Multi-unmanned aerial vehicle multi-target task planning method and system and electronic equipment

Abstract

The embodiment of the invention provides a multi-unmanned aerial vehicle multi-target task planning method and system and electronic equipment, and relates to the field of unmanned aerial vehicles, and the method comprises the steps: generating a multi-unmanned aerial vehicle task distribution diagram; state information and environment information of the unmanned aerial vehicle are collected and preprocessed; obtaining constraint conditions, and integrating the constraint conditions into the task allocation model; on the basis of the constraint conditions, node feature updating is carried out on the multi-unmanned aerial vehicle task allocation graph by using a graph neural network, and a first task allocation strategy of each unmanned aerial vehicle is generated; a reinforcement learning algorithm is adopted to optimize the first task allocation strategy, and a second task allocation strategy of each unmanned aerial vehicle is generated; and performing real-time adjustment on the first task allocation strategy and / or the second task allocation strategy according to the task execution condition and the environment change of each unmanned aerial vehicle, and generating a third task allocation strategy of each unmanned aerial vehicle. According to the embodiment of the invention, efficient and flexible multi-unmanned aerial vehicle task planning can be realized.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicles, and in particular to a multi-unmanned aerial vehicle multi-objective task planning method, system and electronic equipment. Background Art

[0002] With the rapid development of science and technology, unmanned aerial vehicles are increasingly used in all walks of life. Especially in the field of low-altitude economy, multi-UAV systems have become an important tool for achieving multi-target tasks due to their high efficiency, flexibility and low cost. Low-altitude economy covers multiple application scenarios such as environmental monitoring, logistics distribution, agricultural spraying, urban security, infrastructure inspection, disaster emergency response and entertainment activities. In these applications, the collaborative work of multiple UAVs can significantly improve the efficiency and effectiveness of task execution and meet complex and changing work needs.

[0003] Existing multi-UAV mission planning methods mainly focus on task allocation, path planning, and resource management. However, with the increase in mission complexity and the number of UAVs, traditional methods face many challenges: 1) Complexity of task allocation: In multi-UAV multi-objective tasks, how to efficiently allocate tasks to different UAVs to fully utilize their respective resources and capabilities is a complex problem. Traditional task allocation methods often rely on predefined rules or simple optimization algorithms, which are difficult to adapt to dynamically changing task requirements and environmental conditions.

[0004] 2) Dynamic nature of path planning: The low-altitude environment is complex and changeable, such as weather changes, terrain obstacles and other dynamic obstacles, which make path planning more difficult. Existing path planning algorithms such as the A* algorithm and the Dijkstra algorithm, although they perform well in static environments, often lack the ability to adjust in real time in dynamic environments and cannot effectively respond to emergencies.

[0005] 3) Optimization of resource management: Multi-UAV systems need to reasonably manage resources such as battery power, communication bandwidth, and load capacity to ensure the successful completion of the mission. Traditional methods often lack comprehensiveness and real-time performance in resource management, making it difficult to achieve optimal resource allocation in complex tasks.

[0006] 4) Real-time collaboration: In a multi-UAV system, collaboration between UAVs is crucial to the success of the mission. However, existing methods are insufficient in achieving real-time collaboration and information sharing, resulting in delays or conflicts during mission execution.

[0007] In order to solve the above problems, advanced artificial intelligence technologies such as graph neural networks and deep reinforcement learning have been introduced into multi-UAV mission planning in recent years. These technologies achieve more efficient and flexible task allocation and path planning by modeling the complex relationship between UAVs and tasks and learning optimization strategies. However, existing research still has the following shortcomings: 1) Insufficient comprehensive consideration of constraints: Existing methods often fail to fully consider constraints such as the battery power, environmental information, and resource consumption of the drone during task allocation and path planning, resulting in feasibility and efficiency issues in practical applications of task allocation schemes.

[0008] 2) Limited real-time feedback adjustment capabilities: In a dynamic environment, task execution and environmental conditions will continue to change. Existing methods lack an effective real-time feedback mechanism and are unable to dynamically adjust task allocation plans based on real-time information, affecting the overall execution effect of the task.

[0009] 3) Insufficient multi-agent collaborative optimization: Although multi-agent reinforcement learning can theoretically optimize task allocation strategies through the collaboration and competition of multiple agents, in practical applications, how to efficiently achieve collaborative optimization among multiple drones remains a problem that needs to be solved. Summary of the invention

[0010] The purpose of the embodiments of the present invention is to provide a multi-UAV multi-objective mission planning method, system and electronic device to achieve efficient and flexible multi-UAV mission planning. The specific technical solution is as follows: In a first aspect, an embodiment of the present invention provides a multi-UAV multi-objective mission planning method, the method comprising: Generate a multi-UAV task allocation graph; wherein the nodes in the multi-UAV task allocation graph represent multiple UAVs and multiple tasks, and the edges in the multi-UAV task allocation graph represent the relationship between tasks and UAVs; Collect and pre-process the status information and environmental information of the drone, the status information includes: battery power, location information, and the environmental information includes: weather conditions, terrain characteristics, and obstacle distribution; Obtaining constraint conditions and integrating the constraint conditions into a task allocation model; Based on the constraints integrated in the task allocation model, the state information of each UAV and the environmental information, a graph neural network is used to update the node features of the multi-UAV task allocation graph to generate a first task allocation strategy for each UAV; Based on the feedback of rewards and penalties for task execution by each drone, the first task allocation strategy is optimized using a reinforcement learning algorithm to generate a second task allocation strategy for each drone; According to the environmental information and the status information of each UAV, the first task allocation strategy and / or the second task allocation strategy are adjusted in real time to generate a third task allocation strategy for each UAV.

[0011] In one embodiment of the present invention, the constraint conditions include battery power constraint, environmental information constraint and resource consumption constraint; The battery power constraint is used to ensure that the task allocation will not cause the drone to run out of power. The battery power constraint is expressed by the following formula: ; in, Indicates The remaining battery power of the drone, The minimum remaining power required for the drone to perform a mission; The environmental information constraints are used to optimize the flight path and mission execution plan of the UAV, wherein the flight path of the UAV is optimized using the following path planning algorithm: ; in, For the The flight path of the drone, for Star algorithm, used to calculate the optimal flight path, Indicates the current environment information. Indicates missions performed by drones; The resource consumption constraint is used to optimize the task allocation of the UAV to improve the resource utilization efficiency. The resource utilization efficiency is expressed by the following formula: ; in, For resource efficiency, is the number of drones, For the The number of missions performed by each drone, For the The total amount of resources consumed by each drone.

[0012] In one embodiment of the present invention, the graph neural network is a graph attention network, and the step of updating node features of the multi-UAV task allocation graph using the graph neural network includes: Initialize each node feature in the multi-UAV task allocation graph according to the state information or task requirements of the UAV; For each neighbor node of any target node in the multi-UAV task allocation graph, calculating the attention weight of the neighbor node to the target node; Use the graph attention network to update node features.

[0013] In one embodiment of the present invention, the graph neural network is one of a graph convolutional network, a graph isomorphism network, a graph sampling and aggregation generator GraphSAGE.

[0014] In one embodiment of the present invention, the reinforcement learning algorithm is a deep reinforcement learning algorithm, and the reward and punishment of task execution are fed back by each drone, and the first task allocation strategy is optimized by the reinforcement learning algorithm, including: Train each drone as an independent agent, selecting actions based on the drone’s current state when performing a task; Determine the reward value and next state of the drone after executing the action; Record and construct trajectories based on the current state, action, reward, and next state for each time period; Determine the cumulative return based on the trajectory corresponding to each time period; The first task allocation strategy is updated according to the accumulated reward and the maximized expected reward.

[0015] In one embodiment of the present invention, real-time adjustment of the first task allocation strategy and / or the second task allocation strategy includes: The first task allocation strategy and / or the second task allocation strategy are adjusted in real time in at least one of the following ways: When a preset condition is triggered during the execution of a task by the first drone, the task of the first drone is assigned to the second drone, wherein the preset condition indicates that the execution of the task by the drone is blocked; Adjust task priorities based on the status of the drone, the drone’s resources, and the urgency of the task; Adjust the drone’s flight path and / or mission based on current environmental information.

[0016] In a second aspect, an embodiment of the present invention provides a multi-UAV multi-objective mission planning system, the system comprising a server and a UAV; The server is used to perform multi-UAV multi-objective task planning according to any of the multi-UAV multi-objective task planning methods above to generate a task allocation strategy, and send the task allocation strategy to the UAVs; The drone is used to receive the task allocation strategy and execute the task according to the task allocation strategy.

[0017] In a third aspect, an embodiment of the present invention provides a multi-UAV multi-objective mission planning device, the device comprising: A task allocation graph generation module is used to generate a multi-UAV task allocation graph; wherein the nodes in the multi-UAV task allocation graph represent multiple UAVs and multiple tasks, and the edges in the multi-UAV task allocation graph represent the relationship between tasks and UAVs; An information collection and processing module is used to collect and pre-process the state information and environmental information of the drone. The state information includes: battery power, location information, and the environmental information includes: weather conditions, terrain features, and obstacle distribution. The state information and environmental information are used for graph neural network processing; A constraint condition acquisition and integration module, used for acquiring constraint conditions and integrating the constraint conditions into the task allocation model; A first task allocation strategy generation module is used to update node features of the multi-UAV task allocation graph using a graph neural network based on the constraints integrated in the task allocation model, the state information of each UAV and the environmental information, and generate a first task allocation strategy for each UAV; A second task allocation strategy generation module is used to generate a second task allocation strategy for each drone by optimizing the first task allocation strategy using a reinforcement learning algorithm through feedback of rewards and penalties for task execution by each drone; The third task allocation strategy generation module is used to adjust the first task allocation strategy and / or the second task allocation strategy in real time according to the environmental information and the status information of each drone, and generate a third task allocation strategy for each drone.

[0018] In one embodiment of the present invention, the constraint conditions include battery power constraint, environmental information constraint and resource consumption constraint; The battery power constraint is used to ensure that the task allocation will not cause the drone to run out of power. The battery power constraint is expressed by the following formula: ; in, Indicates The remaining battery power of the drone, The minimum remaining power required for the drone to perform a mission; The environmental information constraints are used to optimize the flight path and mission execution plan of the UAV, wherein the flight path of the UAV is optimized using the following path planning algorithm: ; in, For the The flight path of the drone, for Star algorithm, used to calculate the optimal flight path, Indicates the current environment information. Indicates missions performed by drones; The resource consumption constraint is used to optimize the task allocation of the UAV to improve the resource utilization efficiency. The resource utilization efficiency is expressed by the following formula: ; in, For resource efficiency, is the number of drones, For the The number of missions performed by each drone, For the The total amount of resources consumed by each drone.

[0019] Beneficial effects of the embodiments of the present invention: This invention solves the problems of complex task allocation, path planning, resource management and collaborative optimization in multi-UAV multi-objective mission planning by combining graph neural network and deep reinforcement learning technology. Compared with traditional methods, this invention has the following advantages: 1) Efficient task allocation: Through deep modeling of the relationship between tasks and drones through graph neural networks, task allocation plans can be accurately generated, and on this basis, task execution strategies can be optimized through reinforcement learning.

[0020] 2) Strong dynamic adaptability: Through the real-time feedback mechanism, the present invention can dynamically adjust the task allocation plan according to environmental changes and task execution status, ensuring that tasks are completed efficiently in a complex dynamic environment.

[0021] 3) Resource utilization optimization: By integrating constraints such as battery power and environmental information, resource consumption can be effectively optimized to avoid excessive consumption of battery power or other resources by drones, thereby improving the feasibility and efficiency of mission execution.

[0022] 4) Flexibility and scalability: The task planning method and system of the present invention can be flexibly applied to a variety of low-altitude economic scenarios, such as logistics distribution, environmental monitoring, agricultural spraying, security inspection, etc., and can adapt to changes in the complexity of different tasks and the number of drones. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0024] Figure 1 A schematic diagram of a process flow of a multi-UAV multi-objective mission planning method provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a multi-UAV multi-objective mission planning system provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a multi-UAV multi-objective mission planning device provided by an embodiment of the present invention; Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field based on the present invention belong to the scope of protection of the present invention.

[0026] See also Figure 1 , which is a flow chart of a multi-UAV multi-objective mission planning method provided in an embodiment of the present application. The method can be applied to electronic devices such as servers and smart terminals. The method includes steps S11-S16.

[0027] S11, generating a multi-UAV task allocation graph.

[0028] The nodes in the multi-UAV task allocation graph represent multiple UAVs and multiple tasks, and the edges in the multi-UAV task allocation graph represent the relationship between tasks and UAVs.

[0029] This process is the process of graph modeling. In an embodiment of the present invention, it is first necessary to construct a multi-UAV task allocation graph, in which UAVs and tasks are represented as nodes in the graph respectively, and the relationship between tasks and UAVs is connected through edges in the graph.

[0030] The structure of the graph model is as follows: Drone node: Each drone node contains the current status information of the drone, such as location information, battery power, flight speed, etc.

[0031] Task node: Each task node contains basic information about the task, such as task location, task priority, task type, etc.

[0032] Edge establishment: The edge between the task node and the drone node represents the potential correlation between the task and the drone. The weight of the edge can be calculated based on factors such as distance and task priority.

[0033] S12, collects and preprocesses environmental information, drone status information and mission information.

[0034] In the embodiment of the present invention, the following data is collected and processed to ensure that all data can be correctly input into the graph neural network: Drone status data: battery power, location information, speed information, etc. Environmental information data: weather conditions, obstacle locations, and terrain information; Task information data: task location, task priority, task urgency, etc.

[0035] These data need to be cleaned, denoised, and standardized to be converted into a format suitable for graph neural network input. For example, the battery power of a drone can be normalized based on the maximum power to ensure the consistency of data input.

[0036] In particular, in order to ensure the accuracy of the data, a data matrix can be constructed based on the normalized data, and a standard data matrix can also be constructed for the data under normal circumstances. Then, the correlation values ​​between the data matrix and the standard data matrix are calculated respectively according to the Pearson coefficient. When the correlation value is within the preset range, it is determined that the normalized data meets the requirements, otherwise further processing is required, such as retrospective inspection, etc.; or based on the data matrix, the normalized data is used as grayscale values, and the pixel points corresponding to the grayscale values ​​are filled according to the positions of these data in the data matrix, thereby obtaining a grayscale pixel map of these data, and then the grayscale pixel map is input into the neural network model, and its confidence is calculated by the softmax function. When the confidence is within the preset range, it is determined that the data meets the requirements, otherwise further processing is performed, and so on, without limitation here.

[0037] S13, obtaining constraint conditions and integrating the constraint conditions into the graph neural network.

[0038] In the embodiment of the present invention, the following constraints are considered to ensure that the task allocation scheme is feasible and efficient in actual execution: 1) Battery power constraint, based on the remaining power of each drone , to ensure that the task allocation will not cause the UAV to run out of power. The constraint is expressed as: ; in, Indicates The remaining battery power of the drone, The minimum remaining power required for the drone to perform a mission; 2) Environmental information constraints: optimize the flight path and mission execution plan of the UAV based on the current environmental information (such as weather, terrain, obstacles, etc.), and the path planning is based on an appropriate path planning algorithm; ; in, For the The flight path of the drone, for Star algorithm, used to calculate the optimal flight path, Indicates the current environment information. Indicates missions performed by drones; 3) Resource consumption constraints: Considering the resource consumption of the UAV (such as fuel, communication bandwidth, etc.), optimize task allocation to maximize resource utilization efficiency. It is expressed as: ; in, is the number of drones, For the The number of missions performed by each drone, For the The total amount of resources consumed by each drone.

[0039] S14, based on the constraint conditions, the status information and task information of each drone, and the environmental information, the graph neural network is used to update the node features of the multi-drone task allocation graph to generate a first task allocation strategy for each drone.

[0040] This step is a graph neural network processing step, in which the multi-UAV task allocation graph is processed using the graph neural network. Specifically, the graph neural network can be a graph attention network, and step S14 can be specifically performed through the following steps.

[0041] S141, initializing the characteristics of each node in the multi-UAV task allocation graph according to the status of the UAV (such as battery power, location, task type, etc.) or the requirements of the task (such as task priority, resource requirements, etc.); S142. For each neighbor node , calculate its pair node The attention weight , specifically including: First, the node The characteristics and remaining power are combined: ; in, Respectively represent nodes The characteristic vector of Representation Node The remaining power, Represents a vector concatenation operation; Then, using a weight matrix Perform linear transformation on the concatenated features: ; in, is a learnable weight matrix, and a is a learnable parameter vector, which represents the process of converting the concatenated features into a scalar.

[0042] Get the raw attention score After that, the LeakyReLU activation function is usually used to perform nonlinear transformation to obtain the final attention weight: ; Next, we adjust the weights of the attention scores to ensure that drones (nodes) with lower battery power have less impact on task allocation. Specifically, we use a weight factor To weight the attention score: ; in, is the adjustment factor of the battery power, which indicates the influence of the battery power on the attention weight, and can be defined as: ; in, is the maximum battery capacity of this type of drone (i.e. the maximum capacity under normal circumstances); Finally, the softmax function is used to normalize the attention weights of all neighbor nodes: ;

[0043] S143. Update node features using graph structure, attention mechanism and actual constraints: ; in, It is a function that combines the environment information with the characteristics of the current node and adjusts the update of node characteristics. is also a function that takes into account the resource consumption of each node and adjusts the update of node characteristics. and is the weight coefficient, which is used to adjust the influence of each constraint condition in feature update. Representation Node In the The characteristics of the layer, Representation Node The set of neighbor nodes of For Node For Node In the The attention weights of the layers, For the The weight matrix of the layer is used to transform the features of neighbor nodes. Representation Node In the The characteristics of the layer, Represents the activation function, which is used to perform nonlinear transformation on features.

[0044] Through the graph attention operation, each node in the graph will update its features according to the information of adjacent nodes, thereby capturing the complex relationship between drones and tasks and generating a preliminary task allocation plan for each drone, namely the first task allocation strategy.

[0045] In some embodiments of the present invention, the graph neural network may also be one of a graph convolutional network, a graph isomorphism network, a graph sampling and aggregation generator GraphSAGE.

[0046] S15, based on the feedback of rewards and penalties of each UAV for task execution, the first task allocation strategy is optimized by using a reinforcement learning algorithm to generate a second task allocation strategy.

[0047] This step is the process of reinforcement learning optimization. Specifically, the task allocation scheme is optimized through deep reinforcement learning. Under the multi-agent reinforcement learning framework, a reward function is designed to guide the drone to select the optimal task allocation strategy. The specific steps of this process are as follows: S151. Define the environment, state, and action space of reinforcement learning: 1) Status : The state of each drone at each time step t 2) Action : Actions the drone can take 3) Rewards : Give corresponding rewards according to the action results of each drone S152, Initialize strategy, define a parameterized strategy , such as defining the first task allocation strategy as a parameterized strategy, using a deep neural network architecture to approximate the strategy, and the input of the neural network is the current state , the output is the probability distribution of each action in this state; deep neural network architectures include but are not limited to multi-layer perceptrons and convolutional neural networks.

[0048] S153, multi-agent interaction and collaboration. Each UAV is trained as an independent agent and selects corresponding actions based on the current state when performing tasks. There is both cooperation and competition among multiple UAVs. When selecting actions, each UAV takes into account the strategies and mission status of other UAVs.

[0049] S154. Interact with the environment and collect experience, including: 1) Each drone has a strategy at the current moment Select an action .

[0050] 2) After the drone performs the action, the environment feedbacks the new state and rewards .

[0051] 3) Record the state, action, reward, and next state of each time step to form a trajectory:

[0052] S155. Calculate cumulative returns , at each time step , calculate the cumulative return starting from this time step based on the reward obtained : ; in, is a discount factor that controls the impact of future rewards on current decisions, For the The reward value at that moment, The task termination time; S156. Use the policy gradient method to update policy parameters , the goal is to maximize the expected return , the update rule of the strategy is: ; in, is the expected return of the strategy, is the strategy parameter, For strategy, The trajectory of the drone; S156. Continuously optimize strategies through multiple interactions and training .

[0053] S16, adjusting the first task allocation strategy or the second task allocation strategy in real time according to the environmental information, the status information of each UAV, and the task information, to generate a third task allocation strategy.

[0054] This step is the process of real-time feedback adjustment. During the task execution, the system adjusts the task allocation plan based on real-time feedback information. For example, the drone's battery power, power consumption, task execution progress, environmental changes and other data will be continuously monitored to adjust the task allocation in real time. The specific adjustment strategies are as follows: Dynamically optimize the task allocation strategy. Based on real-time feedback information, dynamically adjust the task allocation strategy including: 1) Reassign tasks: If the execution of tasks of some drones is hindered (such as low battery, flight path blocked by obstacles, etc.), reallocate the tasks to other drones.

[0055] 2) Adjust priority: Dynamically adjust the priority of the task according to the urgency of the task, resource conditions or the status of the drone.

[0056] 3) Path and action adjustment: If the environment changes (such as weather changes, obstacle movement, etc.), adjust the drone's flight path or action plan based on the new environmental information.

[0057] In one embodiment of the present invention, further optimization can be performed, and the real-time feedback adjustment is usually achieved in the following manner: 1) Dynamic adjustment of reinforcement learning: Through real-time reward and punishment mechanisms, the task allocation strategy is continuously adjusted using reinforcement learning algorithms. After each adjustment, feedback is provided based on task completion and efficiency to further optimize decisions.

[0058] 2) Prediction and real-time correction: The system can use predictive models to estimate the completion time and resource requirements of tasks, and correct these predictions based on real-time feedback.

[0059] 3) Heuristic algorithm: Quickly adjust task allocation through simple rules or heuristic methods (for example, based on information such as remaining mission time and drone status).

[0060] The following describes the multi-UAV multi-objective mission planning method provided by an embodiment of the present invention in combination with three different application scenarios.

[0061] In the following examples, three specific application scenarios of the multi-UAV multi-objective mission planning method are demonstrated. These scenarios involve different mission types, environmental conditions, and resource constraints, illustrating the wide adaptability and high efficiency of the method of the present invention.

[0062] Scenario 1: Drone swarms perform mapping tasks Mission description: In this scenario, multiple drones need to collaborate to perform mapping tasks, with the goal of high-precision terrain mapping of a certain area. There are multiple target locations distributed in the mission area, and each target location needs to be scanned in detail by a drone.

[0063] Specific steps: Graph modeling: The system first builds a task allocation graph. The drone node contains the location information, battery power and other status information of each drone, and the task node contains information such as the target location and task priority.

[0064] Data collection and preprocessing: Collect information about each drone’s battery level, flight speed, location, mission priority, and terrain features. Adjust the weight of mission nodes based on weather conditions (such as wind speed and visibility).

[0065] Constraint analysis and integration: Integrate battery power constraints and flight path constraints to ensure that the UAV task allocation is consistent with the endurance. The priority of the task node is dynamically adjusted according to the complexity of the target terrain.

[0066] Graph neural network processing: Updates the connection between drones and tasks through graph convolution operations and generates a preliminary task allocation plan. The system will optimize the allocation results by considering task priorities, flight paths, and environmental constraints.

[0067] Reinforcement learning optimization: Multi-agent reinforcement learning is used to optimize the task allocation scheme. Each drone adjusts its flight path based on the status feedback of task execution to maximize the overall task completion efficiency.

[0068] Real-time feedback adjustment: Adjust task allocation based on real-time battery power and flight progress. When a drone’s battery is close to running out, the system will reallocate tasks to ensure that each drone completes the appropriate task.

[0069] Technical effect: Through the method of the present invention, the UAV can efficiently complete surveying and mapping tasks in complex terrain and dynamic environments, adjust the task allocation plan in real time, and ensure the efficiency and accuracy of surveying and mapping.

[0070] Scenario 2: Mission planning of drone swarms in post-disaster rescue Mission Description: After a natural disaster, multiple drones need to work together to perform post-disaster search and rescue missions. The mission includes assessing key areas in the disaster area, searching for trapped people, and providing real-time data feedback to the ground command center.

[0071] Specific steps: Graph modeling: The drone node includes information such as the drone’s battery level, flight time, and current location; the mission node includes key locations in the disaster area and mission priorities (such as rescue personnel, dangerous area assessment, etc.).

[0072] Data collection and preprocessing: Collect environmental information (such as wind speed, weather conditions, obstacles in the area, etc.) and preprocess this data into a format suitable for graph neural networks. Dynamically adjust the weights of task nodes based on factors such as weather, building distribution, and disaster severity.

[0073] Constraint analysis and integration: Battery power, power consumption, communication bandwidth, etc. will be analyzed as constraints. The flight path is also affected by obstacles and weather conditions, so environmental information needs to be integrated into the task allocation model to ensure that the UAV can successfully perform tasks in complex environments.

[0074] Graph Neural Network Processing: The task allocation plan for each drone is updated through a graph neural network, dynamically taking into account the drone’s battery level, task priority, and environmental changes. The task allocation plan is adjusted in real time to ensure that all tasks can be completed as efficiently as possible in a post-disaster environment.

[0075] Reinforcement learning optimization: The system optimizes the task allocation strategy through multi-agent reinforcement learning. Each drone, as an agent, adjusts its task execution strategy according to its own task situation, environmental changes and feedback information to maximize the efficiency of task completion.

[0076] Real-time feedback adjustment: According to the real-time situation in the disaster area (such as changes in the disaster area environment, changes in the task execution status, etc.), the system will adjust the task allocation plan in real time. Through the feedback mechanism, the drone can dynamically optimize the task allocation according to the real-time battery power, task urgency and environmental changes.

[0077] Technical effect: Through the method of the present invention, multiple drones can collaborate efficiently in post-disaster rescue scenarios, ensuring that each task is completed according to priority and resource constraints, improving rescue efficiency, and helping the command center obtain key data from the disaster area in a timely manner.

[0078] Scenario 3: Drone swarms perform cargo delivery tasks Task description: In the urban cargo delivery scenario, multiple drones need to collaborate to complete multiple cargo delivery tasks, including transporting cargo from the distribution center to different destinations according to real-time order requirements.

[0079] Specific steps: Graph modeling: The system builds a task allocation graph, where each node represents a drone or a delivery task. The task node includes information such as the target address, cargo weight, and task time requirements. The drone node includes information such as the drone’s battery power, flight speed, and location.

[0080] Data collection and preprocessing: Collect order information, delivery requirements, real-time traffic conditions, etc., and preprocess them into a format suitable for graph neural network input. Preprocess the resource consumption of drones, traffic conditions, etc. to ensure the real-time and accuracy of the data.

[0081] Constraint analysis and integration: Multiple constraints need to be considered during task allocation, such as battery power, traffic restrictions, time constraints, etc. The weight of each task node will be dynamically adjusted according to the delivery requirements (such as timeliness, cargo type, etc.).

[0082] Graph neural network processing: Through graph convolution operations, the graph neural network generates a task allocation plan for each drone. Each drone selects and allocates tasks based on its own status information and environmental information to minimize task delays and power consumption.

[0083] Reinforcement learning optimization: Optimize the task allocation strategy through reinforcement learning. Each drone continuously adjusts its strategy based on historical task execution feedback to maximize delivery efficiency and reduce resource consumption.

[0084] Real-time feedback adjustment: Real-time monitoring of the drone’s battery power, mission completion progress, traffic conditions and other information. The system will dynamically adjust task allocation based on these real-time data to ensure that each task can be completed on time.

[0085] Technical effect: Through the method of the present invention, drones can efficiently complete cargo delivery tasks in urban environments, optimize path planning and task allocation, reduce delivery time, improve resource utilization, reduce task conflicts, and maximize system efficiency.

[0086] The present invention shows broad application potential in multiple fields and is widely applicable to multiple application scenarios in the low-altitude economy, such as environmental monitoring, logistics distribution, agricultural spraying, urban security, infrastructure inspection, disaster emergency response, etc., including but not limited to the above-mentioned application scenarios.

[0087] Based on the same inventive concept, the embodiment of the present invention also provides a multi-UAV multi-objective mission planning system, such as Figure 2 As shown, the system includes a server 21 and a drone 22; The server 21 is used to perform multi-UAV multi-objective task planning according to the multi-UAV multi-objective task planning method provided in any of the above embodiments to generate a task allocation strategy, and send the task allocation strategy to the UAVs; The drone 22 is used to receive the task allocation strategy and execute the task according to the task allocation strategy.

[0088] Based on the same inventive concept, the embodiment of the present invention also provides a multi-UAV multi-objective mission planning device, such as Figure 3 As shown, the device comprises: The task allocation graph generation module 31 is used to generate a multi-UAV task allocation graph; wherein the nodes in the multi-UAV task allocation graph represent multiple UAVs and multiple tasks, and the edges in the multi-UAV task allocation graph represent the relationship between tasks and UAVs; The information collection and processing module 32 is used to collect and pre-process the state information and environmental information of the drone, wherein the state information includes: battery power, location information, and the environmental information includes: weather conditions, terrain characteristics, and obstacle distribution. The state information and environmental information are used for graph neural network processing; A constraint condition acquisition and integration module 33, used to acquire constraint conditions and integrate the constraint conditions into the task allocation model; A first task allocation strategy generating module 34 is used to update node features of the multi-UAV task allocation graph using a graph neural network based on the constraints integrated in the task allocation model, the state information of each UAV and the environmental information, and generate a first task allocation strategy; A second task allocation strategy generating module 35 is used to generate a second task allocation strategy by optimizing the first task allocation strategy using a reinforcement learning algorithm through feedback of rewards and penalties for task execution by each UAV; The third task allocation strategy generating module 36 is used to adjust the first task allocation strategy and / or the second task allocation strategy in real time according to the environmental information and the status information of each UAV to generate a third task allocation strategy.

[0089] The embodiment of the present invention further provides an electronic device, such as Figure 4 As shown, it includes a processor 41 , a communication interface 42 , a memory 43 and a communication bus 44 , wherein the processor 41 , the communication interface 42 , and the memory 43 communicate with each other via the communication bus 44 .

[0090] A memory 43, for storing computer programs; The processor 41 is used to implement any of the steps of the above-mentioned multi-UAV multi-objective mission planning method when executing the program stored in the memory 43. The communication bus mentioned in the above-mentioned electronic device can be a peripheral component interconnect standard (Peripheral Component Interconnect, PCI) bus or an extended industry standard architecture (ExtendedIndustryStandardArchitecture, EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0091] The communication interface is used for communication between the above electronic device and other devices.

[0092] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0093] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0094] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive Solid State Disk (SSD)), etc.

[0095] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0096] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0097] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A multi-unmanned aerial vehicle multi-objective mission planning method, characterized in that: The method comprises: Generate a multi-UAV task allocation graph; wherein the nodes in the multi-UAV task allocation graph represent multiple UAVs and multiple tasks, and the edges in the multi-UAV task allocation graph represent the relationship between tasks and UAVs; Collect and pre-process environmental information, drone status information, and mission information; Obtaining constraints and integrating the constraints into the graph neural network; Based on the constraint conditions, the state information and task information of each drone, and the environmental information, a graph neural network is used to update node features of the multi-drone task allocation graph to generate a first task allocation strategy; Based on the feedback of rewards and penalties for task execution by each UAV, the first task allocation strategy is optimized by using a reinforcement learning algorithm to generate a second task allocation strategy; According to the environmental information, the status information and the task information of each UAV, the first task allocation strategy or the second task allocation strategy is adjusted in real time to generate a third task allocation strategy.

2. The method according to claim 1, characterized in that The constraints include battery power constraints, environmental information constraints and resource consumption constraints; The battery power constraint is used to ensure that the task allocation will not cause the drone to run out of power. The battery power constraint is expressed by the following formula: ； in, Indicates The remaining battery power of the drone, The minimum remaining power required for the drone to perform a mission; The environmental information constraints are used to optimize the flight path and mission execution plan of the UAV, wherein the flight path of the UAV is optimized using the following path planning algorithm: ； in, For the The flight path of the drone, for Star algorithm, used to calculate the optimal flight path, Indicates the current environment information. Indicates missions performed by drones; The resource consumption constraint is used to optimize the task allocation of the UAV to improve the resource utilization efficiency. The resource utilization efficiency is expressed by the following formula: ； in, For resource efficiency, is the number of drones, For the The number of missions performed by each drone, For the The total amount of resources consumed by each drone.

3. The method according to claim 1, characterized in that The graph neural network is a graph attention network, and the use of the graph neural network to update node features of the multi-UAV task allocation graph includes: Initialize each node feature in the multi-UAV task allocation graph according to the state information or task requirements of the UAV; For each neighbor node of any target node in the multi-UAV task allocation graph, calculating the attention weight of the neighbor node to the target node; Use the graph attention network to update node features.

4. The method according to claim 1, characterized in that The graph neural network is one of a graph convolutional network, a graph isomorphism network, a graph sampling and aggregation generator GraphSAGE.

5. The method according to claim 1, characterized in that The reinforcement learning algorithm is a deep reinforcement learning algorithm, and the first task allocation strategy is optimized by using the reinforcement learning algorithm to provide feedback on rewards and penalties for task execution by each drone, including: Train each drone as an independent agent, selecting actions based on the drone’s current state when performing a task; Determine the reward value and next state of the drone after executing the action; Record and construct trajectories based on the current state, action, reward, and next state for each time period; Determine the cumulative return based on the trajectory corresponding to each time period; The first task allocation strategy is updated according to the accumulated reward and the maximized expected reward.

6. The method according to claim 1, characterized in that The first task allocation strategy and / or the second task allocation strategy are adjusted in real time, including: The first task allocation strategy and / or the second task allocation strategy are adjusted in real time in at least one of the following ways: When a preset condition is triggered during the execution of a task by the first drone, the task of the first drone is assigned to the second drone, wherein the preset condition indicates that the execution of the task by the drone is blocked; Adjust task priorities based on the status of the drone, the drone’s resources, and the urgency of the task; Adjust the drone’s flight path and / or mission based on current environmental information.

7. A multi-UAV multi-objective mission planning system, characterized in that: The system includes a server and a drone; The server is used to perform multi-UAV multi-objective mission planning according to the method according to any one of claims 1 to 6 to generate a task allocation strategy, and send the task allocation strategy to the UAVs; The drone is used to receive the task allocation strategy and execute the task according to the task allocation strategy.

8. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-6 when executing a program stored in a memory.

Images