Intelligent unmanned aerial vehicle cluster cooperative operation system

Abstract

The application provides an intelligent unmanned aerial vehicle cluster cooperative operation system, and belongs to the field of unmanned aerial vehicle cooperative operation, and comprises a task planning module, a cooperative decision module, a resource management module and an execution module.The task planning module is used for receiving task requirements of an unmanned aerial vehicle cluster, decomposing the task requirements into subtasks, and determining the importance of the subtasks.The cooperative decision module is used for calculating the comprehensive capabilities of each unmanned aerial vehicle, and generating a task allocation scheme according to the comprehensive capabilities and the importance of the subtasks.The resource management module is used for determining total task resources according to the task requirements of the unmanned aerial vehicle cluster, and generating a resource allocation result based on the task allocation scheme and the total task resources.The execution module is used for guiding the unmanned aerial vehicle cluster to execute tasks based on the task allocation scheme and the resource allocation result.The application decomposes task requirements, constructs a subtask cooperative network graph, calculates the importance of the subtasks, dynamically sorts the subtasks according to the importance of the subtasks, realizes optimized allocation of the tasks, and improves the overall operation efficiency of the unmanned aerial vehicle cluster.

Description

Technical Field

[0001] This invention relates to the field of drone collaborative operation technology, and in particular to an intelligent drone swarm collaborative operation system. Background Technology

[0002] With the development of drone technology, the functions and performance of a single drone are no longer sufficient to meet the needs of complex tasks. Therefore, some manufacturers are trying to use drone swarms for collaborative operations. In this way, a drone swarm works together with multiple drones to complete complex tasks that a single drone cannot accomplish independently. Drone swarm collaborative operations are becoming a research hotspot and application trend due to their advantages such as high efficiency, high reliability, and system fault tolerance.

[0003] However, in existing technologies, drone swarms mostly adopt small-scale and simple collaboration, failing to fully consider the heterogeneous characteristics between drones and the diverse complexity of tasks. When faced with complex and ever-changing task environments, they cannot carry out continuous operations.

[0004] Chinese invention patent CN115630514A discloses a method and apparatus for collaborative task allocation in a drone swarm. It establishes a collaborative task allocation model by detecting and modeling signals, calculates task execution benefits based on the pose and state information of the drones and targets, and uses an optimization model for task allocation. However, this scheme ignores the impact of communication relationships between drones on task execution efficiency during task allocation and cannot regulate resources during drone operations. Summary of the Invention

[0005] In view of this, the present invention proposes an intelligent drone swarm collaborative operation system, which can solve the defects of existing technologies that cannot regulate resources in a timely manner, realize the optimized allocation of tasks, and improve the overall operational efficiency of drone swarms.

[0006] The technical solution of this invention is implemented as follows: This invention provides an intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system, comprising:

[0007] The task planning module is used to receive task requirements from the drone swarm, decompose the task requirements into sub-tasks, and determine the importance of the sub-tasks.

[0008] The collaborative decision-making module is used to calculate the comprehensive capabilities of each UAV, generate a task allocation scheme based on the comprehensive capabilities of each UAV and the importance of the sub-tasks, and send the task allocation scheme to the resource management module.

[0009] The resource management module is used to determine the total task resources based on the task requirements of the UAV cluster, and generate resource allocation results based on the task allocation scheme and the total task resources.

[0010] The execution module is used to guide the drone cluster to execute tasks based on the task allocation scheme and resource allocation results.

[0011] Based on the above technical solutions, preferably, the task requirements include task content, task scope, target coordinates, and task time constraints.

[0012] Based on the above technical solutions, preferably, the task planning module specifically includes:

[0013] The task decomposition unit is used to divide the task requirements based on spatial, functional, and temporal dimensions to generate multiple sub-tasks.

[0014] A subtask collaboration unit is used to construct a subtask collaboration network graph based on the multiple subtasks;

[0015] The subtask evaluation unit is used to calculate the importance of each subtask based on the subtask collaboration network graph, and to sort the subtasks in descending order according to their importance to form a subtask sequence list. The subtask list includes subtask ID, subtask content, subtask scope, subtask hard constraints, and subtask importance.

[0016] Based on the above technical solutions, the preferred method for calculating the sub-task evaluation unit is as follows:

[0017] The distance between two adjacent subtasks is determined based on the subtask collaboration network graph, and the influence of the subtask on the task operation range is determined based on the distance between the two adjacent subtasks.

[0018] The time urgency of subtasks is determined based on the aforementioned task time constraints;

[0019] The importance of a subtask is calculated based on the aforementioned impact and time urgency, using the following formula:

[0020]

[0021] Among them, I k P represents the importance of a subtask. base，k Let d(k, g) represent the basic importance score, α represent the influence weighting factor, g represent a neighboring subtask in N(k), N(k) represent the set of neighboring subtasks of subtask k in the subtask collaboration network graph, k represent any subtask in the subtask collaboration network graph, and d(k, g) represent the distance between subtask k and subtask g. dist T represents the distance adjustment constant, β represents the time urgency weighting factor, and T represents the distance adjustment constant. d，k T represents the latest completion time of subtask k. c D represents the current time. kIndicates the estimated execution time of subtask k, ∈ time This represents the time adjustment constant.

[0022] Based on the above technical solutions, preferably, the collaborative decision-making module specifically includes:

[0023] The capability assessment unit is used to calculate the comprehensive capability of each drone in the drone swarm based on the drone's communication capability, payload capability, and endurance capability, and to determine the leader drone of the drone swarm based on the comprehensive capability.

[0024] The collaborative communication unit is used to build a drone network topology based on the Navigator drone.

[0025] The task allocation unit is used to allocate tasks based on the comprehensive capabilities of each UAV, the UAV network topology, and the importance of subtasks, and to generate a task allocation scheme.

[0026] The path planning unit is used by each UAV to plan and execute paths based on global collaborative rules.

[0027] Based on the above technical solutions, preferably, the formula for calculating the comprehensive capability is as follows:

[0028] A i =w c ·Norm(C i )+w l ·Norm(L i )+w e ·Norm(E i );

[0029] Among them, A i w represents the overall capability score of the i-th drone. c The weighting coefficients represent communication capabilities, Norm(·) represents the normalization function, and C... i w represents the communication capability of the i-th drone. l L represents the weighting factor for load capacity. i w represents the payload capacity of the i-th drone. e E represents the weighting coefficient for battery life. i This represents the current endurance of the i-th drone.

[0030] Based on the above technical solutions, preferably, the specific steps of the task allocation unit are as follows:

[0031] The drones in the drone cluster are arranged in descending order of their comprehensive capabilities to form a drone queue;

[0032] Based on the aforementioned UAV network topology, determine the connection status and communication link quality between each UAV in the UAV cluster and the navigator UAV.

[0033] The drone queue is initially screened based on the hard constraints of each subtask in the subtask sequence list to form a first candidate drone sequence. The first candidate drone sequence is then screened a second time based on the connection status and communication link quality between each drone and the navigator drone to form a second candidate drone sequence. The hard constraints include communication capability requirements, load capacity requirements, and endurance requirements.

[0034] The subtask sequence list is matched one-to-one with the second candidate UAV sequence to generate a task allocation scheme.

[0035] Based on the above technical solutions, preferably, the global coordination rules specifically include formation rules, motion coordination rules, and task synchronization rules, wherein...

[0036] The formation pattern rules form a formation pattern based on the task allocation scheme and the task operation scope, and calculate the formation pattern parameters, which include the distance between drones in the drone cluster and the formation orientation.

[0037] The motion coordination rule calculates the speed, spacing, and orientation of the drone swarm based on the speed, position, and orientation of the drones within the swarm. The calculation formula is as follows:

[0038]

[0039] in, This represents the adjusted velocity vector of drone i. Let α represent the current velocity vector of drone i. v N represents the speed adjustment factor. i Let j represent the set of neighboring nodes directly connected to UAV i in the communication network topology, and j represent UAV j. This represents the current velocity vector of drone j. Let β represent the position adjustment vector of drone i. p This represents the position adjustment factor. This indicates the current position of drone j. d represents the current position of drone i. des Indicates the desired spacing. This indicates the adjusted orientation of drone i. Indicates the current orientation of drone i, γ θ Indicates the orientation adjustment factor. Indicates the current orientation of drone j;

[0040] The task synchronization rules are used to coordinate the time synchronization and task execution synchronization of the drone cluster based on the speed, spacing, and orientation of the drone cluster.

[0041] Based on the above technical solutions, preferably, the steps of the resource management module specifically include:

[0042] The system receives the task allocation result from the collaborative decision-making module and determines the resource requirements of each UAV in the UAV cluster when performing a task based on the task allocation result; wherein the resource requirements include resource requirement type, resource requirement quantity, and resource requirement time; the resources include power resources, communication bandwidth resources, sensor resources, and payload resources.

[0043] The resource requirements of each UAV during mission execution are summarized to form the total resource requirements;

[0044] Compare the total resource requirements with the total resources required for the task:

[0045] If the total resource requirement does not exceed the total resource requirement of the task, then the resources will be allocated according to the resource requirement.

[0046] If the total resource requirement exceeds the total task resources, the priority of each UAV resource request is determined by the expected importance of the subtasks, and resources are allocated according to the priority of the UAV resource requests.

[0047] More preferably, it also includes a security monitoring module, wherein the security monitoring module includes:

[0048] The status monitoring unit is used to monitor the parameters of the work area and the operating parameters of the UAV cluster in real time. The operating parameters include formation parameters, speed, spacing, orientation of the UAV cluster, and resource allocation results of the UAV cluster.

[0049] The risk warning unit is used to perform abnormal state analysis based on the results of the status monitoring unit, generate warning information, and provide risk warning prompts based on the warning information.

[0050] The intelligent drone swarm collaborative operation system of the present invention has the following advantages over the prior art:

[0051] (1) By decomposing the task requirements, constructing a subtask collaboration network diagram and calculating the importance of subtasks, and dynamically sorting them according to the importance of subtasks, the task allocation is optimized, thereby improving the overall operational efficiency of the UAV cluster.

[0052] (2) The first screening is based on the hard constraints of the sub-tasks to ensure the basic possibility of the task, and the second screening is based on the connection status with the navigator and the quality of the communication link to ensure the reliability of collaborative command and information interaction during the task execution process.

[0053] (3) By determining the priority of UAV resource requests based on the importance of sub-tasks and allocating resources according to the priority of UAV resource requests, the execution module guides and adjusts the behavior of the UAV cluster in real time based on global coordination rules, which not only improves the utilization efficiency of limited resources, but also ensures the overall coordination of the cluster in complex dynamic environments. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a flowchart of the intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system of the present invention;

[0056] Figure 2 This is a flowchart of the collaborative decision-making module of the intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system of the present invention. Detailed Implementation

[0057] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0058] like Figure 1 As shown, the present invention provides an intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system, comprising:

[0059] The task planning module is used to receive task requirements from the UAV swarm, decompose the task requirements into sub-tasks, and determine the importance of the sub-tasks. The task requirements include task content, task operation range, target coordinates, and task time constraints.

[0060] Understandably, mission requirements include mission content (such as high-resolution reconnaissance of a specific area, searching for and identifying specific types of moving targets, and precision strikes against designated targets), mission operation scope (such as a polygonal area defined by latitude and longitude coordinates, a specific airspace range, etc.), target coordinates (such as the precise or approximate geographical coordinates of the target), and mission time constraints (such as specifying the mission start time, end time, latest completion time, or time limits for each stage, etc.).

[0061] This invention decomposes task requirements, constructs a sub-task collaborative network graph, calculates the importance of sub-tasks, and dynamically sorts them according to their importance to achieve optimized task allocation. This ensures that key sub-tasks are executed by the most capable drones, significantly improving the overall operational efficiency of drone swarms.

[0062] Specifically, the task planning module includes:

[0063] The task decomposition unit is used to divide the task requirements based on spatial, functional, and temporal dimensions to generate multiple sub-tasks.

[0064] The subtask coordination unit is used to construct a subtask coordination network graph based on the multiple subtasks, G = (V, E), where V is a set of nodes representing subtasks, and E is a set of edges representing the relationship between any two subtasks; the relationship is the temporal dependency relationship between two adjacent subtasks.

[0065] The subtask evaluation unit is used to calculate the importance of each subtask based on the subtask collaboration network graph, and to sort the subtasks in descending order according to their importance to form a subtask sequence list. The subtask list includes subtask ID, subtask content, subtask scope, subtask hard constraints, and subtask importance.

[0066] In one embodiment of the present invention, spatial dimension-based division refers to task decomposition based on the geographic spatial characteristics of the task's operational range. This involves dividing the entire task area into multiple sub-regions based on factors such as terrain features, target distribution density, and task execution difficulty. For example, a large search area can be divided into multiple smaller search sub-regions based on indicators such as spatial distance, terrain complexity, or target density; or the space can be divided into high-altitude, mid-altitude, and low-altitude operational areas based on altitude information. By spatially dividing the task requirements, it ensures that the operational space of each sub-task is suitable for efficient execution by a single or small number of UAVs, while maintaining the integrity of overall spatial coverage.

[0067] In one embodiment of the present invention, functional-based division refers to decomposing tasks according to their functional types and execution processes. This involves breaking down different functional operations into independent sub-tasks, forming a functional chain for task execution. For example, a complete reconnaissance and strike mission can be decomposed into a preliminary reconnaissance sub-task, a mid-stage target identification and locking sub-task, and a final precision strike sub-task. By dividing task requirements according to functional dimensions, it ensures that each sub-task has a clear functional positioning and measurable execution standards.

[0068] In one embodiment of the present invention, time-based partitioning refers to temporal decomposition based on task time constraints and execution order, that is, dividing the overall task into multiple phased sub-tasks with temporal dependencies. Time-based partitioning is typically based on the following factors: the overall task time window, key time nodes, the logical order of task execution, and the time urgency of each phase. For example, the task can be divided into a preparation phase, an execution phase, and an evaluation phase; or it can be divided into urgent sub-tasks and routine sub-tasks based on the time urgency of the task. By partitioning task requirements according to the time dimension, the temporal continuity of sub-tasks and the rationality of resource scheduling can be ensured.

[0069] In one embodiment of the present invention, the calculation method of the sub-task evaluation unit is as follows:

[0070] The distance between two adjacent subtasks is determined based on the subtask collaboration network graph, and the influence of the subtask on the task operation range is determined based on the distance between the two adjacent subtasks.

[0071] The time urgency of subtasks is determined based on the aforementioned task time constraints;

[0072] The importance of a subtask is calculated based on the aforementioned impact and time urgency, using the following formula:

[0073]

[0074] Among them, I k P represents the importance of a subtask. base，k Let d(k, g) represent the basic importance score, α represent the influence weighting factor, g represent a neighboring subtask in N(k), N(k) represent the set of neighboring subtasks of subtask k in the subtask collaboration network graph, k represent any subtask in the subtask collaboration network graph, and d(k, g) represent the distance between subtask k and subtask g. dist T represents the distance adjustment constant, β represents the time urgency weighting factor, and T represents the distance adjustment constant. d，k T represents the latest completion time of subtask k. c D represents the current time. k Indicates the estimated execution time of subtask k, ∈ time This represents the time adjustment constant.

[0075] Understandable, the basic importance score P base，k It is pre-set based on the task type and the degree of impact of subtasks on the key objectives, and can be set according to the actual use scenario. This invention does not impose specific limitations on this. The degree of impact of subtasks on the task operation scope is determined by... This is reflected in the fact that the influence weighting factor α is an adjustment parameter, and the distance adjustment constant ∈ dist This is to avoid situations where a subtask is isolated or all d(k, g) are very large, causing the denominator to be 0, while the time urgency of the subtask is determined by... This is reflected in, This means that the less time remaining, the higher the time urgency, and the more the time adjustment constant ∈ time This is to avoid situations where there is plenty of time remaining, Approaching 0.

[0076] The collaborative decision-making module is used to calculate the comprehensive capabilities of each UAV, generate a task allocation scheme based on the comprehensive capabilities of each UAV and the importance of the sub-tasks, and send the task allocation scheme to the resource management module.

[0077] like Figure 2 As shown, specifically, the collaborative decision-making module includes:

[0078] The capability assessment unit is used to calculate the comprehensive capability of each drone in the drone swarm based on the drone's communication capability, payload capability, and endurance capability, and to determine the leader drone of the drone swarm based on the comprehensive capability.

[0079] The collaborative communication unit is used to build a drone network topology based on the Navigator drone.

[0080] The task allocation unit is used to allocate tasks based on the comprehensive capabilities of each UAV, the UAV network topology, and the importance of subtasks, and to generate a task allocation scheme.

[0081] The path planning unit is used by each UAV to plan and execute paths based on global collaborative rules.

[0082] Understandably, this invention quantifies the communication capabilities, payload capacity, and endurance of a drone to calculate its overall capabilities. The drone's communication capabilities are typically obtained through its hardware specifications and real-time status monitoring. Static parameters such as the type of communication equipment carried by the drone, maximum communication distance, maximum bandwidth, supported communication protocols, and anti-interference level are configured during drone initialization. Dynamic parameters such as signal strength, signal-to-noise ratio, and data transmission rate during mission execution are obtained through real-time monitoring by sensors.

[0083] The payload capacity of a drone is mainly determined by its physical design and current payload configuration, including its inherent attributes such as maximum takeoff weight, maximum payload capacity, and the type and number of available mount interfaces, as well as the type and weight of the mission payload already mounted on the drone.

[0084] The endurance of a drone is related to its energy system and current consumption status, including static parameters such as total battery capacity, engine type, and maximum flight time or range, as well as dynamic parameters such as the current percentage of remaining battery power, remaining fuel, and estimated remaining flight time.

[0085] In one embodiment of the present invention, constructing a drone network topology based on the Navigator drone specifically includes:

[0086] The navigator drone broadcasts its presence and network information (such as network ID, communication frequency, encryption key, etc.), and other drones listen for these broadcasts and attempt to join the network;

[0087] A regular drone sends a joining request to a navigator. The navigator approves or rejects the request according to preset rules (such as identity verification and capability assessment). After successfully joining, the drone registers its information with the navigator.

[0088] Once connected to the network, drones can also discover neighbors and establish direct communication links (if the communication range allows and the policy supports it), forming a multi-hop network or mesh network structure. All drones communicate only with the navigator, which is responsible for maintaining the topology information of the entire network.

[0089] The drone network topology can be star-shaped (centered on the navigator drone), tree-shaped (hierarchical structure), mesh-shaped, or hybrid topology. It can be specifically configured according to the drone's communication technology, the mission requirements of the drone swarm, and the complexity of the operating environment. This invention does not impose any specific limitations on this.

[0090] Furthermore, due to factors such as drone movement, energy consumption, signal interference, and the addition or removal of new drones, the drone network topology will dynamically change. Navigator drones need to update topology information periodically, for example, through periodic heartbeats and routing update protocols.

[0091] In one embodiment of the present invention, the formula for calculating the comprehensive capability is:

[0092] A i =w c ·Norm(C i )+w l ·Norm(L i )+w e ·Norm(Ei );

[0093] Among them, A i w represents the overall capability score of the i-th drone. c The weighting coefficients represent communication capabilities, Norm(·) represents the normalization function, and C... i w represents the communication capability of the i-th drone. l L represents the weighting factor for load capacity. i w represents the payload capacity of the i-th drone. e E represents the weighting coefficient for battery life. i This represents the current endurance of the i-th drone.

[0094] Understandably, the original units of measurement and numerical ranges for different capabilities vary considerably. For example, communication distance is measured in meters, battery life in minutes, and payload weight in kilograms. Therefore, Norm(·) is used for normalization, mapping them to a unified, dimensionless interval (usually [0,1] or [-1,1]). The weighting coefficient w... c w l and w e The settings can be configured according to actual usage requirements. For example, if the task has high requirements for real-time high-definition video transmission, the weight of communication capability will be higher. This invention does not impose specific limitations on this.

[0095] In one embodiment of the present invention, the specific steps of the task allocation unit are as follows:

[0096] The drones in the drone cluster are arranged in descending order of their comprehensive capabilities to form a drone queue;

[0097] Based on the aforementioned UAV network topology, determine the connection status and communication link quality between each UAV in the UAV cluster and the navigator UAV.

[0098] The drone queue is initially screened based on the hard constraints of each subtask in the subtask sequence list to form a first candidate drone sequence. The first candidate drone sequence is then screened a second time based on the connection status and communication link quality between each drone and the navigator drone to form a second candidate drone sequence. The hard constraints include communication capability requirements, load capacity requirements, and endurance requirements.

[0099] The subtask sequence list is matched one-to-one with the second candidate UAV sequence to generate a task allocation scheme.

[0100] The connection status refers to whether the drone is currently able to establish a valid communication connection with the navigator drone. If a connection cannot be established, it may not be suitable for performing subtasks that require frequent interaction with the navigator or rely on the navigator's commands. Communication link quality includes signal strength, signal-to-noise ratio, bandwidth, latency, and packet loss rate.

[0101] Understandably, communication capability requirements refer to the minimum communication distance, minimum bandwidth, or support for specific communication protocols (such as encrypted communication with ground stations) required for the drone to perform the sub-task. Payload capability requirements refer to the specific sensors (such as infrared cameras or LiDAR), specific tools (such as robotic arms), or minimum payload weight requirements needed to perform the sub-task. Endurance capability requirements refer to the minimum remaining flight time or minimum remaining battery percentage required to perform the sub-task (including round trip).

[0102] This invention performs a first screening based on the hard constraints of sub-tasks to ensure the basic feasibility of the task, and a second screening based on the connection status with the navigator and the quality of the communication link to ensure the reliability of collaborative command and information exchange during the task execution process.

[0103] In one embodiment of the present invention, the acknowledgment (ACK) mechanism can be used to determine whether data has been successfully delivered in the UAV network topology, thereby indirectly reflecting the connection status.

[0104] In one embodiment of the present invention, the communication link quality of the UAV network topology is determined by monitoring the parameters of the UAV physical layer and link layer.

[0105] In one embodiment of the present invention, the global coordination rules specifically include formation rules, motion coordination rules, and task synchronization rules, wherein,

[0106] The formation pattern rules form a formation pattern based on the task allocation scheme and the task operation scope, and calculate the formation pattern parameters, which include the distance between drones in the drone cluster and the formation orientation.

[0107] The motion coordination rule calculates the speed, spacing, and orientation of the drone swarm based on the speed, position, and orientation of the drones within the swarm. The calculation formula is as follows:

[0108]

[0109] in, This represents the adjusted velocity vector of drone i. Let α represent the current velocity vector of drone i. v N represents the speed adjustment factor. i Let j represent the set of neighboring nodes directly connected to UAV i in the communication network topology, and j represent UAV j. This represents the current velocity vector of drone j. Let β represent the position adjustment vector of drone i. p This represents the position adjustment factor. This indicates the current position of drone j. d represents the current position of drone i. des Indicates the desired spacing. This indicates the adjusted orientation of drone i. Indicates the current orientation of drone i, γ θ Indicates the orientation adjustment factor. Indicates the current orientation of drone j;

[0110] The task synchronization rules are used to coordinate the time synchronization and task execution synchronization of the drone cluster based on the speed, spacing, and orientation of the drone cluster.

[0111] Understandably, the formation is the most suitable formation selected from the ground control station based on the mission type and the size and shape of the operational area. The drone spacing refers to the distance between adjacent drones. The formation orientation refers to the direction of movement of the drone swarm. The paths of each drone can be initially planned using the A* algorithm, and then adjusted based on global coordination rules.

[0112] Drone i observes the speed of neighboring drone j and its own speed By comparing the relative motion trends of adjacent drone j with respect to drone i, and summing the velocity differences of all adjacent drones, if the average speed of adjacent drones is faster than that of drone i, then... Point ahead of drone i's current speed. If neighboring drones are on average slower than drone i, then... Point behind the current speed of drone i. Calculate the current distance between drone i and each of its neighboring drones j, and compare it with the expected distance to obtain the distance error. When the distance error is greater than 0, it means that the actual distance between drone i and its neighboring drone is greater than the expected distance, and they need to move closer; when the distance error is less than 0, it means that the actual distance between drone i and its neighboring drone is less than the expected distance, and they need to move further away. Drone i observes the orientation of its neighboring drone j. and towards itself By comparison, the orientation difference is obtained. use Quantify the orientation influence of drone j on drone i, and sum the orientation influences of all adjacent drones. This reflects how neighboring drones as a whole expect drone i to adjust its orientation.

[0113] In one embodiment of the present invention, time synchronization refers to synchronizing the time of each drone in the drone cluster when the speed, spacing and orientation of the drone cluster reach a first threshold, wherein the first threshold is set according to the specific use scenario of the drone cluster.

[0114] In one embodiment of the present invention, task execution synchronization means that when the speed, spacing and orientation of the drone cluster reach a second threshold, each drone in the drone cluster can proceed to the next task stage. The second threshold may be the same as or different from the first threshold, and it is specifically set according to the actual use scenario of the drone cluster.

[0115] The resource management module is used to determine the total task resources based on the task requirements of the UAV cluster, and generate resource allocation results based on the task allocation scheme and the total task resources.

[0116] Specifically, the steps of the resource management module include:

[0117] The system receives the task allocation result from the collaborative decision-making module and determines the resource requirements of each UAV in the UAV cluster when performing a task based on the task allocation result; wherein the resource requirements include resource requirement type, resource requirement quantity, and resource requirement time; the resources include power resources, communication bandwidth resources, sensor resources, and payload resources.

[0118] The resource requirements of each UAV during mission execution are summarized to form the total resource requirements;

[0119] Compare the total resource requirements with the total resources required for the task:

[0120] If the total resource requirement does not exceed the total resource requirement of the task, then the resources will be allocated according to the resource requirement.

[0121] If the total resource requirement exceeds the total task resources, the priority of each UAV resource request is determined by the expected importance of the subtasks, and resources are allocated according to the priority of the UAV resource requests.

[0122] Understandably, the total mission resources are the resource consumption limit set by the UAV swarm command system for the entire UAV swarm. For example, power resources can be represented by the total available flight hours, communication bandwidth resources can be represented by the total available peak bandwidth, sensor resources can be represented by the maximum number of sensors of a specific type that can be used simultaneously, and payload resources can be represented by the number of payloads of a specific type that can be used simultaneously.

[0123] The priority of drone resource requests is primarily based on the importance of the subtasks currently assigned to the drone. Since the subtask sequence list is already arranged in descending order of importance, drones executing higher-importance subtasks receive higher priority for their resource requests. If a drone is assigned multiple subtasks, its priority should be determined by the level of the most important subtask it is executing.

[0124] This invention determines the priority of UAV resource requests based on the importance of subtasks and allocates resources according to the priority of UAV resource requests. At the same time, the execution module guides and adjusts the behavior of the UAV cluster in real time based on global coordination rules. This not only improves the utilization efficiency of limited resources, but also ensures the overall coordination of the cluster in complex dynamic environments.

[0125] In one embodiment of the present invention, the resource requirements of each UAV during mission execution are summarized to form a total resource requirement, which specifically includes:

[0126] The total power requirement is obtained by estimating the power consumption or data transmission volume of each drone for its assigned sub-tasks.

[0127] Analyze the execution time windows of all subtasks, find the sum of the bandwidth requirements of concurrently executed subtasks at all time points, and take the maximum value as the peak total requirement.

[0128] For each specific type of sensor and / or payload, determine the maximum number of drones that need to be used simultaneously at any given moment during mission execution.

[0129] The execution module is used to guide the drone cluster to execute tasks based on the task allocation scheme and resource allocation results.

[0130] Specifically, the execution module transforms the task allocation scheme and resource allocation results into a sequence of underlying navigation, payload control, and other instructions that can be executed by each UAV flight control system. During the guidance of task execution, the module adjusts the instructions in real time according to global coordination rules to maintain the coordination of the cluster and the synchronization of tasks.

[0131] In one embodiment of the present invention, a security monitoring module is further included, wherein the security monitoring module includes:

[0132] The status monitoring unit is used to monitor the parameters of the work area and the operating parameters of the UAV cluster in real time. The operating parameters include formation parameters, speed, spacing, orientation of the UAV cluster, and resource allocation results of the UAV cluster.

[0133] The risk warning unit is used to perform abnormal state analysis based on the results of the status monitoring unit, generate warning information, and provide risk warning prompts based on the warning information.

[0134] Understandably, parameters of the operational area are obtained through ground monitoring stations, including weather conditions (wind speed, precipitation, visibility), electromagnetic environment (interference intensity), and whether there are unknown obstacles or non-cooperative aircraft in the airspace.

[0135] The abnormal state analysis specifically includes:

[0136] A safety threshold is set, and the results of the status monitoring unit are compared with the safety threshold. If the safety threshold is exceeded, the drone swarm is judged to be in abnormal flight, and an early warning message is generated. The early warning message includes an early warning ID, a timestamp, and a risk level. The risk level can be set according to the specific use case of the drone swarm, and this invention does not impose any specific restrictions on it.

[0137] This invention sets up a safety monitoring module to monitor environmental parameters of the work area and various operating parameters of the drone swarm in real time. By comparing these parameters with preset safety thresholds to analyze abnormal states, it can detect and identify potential dangers in advance, effectively reducing the probability of safety accidents. This significantly improves the safety and robustness of the entire drone swarm collaborative operation system in complex environments.

[0138] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system, characterized in that, include: The task planning module is used to receive task requirements from the drone swarm, decompose the task requirements into sub-tasks, and determine the importance of the sub-tasks. The task planning module specifically includes: The task decomposition unit is used to divide the task requirements based on spatial, functional, and temporal dimensions to generate multiple sub-tasks. A subtask collaboration unit is used to construct a subtask collaboration network graph based on the multiple subtasks; The subtask evaluation unit is used to calculate the importance of each subtask based on the subtask collaboration network graph, and to sort each subtask in descending order according to the importance of each subtask to form a subtask sequence list, wherein the subtask list includes subtask ID, subtask content, subtask operation scope, subtask hard constraints and subtask importance. The calculation method for the sub-task evaluation unit is as follows: The distance between two adjacent subtasks is determined based on the subtask collaboration network graph, and the influence of the subtask on the task operation range is determined based on the distance between the two adjacent subtasks. Determine the time urgency of subtasks based on task time constraints; The importance of a subtask is calculated based on the aforementioned impact and time urgency, using the following formula: ； in, Indicates the importance of subtasks. This indicates the basic importance score. Indicates the influence weight factor. express One of the adjacent subtasks, Let k represent the set of adjacent subtasks of subtask k in the subtask collaboration network graph. This represents any subtask in the subtask collaboration network graph. This represents the distance between subtasks k and g. Represents the distance adjustment constant. Indicates the weighting factor for time urgency. This represents the latest completion time of subtask k. Indicates the current time. This indicates the estimated execution time of subtask k. Represents the time adjustment constant; A collaborative decision-making module is used to calculate the overall capabilities of each UAV, generate a task allocation scheme based on the overall capabilities of each UAV and the importance of sub-tasks, and send the task allocation scheme to the resource management module; the collaborative decision-making module specifically includes: The capability assessment unit is used to calculate the comprehensive capability of each drone in the drone swarm based on the drone's communication capability, payload capability, and endurance capability, and to determine the leader drone of the drone swarm based on the comprehensive capability. The collaborative communication unit is used to build a drone network topology based on the Navigator drone. The task allocation unit is used to allocate tasks based on the comprehensive capabilities of each UAV, the UAV network topology, and the importance of subtasks, and to generate a task allocation scheme. The path planning unit is used by each UAV to plan its execution path based on global collaborative rules. The specific steps of the task allocation unit are as follows: The drones in the drone cluster are arranged in descending order of their comprehensive capabilities to form a drone queue; Based on the aforementioned UAV network topology, determine the connection status and communication link quality between each UAV in the UAV cluster and the navigator UAV. The drone queue is initially screened based on the hard constraints of each subtask in the subtask sequence list to form a first candidate drone sequence. A second screening is then conducted based on the connection status and communication link quality between each drone and the navigator drone to form a second candidate drone sequence. The hard constraints include communication capability requirements, load capacity requirements, and endurance requirements. Connection status refers to whether the drone can currently establish a valid communication connection with the navigator drone. If a connection cannot be established, the drone may not be suitable for performing subtasks requiring frequent interaction with the navigator or relying on navigator commands. Communication link quality includes signal strength, signal-to-noise ratio, bandwidth, latency, and packet loss rate. The subtask sequence list is matched one-to-one with the second candidate UAV sequence to generate a task allocation scheme; The first screening is based on the hard constraints of the sub-tasks to ensure the basic feasibility of the task, and the second screening is based on the connection status with the navigator and the quality of the communication link to ensure the reliability of coordinated command and information exchange during the task execution process. The global coordination rules specifically include formation rules, motion coordination rules, and task synchronization rules, among which... The formation pattern rules form a formation pattern based on the task allocation scheme and the task operation scope, and calculate the formation pattern parameters, which include the distance between drones in the drone cluster and the formation orientation. The motion coordination rule calculates the speed, spacing, and orientation of the drone swarm based on the speed, position, and orientation of the drones within the swarm. The calculation formula is as follows: ； ； ； in, This represents the adjusted velocity vector of drone i. This represents the current velocity vector of drone i. Indicates the speed adjustment coefficient. Let represent the set of neighboring nodes directly connected to UAV i in the communication network topology. Indicates drone j, This represents the current velocity vector of drone j. This represents the position adjustment vector of drone i. This represents the position adjustment factor. This indicates the current position of drone j. This indicates the current position of drone i. Indicates the desired spacing. This indicates the adjusted orientation of drone i. Indicates the current orientation of drone i. Indicates the orientation adjustment factor. Indicates the current orientation of drone j; The task synchronization rules are used to coordinate the time synchronization and task execution synchronization of the drone cluster based on the speed, spacing and orientation of the drone cluster. The resource management module is used to determine the total task resources based on the task requirements of the UAV cluster, and generate resource allocation results based on the task allocation scheme and the total task resources. The specific steps of the resource management module include: The system receives the task allocation result from the collaborative decision-making module and determines the resource requirements of each UAV in the UAV cluster when performing a task based on the task allocation result; wherein the resource requirements include resource requirement type, resource requirement quantity, and resource requirement time; the resources include power resources, communication bandwidth resources, sensor resources, and payload resources. The resource requirements of each UAV during mission execution are summarized to form the total resource requirements; Compare the total resource requirements with the total resources required for the task: If the total resource requirement does not exceed the total resource requirement of the task, then the resources will be allocated according to the resource requirement. If the total resource requirement exceeds the total task resources, the priority of each UAV resource request is determined by the expected importance of the subtasks, and resources are allocated according to the priority of the UAV resource requests. The execution module is used to guide the drone cluster to execute tasks based on the task allocation scheme and resource allocation results.

2. The intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system as described in claim 1, characterized in that: The task requirements include task content, task scope, target coordinates, and task time constraints.

3. The intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system as described in claim 2, characterized in that: The formula for calculating the comprehensive capability is as follows: ； in, This represents the overall capability score of the i-th drone. Weighting coefficients representing communication capabilities This represents the normalization processing function. This represents the communication capability of the i-th drone. Weighting coefficients representing load capacity This represents the payload capacity of the i-th drone. The weighting coefficients representing battery life. This represents the current endurance of the i-th drone.

4. The intelligent unmanned aerial vehicle (UAV) swarm collaborative operation system as described in claim 1, characterized in that: It also includes a security monitoring module, wherein the security monitoring module includes: The status monitoring unit is used to monitor the parameters of the work area and the operating parameters of the drone swarm in real time. The operating parameters include formation parameters, speed, spacing, orientation of the drone swarm, and resource allocation results of the drone swarm. The risk warning unit is used to perform abnormal state analysis based on the results of the status monitoring unit, generate warning information, and provide risk warning prompts based on the warning information.

Images