Multi-source data fusion and collaborative decision-making system for one-network unified flight of unmanned aerial vehicle

Abstract

The invention relates to unmanned aerial vehicle cluster collaboration, in particular to a multi-source data fusion and collaborative decision-making system for unmanned aerial vehicle one-network unified flight, which comprises a control unit, and the control unit searches a plurality of initial global paths for an unmanned aerial vehicle cluster in a centralized manner through an initial global path search module; a safety corridor construction module is used for constructing a safety corridor in a future period of time for each unmanned aerial vehicle according to the initial global path, and the control unit clusters the unmanned aerial vehicle cluster according to the safety corridor through an unmanned aerial vehicle clustering module; a collaborative planning module is utilized to carry out collaborative planning on the unmanned aerial vehicles of the same kind according to the safety corridor and the clustering result, and a control unit sends a collaborative planning path to each unmanned aerial vehicle through a wireless communication module; the technical scheme provided by the invention can effectively overcome the defects of lack of global coordination and difficulty in planning a safe and reasonable outlier path for the outlier unmanned aerial vehicle during path planning for each unmanned aerial vehicle in the prior art.

Description

Technical Field

[0001] This invention relates to drone swarm collaboration, specifically to a multi-source data fusion and collaborative decision-making system for unified drone flight. Background Technology

[0002] With the rapid development of drone technology, drone swarms, with their advantages of high efficiency and wide mission coverage, have shown great application potential in many fields, such as disaster monitoring, logistics delivery, and military reconnaissance. However, achieving efficient and safe operation of drone swarms faces many challenges in practical applications.

[0003] In traditional drone swarm control, each drone typically plans its path independently, lacking global coordination and unified scheduling. This can easily lead to path conflicts, resource waste, and reduced mission execution efficiency of the drone swarm. Furthermore, in complex and ever-changing real-world environments, drone flight is affected by various dynamic factors, such as the sudden appearance of obstacles and changes in airflow. Existing technologies struggle to create a safe and reliable flight space for each drone in real time, ensuring its safe flight.

[0004] Furthermore, during mission execution, drone swarms need to flexibly adjust their formation and spacing according to actual conditions to adapt to different mission requirements and environmental changes. However, currently, there is a lack of effective collaborative adjustment mechanisms, making it difficult to achieve precise and rapid formation adjustments. At the same time, how to plan safe and reasonable departure paths for drones that need to leave the swarm is also an urgent problem to be solved. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a multi-source data fusion and collaborative decision-making system for unmanned aerial vehicles (UAVs) to fly in a unified network. It can effectively overcome the shortcomings of the existing technology, such as the lack of global coordination when planning paths for each UAV and the difficulty in planning safe and reasonable out-of-group paths for out-of-group UAVs.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A multi-source data fusion and collaborative decision-making system for unified flight of unmanned aerial vehicles (UAVs) includes a control unit. The control unit centrally searches for multiple initial global paths for the UAV swarm through an initial global path search module, and constructs a safe corridor for each UAV for a future period based on the initial global paths using a safe corridor construction module. The control unit clusters the UAV swarm based on the safe corridors using a UAV clustering module, and performs collaborative planning for UAVs of the same type based on the safe corridors and clustering results using a collaborative planning module. The control unit sends the collaboratively planned paths to each UAV through a wireless communication module.

[0010] The control unit sends a status information acquisition request to the drone cluster through the status information acquisition module, and simultaneously receives the feedback status information. It also uses the cluster formation adjustment module to coordinately adjust the formation and spacing of the drone cluster based on the status information. The control unit then uses the out-of-group path planning module to plan the path for the stray drones that need to leave the drone cluster based on the adjusted formation and spacing, and sends the out-of-group path to the stray drones through the wireless communication module.

[0011] Preferably, the initial global path search module centrally searches for multiple initial global paths for the UAV swarm, including:

[0012] The map collected by the drone swarm is converted into a raster map, and the raster map is expanded according to the drone radius;

[0013] By combining the expanded grid map with the UAV kinematics model, the Hybrid Astar algorithm is used to independently search an initial global path for each UAV.

[0014] Detect whether there are spatiotemporal conflicts between all initial global paths. If there are spatiotemporal conflicts, apply conflict constraints to the UAVs corresponding to the initial global paths with spatiotemporal conflicts and then re-search for global paths until there are no spatiotemporal conflicts between all initial global paths.

[0015] Preferably, the safe corridor construction module constructs a safe corridor for each UAV for a future period of time based on the initial global path, including:

[0016] Based on the initial global path, a safe corridor is generated between two adjacent path points for each UAV, starting from the first path point in the future and ending at the optimal path point.

[0017] An initial cuboid safety corridor is generated using two adjacent path points as diagonals, and the initial cuboid safety corridor is expanded outward until it encounters an obstacle or reaches a preset maximum size, thus constructing a safety corridor for each drone for a future period of time.

[0018] Preferably, the drone clustering module clusters drone swarms based on security corridors, including:

[0019] Detect whether there is any overlap between the safety corridors of all drones in the future, group the drones corresponding to the overlapping safety corridors into a separate category, and cluster the remaining drones according to the safety corridors.

[0020] In this context, the safety corridors between any one drone in one class and all drones in other classes will not overlap in the near future.

[0021] Preferably, detecting whether there is overlap between the safety corridors of all drones over a future period includes:

[0022] The AABB collision detection algorithm is used to project the two safe corridors to be detected onto the X-axis, Y-axis, and Z-axis as line segments, respectively. The algorithm detects whether there is overlap between the line segments on the X-axis, Y-axis, and Z-axis. If there is overlap, it is determined that there is overlap between the two safe corridors to be detected, that is, a collision will occur between the drones.

[0023] Preferably, the status information acquisition module sends a status information acquisition request to the UAV cluster and simultaneously receives feedback status information, including:

[0024] The status information includes the individual status information of the stray drone that needs to leave the drone swarm, as well as the swarm status information of the drone swarm to which the stray drone belongs.

[0025] Individual status information includes information on the reason for leaving the group and the formation position information of the stray drone in the drone swarm. The swarm status information includes information on the number of drones, as well as formation information and spacing information.

[0026] Preferably, the out-of-group cause information includes system failure and execution of other tasks. When the out-of-group cause information is system failure, it is also necessary to obtain the directional range in which the out-of-group UAV can continue to fly.

[0027] Preferably, the cluster formation adjustment module coordinates the formation and spacing of the UAV cluster based on status information, including:

[0028] When the out-of-group reason information is system failure:

[0029] If there is only one stray drone that needs to leave the drone swarm, the positions of other drones in the drone swarm are dynamically adjusted based on the stray drone, and the formation and spacing are adjusted so that the stray drone has at least one passable stray path within the direction range where it can continue to fly.

[0030] If there are multiple stray drones that need to leave the drone swarm, the positions of other drones in the drone swarm are dynamically adjusted based on each stray drone, and the formation and spacing are adjusted so that each stray drone has at least one passable stray path within the direction in which it can continue to fly.

[0031] Preferably, the cluster formation adjustment module coordinates the formation and spacing of the UAV cluster based on status information, including:

[0032] When the reason for leaving the group is performing other tasks:

[0033] With the premise of minimizing adjustments, the positions of all drones in the drone swarm are dynamically adjusted, and the formation and spacing are adjusted so that any stray drone that needs to leave the drone swarm has at least one passable stray path.

[0034] Preferably, the out-of-group path planning module performs path planning for out-of-group drones that need to leave the drone swarm based on the adjusted formation and spacing, including:

[0035] Based on the adjusted formation and spacing, path planning is performed for stray drones that need to leave the drone swarm, resulting in a stray path that meets the requirements of the direction range for the stray drones to continue flying and maintains a safe distance from other drones.

[0036] (III) Beneficial Effects

[0037] Compared with existing technologies, the multi-source data fusion and collaborative decision-making system for unified flight of unmanned aerial vehicles (UAVs) provided by this invention has the following beneficial effects:

[0038] 1) The initial global path search module centrally searches for multiple initial global paths for the UAV swarm. The safe corridor construction module constructs a safe corridor for each UAV for a future period based on the initial global paths. The UAV clustering module clusters the UAV swarm based on the safe corridors. The collaborative planning module performs collaborative planning for UAVs of the same type based on the safe corridors and clustering results. The collaboratively planned path is sent to each UAV through the wireless communication module. This allows for centralized searching of multiple initial global paths for the UAV swarm, construction of safe corridors for each UAV for a future period based on the initial global paths and clustering, and finally collaborative planning for UAVs of the same type based on the safe corridors and clustering results. This enables path planning for each UAV on a global collaborative basis, effectively improving the task execution efficiency of the UAV swarm. At the same time, it can construct a safe and reliable flight space for each UAV in real time to ensure its safe flight.

[0039] 2) The status information acquisition module sends a status information acquisition request to the UAV cluster and receives the feedback status information. The cluster formation adjustment module adjusts the formation and spacing of the UAV cluster in coordination based on the status information. The out-of-staff path planning module plans the path for the out-of-staff UAVs that need to leave the UAV cluster based on the adjusted formation and spacing. The out-of-staff path is sent to the out-of-staff UAVs through the wireless communication module. This allows for flexible adjustment of the formation and spacing based on the actual situation on the basis of multi-source data fusion, in order to adapt to different mission requirements and environmental changes. When a UAV needs to leave the UAV cluster, it can plan a safe and reasonable out-of-staff path for it, effectively improving the flight stability and reliability of the UAV cluster in the face of emergencies while ensuring mission execution efficiency. Attached Figure Description

[0040] 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.

[0041] Figure 1 This is a schematic diagram of the system of the present invention;

[0042] Figure 2 This is a schematic diagram of the process for collaborative planning of drone swarms in this invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0044] The following describes the specific functional modules of the multi-source data fusion and collaborative decision-making system for unified UAV flight provided by this invention, using specific examples (such as...). Figure 1 As shown in the figure, and considering the technical effects, the system functional modules include:

[0045] The control unit centrally searches for multiple initial global paths for the drone swarm through the initial global path search module, and uses the safe corridor construction module to construct a safe corridor for each drone for a future period of time based on the initial global paths. The control unit clusters the drone swarm based on the safe corridors through the drone clustering module, and uses the collaborative planning module to collaboratively plan the drones of the same type based on the safe corridors and clustering results. The control unit sends the collaboratively planned path to each drone through the wireless communication module.

[0046] The control unit sends a status information acquisition request to the drone cluster through the status information acquisition module, and receives the feedback status information. It also uses the cluster formation adjustment module to adjust the formation and spacing of the drone cluster according to the status information. The control unit uses the out-of-group path planning module to plan the path for the out-of-group drones that need to leave the drone cluster according to the adjusted formation and spacing, and sends the out-of-group path to the out-of-group drones through the wireless communication module.

[0047] The specific functions of each system module are described below:

[0048] I. Initial Global Path Search Module

[0049] The initial global path search module centrally searches for multiple initial global paths for the UAV swarm, such as... Figure 2 As shown, it includes:

[0050] The map collected by the drone swarm is converted into a raster map, and the raster map is expanded according to the drone radius;

[0051] By combining the expanded grid map with the UAV kinematics model, the Hybrid Astar algorithm is used to independently search an initial global path for each UAV.

[0052] Detect whether there are spatiotemporal conflicts between all initial global paths. If there are spatiotemporal conflicts, apply conflict constraints to the UAVs corresponding to the initial global paths with spatiotemporal conflicts and then re-search for global paths until there are no spatiotemporal conflicts between all initial global paths.

[0053] II. Safe Corridor Construction Module

[0054] The safe corridor construction module builds a safe corridor for each drone over a future period based on the initial global path, such as... Figure 2 As shown, it includes:

[0055] Based on the initial global path, a safe corridor is generated between two adjacent path points for each UAV, starting from the first path point in the future and ending at the optimal path point.

[0056] An initial cuboid safety corridor is generated using two adjacent path points as diagonals, and the initial cuboid safety corridor is expanded outward until it encounters an obstacle or reaches a preset maximum size, thus constructing a safety corridor for each drone for a future period of time.

[0057] III. Unmanned Aerial Vehicle (UAV) Clustering Module

[0058] The drone clustering module clusters drone swarms based on safety corridors, such as... Figure 2 As shown, it includes:

[0059] Detect whether there is any overlap between the safety corridors of all drones in the future, group the drones corresponding to the overlapping safety corridors into a separate category, and cluster the remaining drones according to the safety corridors.

[0060] In this context, the safety corridors between any one drone in one class and all drones in other classes will not overlap in the near future.

[0061] Specifically, this involves detecting whether there is overlap between the safety corridors of all drones over a future period of time, including:

[0062] The AABB collision detection algorithm is used to project the two safe corridors to be detected onto the X-axis, Y-axis, and Z-axis as line segments, respectively. The algorithm detects whether there is overlap between the line segments on the X-axis, Y-axis, and Z-axis. If there is overlap, it is determined that there is overlap between the two safe corridors to be detected, that is, a collision will occur between the drones.

[0063] The above technical solution involves a centralized initial global path search module that searches multiple initial global paths for the UAV swarm, a safe corridor construction module that constructs a safe corridor for each UAV for a future period based on the initial global paths, a UAV clustering module that clusters the UAV swarm based on the safe corridors, and a collaborative planning module that performs collaborative planning for UAVs of the same type based on the safe corridors and clustering results. The collaboratively planned paths are then sent to each UAV via a wireless communication module. This approach enables centralized searching of multiple initial global paths for the UAV swarm, construction of safe corridors for each UAV for a future period based on the initial global paths, clustering, and finally collaborative planning for UAVs of the same type based on the safe corridors and clustering results. This allows for path planning for each UAV on a global collaborative basis, effectively improving the task execution efficiency of the UAV swarm and simultaneously constructing a safe and reliable flight space for each UAV in real time, ensuring its safe flight.

[0064] IV. Status Information Acquisition Module

[0065] The status information acquisition module sends a status information acquisition request to the drone cluster and simultaneously receives feedback status information, including:

[0066] The status information includes the individual status information of the stray drone that needs to leave the drone swarm, as well as the swarm status information of the drone swarm to which the stray drone belongs.

[0067] Individual status information includes information on the reason for leaving the group and the formation position information of the stray drone in the drone swarm. The swarm status information includes information on the number of drones, as well as formation information and spacing information.

[0068] Specifically, the reasons for leaving the group include system failure and performing other tasks. When the reason for leaving the group is system failure, it is also necessary to obtain the range of directions in which the stray drone can continue to fly.

[0069] V. Cluster Formation Adjustment Module

[0070] The cluster formation adjustment module coordinates the formation and spacing of the drone cluster based on status information, including:

[0071] 1) When the outgoing traffic cause information indicates a system malfunction:

[0072] If there is only one stray drone that needs to leave the drone swarm, the positions of other drones in the drone swarm are dynamically adjusted based on the stray drone, and the formation and spacing are adjusted so that the stray drone has at least one passable stray path within the direction range where it can continue to fly.

[0073] If there are multiple stray drones that need to leave the drone swarm, the positions of other drones in the drone swarm are dynamically adjusted based on each stray drone, and the formation and spacing are adjusted so that each stray drone has at least one passable stray path within the direction in which it can continue to fly.

[0074] 2) When the reason for leaving the group is "performing other tasks":

[0075] With the premise of minimizing adjustments, the positions of all drones in the drone swarm are dynamically adjusted, and the formation and spacing are adjusted so that any stray drone that needs to leave the drone swarm has at least one passable stray path.

[0076] VI. Outlier Path Planning Module

[0077] The out-of-group path planning module plans paths for out-of-group drones that need to leave the drone swarm based on the adjusted formation and spacing, including:

[0078] Based on the adjusted formation and spacing, path planning is performed for stray drones that need to leave the drone swarm, resulting in a stray path that meets the requirements of the direction range for the stray drones to continue flying and maintains a safe distance from other drones.

[0079] The above technical solution involves a status information acquisition module sending a status information acquisition request to the drone swarm and simultaneously receiving feedback status information. A swarm formation adjustment module collaboratively adjusts the formation and spacing of the drone swarm based on the status information. An out-of-swarm path planning module plans paths for drones needing to leave the swarm based on the adjusted formation and spacing. These out-of-swarm paths are then transmitted to the drones via a wireless communication module. This allows for flexible adjustment of formation and spacing based on multi-source data fusion and actual conditions to adapt to different mission requirements and environmental changes. Furthermore, when a drone needs to leave the swarm, a safe and reasonable out-of-swarm path can be planned for it, effectively improving the flight stability and reliability of the drone swarm in the face of unforeseen circumstances while ensuring mission execution efficiency.

[0080] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-source data fusion and collaborative decision-making system for unified flight of unmanned aerial vehicles (UAVs), characterized in that: The system includes a control unit. The control unit centrally searches for multiple initial global paths for the UAV swarm through an initial global path search module, and constructs a safe corridor for each UAV for a future period based on the initial global paths using a safe corridor construction module. The control unit clusters the UAV swarm based on the safe corridors using a UAV clustering module, and performs collaborative planning for UAVs of the same type based on the safe corridors and clustering results using a collaborative planning module. The control unit sends the collaboratively planned paths to each UAV through a wireless communication module. The control unit sends a status information acquisition request to the drone cluster through the status information acquisition module, and simultaneously receives the feedback status information. It also uses the cluster formation adjustment module to coordinately adjust the formation and spacing of the drone cluster based on the status information. The control unit then uses the out-of-group path planning module to plan the path for the stray drones that need to leave the drone cluster based on the adjusted formation and spacing, and sends the out-of-group path to the stray drones through the wireless communication module.

2. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 1, characterized in that: The initial global path search module centrally searches for multiple initial global paths for the UAV swarm, including: The map collected by the drone swarm is converted into a raster map, and the raster map is expanded according to the drone radius; By combining the expanded grid map with the UAV kinematics model, the Hybrid Astar algorithm is used to independently search an initial global path for each UAV. Detect whether there are spatiotemporal conflicts between all initial global paths. If there are spatiotemporal conflicts, apply conflict constraints to the UAVs corresponding to the initial global paths with spatiotemporal conflicts and then re-search for global paths until there are no spatiotemporal conflicts between all initial global paths.

3. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 2, characterized in that: The safe corridor construction module constructs a safe corridor for each UAV for a future period of time based on the initial global path, including: Based on the initial global path, a safe corridor is generated between two adjacent path points for each UAV, starting from the first path point in the future and ending at the optimal path point. An initial cuboid safety corridor is generated using two adjacent path points as diagonals, and the initial cuboid safety corridor is expanded outward until it encounters an obstacle or reaches a preset maximum size, thus constructing a safety corridor for each drone for a future period of time.

4. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 3, characterized in that: The drone clustering module clusters drone swarms based on security corridors, including: Detect whether there is any overlap between the safety corridors of all drones in the future, group the drones corresponding to the overlapping safety corridors into a separate category, and cluster the remaining drones according to the safety corridors. In this context, the safety corridors between any one drone in one class and all drones in other classes will not overlap in the near future.

5. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 4, characterized in that: The detection of whether there is overlap between the safety corridors of all drones over a future period includes: The AABB collision detection algorithm is used to project the two safe corridors to be detected onto the X-axis, Y-axis, and Z-axis as line segments, respectively. The algorithm detects whether there is overlap between the line segments on the X-axis, Y-axis, and Z-axis. If there is overlap, it is determined that there is overlap between the two safe corridors to be detected, that is, a collision will occur between the drones.

6. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 1, characterized in that: The status information acquisition module sends a status information acquisition request to the UAV cluster and simultaneously receives feedback status information, including: The status information includes the individual status information of the stray drone that needs to leave the drone swarm, as well as the swarm status information of the drone swarm to which the stray drone belongs. Individual status information includes information on the reason for leaving the group and the formation position information of the stray drone in the drone swarm. The swarm status information includes information on the number of drones, as well as formation information and spacing information.

7. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 6, characterized in that: The reasons for leaving the group include system failure and execution of other tasks. When the reason for leaving the group is system failure, it is also necessary to obtain the range of directions in which the stray drone can continue to fly.

8. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 7, characterized in that: The cluster formation adjustment module coordinates the formation and spacing of the UAV cluster based on status information, including: When the out-of-group reason information is system failure: If there is only one stray drone that needs to leave the drone swarm, the positions of other drones in the drone swarm are dynamically adjusted based on the stray drone, and the formation and spacing are adjusted so that the stray drone has at least one passable stray path within the direction range where it can continue to fly. If there are multiple stray drones that need to leave the drone swarm, the positions of other drones in the drone swarm are dynamically adjusted based on each stray drone, and the formation and spacing are adjusted so that each stray drone has at least one passable stray path within the direction in which it can continue to fly.

9. The multi-source data fusion and collaborative decision-making system for unified flight of unmanned aerial vehicles according to claim 8, characterized in that: The cluster formation adjustment module coordinates the formation and spacing of the UAV cluster based on status information, including: When the reason for leaving the group is "performing other tasks": With the premise of minimizing adjustments, the positions of all drones in the drone swarm are dynamically adjusted, and the formation and spacing are adjusted so that any stray drone that needs to leave the drone swarm has at least one passable stray path.

10. The multi-source data fusion and collaborative decision-making system for unified UAV flight according to claim 9, characterized in that: The out-of-group path planning module performs path planning for out-of-group drones that need to leave the drone swarm based on the adjusted formation and spacing, including: Based on the adjusted formation and spacing, path planning is performed for stray drones that need to leave the drone swarm, resulting in a stray path that meets the requirements of the direction range for the stray drones to continue flying and maintains a safe distance from other drones.

Images