Multi-unmanned aerial vehicle cooperative positioning method and application

Abstract

The invention discloses a multi-unmanned aerial vehicle cooperative positioning method and application, and relates to the technical field of unmanned aerial vehicle control and navigation, and the method comprises the steps: constructing a three-dimensional visual simulation platform, including dynamic environment modeling, sensor data simulation and virtual-real combination verification; identifying and tracking a target through a multi-sensor data fusion and deep learning algorithm of the three-dimensional visual simulation platform; based on a multi-unmanned aerial vehicle cooperative positioning algorithm, realizing real-time estimation of the position and state of the target through geometric modeling and interactive multi-model unscented Kalman filtering; information fusion and collaborative decision among multiple unmanned aerial vehicles are realized through a wireless communication network; and optimizing the flight path, task allocation and target identification strategy of the unmanned aerial vehicle based on simulation feedback. The problems that a traditional unmanned aerial vehicle cluster is low in positioning precision and poor in target recognition and tracking capacity in a complex environment can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) control and navigation, and particularly to a multi-UAV collaborative positioning method and application. Background Art

[0002] With the rapid development of UAV technology, especially its applications in fields such as UAV swarm operations, collaborative positioning, and environmental perception, traditional target recognition and positioning technologies can no longer meet the increasingly complex mission requirements. In urban canyons, mountainous areas, indoor environments, and other areas with weak GPS signals, traditional GPS-dependent positioning methods often cannot provide sufficient accuracy or even completely fail. Therefore, the positioning accuracy of UAVs is often limited by environmental interference, sensor accuracy, and system computing power, resulting in deviations in multi-UAV collaborative positioning.

[0003] As a virtualization technology, the three-dimensional visual simulation platform has been widely used in the flight testing and mission simulation of UAVs in recent years, and can simulate various environments and dynamic changes in flight missions. However, existing three-dimensional simulation platforms mostly focus on flight control and path planning, and the integrated application of target recognition and multi-UAV collaborative positioning is not yet perfect. Therefore, it is particularly important to develop new UAV positioning and target recognition methods.

[0004] In complex three-dimensional environments, there are great challenges in the rapid recognition, classification, and tracking of targets. Especially when multiple UAVs simultaneously perform target recognition, misrecognition or recognition delay is likely to occur. Moreover, when multiple UAVs execute tasks, they need to coordinate actions, share data, and adjust flight paths. How to ensure that multiple UAVs work together to improve efficiency and mission completion rate is still an urgent problem to be solved. Summary of the Invention

[0005] In view of the above defects or improvement requirements of the prior art, the present invention provides a multi-UAV collaborative positioning method and application, which can solve the problems of low positioning accuracy and poor target recognition and tracking ability of traditional UAV swarms in complex environments.

[0006] On the one hand, an embodiment of the present invention provides a multi-UAV collaborative positioning method, including: constructing a three-dimensional visual simulation platform, including dynamic environment modeling, sensor data simulation, and virtual-real combination verification; identifying and tracking a target through multi-sensor data fusion and deep learning algorithms of the three-dimensional visual simulation platform; based on a multi-UAV collaborative positioning algorithm, realizing real-time estimation of the position and state of the target through geometric modeling and interactive multi-model unscented Kalman filtering; realizing information fusion and collaborative decision-making among multiple UAVs through a wireless communication network; and optimizing the flight path, task allocation, and target recognition strategy of the UAVs based on simulation feedback.

[0007] In one embodiment of the present invention, the dynamic environment modeling includes: constructing a three-dimensional virtual environment through point clouds, meshes, and textures to simulate terrains, buildings, dynamic targets, and climate changes; integrating virtual sensors to simulate data feedback of cameras, LiDAR, millimeter-wave radars, and IMUs in different environments; and verifying the stability of the algorithm in the actual environment through virtual-real data interaction.

[0008] In one embodiment of the present invention, the identification and tracking of targets include: designing a lightweight classification head using multi-scale feature fusion and depthwise separable convolution; extracting features by combining target color, shape, and motion trajectory; and using the YOLO or Faster R-CNN model to achieve target classification, detection, and tracking.

[0009] In one embodiment of the present invention, the UAV cooperative positioning algorithm includes: defining the relative positions of each UAV and the target by establishing multiple coordinate systems; calculating the initial position of the target based on the least squares optimization method; and updating the position and covariance matrix of the target through IMM-UKF fusion of the multi-motion mode state estimation of the target.

[0010] In one embodiment of the present invention, calculating the initial position of the target based on the least squares optimization method includes: converting the image coordinate system to the camera coordinate system through the pinhole camera model; converting the camera coordinate system to the world coordinate system by combining the UAV attitude angle; and calculating the world coordinate position of the target by minimizing the foot point error.

[0011] In one embodiment of the present invention, the information fusion and collaborative decision-making include: implementing real-time data communication between UAVs using the MavLink protocol; fusing the positioning data of multiple UAVs through extended Kalman filtering; and dynamically adjusting the task priority and path planning to achieve decentralized collaborative control.

[0012] In one embodiment of the present invention, optimizing the flight path, task allocation, and target recognition strategy of UAVs based on simulation feedback includes: generating a dynamic obstacle avoidance path based on the A* or RRT algorithm; predicting the target trajectory through particle filtering and real-time correcting the flight path; and performing adaptive parameter adjustment by combining the task execution efficiency and positioning error.

[0013] On the other hand, an embodiment of the present invention provides a multi-UAV collaborative positioning device, including: a simulation platform construction module for constructing a three-dimensional visual simulation platform, including dynamic environment modeling, sensor data simulation, and virtual-real combination verification; a target recognition and tracking module for recognizing and tracking a target through multi-sensor data fusion and deep learning algorithms of the three-dimensional visual simulation platform; a position and state estimation module for realizing real-time estimation of the position and state of the target through geometric modeling and interactive multi-model unscented Kalman filtering based on a multi-UAV collaborative positioning algorithm; an information fusion and collaborative decision-making module for realizing information fusion and collaborative decision-making among multiple UAVs through a wireless communication network; and a strategy optimization module for optimizing the flight path, task allocation, and target recognition strategy of the UAVs based on simulation feedback.

[0014] On yet another aspect, an embodiment of the present invention provides an electronic device, including: a memory and one or more processors connected to the memory, where the memory stores a computer program, and the processors are configured to execute the computer program to implement the multi-UAV collaborative positioning method according to any one of the above embodiments.

[0015] On still another aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer-executable instructions for executing the multi-UAV collaborative positioning method according to any one of the above embodiments.

[0016] As can be seen from the above, compared with the prior art, the above solution conceived by the present invention may have one or more of the following beneficial effects:

[0017] The multi-UAV collaborative positioning method proposed by the embodiment of the present invention provides a realistic simulation environment for the flight mission based on the establishment of a three-dimensional visual simulation platform, which helps to verify the execution effect of the target recognition and positioning algorithm and reduce the risk in actual operation; through deep learning and multi-sensor data fusion, the accuracy and robustness of target recognition in complex environments are significantly improved. Especially in dynamic environments, low light, or obstacle occlusion conditions, efficient target recognition can be ensured; the decentralized collaborative control algorithm is used to solve the communication and path planning problems among multiple UAVs, ensuring that the UAV group can cooperate efficiently and complete difficult tasks in complex environments; through dynamically adjusting the simulation environment and flight mission, accurate target recognition and positioning can be carried out under different environmental conditions, providing technical support for UAV applications in complex scenarios; the real-time evaluation and feedback mechanism during task execution can continuously optimize the target recognition and positioning algorithm to ensure the efficient execution of the algorithm in different tasks and environments.

[0018] The features of other aspects of the present invention become apparent from the following detailed description with reference to the accompanying drawings. However, it should be understood that the drawings are only designed for the purpose of explanation and not as a limitation of the scope of the present invention. It should also be understood that, unless otherwise indicated, the drawings are not necessarily drawn to scale and they only attempt to conceptually illustrate the structures and processes described herein. Description of the Drawings

[0019] The drawings described herein are used to provide a further understanding of the present invention and form a part of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 is a flowchart of a multi-UAV collaborative positioning method provided by an embodiment of the present invention;

[0021] Figure 2 is a schematic diagram of the specific execution steps of a multi-UAV collaborative positioning method provided by an embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of a typical application scenario of a multi-UAV collaborative positioning method provided by an embodiment of the present invention;

[0023] Figure 4 is a schematic diagram of the structure of a multi-UAV collaborative positioning device provided by an embodiment of the present invention;

[0024] Figure 5 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;

[0025] Figure 6 is a schematic diagram of the structure of a computer-readable storage medium provided by an embodiment of the present invention. Detailed Embodiments

[0026] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described below with reference to the drawings and in conjunction with the embodiments.

[0027] In order to enable those of ordinary skill in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments, and all should fall within the protection scope of the present invention.

[0028] It should be noted that the terms "first", "second", etc. in the description, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such terms can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0029] It should also be noted that the division of multiple embodiments in the present invention is only for convenience of description and should not constitute a special limitation. The features in various embodiments can be combined and cross-referenced without conflict.

[0030]

First Embodiment

[0031] As Figure 1 shown, the first embodiment of the present invention proposes a multi-UAV collaborative positioning method, including the following steps: Step S1, constructing a three-dimensional visual simulation platform, including dynamic environment modeling, sensor data simulation and virtual-real combination verification; Step S2, identifying and tracking a target through multi-sensor data fusion and deep learning algorithm of the three-dimensional visual simulation platform; Step S3, based on the multi-UAV collaborative positioning algorithm, realizing real-time estimation of the position and state of the target through geometric modeling and interactive multi-model unscented Kalman filtering; Step S4, realizing information fusion and collaborative decision-making among multiple UAVs through a wireless communication network; Step S5, optimizing the flight path, task allocation and target recognition strategy of the UAV based on simulation feedback.

[0032] Combined with Figure 2 shown, in Step S1, a highly simulated three-dimensional visual platform is constructed. For example, by accurately modeling environmental factors such as terrain, buildings, dynamic targets, obstacles, and climate change, a flight scene in the real world is simulated. Specifically, the core of the simulation platform describes various elements in the physical world through data structures (such as point clouds, meshes, textures, etc.), and combines the sensor data with real-time feedback to generate a simulation environment highly consistent with the actual environment.

[0033] Furthermore, the simulation platform supports dynamic environment simulation, including climate change, relative movement of obstacles, dynamic changes of targets, etc. Factors such as wind speed, air flow, temperature and humidity in the environment can also be accurately simulated, providing multi-dimensional support for flight control and target recognition. Moreover, various sensors (such as cameras, LIDAR, millimeter-wave radars, IMUs, etc.) are integrated, and their working states and data feedback under different environmental conditions are simulated. For example, in low-light or high-reflection environments, the working effects of cameras and LIDAR are simulated to ensure the robustness and accuracy of the recognition algorithm.

[0034] Furthermore, through the data interaction between the constructed three-dimensional visual simulation platform and the actual UAV system, the combination of the virtual environment and the real environment can be realized, ensuring that the algorithm can be executed stably and accurately in the real world. The simulation platform can not only support the process of target recognition and positioning, but also simulate the execution of complex tasks (such as patrol, search and rescue, reconnaissance, etc.), so as to verify the performance of the UAV under various tasks.

[0035] In step S2, for example, based on image processing and computer vision technologies, target recognition is carried out by combining the virtual environment data provided by the three-dimensional visual simulation platform. Specifically, the camera or sensor carried by the UAV is used to collect real-time images, and target classification, target detection and target tracking are carried out through deep learning algorithms (such as convolutional neural network CNN) and target detection algorithms. During the recognition process, multi-scale feature fusion is adopted to enable the model to detect target objects of different sizes at the same time. A depthwise separable convolution is used to design a lightweight classification head, which reduces the computational overhead while improving the classification accuracy. The classification head can efficiently predict the target category, ensuring the recognition ability for complex targets, and target screening is carried out by combining environmental information. By extracting target features (such as color, shape, motion trajectory, etc.), the accuracy and real-time performance of recognition are improved.

[0036] In step S3, a multi-UAV cooperative positioning algorithm is adopted to solve the positioning error problem of multiple UAVs in a dynamic environment, and the virtual data in the three-dimensional simulation environment is combined through sensor fusion technologies (such as GPS, IMU, vision sensors, etc.) to update the positions of the UAVs in real time. Multiple UAVs share position information and target recognition results through the cooperative positioning algorithm to form a shared positioning system, ensuring the unified positioning accuracy of the entire UAV group.

[0037] Specifically, the relative position relationship and sensor data between UAVs are used to minimize the error. For example, it includes the following steps:

[0038] Step S3.1: Establishment of coordinate systems;

[0039] Multiple coordinate systems are established to define the relative positions between the UAVs and the targets;

[0040] Step S3.2: Calculation of the target initial position;

[0041] Based on the geometric model, using the visual observation data and coordinate transformation of multiple UAVs, unify the visual axis of each UAV to the world coordinate system;

[0042] Calculate the initial position of the target through the least squares optimization method;

[0043] Step S3.2.1: Conversion from image to camera coordinate system;

[0044] Using the pinhole camera model, calculate the unit vector of the visual axis from the camera optical center to the target: where: u, v are the pixel coordinates of the target in the image coordinate system; u 0 , v 0 is the position of the optical axis at the center of the image; f is the camera focal length;

[0045] Step S3.2.2: Conversion from camera coordinate system to world coordinate system;

[0046] Through the attitude angles (pitch, yaw, roll angle) of the UAV, convert the unit visual axis vector from the camera coordinate system to the world coordinate system: LOS w = R c→w · LOS c ;

[0047] where, R c→w is the rotation matrix, describing the transformation from the camera coordinate system to the world coordinate system;

[0048] Step S3.2.3: Geometric modeling of the target position;

[0049] The geometric model of the collaborative observation of multiple UAVs assumes that the target is at the intersection of the visual axes of each UAV, but due to sensor errors and noise, the actual visual axes do not completely converge;

[0050] The estimation of the target position P in the world coordinate system is calculated by minimizing the foot point error:

[0051]

[0052] where, is the foot position of the target to the i-th visual axis;

[0053] The solution through the least squares method is:

[0054] Step S3.3: Target position and state estimation;

[0055] The target state estimation is achieved through the Interacting Multiple Model Unscented Kalman Filter (IMM-UKF), which estimates the end-to-end estimation model of the target in the world coordinate system from the image coordinates of the target, and continuously updates the state of the target using dynamic observations and filtering.

[0056] Step S3.4: Multi-model fusion (IMM);

[0057] For various motion modes of the target (such as uniform motion, uniform acceleration, and uniform turning), the Interacting Multiple Model (IMM) method is used to fuse the state estimations of different models, specifically including:

[0058] Model probability update formula:

[0059] where, π ij represents the transition probability from model i to model j;

[0060] Joint state estimation formula:

[0061] Joint covariance matrix:

[0062] In step S4, the UAVs share information through a wireless communication network (such as the MavLink protocol), and transmit target recognition data and positioning information in real time. Information fusion technology enables the UAV swarm to make collaborative decisions, share target positions, flight paths, and task assignments.

[0063] In step S5, for example, through the feedback mechanism of the simulation environment, the flight paths, task assignments, and target recognition strategies of the multi-UAVs are optimized to improve the task execution efficiency and accuracy of the UAV swarm. The optimization algorithm is based on the target recognition and positioning accuracy, and adaptively adjusts the flight trajectory and collaborative strategy to minimize the task execution time and positioning error.

[0064] Such as Figure 2 shows in detail the detailed steps from the initialization of the 3D simulation environment, sensor configuration, to target recognition and multi-UAV collaborative positioning process, specifically including:

[0065] 3D environment modeling and target generation: According to the sensor and environment models, generate and update the target position, speed, direction, etc. in real time; including:

[0066] 3D environment update: Dynamically update the 3D models of buildings, roads, obstacles, etc.;

[0067] Target generation: Generate dynamic or static targets through algorithms and update the motion trajectories of the targets;

[0068] Data flow: The generated target information is used for subsequent target recognition and tracking.

[0069] Sensor Data Fusion and Target Detection: Fuse data from different sensors to improve the accuracy of target recognition and classification; including:

[0070] Data Fusion: Fuse data from different sensors (such as vision, LiDAR, IMU) to generate high-quality target detection results;

[0071] Deep Learning Models: Use models such as YOLO, Faster R-CNN to identify and classify targets;

[0072] Data Stream: The identified target information (such as target type, location, speed, etc.) is transmitted to the target tracking and positioning module.

[0073] Multi-UAV Cooperative Localization and Information Sharing: Share location information among multiple UAVs to perform cooperative localization to improve accuracy; including:

[0074] Localization Technology: Use SLAM (Simultaneous Localization and Mapping) or GPS for UAV localization;

[0075] Information Sharing: Real-time share location and sensor data through wireless communication protocols (such as Wi-Fi, LoRa);

[0076] Data Fusion: Use Extended Kalman Filter (EKF) for multi-UAV position fusion;

[0077] Data Stream: The shared location information is used for further path planning and cooperative localization.

[0078] 3D Target Trajectory Prediction and Dynamic Tracking: Predict the future trajectory of the target based on its current state and historical data, and perform dynamic tracking; including:

[0079] Kalman Filter: Used for target state prediction and trajectory estimation;

[0080] Particle Filter: Improve the tracking accuracy of the target in complex environments;

[0081] Data Stream: The tracking data and predicted trajectory are transmitted to the path planning module for dynamic adjustment.

[0082] 3D Path Planning and Task Scheduling: Perform path planning and task scheduling according to the target location and environmental information; including:

[0083] Path Planning: Use path planning algorithms such as A*, RRT to calculate the shortest path;

[0084] Dynamic Obstacle Avoidance: Avoid obstacles and dangerous areas that appear in real time;

[0085] Task scheduling: Adjust task allocation based on task priorities and UAV status;

[0086] Data flow: The generated path information and task scheduling results are sent to each UAV for execution.

[0087] Real-time feedback and path adjustment: During the execution process, adjust the flight path according to real-time monitoring data; including:

[0088] Real-time status monitoring: Monitor the status of each UAV, such as position, speed, and sensor status;

[0089] Path adjustment: Dynamically adjust the flight path according to environmental changes or targets;

[0090] Data flow: Real-time feedback data is used to correct the flight path to ensure the smooth completion of tasks.

[0091] Task completion and data recording: After the task is completed, record the data and generate a task report; including:

[0092] Task data recording: Record the execution data of each task, such as target type, path, time, etc.

[0093] Report generation: Generate a detailed report based on the task data for subsequent analysis.

[0094] Data flow: Task data is uploaded to the control platform, and the task report is generated and submitted.

[0095] Figure 3 Shows a design diagram of a typical application scenario, which is based on Figure 2 the key links of target recognition and multi-UAV collaborative positioning in the 3D visual simulation platform described in

[0096] In summary, the multi-UAV collaborative positioning method proposed in the embodiment of the present invention provides a realistic simulation environment for flight missions based on the establishment of a three-dimensional visual simulation platform, which helps to verify the execution effect of the target recognition and positioning algorithm and reduce the risks in actual operations; through deep learning and multi-sensor data fusion, the accuracy and robustness of target recognition in complex environments are significantly improved. Especially in dynamic environments, low-light or obstacle occlusion conditions, it can ensure the efficient recognition of targets; the decentralized collaborative control algorithm is used to solve the communication and path planning problems between multiple UAVs, ensuring that the UAV group can cooperate efficiently and complete difficult tasks in complex environments; by dynamically adjusting the simulation environment and flight missions, accurate target recognition and positioning can be carried out under different environmental conditions, providing technical support for UAV applications in complex scenarios; the real-time evaluation and feedback mechanism during task execution can continuously optimize the target recognition and positioning algorithm to ensure the efficient execution of the algorithm in different tasks and environments.

[0097]

Second Embodiment

[0098] As Figure 4 shown, the second embodiment of the present invention proposes a multi-UAV collaborative positioning device 20, which, for example, includes: a simulation platform construction module 201, a target recognition and tracking module 202, a position and state estimation module 203, an information fusion and collaborative decision-making module 204, and a strategy optimization module 205.

[0099] Among them, the simulation platform construction module 201 is used to construct a three-dimensional visual simulation platform, including dynamic environment modeling, sensor data simulation, and virtual-real combination verification. The target recognition and tracking module 202 is used to recognize and track the target through multi-sensor data fusion and deep learning algorithms of the three-dimensional visual simulation platform. The position and state estimation module 203 is used to realize real-time estimation of the position and state of the target based on the multi-UAV collaborative positioning algorithm through geometric modeling and interactive multi-model unscented Kalman filtering. The information fusion and collaborative decision-making module 204 is used to realize information fusion and collaborative decision-making between multiple UAVs through a wireless communication network. The strategy optimization module 205 is used to optimize the flight path, task allocation, and target recognition strategy of the UAV based on simulation feedback.

[0100] The multi-UAV collaborative positioning method implemented by the multi-UAV collaborative positioning device 20 disclosed in the second embodiment of the present invention is as described in the foregoing first embodiment, so it will not be elaborated in detail here. Optionally, each module in the second embodiment and the above other operations or functions are respectively for implementing the multi-UAV collaborative positioning method described in the first embodiment, and the beneficial effects of this embodiment are the same as those of the foregoing first embodiment. For the sake of brevity, they will not be repeated here.

[0101]

Third Embodiment

[0102] As Figure 5 shown, a third embodiment of the present invention provides an electronic device 30, which includes, for example: a memory 32 and one or more processors 31 connected to the memory 32. The memory 32 stores a computer program, and the processor 31 is configured to execute the computer program to implement the multi-UAV collaborative positioning method as described in the first embodiment. For specific details, reference can be made to the method described in the first embodiment, which will not be elaborated here for the sake of brevity. Moreover, the beneficial effects of the electronic device 30 provided in this embodiment are the same as those of the multi-UAV collaborative positioning method provided in the first embodiment.

[0103]

Fourth Embodiment

[0104] As Figure 6 shown, a fourth embodiment of the present invention provides a computer-readable storage medium 40. The computer-readable storage medium 40 is a non-volatile memory and stores computer-readable instructions. When the computer-readable instructions are executed by one or more processors, for example, it causes the one or more processors to execute the multi-UAV collaborative positioning method described in the foregoing first embodiment. For the specific method, reference can be made to the method described in the first embodiment, which will not be elaborated here for the sake of brevity. Moreover, the beneficial effects of the computer-readable storage medium 40 provided in this embodiment are the same as those of the multi-UAV collaborative positioning method provided in the first embodiment.

[0105] In addition, it can be understood that the foregoing various embodiments are only illustrative descriptions of the present invention. On the premise that there is no conflict in technical features, no contradiction in structure, and no violation of the invention purpose of the present invention, the technical solutions of the various embodiments can be arbitrarily combined and used in combination.

[0106] In several embodiments provided by the present invention, it should be understood that the disclosed system, device, and / or method can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units / modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0107] The units / modules described as separate components may or may not be physically separated. The components shown as units / modules may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple network units. Some or all of the units / modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0108] In addition, each functional unit / module in various embodiments of the present invention may be integrated into one processing unit / module, or each unit / module may exist physically alone, or two or more units / modules may be integrated into one unit / module. The above-mentioned integrated unit / module may be implemented in the form of hardware, or in the form of a combination of hardware and software functional units / modules.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A multi-UAV collaborative positioning method, characterized in that: include: Build a 3D visual simulation platform, including dynamic environment modeling, sensor data simulation, and virtual-reality combination verification; Identify and track targets through multi-sensor data fusion and deep learning algorithms of the three-dimensional visual simulation platform; Based on the multi-UAV collaborative positioning algorithm, the real-time estimation of the position and state of the target is achieved through geometric modeling and interactive multi-model unscented Kalman filtering; Realize information fusion and collaborative decision-making among multiple UAVs through wireless communication networks; Optimize the UAV's flight path, task allocation, and target identification strategy based on simulation feedback.

2. The multi-UAV collaborative positioning system according to claim 1, characterized in that: The dynamic environment modeling includes: Build a 3D virtual environment through point clouds, meshes and textures to simulate terrain, buildings, dynamic targets and climate change; Integrated virtual sensors simulate data feedback from cameras, LiDAR, millimeter-wave radar and IMU in different environments; The stability of the algorithm in the actual environment is verified through the interaction of virtual and real data.

3. The multi-UAV collaborative positioning method according to claim 1, characterized in that: The target identification and tracking includes: A lightweight classification head is designed using multi-scale feature fusion and depth-wise separable convolution; Combine target color, shape and motion trajectory to extract features; Use YOLO or Faster R-CNN models to achieve target classification, detection and tracking.

4. The multi-UAV collaborative positioning method according to claim 1, characterized in that: The UAV collaborative positioning algorithm includes: Define the relative positions of each UAV and the target by establishing a multi-coordinate system; Calculate the initial position of the target based on the least squares optimization method; The multi-motion mode state estimation of the target is fused through IMM-UKF to update the target position and covariance matrix.

5. The multi-UAV collaborative positioning method according to claim 4, characterized in that: The method of calculating the initial position of the target based on the least squares optimization method includes: The image coordinate system is converted into the camera coordinate system through the pinhole camera model; The camera coordinate system is converted to the world coordinate system based on the drone attitude angle; The world coordinate position of the target is calculated by minimizing the perpendicular point error.

6. The multi-UAV collaborative positioning method according to claim 1, characterized in that: The information fusion and collaborative decision-making include: The MavLink protocol is used to achieve real-time data communication between drones; Fusion of multi-UAV positioning data through extended Kalman filtering; Dynamically adjust task priorities and path planning to achieve decentralized collaborative control.

7. The multi-UAV collaborative positioning method according to claim 1, characterized in that: The flight path, task allocation and target identification strategy of the UAV based on simulation feedback optimization includes: Generate dynamic obstacle avoidance paths based on A* or RRT algorithms; Predict target trajectory through particle filtering and correct flight path in real time; Adaptive parameter adjustment is performed based on task execution efficiency and positioning error.

8. A multi-UAV collaborative positioning device, characterized in that: include: The simulation platform construction module is used to build a 3D scene simulation platform, including dynamic environment modeling, sensor data simulation, and virtual-reality combination verification; A target recognition and tracking module is used to recognize and track targets through multi-sensor data fusion and deep learning algorithms of the three-dimensional visual simulation platform; A position and state estimation module is used to achieve real-time estimation of the position and state of the target through geometric modeling and interactive multi-model unscented Kalman filtering based on a multi-UAV collaborative positioning algorithm; Information fusion and collaborative decision-making module, used to realize information fusion and collaborative decision-making among multiple UAVs through wireless communication networks; The strategy optimization module is used to optimize the flight path, task allocation and target identification strategy of the UAV based on simulation feedback.

9. An electronic device, characterized in that: include: A memory and one or more processors connected to the memory, the memory storing a computer program, and the processor being used to execute the computer program to implement the multi-UAV collaborative positioning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to execute the multi-UAV collaborative positioning method as described in any one of claims 1 to 7.

Images