A method, device, and medium for task scheduling based on a UAV cluster

Abstract

The present application relates to a kind of based on unmanned aerial vehicle cluster task scheduling method, device, equipment and medium, the method includes: constructing flight task model;To the unmanned aerial vehicle cluster to be distributed is scored and obtains first comprehensive analysis table, according to first comprehensive analysis table determines the best unmanned aerial vehicle cluster combination for executing task;Obtain the historical data of the best unmanned aerial vehicle cluster combination for executing task when executing task, according to historical data and evaluation function determines task execution score;According to task execution score, update first comprehensive analysis table and obtain second comprehensive analysis table;Obtain the log data of unmanned aerial vehicle cluster task execution and unmanned aerial vehicle cluster task allocation characteristics, according to the log data of unmanned aerial vehicle cluster task execution, unmanned aerial vehicle cluster task allocation characteristics and feature extraction model determine task matching degree score;According to task matching degree score, update second comprehensive analysis table and obtain third comprehensive analysis table;From third comprehensive analysis table, determine the unmanned aerial vehicle cluster with highest comprehensive score executes task.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) scheduling technology, and in particular to a method, apparatus, equipment, and medium for UAV swarm task scheduling. Background Technology

[0002] Drone swarm missions refer to tasks that require multiple drones to work together. These missions have sufficient time resources to allow multiple drones to work sequentially or sufficient space resources to accommodate multiple drones cooperating simultaneously. Different missions require different strategies from drones. Currently, there is no solution to the problem of insufficient intelligent scheduling in the process of drones selecting different strategies for different missions. Summary of the Invention

[0003] To address the aforementioned issues, this invention provides a method, apparatus, device, and medium for scheduling unmanned aerial vehicle (UAV) swarm tasks.

[0004] In a first aspect, embodiments of the present invention provide a method for task scheduling based on unmanned aerial vehicle (UAV) swarms, including: Obtain the maximum range between the UAV and the target point, and construct a flight mission model based on the UAV's maximum range, mission time window constraints, mission execution sequence, UAV payload, and mission computational power metrics. The first comprehensive analysis table is obtained by scoring the drone clusters to be assigned, and the optimal drone cluster combination for performing the task is determined based on the first comprehensive analysis table. Obtain historical data on the optimal drone swarm combination for mission execution, and determine the mission execution score based on the historical data and evaluation function; The second comprehensive analysis table is obtained by updating the first comprehensive analysis table based on the task execution score; Obtain log data of drone cluster tasks and drone cluster task allocation characteristics, and determine task matching degree score based on drone cluster task log data, drone cluster task allocation characteristics and feature extraction model; The second comprehensive analysis table is updated based on the task matching score to obtain the third comprehensive analysis table; the drone cluster with the highest comprehensive score is selected from the third comprehensive analysis table to perform the task.

[0005] Furthermore, in the above-mentioned drone swarm-based task scheduling method, a first comprehensive analysis table is obtained by scoring the drone swarms to be assigned, and the optimal drone swarm combination for executing the task is determined based on the first comprehensive analysis table, including: The task computational power score is determined based on the computational power optimization model and task computational power optimization indicators. The first comprehensive analysis table should include at least the following: the highest endurance score of the drone cluster to be assigned, the task time window constraint score, the task execution time sequence score, the drone's payload score, the task computing power score, and the comprehensive score. The drone cluster with the highest comprehensive score in the first comprehensive analysis table is determined as the optimal drone cluster combination for mission execution.

[0006] Furthermore, in the aforementioned method for scheduling drone swarms, obtaining historical data on the optimal drone swarm combination for task execution, and determining the task execution score based on the historical data and an evaluation function, includes: Multiple pre-defined behaviors are used when performing tasks; Each behavior corresponds to a different model; The training results are obtained by feeding the data corresponding to each behavior in the historical data into the corresponding model and training it. The training results are scored using an evaluation function to obtain a task performance score.

[0007] Furthermore, in the aforementioned method for scheduling tasks based on UAV swarms, the various behaviors include at least: target behavior, formation maintenance behavior, internal collision avoidance behavior, obstacle avoidance behavior, and random behavior.

[0008] Furthermore, in the aforementioned method for scheduling tasks based on UAV swarms, a task execution score is obtained by scoring the training results using an evaluation function, including: The evaluation functions include eight types: accuracy, error rate, hit rate, true negative rate, precision, false positive rate, negative term accuracy, and positive term error rate. The evaluation function is divided into four groups of two. The training results are scored using the evaluation function to obtain the sum of the normal and outlier values ​​for each group. The sum of the normal and outlier values ​​for each group determines the task execution score.

[0009] Furthermore, in the aforementioned method for scheduling tasks based on UAV swarms, the process of acquiring log data of UAV swarm tasks and the characteristics of UAV swarm task allocation, and determining a task matching score based on the log data of UAV swarm tasks, the characteristics of UAV swarm task allocation, and a feature extraction model, includes: Based on the characteristics of drone cluster task allocation and the feature extraction model, a classification dataset is obtained from the logs. The number of times the extracted feature keywords appear in the classification dataset is then analyzed. The task matching score is determined based on the number of times it occurs.

[0010] Furthermore, in the above-mentioned UAV swarm task scheduling method, the characteristics of UAV swarm task allocation include: complexity, accuracy, and real-time performance.

[0011] Secondly, embodiments of the present invention also provide a drone swarm task scheduling device, comprising: Acquisition module and construction module: used to acquire the maximum range between the UAV and the target point, and construct a flight mission model based on the UAV's maximum range, mission time window constraints, mission execution sequence, UAV's payload and mission computing power quantitative indicators; The first determining module is used to score the drone clusters to be assigned to obtain a first comprehensive analysis table, and to determine the optimal drone cluster combination for performing the task based on the first comprehensive analysis table. The second determination module is used to obtain historical data on the optimal drone swarm combination for performing the task, and to determine the task execution score based on the historical data and the evaluation function. Update module: Used to update the first comprehensive analysis table to obtain the second comprehensive analysis table based on the task execution score; The third determining module is used to obtain log data of the drone cluster performing tasks and the characteristics of drone cluster task allocation, and to determine the task matching degree score based on the log data of the drone cluster performing tasks, the characteristics of drone cluster task allocation, and the feature extraction model. The fourth determination module is used to update the second comprehensive analysis table based on the task matching score to obtain the third comprehensive analysis table; and to determine the drone cluster with the highest comprehensive score from the third comprehensive analysis table to execute the task.

[0012] Thirdly, embodiments of the present invention also provide an electronic device, including: a processor and a memory; The processor executes any of the above-described drone swarm task scheduling methods by calling programs or instructions stored in the memory.

[0013] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a program or instructions that cause a computer to execute any of the above-described unmanned aerial vehicle (UAV) swarm task scheduling methods.

[0014] The advantages of this invention are as follows: This invention obtains the maximum flight distance between the UAV and the target point, and constructs a flight mission model based on the UAV's maximum flight distance, mission time window constraints, mission execution sequence, UAV payload, and mission computational power metrics. It then scores the UAV clusters to be assigned to obtain a first comprehensive analysis table, and determines the optimal UAV cluster combination for mission execution based on this table. Historical data of the optimal UAV cluster combination during mission execution is obtained, and a mission execution score is determined based on this historical data and an evaluation function. The first comprehensive analysis table is updated based on the mission execution score to obtain a second comprehensive analysis table. Log data of UAV cluster mission execution and UAV cluster mission allocation characteristics are obtained, and a mission matching score is determined based on this log data, mission allocation characteristics, and a feature extraction model. The second comprehensive analysis table is updated based on the mission matching score to obtain a third comprehensive analysis table. Finally, the UAV cluster with the highest comprehensive score is selected from the third comprehensive analysis table to execute the mission. This invention not only determines the mission execution status of the UAV cluster but also determines the mission execution score and mission matching score, thus improving the accuracy of UAV cluster-based mission scheduling. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the conventional technology, the drawings used in the description of the embodiments or the conventional technology 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.

[0016] Figure 1 This invention provides an illustration of a task scheduling method based on a drone swarm. Figure 1 ; Figure 2 This invention provides an illustration of a task scheduling method based on a drone swarm. Figure 2 ; Figure 3 This invention provides an illustration of a task scheduling method based on a drone swarm. Figure 3 ; Figure 4 This is a schematic diagram of a drone swarm task scheduling device provided in an embodiment of the present invention; Figure 5 This is a schematic block diagram of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0019] Figure 1 This invention provides an illustration of a task scheduling method based on a drone swarm. Figure 1 .

[0020] Firstly, embodiments of the present invention provide a method for task scheduling based on unmanned aerial vehicle (UAV) swarms, combined with Figure 1 ,include: S101: Obtain the maximum range between the UAV and the target point, and construct a flight mission model based on the UAV's maximum range, mission time window constraints, mission execution sequence, UAV payload, and mission computational metrics.

[0021] Specifically, in this embodiment of the invention, a two-dimensional or three-dimensional variable is used to define the position of a UAV and a target point. By obtaining the positions of the UAV and the target point, the maximum flight distance between the UAV and the target point can be obtained. The task time window constraint, the order of task execution, the UAV's payload, and the task computing power quantitative indicators are the set constraints. The flight mission model is constructed by the maximum flight distance between the UAV and the target point and the constraints.

[0022] S102: The drone clusters to be assigned are scored to obtain the first comprehensive analysis table. Based on the first comprehensive analysis table, the optimal drone cluster combination for performing the task is determined.

[0023] Specifically, in this embodiment of the invention, after constructing the flight mission model, the drone clusters to be assigned are scored to obtain a first comprehensive analysis table. The method for determining the optimal drone cluster combination for performing the mission based on the first comprehensive analysis table is described in detail below.

[0024] S103: Obtain historical data on the optimal drone swarm combination for mission execution, and determine the mission execution score based on the historical data and evaluation function.

[0025] Specifically, in this embodiment of the invention, the method for obtaining historical data of the optimal drone swarm combination for performing the task and determining the task execution score based on the historical data and the evaluation function is described in detail below.

[0026] S104: Update the first comprehensive analysis table based on the task execution score to obtain the second comprehensive analysis table.

[0027] Specifically, in this embodiment of the invention, the task execution score is updated in the first comprehensive analysis table to obtain the second comprehensive analysis table.

[0028] S105: Obtain log data of drone cluster tasks and drone cluster task allocation characteristics, and determine the task matching degree score based on the drone cluster task log data, drone cluster task allocation characteristics and feature extraction model.

[0029] Specifically, in this embodiment of the invention, the method for obtaining log data of drone clusters performing tasks and the characteristics of drone cluster task allocation, and determining the task matching degree score based on the log data of drone clusters performing tasks, the characteristics of drone cluster task allocation, and the feature extraction model is described in detail below.

[0030] S106: Update the second comprehensive analysis table based on the task matching score to obtain the third comprehensive analysis table; determine the drone cluster with the highest comprehensive score from the third comprehensive analysis table to execute the task.

[0031] Specifically, in this embodiment of the invention, the task matching score is updated in the second comprehensive analysis table to obtain the third comprehensive analysis table, and the drone cluster with the highest comprehensive score is determined from the third comprehensive analysis table to perform the task.

[0032] It should be understood that this invention not only determines the task execution status of the drone cluster, but also determines the task execution score and task matching degree score of the drone cluster, thereby improving the accuracy of drone cluster-based task scheduling.

[0033] Furthermore, in the above-mentioned drone swarm-based task scheduling method, a first comprehensive analysis table is obtained by scoring the drone swarms to be assigned, and the optimal drone swarm combination for executing the task is determined based on the first comprehensive analysis table, including: The task computational power score is determined based on the computational power model and task computational power metrics.

[0034] The first comprehensive analysis table should include at least the following: the highest endurance score of the drone cluster to be assigned, the task time window constraint score, the task execution time sequence score, the drone's payload score, the task computing power score, and the comprehensive score. The drone cluster with the highest comprehensive score in the first comprehensive analysis table is determined as the optimal drone cluster combination for mission execution.

[0035] Specifically, in this embodiment of the invention, the task computing power score is first determined based on the computing power quantification model and the task computing power quantification index. Then, the maximum flight distance and constraints between the UAV and the target point are scored to obtain the first comprehensive analysis table.

[0036] Specifically, the following table is the first comprehensive analysis table:

[0037] As can be seen from the table, Drone 2 has the highest overall score, making it the best drone for performing the mission.

[0038] Figure 2 This invention provides an illustration of a task scheduling method based on a drone swarm. Figure 2 .

[0039] Furthermore, in the aforementioned method for scheduling drone swarms, historical data on the optimal drone swarm combination for task execution is obtained. Based on this historical data and an evaluation function, a task execution score is determined, and this score is then combined with… Figure 2 It includes four steps, S201 to S204: S201: Preset multiple behaviors when performing tasks.

[0040] The various behaviors include at least: target behavior, formation maintenance behavior, internal collision avoidance behavior, obstacle avoidance behavior, and random behavior.

[0041] S202: Each behavior corresponds to one model; S203: Input the data corresponding to each behavior in the historical data into the corresponding model for training to obtain the training results; S204: The training results are scored using an evaluation function to obtain a task execution score.

[0042] Specifically, in this embodiment of the application, in order to make more accurate selection of drones or swarms performing tasks, an evaluation function is constructed to train the model corresponding to different behaviors in the historical task execution process of the drones or swarms by inputting various behavior classification data into the model, and the training results are scored to obtain the task execution score.

[0043] Furthermore, in the aforementioned UAV swarm-based task scheduling method, the task execution score is obtained by scoring the training results through an evaluation function, including: The evaluation functions include eight types: accuracy, error rate, hit rate, true negative rate, precision, false positive rate, negative term accuracy, and positive term error rate. The evaluation functions are divided into four groups of two. The training results are scored using the evaluation functions to obtain the sum of the normal and outlier values ​​for each group. The sum of the normal and outlier values ​​for each group determines the task execution score.

[0044] Specifically, in this embodiment of the invention, the evaluation function consists of eight indicators divided into four groups, each accounting for 25% of the model evaluation score. Since all four groups of indicators are binary indicators (0 or 1), a correct result for each indicator is considered normal, while a result that is incorrect is considered abnormal. Finally, the proportions of normal and abnormal results are summed to obtain the final predicted value of the model evaluation.

[0045] Model evaluation accuracy: Each of the four indicators accounts for 25% of the model evaluation score. The final predicted value of the model is obtained by summing the proportions of normal and abnormal results. For example, the model accuracy is 75%, and the abnormality rate is 25%.

[0046] Assumption: False alarm rate 10%; Overall accuracy (model evaluation accuracy 75% + data evaluation accuracy 90%) / 2 = 82.5%; The overall anomaly rate is (25% + 10%) / 2 = 17.5%.

[0047] Specifically, the table below is the second comprehensive analysis table for updating task execution scores:

[0048] As can be seen from the table, Drone 2 has the highest overall score. Drone 2 is not only the best drone in terms of mission execution, but also the drone with the highest mission execution score.

[0049] Figure 3 This invention provides an illustration of a task scheduling method based on a drone swarm. Figure 3 .

[0050] Furthermore, in the aforementioned method for scheduling tasks based on UAV swarms, log data of UAV swarm tasks and characteristics of UAV swarm task allocation are obtained. Based on the log data of UAV swarm tasks, characteristics of UAV swarm task allocation, and a feature extraction model, a task matching score is determined, combined with... Figure 3 It includes two steps, S301 to S302: S301: Obtain a classification dataset from the logs based on the characteristics of drone cluster task allocation and a feature extraction model, and determine the number of times the extracted feature keywords appear in the classification dataset; S302: Determine the task matching score based on the number of occurrences.

[0051] Furthermore, in the above-mentioned UAV swarm task scheduling method, the characteristics of UAV swarm task allocation include: complexity, accuracy, and real-time performance.

[0052] Specifically, in this embodiment of the invention, for example, the keywords obtained from the training results of the feature extraction model in the task execution log data of UAV 1, which are complexity log data, accuracy log data, and real-time log data, appear 6 times in complexity log data, 5 times in accuracy log data, and 4 times in real-time log data, respectively. This determines that the current UAV is involved in a lot of complex tasks, thereby determining the task matching degree score.

[0053] Specifically, the table below is the third analysis table for updating the task matching score:

[0054] As can be seen from the table, drones 1 and 2 have the highest overall scores.

[0055] In some embodiments, the drone with the highest task performance score is given priority, so drone 1 is preferred.

[0056] Figure 4 This is a schematic diagram of a drone cluster task scheduling device provided in an embodiment of the present invention.

[0057] Secondly, embodiments of the present invention also provide a drone swarm task scheduling device, combined with Figure 4 ,include: The acquisition module 401 and the construction module 402 are used to acquire the maximum range between the UAV and the target point, and to construct a flight mission model based on the maximum range of the UAV, the mission time window constraint, the order of mission execution, the UAV's payload, and the quantitative indicators of mission computing power.

[0058] Specifically, in this embodiment of the invention, the acquisition module 401 defines the position of a UAV and a target point using a two-dimensional or three-dimensional variable. By obtaining the positions of the UAV and the target point, the maximum flight distance between the UAV and the target point can be obtained. The acquisition module 401 acquires the set constraints such as the task time window constraint, the order of task execution, the UAV's payload, and the task computing power quantitative indicators, and constructs a flight mission model through the construction module 402.

[0059] First determining module 403: used to score the drone clusters to be assigned to obtain a first comprehensive analysis table, and to determine the best drone cluster combination to perform the task based on the first comprehensive analysis table.

[0060] Specifically, in this embodiment of the invention, after constructing the flight mission model, the drone clusters to be assigned are scored to obtain a first comprehensive analysis table. The method by which the first determining module 403 determines the optimal drone cluster combination for performing the mission based on the first comprehensive analysis table has been described in detail above.

[0061] The second determining module 404 is used to obtain historical data of the optimal drone swarm combination for performing the task, and to determine the task execution score based on the historical data and the evaluation function.

[0062] Specifically, in this embodiment of the invention, the method for obtaining historical data of the optimal drone swarm combination for performing the task and determining the task execution score based on the historical data and the evaluation function by the second determining module 404 has been described in detail.

[0063] Update module 405: Used to update the first comprehensive analysis table to obtain the second comprehensive analysis table based on the task execution score.

[0064] Specifically, in this embodiment of the invention, the update module 405 updates the task execution score to the first comprehensive analysis table to obtain the second comprehensive analysis table.

[0065] The third determining module 406 is used to obtain the log data of the drone cluster performing tasks and the characteristics of the drone cluster task allocation, and to determine the task matching degree score based on the log data of the drone cluster performing tasks, the characteristics of the drone cluster task allocation, and the feature extraction model.

[0066] Specifically, in this embodiment of the invention, the third determining module 406 obtains the log data of the UAV cluster performing tasks and the task allocation characteristics of the UAV cluster. The method for determining the task matching degree score based on the log data of the UAV cluster performing tasks, the task allocation characteristics of the UAV cluster, and the feature extraction model has been described in detail above.

[0067] The fourth determining module 407 is used to update the second comprehensive analysis table based on the task matching score to obtain the third comprehensive analysis table; and to determine the drone cluster with the highest comprehensive score from the third comprehensive analysis table to execute the task.

[0068] Specifically, in this embodiment of the invention, the fourth determining module 407 updates the task matching degree score to the second comprehensive analysis table to obtain the third comprehensive analysis table, and determines the drone cluster with the highest comprehensive score from the third comprehensive analysis table to execute the task.

[0069] Thirdly, embodiments of the present invention also provide an electronic device, including: a processor and a memory; The processor executes any of the above-described drone swarm task scheduling methods by calling programs or instructions stored in the memory.

[0070] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a program or instructions that cause a computer to execute any of the above-described unmanned aerial vehicle (UAV) swarm task scheduling methods.

[0071] Figure 5 This is a schematic block diagram of an electronic device provided in an embodiment of this disclosure.

[0072] like Figure 5 As shown, the electronic device includes at least one processor 501, at least one memory 502, and at least one communication interface 503. The various components in the electronic device are coupled together via a bus system 504. The communication interface 503 is used for information transmission with external devices. It is understood that the bus system 504 is used to implement communication between these components. In addition to a data bus, the bus system 504 also includes a power bus, a control bus, and a status signal bus. However, for clarity, ... Figure 5 The general designated all buses as Bus System 504.

[0073] It is understood that the memory 502 in this embodiment can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.

[0074] In some implementations, memory 502 stores elements such as executable units or data structures, or subsets thereof, or extended sets thereof: operating systems and applications.

[0075] The operating system includes various system programs, such as the framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. The application programs include various applications, such as media players and browsers, used to implement various application functions. A program implementing any method in the UAV swarm task scheduling method provided in this embodiment of the invention can be included in the application programs.

[0076] In this embodiment of the invention, the processor 501 executes the steps of various embodiments of the UAV cluster task scheduling method provided in this embodiment of the invention by calling the program or instructions stored in the memory 502, specifically, the program or instructions stored in the application program.

[0077] Obtain the maximum range between the UAV and the target point, and construct a flight mission model based on the UAV's maximum range, mission time window constraints, mission execution sequence, UAV payload, and mission computational power metrics. The first comprehensive analysis table is obtained by scoring the drone clusters to be assigned, and the optimal drone cluster combination for performing the task is determined based on the first comprehensive analysis table. Obtain historical data on the optimal drone swarm combination for mission execution, and determine the mission execution score based on the historical data and evaluation function; The second comprehensive analysis table is obtained by updating the first comprehensive analysis table based on the task execution score; Obtain log data of drone cluster tasks and drone cluster task allocation characteristics, and determine task matching degree score based on drone cluster task log data, drone cluster task allocation characteristics and feature extraction model; The second comprehensive analysis table is updated based on the task matching score to obtain the third comprehensive analysis table; the drone cluster with the highest comprehensive score is selected from the third comprehensive analysis table to perform the task.

[0078] Any method in the UAV swarm task scheduling method provided in this embodiment of the invention can be applied to, or implemented by, the processor 501. The processor 501 can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of the processor 501 or by instructions in software form. The processor 501 can be a general-purpose processor, 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, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional device.

[0079] The steps of any method in the UAV swarm task scheduling method provided in this embodiment of the invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software units in the decoding processor. The software units can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory 502. The processor 501 reads the information in memory 502 and, in conjunction with its hardware, completes the steps of the UAV swarm task scheduling method.

[0080] Those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0081] Those skilled in the art will understand that the descriptions of the various embodiments have different focuses, and for parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0082] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for task scheduling based on unmanned aerial vehicle (UAV) swarms, characterized in that, include: Obtain the maximum range between the UAV and the target point, and construct a flight mission model based on the UAV's maximum range, mission time window constraints, mission execution sequence, UAV's payload, and mission computing power metrics. The first comprehensive analysis table is obtained by scoring the drone clusters to be assigned, and the optimal drone cluster combination for performing the task is determined based on the first comprehensive analysis table. Obtain historical data on the optimal drone swarm combination for mission execution, and determine the mission execution score based on the historical data and the evaluation function; The first comprehensive analysis table is updated based on the task execution score to obtain the second comprehensive analysis table; Obtain log data of drone cluster tasks and drone cluster task allocation characteristics, and determine task matching degree score based on the drone cluster task log data, drone cluster task allocation characteristics and feature extraction model; The second comprehensive analysis table is updated based on the task matching score to obtain the third comprehensive analysis table; the drone cluster with the highest comprehensive score is selected from the third comprehensive analysis table to execute the task. The process of scoring the drone clusters to be assigned yields a first comprehensive analysis table. Based on this table, the optimal drone cluster combination for mission execution is determined, including: The task computational power score is determined based on the computational power optimization model and task computational power optimization indicators. The first comprehensive analysis table includes at least: the highest endurance score of the drone cluster to be assigned, the task time window constraint score, the task execution time sequence score, the drone's payload score, the task computing power score, and the comprehensive score. The drone cluster with the highest comprehensive score in the first comprehensive analysis table is determined as the optimal drone cluster combination for mission execution; The process of obtaining historical data on the optimal drone swarm combination for task execution, and determining the task execution score based on the historical data and the evaluation function, includes: Multiple pre-defined behaviors are used when performing tasks; Each behavior corresponds to a different model; The training results are obtained by feeding the data corresponding to each behavior in the historical data into the corresponding model and training it. The training results are scored using an evaluation function to obtain a task performance score.

2. The method for scheduling unmanned aerial vehicle (UAV) swarm tasks according to claim 1, characterized in that, The various behaviors include at least: target behavior, formation maintenance behavior, internal collision avoidance behavior, obstacle avoidance behavior, and random behavior.

3. The method for scheduling unmanned aerial vehicle (UAV) swarm tasks according to claim 1, characterized in that, The task execution score is obtained by scoring the training results using an evaluation function, including: The evaluation functions include eight types: accuracy, error rate, hit rate, true negative rate, precision, false positive rate, negative term accuracy, and positive term error rate. The evaluation function is divided into four groups of two. The training results are scored using the evaluation function to obtain the sum of the normal and outlier values ​​for each group. The sum of the normal and outlier values ​​for each group determines the task execution score.

4. The method for scheduling unmanned aerial vehicle (UAV) swarm tasks according to claim 1, characterized in that, The process of acquiring log data of drone swarm tasks and drone swarm task allocation characteristics, and determining a task matching score based on the drone swarm task execution log data, drone swarm task allocation characteristics, and feature extraction model, includes: Based on the characteristics of drone cluster task allocation and the feature extraction model, a classification dataset is obtained from the logs. The number of times the extracted feature keywords appear in the classification dataset is then analyzed. The task matching score is determined based on the number of occurrences.

5. The method for scheduling unmanned aerial vehicle (UAV) swarm tasks according to claim 1, characterized in that, The characteristics of the drone swarm task allocation include: complexity, accuracy, and real-time performance.

6. A task scheduling device based on unmanned aerial vehicle (UAV) swarms, characterized in that, include: Acquisition module and construction module: used to acquire the maximum range between the UAV and the target point, and construct a flight mission model based on the UAV's maximum range, mission time window constraints, mission execution sequence, UAV's payload and mission computing power quantitative indicators; The first determining module is used to score the drone clusters to be assigned to obtain a first comprehensive analysis table, and to determine the optimal drone cluster combination for performing the task based on the first comprehensive analysis table. The second determining module is used to acquire historical data of the optimal drone swarm combination for performing the task, and to determine the task execution score based on the historical data and the evaluation function. Update module: used to update the first comprehensive analysis table to obtain the second comprehensive analysis table based on the task execution score; The third determining module is used to obtain log data of the drone cluster performing tasks and the characteristics of drone cluster task allocation, and to determine the task matching degree score based on the log data of the drone cluster performing tasks, the characteristics of drone cluster task allocation, and the feature extraction model. The fourth determination module is used to update the second comprehensive analysis table based on the task matching score to obtain the third comprehensive analysis table; and to determine the drone cluster with the highest comprehensive score from the third comprehensive analysis table to execute the task. The process of scoring the drone clusters to be assigned yields a first comprehensive analysis table. Based on this table, the optimal drone cluster combination for mission execution is determined, including: The task computational power score is determined based on the computational power optimization model and task computational power optimization indicators. The first comprehensive analysis table includes at least: the highest endurance score of the drone cluster to be assigned, the task time window constraint score, the task execution time sequence score, the drone's payload score, the task computing power score, and the comprehensive score. The drone cluster with the highest comprehensive score in the first comprehensive analysis table is determined as the optimal drone cluster combination for mission execution; The process of obtaining historical data on the optimal drone swarm combination for task execution, and determining the task execution score based on the historical data and the evaluation function, includes: Multiple pre-defined behaviors are used when performing tasks; Each behavior corresponds to a different model; The training results are obtained by feeding the data corresponding to each behavior in the historical data into the corresponding model and training it. The training results are scored using an evaluation function to obtain a task performance score.

7. An electronic device, characterized in that, include: Processor and memory; The processor executes a drone swarm task scheduling method as described in any one of claims 1 to 5 by calling the program or instructions stored in the memory.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program or instructions that cause a computer to execute a drone swarm-based task scheduling method as described in any one of claims 1 to 5.

Images