UAV scheduling method and system based on multi-level collaborative decision-making

Abstract

The present invention discloses a UAV scheduling method and system based on multi-level collaborative decision-making, which belongs to the field of UAV intelligent scheduling technology. It includes: obtaining the flight mission of the UAV cluster and decomposing it into multiple subtasks, combining the performance of the individual UAV and the subtask requirements to determine the subtask allocation strategy; according to the subtask allocation strategy, combined with the characteristic importance of flight influencing factors, constructing an objective function, and iteratively optimizing through a particle swarm optimization algorithm to determine the optimal flight mission execution plan for the individual UAV; according to the optimal flight mission execution plan, multi-UAV collaborative path planning is performed through a path planning algorithm to generate the optimal task execution path. It can improve the task execution efficiency of the entire UAV system, and also enhance the UAV system's ability to cope with dynamic and complex environments, solving the problems of slow response speed, low accuracy, poor environmental adaptability, etc. in task allocation and resource scheduling of existing UAVs.

Description

Technical Field

[0001] The present invention relates to the technical field of drone intelligent scheduling, and in particular to a drone scheduling method and system based on multi-level collaborative decision-making. Background Art

[0002] The statements in this section merely mention background art related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid development of drone technology, drones have been widely used in scenarios including power transmission line inspection, environmental monitoring, and disaster relief. In complex and dynamic environments, the mission execution capability and resource scheduling efficiency of drones have a significant impact on the overall performance of drone systems.

[0004] Faced with complex and changeable environmental conditions such as bad weather, communication interruptions, and dense obstacles, drones still have problems such as slow response speed, low accuracy, and poor environmental adaptability in task allocation and resource scheduling. These problems seriously restrict the widespread promotion and effectiveness of drone technology in practical applications.

[0005] In addition, with the increase in mission complexity and the expansion of the size of drone groups, traditional single optimization methods are difficult to meet the task allocation and scheduling needs under multi-objective and multi-constraint conditions, affecting the efficiency of drone system mission execution. Summary of the Invention

[0006] In order to address the shortcomings of the existing technology, the present invention provides a drone scheduling method, system, electronic device, computer-readable storage medium and computer program product based on multi-level collaborative decision-making. By constructing a multi-level collaborative decision-making framework, the organic linkage between individual drone optimization and group collaborative management is achieved.

[0007] In a first aspect, the present invention provides a method for dispatching drones based on multi-level collaborative decision-making;

[0008] A drone scheduling method based on multi-level collaborative decision-making, comprising:

[0009] Obtain the flight mission of the drone swarm and decompose it into multiple subtasks. Combine the performance of individual drones and the subtask requirements to determine the subtask allocation strategy.

[0010] Based on the subtask allocation strategy and the importance of the characteristics of flight influencing factors, an objective function is constructed and iteratively optimized using the particle swarm optimization algorithm to determine the optimal flight mission execution plan for a single UAV.

[0011] According to the optimal flight mission execution plan, multi-UAV collaborative path planning is carried out through the path planning algorithm to generate the optimal mission execution path.

[0012] In a second aspect, the present invention provides a drone scheduling system based on multi-level collaborative decision-making;

[0013] A drone dispatching system based on multi-level collaborative decision-making, including:

[0014] The task allocation module is configured to: obtain the flight mission of the drone cluster and decompose it into multiple subtasks, determine the subtask allocation strategy based on the performance of individual drones and the subtask requirements; construct an objective function based on the subtask allocation strategy and the importance of the characteristics of flight influencing factors, and iteratively optimize it using the particle swarm optimization algorithm to determine the optimal flight mission execution plan for each individual drone;

[0015] The path planning module is configured to: perform multi-UAV collaborative path planning through the path planning algorithm according to the optimal flight mission execution plan, and generate the optimal mission execution path.

[0016] In a third aspect, the present invention provides an electronic device;

[0017] An electronic device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-mentioned drone scheduling method based on multi-level collaborative decision-making.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium;

[0019] A computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the above-mentioned drone scheduling method based on multi-level collaborative decision-making.

[0020] In a fifth aspect, the present invention provides a computer program product;

[0021] A computer program product includes a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned drone scheduling method based on multi-level collaborative decision-making.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] 1. The technical solution provided by the present invention designs a method for optimizing the performance and executing tasks of individual drones. By studying the performance optimization and task execution methods of individual drones, the adaptive capability and efficient execution of drones in complex mission environments are realized; the mission success rate of drones under different environmental conditions is improved, and the service life of drones is extended, laying a solid foundation for the application of drones in various mission scenarios.

[0024] 2. The technical solution provided by the present invention designs and provides a collaborative optimization and task allocation strategy at the drone group level, which improves the overall execution efficiency and task completion rate of the drone cluster, effectively solves problems such as uneven task allocation and resource waste, and promotes the development of drone cluster technology in complex application scenarios.

[0025] 3. The technical solution provided by this invention, through the construction of a multi-level collaborative decision-making framework, achieves an organic linkage between individual drone optimization and group collaborative management. This multi-level collaborative design not only improves the mission execution efficiency of the entire drone system, but also enhances the system's ability to cope with dynamic and complex environments, providing new ideas for the intelligent and clustered development of drone systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0027] Figure 1 A flowchart of a drone scheduling method based on multi-level collaborative decision-making provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0029] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0030] Example 1

[0031] Existing drones have problems such as slow response speed, low accuracy and difficulty adapting to complex dynamic environments during the execution of flight missions. Therefore, the present invention provides a drone scheduling method based on multi-level collaborative decision-making. By constructing a multi-level collaborative decision-making framework, the drone system's mission execution efficiency and its ability to cope with dynamic and complex environments are improved.

[0032] Next, combine Figure 1 , a drone scheduling method based on multi-level collaborative decision-making disclosed in this embodiment is described in detail. The drone scheduling method based on multi-level collaborative decision-making includes the following steps:

[0033] S1. Obtain the flight mission of the drone cluster and decompose it into multiple subtasks. Combine the performance of individual drones and subtask requirements to determine the subtask allocation strategy.

[0034] During the task execution process of existing drone clusters, there are problems such as uneven task allocation and resource waste. Therefore, in this embodiment, task allocation and collaborative optimization of multiple individual drones are first performed at the drone group level, thereby improving the overall execution efficiency and task completion rate of the drone cluster in a multi-task, multi-target environment.

[0035] As an implementation method, S1 specifically includes:

[0036] S101. Initialize the allocation of flight missions for the drone cluster, decompose the flight missions into multiple subtasks, and preliminarily assign the subtasks to different individual drones, taking into account the subtask requirements and the advantages of individual drones.

[0037] In this embodiment, when performing the decomposition and preliminary division of subtasks, the operating personnel manually match them based on the most important requirements of the flight mission and the greatest advantages of the individual UAVs.

[0038] S102. Based on the preliminary division of subtasks, combined with the performance of individual UAVs and subtask requirements, determine the subtask allocation strategy.

[0039] If the flight mission is a monitoring mission, for example, an auction mechanism is used to evaluate each subtask. Based on the load of the individual drone, the mission execution cost of the individual drone, and the mission distance of the individual drone, the bid price corresponding to each individual drone is obtained, and the subtask is assigned to the individual drone with the lowest bid price.

[0040] The bidding prices are as follows:

[0041] ;

[0042] in, It is a single drone Subtasks the bid price; It is a single drone Complete subtasks implementation costs; It is a single drone To subtask distance, It's a drone The load, α and β It is the coefficient that adjusts the influence of distance and load on the bidding price.

[0043] If all bids for a subtask are too high, you may need to adjust your bidding parameters or bid again.

[0044] In this embodiment, an auction mechanism is used to evaluate each subtask, which can ensure that the most suitable single drone for executing this subtask is selected according to the priority of the subtask.

[0045] If the flight mission is a rescue mission, in some embodiments, for example, a greedy algorithm is used to determine the subtask allocation strategy based on the urgency of the subtasks and the mission distance of the individual UAV, and gradually optimize the mission execution of the individual UAV; the specific process is as follows:

[0046] (1) Based on the urgency of the subtask and the mission distance of the single UAV, the task score of each single UAV corresponding to each subtask is determined, and then the single UAV is evaluated to determine which single UAV is most suitable for performing that subtask. The task score is expressed as follows:

[0047] ;

[0048] in, Indicates a single drone Subtasks Rating, Indicates a single drone To subtask distance; Represents a subtask The urgency of The coefficient that regulates the influence of the urgency of the subtask.

[0049] (2) According to the task score, the single UAV with the highest score is selected to perform the subtask, and the subtask is removed from the task list. At the same time, the status of the single UAV (such as load, energy, etc.) is updated to ensure that the UAV does not repeatedly undertake tasks that exceed its capabilities in task allocation.

[0050] Here, the single drone with the highest mission score means that it has the best overall performance after considering the mission distance and urgency.

[0051] S2. Based on the subtask allocation strategy and the characteristic importance of flight influencing factors, the objective function is constructed and iteratively optimized through the particle swarm optimization algorithm to determine the optimal flight mission execution plan for a single UAV.

[0052] At the individual level of a single UAV, in this embodiment, the performance and mission execution method of the single UAV are optimized to improve the adaptability and efficient execution of the single UAV in a complex mission environment.

[0053] As an implementation method, S2 specifically includes:

[0054] S201. Obtain historical feature data and the corresponding mission success rate of the current single UAV, perform feature importance evaluation on the flight influencing factors in the historical feature data based on the mission success rate, and obtain the relative importance weight of each flight influencing factor; and construct an objective function based on the flight influencing factors and the corresponding relative importance weights.

[0055] Specifically, historical feature data includes individual drone locations, endurance, performance parameters, and environmental conditions. Individual drone locations include the current and target locations of the individual drone, as well as the impact of that location on mission execution. Endurance includes the drone's remaining battery life, estimated flight time, and the limits imposed by endurance on mission execution. Performance parameters include performance indicators such as the drone's flight speed, payload capacity, and sensor accuracy. Environmental conditions include weather conditions (wind speed, precipitation, etc.), obstacle distribution, and airspace restrictions. For example, linear regression is used to analyze the impact of individual drone characteristics on mission success rate. The regression coefficient reflects the impact of individual drone characteristics on subtask success. The resulting objective function is expressed as follows:

[0056] ;

[0057] in, is the objective function, i.e., the task success rate; is the independent variable, which represents the characteristics of the single UAV, such as the position and endurance; is the regression coefficient, which indicates the contribution of each single UAV feature to the mission success rate, that is, the relative importance weight of each flight influencing factor; is the intercept term, is the error term.

[0058] In some embodiments, a machine learning model (such as a random forest or a decision tree) may be used to automatically learn the relationship between features and target variables through training data, and output the relative importance weights of individual drone features.

[0059] S202: Based on the above objective function, with the goal of maximizing the mission success rate, and with the flight time, remaining battery power, and endurance range of a single drone as constraints, an iterative optimization is performed using a particle swarm optimization algorithm to determine the optimal mission execution plan. The specific process is as follows:

[0060] S2031. Initialize the particle swarm. Represent each individual drone's mission execution plan as a particle. The position of each particle represents the parameter set for the individual drone's mission (such as flight path, speed, and mission sequence), while the velocity of each particle represents the rate of change of the parameter set.

[0061] S2032. Initialize particle position and velocity. Specifically include:

[0062] (1) Position initialization: Generate an initial position for each particle based on historical data or heuristic rules. These positions are the preliminary mission plan of the single UAV, that is, the subtask allocation strategy obtained in S1 to cover a larger search space.

[0063] (2) Speed ​​initialization: Assign an initial speed to each particle. The speed determines the direction and amplitude of the particle's movement in the search space. The initial speed is usually a random value.

[0064] (3) Iterative optimization. Calculate the fitness value: Use the previously determined objective function to calculate the fitness value of each particle. The higher the fitness value, the better the current particle position (i.e., the drone mission plan).

[0065] (4) Update the individual best position: For each particle, if the current fitness value is better than its historical best position, the current position is updated to the individual best position of the particle; Update the global best position: In the entire particle group, find the particle position with the highest fitness value and update it to the global best position.

[0066] The formula of the update process is as follows:

[0067] ;

[0068] in, It is a particle Speed ​​in the next step; Indicates the inertia weight, which is used to control the particle's dependence on the current velocity; and represents the learning factor, which controls the degree to which the particle follows the individual best position and the global best position respectively; and Represents a random number between [0,1], used to increase the randomness of particles.

[0069] The formula for position update is as follows:

[0070] ;

[0071] in, It is a particle In the next step the new location.

[0072] S2033: Determine whether the maximum number of iterations has been reached or whether the change in the global optimal position is small. If the convergence condition is met, stop the iteration; otherwise, return to S2033 to continue optimization.

[0073] Once the algorithm converges, determine the global optimal position in the particle swarm gbest , as the optimal mission execution plan for UAVs.

[0074] In this embodiment, the constraint condition is expressed as:

[0075] ;

[0076] ;

[0077] ;

[0078] Where, Indicates the flight time of a single drone, Indicates the maximum flight time of a single drone. Indicates the remaining battery power of a single drone. Indicates the minimum residual current of a single drone. Indicates the endurance distance of a single drone. Indicates the maximum endurance distance of a single drone.

[0079] S3. Based on the optimal UAV mission execution plan, multi-UAV collaborative path planning is performed through the path planning algorithm to generate the optimal path. Specifically, it includes:

[0080] S301. Based on the optimal mission execution plan for the UAV, with the constraints of not touching obstacles and meeting flight conditions, use the A* path planning algorithm to generate a path for each individual UAV, ensuring that the path of each UAV is optimal while meeting the constraints.

[0081] As an implementation method, S301 specifically includes:

[0082] S3011: Set the current location of the individual drone as the starting point and add it to the OpenList, which stores the path nodes to be evaluated. Set the destination point to be reached by the individual drone. Initialize the Closed List, which stores already evaluated nodes to avoid repeated evaluation and path backtracking.

[0083] S3012, select the current node: select the node with the lowest cost estimate f(n) from the open list as the current node. The cost estimate f(n) represents:

[0084] ;

[0085] in, Represents the actual cost of a single UAV flying from the starting point to the current node, usually calculated based on the flight distance of the path or other cost functions; Represents the heuristic estimated cost from the current node to the target point, which can be estimated by Euclidean distance or Manhattan distance.

[0086] S3013, expand the current node: Evaluate all adjacent nodes of the current node, that is, the possible flight directions of a single UAV near the current flight path. For each adjacent node:

[0087] If an adjacent node is already in the closed list, skip the node. If an adjacent node is not in the open list, add it to the open list and set the current node as the parent node of the adjacent node. Calculate and record the adjacent node's If an adjacent node is already in the open list, but the cost of reaching the adjacent node through the current node is If it is lower, update the parent node of the adjacent node to the current node and recalculate its value.

[0088] S3014. Remove the current node from the open list and add it to the closed list, indicating that the node has been evaluated and will not be considered again.

[0089] S3015, Check the destination: If the current node is the target point, the search stops, the path planning is complete, and the drone can fly along the planned path. If the open list is empty, it means that no path from the starting point to the target point has been found, and the path planning has failed.

[0090] S3016, Path Backtracking and Generation: Once the target point is selected as the current node, the path planning process concludes. The system then begins backtracking from the target point, following the parent node pointers of each node until it returns to the starting point. This backtracking process generates the optimal path for the individual drone, ensuring it can efficiently fly from the starting point to the target point. The resulting path serves as the flight path for the individual drone to execute its mission.

[0091] Furthermore, it also includes:

[0092] S4. Obtain the real-time flight status data and real-time path data of each individual UAV, and determine whether a sudden environmental change has occurred based on the real-time flight status data and real-time path data. If a sudden environmental change has occurred, use the adaptive ant colony algorithm to check whether the remaining power of the individual UAV can complete each task in the current task sequence. If so, no adjustment is required. If not, an early warning is fed back to the control center.

[0093] In this embodiment, the real-time flight status data includes weather conditions. If the weather conditions are strong winds, rain, etc., it is considered that a sudden environmental change has occurred.

[0094] As an implementation method, the adaptive ant colony algorithm is used to check whether the remaining power of a single drone is sufficient to complete each task in the current task sequence. Specifically, the method includes:

[0095] S401: Based on the distances between the unfinished mission points of a single UAV, an adaptive ant colony algorithm is used to determine the order in which the missions are to be executed, taking only the distance factor into account. The specific process is as follows:

[0096] S4011. Initialize pheromone distribution, expressed as:

[0097] ;

[0098] Where, is the initial pheromone concentration of road section i, j, is the basic unit for pheromone initialization, M is the number of iterations; A is the path obtained by the nearest neighbor algorithm, that is, finding the task point closest to the current task point of the single UAV, then finding the task point closest to this task point, and so on to obtain the initial path.

[0099] S4012. Calculate the transfer probability and randomly select the next segment using the roulette wheel method, where the transfer probability is expressed as:

[0100] ;

[0101] Where, for time The transition probability of the road segment, for time pheromone concentrations in road sections; for The path length between each point of the road segment according to A; is the set of paths that are not included in the taboo table along with the current node; It is a pheromone-inducing factor; is the heuristic value heuristic factor; s is the current path to be calculated.

[0102] S4013, pheromone update, the updated pheromone concentration is expressed as:

[0103] ;

[0104] ;

[0105] Where, is the current iteration number, is the set of all road segments walked by all ants in the last iteration, is the set of optimal paths traveled by all ants in the previous iteration; It is the basic unit of pheromone.

[0106] The pheromone concentration along the path is updated. The classic ant colony algorithm's update method results in a significant increase in the pheromone concentration along the optimal path selected with each iteration. This can lead to overly concentrated selection preferences in the next iteration, increasing the likelihood of the algorithm falling into a local optimum. This embodiment uses a phased pheromone update method, dynamically adjusting the pheromone update. In the early stages of the algorithm, the pheromone concentration along the optimal path is weakened to ensure the ant colony has a wider search range.

[0107] After the iteration is completed, the task ranking considering only the cruising distance factor is obtained.

[0108] S402. Based on the task sorting considering only the cruise distance factor, the distances between tasks are summed to obtain the total task distance, and it is determined whether the remaining cruise power can meet the total task distance requirement. If so, the task is executed accordingly; if not, an early warning is fed back to the control center.

[0109] Here, the distance between each uncompleted mission point of a single UAV is determined based on the executed mission points in the real-time path data and the UAV path obtained by the A* path planning algorithm in S3.

[0110] In some embodiments, this also includes: analyzing the results of task execution, identifying potential problems, and upgrading and adjusting the system through a complete task feedback mechanism to improve the execution capabilities of future tasks. Specifically, it includes:

[0111] (1) After the mission is completed, each individual drone uploads the execution data (such as cruising time, path information, remaining battery power, etc.) to the group control center. The group layer analyzes this data to evaluate the effectiveness of the mission execution and summarize potential problems and improvement points in the mission execution.

[0112] (2) Based on the feedback data, the system can adjust the optimization methods at the individual level and the coordination strategies at the group level to adapt to future mission requirements. For example, by analyzing past mission data, the system can better balance the load of drones in future mission allocations to avoid performance degradation of some drones due to overwork.

[0113] For example, first, after the mission is completed, each individual drone uploads the remaining power and cruising time to the group control center; then, the group control center calculates the variance of the remaining power of the individual drones and the variance of the cruising time, and compares the variance of the remaining power with the variance of the cruising time. If it is closer to 1, it means that the task distribution is more even and less adjustment is needed; the further away from 1, the more adjustment is needed. It is necessary to record the single drone tasks with less remaining power and longer cruising time, and increase their impact coefficient during the next planning.

[0114] Example 2

[0115] This embodiment discloses a drone scheduling system based on multi-level collaborative decision-making, including:

[0116] The task allocation module is configured to: obtain the flight mission of the drone cluster and decompose it into multiple subtasks, and determine the subtask allocation strategy based on the performance of individual drones and subtask requirements;

[0117] Based on the subtask allocation strategy and the importance of the characteristics of flight influencing factors, an objective function is constructed and iteratively optimized using the particle swarm optimization algorithm to determine the optimal flight mission execution plan for a single UAV.

[0118] The path planning module is configured to: perform multi-UAV collaborative path planning through the path planning algorithm according to the optimal flight mission execution plan, and generate the optimal mission execution path.

[0119] It should be noted that the task assignment module and path planning module described above correspond to the steps in Example 1. The examples and application scenarios implemented by the modules and the corresponding steps are the same, but are not limited to the contents disclosed in Example 1. It should be noted that the modules described above, as part of the system, can be executed in a computer system, such as a set of computer-executable instructions.

[0120] Example 3

[0121] Embodiment 3 of the present invention provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, the steps of the above-mentioned drone scheduling method based on multi-level collaborative decision-making are completed.

[0122] Example 4

[0123] A fourth embodiment of the present invention provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the above-mentioned drone scheduling method based on multi-level collaborative decision-making are completed.

[0124] Example 5

[0125] Embodiment 5 of the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned drone scheduling method based on multi-level collaborative decision-making.

[0126] The descriptions of the various embodiments in the above embodiments have different focuses. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A drone scheduling method based on multi-level collaborative decision-making, characterized by: include: The flight mission of the drone swarm is obtained and decomposed into multiple subtasks. The subtask allocation strategy is determined based on the performance of individual drones and the subtask requirements. Specifically, if the flight mission is a monitoring mission, an auction mechanism is used to evaluate each subtask. The bid price corresponding to each individual drone is obtained based on the payload, mission distance, and execution cost of each individual drone, and the subtask is assigned to the individual drone with the lowest bid price. If the flight mission is a rescue mission, a greedy algorithm is used to determine the task score of each individual drone corresponding to each subtask. Select the single drone with the highest score to perform the subtask and remove the subtask from the task list; Based on the subtask allocation strategy and the importance of the characteristics of flight influencing factors, an objective function is constructed and iteratively optimized using the particle swarm optimization algorithm to determine the optimal flight mission execution plan for a single UAV. According to the optimal flight mission execution plan, multi-UAV collaborative path planning is carried out through the path planning algorithm to generate the optimal mission execution path; The system obtains real-time flight status and path data of each individual drone to determine whether a sudden environmental change has occurred. If so, it uses an adaptive ant colony algorithm to verify whether the individual drone's battery life is sufficient to complete the mission. It also uses a phased pheromone update method to dynamically adjust the pheromone update. In the early stages of the algorithm calculation, the pheromone concentration on the optimal path is weakened to ensure that the ant colony has a wider search range. After the mission is completed, each individual drone uploads its remaining power and cruising time to the group control center. The group control center then calculates the variance of the remaining power and the variance of the cruising time of the individual drones, and compares the variance of the remaining power with the variance of the cruising time. If the variance is closer to 1, the task distribution is more even and the less adjustment is needed; the further away from 1, the more adjustment is needed. The group control center records the individual drone tasks with less remaining power and longer cruising time, and increases the impact coefficient during the next planning.

2. The method for dispatching drones based on multi-level collaborative decision-making according to claim 1, characterized in that: The multi-UAV collaborative path planning according to the optimal flight mission execution plan is specifically as follows: according to the optimal mission execution plan of the UAV, with the constraints of not touching obstacles and meeting flight conditions, the A* path planning algorithm is used to generate the optimal mission execution path for each single UAV.

3. The method for dispatching drones based on multi-level collaborative decision-making according to claim 1, characterized in that: The objective function constructed according to the subtask allocation strategy and the characteristic importance of flight influencing factors specifically includes: Obtain the historical feature data and corresponding mission success rate of the current single UAV, perform feature importance evaluation on the flight influencing factors in the historical feature data based on the mission success rate, and obtain the relative importance weight of each flight influencing factor; The objective function is constructed based on the flight influencing factors and the corresponding relative importance weights.

4. A drone scheduling system based on multi-level collaborative decision-making, which adopts a drone scheduling method based on multi-level collaborative decision-making as described in any one of claims 1 to 3, characterized in that: include: The task allocation module is configured to: obtain the flight mission of the drone cluster and decompose it into multiple subtasks, and determine the subtask allocation strategy based on the performance of individual drones and subtask requirements; Based on the subtask allocation strategy and the importance of the characteristics of flight influencing factors, an objective function is constructed and iteratively optimized using the particle swarm optimization algorithm to determine the optimal flight mission execution plan for a single UAV. The path planning module is configured to: perform multi-UAV collaborative path planning through the path planning algorithm according to the optimal flight mission execution plan, and generate the optimal mission execution path.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the drone scheduling method based on multi-level collaborative decision-making as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the drone scheduling method based on multi-level collaborative decision-making described in any one of claims 1-3 are implemented.

7. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the drone scheduling method based on multi-level collaborative decision-making described in any one of claims 1-3 are implemented.

Images