file also supports function calls and command inputs, further enhancing the flexibility and customizability of the system.

[0049] S3. Implementation of the UAV cluster verification system. The UAV cluster verification system of the present invention, as the core component of the cluster behavior integrated verification platform, aims to provide comprehensive and practical UAV cluster behavior verification. The system covers physical experiment verification, advanced simulation verification, and virtual-real combination verification functions. To achieve this goal, the design of the system fully considers the interaction requirements with actual UAVs and their control systems, and ensures high compatibility and collaborative working ability with the prototype simulation system. The overall framework of the UAV cluster verification system is as Figure 4 shown.

[0050] Specifically, the overall architecture of the UAV cluster verification system is divided into three major parts: UAV cluster ground control station (ground station): Responsible for coordinating the overall behavior of the cluster UAVs, issuing control instructions, and receiving the status information of the UAVs.

[0051] Communication forwarding module: Used to ensure effective communication between the ground station and external systems, forward data from the ground station to external systems, and receive the data feedback from external systems.

[0052] External systems: Include physical systems (such as UAVs and sensors) and simulation systems (such as advanced simulators, basic simulators).

[0053] The ground station is the core of the system, supporting communication and collaborative work with different external systems. The ground station shares modules and parameter settings with the prototype simulation system to ensure high consistency and module reusability during the process from the simulation stage to the verification stage. The ground station includes the following six functional modules: Visualization module: Inherited from the prototype simulator, used to display simulation results in real time, such as cluster behavior, UAV movement trajectories, etc.

[0054] Map update module: Inherited from the prototype simulator, used to update the information of dynamic obstacles and transmit it to the cluster control system.

[0055] Instruction module: Generates environmental perception data based on the cluster module and sensor module and sends it to the UAV cluster.

[0056] Communication module: Responsible for data transmission between the external system and the ground station, supporting the sending of control commands and the reception of status information.

[0057] Real-time module: Ensures the time scheduling ability of the ground station and guarantees that each module operates under synchronized time.

[0058] Knob module: Provides a user interface, enabling users to control tasks such as unlocking, taking off, hovering, and landing of the UAV cluster through the knob.

[0059] To meet diverse verification requirements, the ground station needs to be able to establish communications with multiple external systems. Due to the differences in communication protocols and interfaces between external systems, the ground station solves this problem through a communication forwarding module, which communicates with external systems using standard protocols. When designing the communication forwarding module, independent communication frameworks and interfaces are adopted according to the requirements of different external systems. Through this design, the ground station can flexibly cooperate with different external systems to achieve joint verification of multiple systems, while ensuring the standardization of communication protocols and supporting the simultaneous cooperation of multiple external systems.

[0060] Detailed design and implementation of the UAV cluster verification system. To achieve a tight integration between the UAV cluster ground control station and the prototype simulation system, the ground station is developed based on the same programming environment (MATLAB & Simulink). All verification function modules are implemented using the Simulink modeling tool, and the main functions of the visualization module, map update module, and instruction module in the prototype simulator are called respectively through the MATLAB Function module. The design of the ground station takes into account the requirements of interactivity and real-time performance, ensuring the stability and efficiency of the system during operation. In particular, the real-time module adjusts the difference between the simulation time and the actual time through the Simulation Pace module of Simulink to ensure the synchronization and real-time performance of the simulation process.

[0061] The communication module is a key part of the UAV cluster verification system. Its main task is to ensure the effective transmission of data between the ground station, the communication forwarding module, and the external system. The communication block diagram of the UAV cluster verification system is as Figure 5 shown. For this reason, the User Datagram Protocol (UDP) is selected as the communication method between components in this system. The UDP protocol has the following advantages: Low latency and high real-time performance: No connection establishment is required, reducing the state maintenance during communication, which is especially suitable for real-time applications.

[0062] Simple and efficient: The UDP protocol is relatively lightweight and suitable for resource-constrained embedded systems or high-density device deployment scenarios.

[0063] Packet Tolerance: UDP has good tolerance for packet loss, which is particularly important in the real-time data transmission of UAV swarms.

[0064] To ensure that multiple external systems can work in parallel, the ground station assigns independent sockets and port numbers to each UAV. Through this solution, multiple communication channels can run in parallel between the ground station and each external system, improving the efficiency of data transmission.

[0065] The design of communication data packets is crucial for the reliable transmission of data. Each data packet contains the following fields: Start Bit: Constantly 0xFA, marking the start of the data packet.

[0066] Payload Length: Related to the data type of the payload within the data packet.

[0067] Sequence Number, Number, Payload Data: Identifying the uniqueness and specific data of the data packet.

[0068] Timestamp, Check Bit: Ensuring the validity of the data and performing data integrity checks.

[0069] The standardized design of the data packet ensures that different types of control instructions can adapt to various application scenarios and facilitates future expansion.

[0070] This invention uses the Tello EDU UAV and the OptiTrack motion capture system as physical systems for verification. The Tello EDU UAV receives control instructions through the UDP protocol, and the control instructions include yaw rate, pitch angle, roll angle, and speed in the height direction, etc. Since the Tello UAV does not have precise positioning function, the OptiTrack motion capture system is used to obtain the position information and attitude of the UAV.

[0071] The communication forwarding module is written in C / C++ language, processes instructions from the ground station and converts them into low-level instructions suitable for UAV control. A cascade PID controller is used to ensure the precise control and stability of the UAV.

[0072] Advanced Simulation and Virtual-Reality Verification. This system uses the RFlySim platform for advanced simulation, uses the CopterSim software to simulate the dynamics of a quadrotor UAV, and constructs a realistic three-dimensional environment through UE4. Since the RFlySim platform provides upper-layer control interfaces and status receiving interfaces, there is no need to build an additional controller, and the system can perform advanced simulation more efficiently.

[0073] In addition, this system also supports virtual-real combination verification. The basic emulator collaborates with the physical system simultaneously through the MATLAB Function module of Simulink to ensure that the virtual and real UAV swarms can work synchronously. By configuring different IP addresses and port numbers, the ground station and the communication forwarding module can achieve simultaneous control of virtual UAVs and physical UAVs, providing more realistic verification of swarm behavior.

Claims

1. An integrated and adaptive general virtual-real combined platform for unmanned aerial vehicle (UAV) swarm flight, characterized in that Including: A prototype simulation system for rapid prototyping simulation of swarm behavior algorithms, supporting functions such as manual parameter adjustment, automatic parameter tuning, and parameter batch processing, and displaying the evolution of swarm behavior in a real-time visualization manner; An unmanned aerial vehicle (UAV) swarm verification system, including basic simulation, advanced simulation verification, physical experiment verification, and virtual-real combination verification, for comprehensively verifying swarm behavior algorithms, supporting interaction with actual UAV control systems, and providing swarm behavior verification in virtual and real environments.

2. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, wherein The prototype simulation system includes: A rapid prototyping simulation module for real-time simulation based on parameters input by users and providing visualization results; An automatic parameter tuning module that automatically adjusts the parameters of the swarm behavior algorithm using optimization algorithms and outputs optimized parameters; A parameter batch processing module that supports users to input multiple different parameter combinations and automatically generates the swarm performance under each parameter combination.

3. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, characterized in that, The prototype simulation system is developed through the MATLAB & Simulink programming environment. Users deploy swarm behavior algorithms in the same programming environment, and all functional modules within the platform support seamless integration of this programming environment.

4. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, characterized in that, The verification system includes: A basic simulation verification module for verifying the correct operation of the UAV swarm ground control station; An advanced simulation verification module for simulating and verifying algorithms based on a three-dimensional environment and UAV dynamics models; A physical experiment verification module that works in coordination with actual UAVs and their control systems to verify the performance of swarm behavior algorithms in the physical world; A virtual-real combination verification module that combines a virtual environment with an actual system for joint verification through the coordinated operation of synchronous simulation and the physical system.

5. The UAV swarm flight virtual-real combined platform according to claim 1, characterized in that, The described UAV swarm verification system includes a ground control station and a communication forwarding module. Among them, the ground control station is responsible for coordinating the behavior of the UAV swarm, sending control instructions, and receiving status information; the communication forwarding module is used for data protocol conversion between the ground control station and external systems to ensure that the platform can communicate effectively with various external hardware systems.

6. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, wherein The ground control station shares modules and parameter settings with the prototype simulation system to ensure the coordinated consistency of modules from the simulation stage to the verification stage and achieve code reuse.

7. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, characterized in that, The platform supports low-latency data transmission based on the UDP communication protocol and can achieve parallel communication and data transmission of multiple external systems.

8. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, wherein The platform communicates with external systems through a standardized data interface to ensure efficient data transmission between systems and supports real-time processing and storage of data.

9. The virtual-real combined platform for the swarm flight of unmanned aerial vehicles according to claim 1, characterized in that Users operate the platform through a graphical user interface (GUI), intuitively set simulation parameters, observe simulation results, and edit a three-dimensional obstacle environment through a visualization interface.

10. The virtual-real combined platform for cluster flight of unmanned aerial vehicles according to claim 1, wherein The platform can support the research on the behavior of UAV swarms of multiple scales and complexities. Whether it is a small experiment or a large-scale system, it can be expanded and adjusted according to actual needs.

Images