Multi-unmanned aerial vehicle autonomous cooperative task allocation method based on GWCA algorithm

Abstract

The application discloses a multi-unmanned aerial vehicle autonomous cooperative task allocation method based on a GWCA algorithm, and particularly relates to the technical field of task allocation, which comprises the following steps: training unmanned aerial vehicles in a fixed scene based on the GWCA algorithm, determining the task execution sequence and route planning of the unmanned aerial vehicles at different positions in the fixed scene, determining the flight path information and flight risk information of a single unmanned aerial vehicle according to the current flight process data of each unmanned aerial vehicle and the historical training data of the unmanned aerial vehicle, obtaining a problematic unmanned aerial vehicle, and re-distributing tasks to the remaining unmanned aerial vehicles by using the GWCA algorithm; in the process of cooperative work of the multiple unmanned aerial vehicles, the overall analysis of the multiple unmanned aerial vehicles is performed, the matching degree information and risk information of the cooperative work of the multiple unmanned aerial vehicles are collected, and the performance of the cooperative work of the multiple unmanned aerial vehicles is quantified; and the application is helpful for making the task scheduling of the multiple unmanned aerial vehicles more flexible and intelligent, and capable of coping with complex environments and task changes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of task allocation, more particularly, the present application relates to a multi-unmanned aerial vehicle autonomous cooperative task allocation method based on a GWCA algorithm. BACKGROUND

[0002] With the rapid development of unmanned aerial vehicle technology, multi-unmanned aerial vehicle systems (UAVs) are increasingly widely used in various application scenarios, especially in the fields of logistics, environmental monitoring, agricultural inspection, etc. The cooperative work of multi-unmanned aerial vehicle systems can improve task execution efficiency, reduce costs, and increase flexibility. In the prior art, the actual flight track distance is used to replace the straight-line distance through the GWCA algorithm, so that the unmanned aerial vehicle can stably obtain a better task allocation result. However, it is not easy to find out the unmanned aerial vehicle with problems in the multi-unmanned aerial vehicle, which may lead to a decrease in the success rate of task execution, and it is not easy to analyze the reliability of the cooperative work of the unmanned aerial vehicles in the process of cooperative work of the multi-unmanned aerial vehicles.

[0003] In order to solve the above two defects, a technical solution is provided. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a safety chip based on unmanned driving and an information interaction method thereof to solve the problems raised in the background art.

[0005] To achieve the above object, the present application provides the following technical scheme:

[0006] The multi-unmanned aerial vehicle autonomous cooperative task allocation method based on the GWCA algorithm specifically comprises the following steps:

[0007] S1: training the unmanned aerial vehicle in a fixed scene based on the GWCA algorithm to determine the task execution order and route planning of the unmanned aerial vehicle at different positions in the fixed scene;

[0008] S2: determining the flight path information and flight risk information of a single unmanned aerial vehicle according to the current flight process data of each unmanned aerial vehicle and the historical training data of the unmanned aerial vehicle;

[0009] S3: comprehensively analyzing the flight path information and flight risk information of the single unmanned aerial vehicle to determine the unmanned aerial vehicle with problems and re-distributing tasks to the remaining unmanned aerial vehicles using the GWCA algorithm;

[0010] S4: in the process of cooperative work of the multi-unmanned aerial vehicles, collecting the matching degree information and risk information of the cooperative work of the multi-unmanned aerial vehicles through overall analysis of the multi-unmanned aerial vehicles, quantifying the performance of the cooperative work of the multi-unmanned aerial vehicles, and generating an alarm signal.

[0011] In a preferred embodiment, the flight path information of the single unmanned aerial vehicle is determined, comprising:

[0012] The flight path information of the single unmanned aerial vehicle is represented by an actual time deviation coefficient;

[0013] The acquisition logic of the actual time deviation coefficient is: determining a preset flight path of the unmanned aerial vehicle when performing a task by the GWCA algorithm, setting a special marker point in the preset flight path, setting a monitoring time period, obtaining the special marker point passed by the unmanned aerial vehicle in the monitoring time period, determining a preset time point when the unmanned aerial vehicle arrives at the special marker point, marking the preset time point when the unmanned aerial vehicle arrives at the special marker point as: , setting a monitoring time period, obtaining the special marker point passed by the unmanned aerial vehicle in the monitoring time period, and determining a time point when the unmanned aerial vehicle actually arrives at the special marker point, marking the time point when the unmanned aerial vehicle actually arrives at the special marker point as: ;

[0014] By comparing the preset time point and the actual time point, the actual time deviation coefficient is determined, and the calculation formula of the actual time deviation coefficient is: ; wherein, is the actual time deviation coefficient.

[0015] In a preferred embodiment, the flight risk information of the single unmanned aerial vehicle is determined, comprising:

[0016] The flight risk information of the single unmanned aerial vehicle is represented by a Bayesian likelihood probability coefficient;

[0017] The acquisition logic of the Bayesian likelihood probability coefficient is: determining flight characteristics of the unmanned aerial vehicle in the flight process, the flight characteristics including power, speed, height, and wind speed, matching the flight characteristics with historical training data of the unmanned aerial vehicle in the scene, determining historical similar flight data by cosine similarity, and the calculation formula is: , wherein A is a flight characteristic vector of the current unmanned aerial vehicle, is a flight characteristic vector in the historical training data, and XS is a flight characteristic vector of the historical similar flight data;

[0018] According to the historical similar flight data, a prior probability of completing the task of the unmanned aerial vehicle is determined, and the prior probability of completing the task of the unmanned aerial vehicle is marked as: , a probability of completing the task in the historical similar flight data is determined, and the probability of completing the task in the historical similar flight data is marked as: ;

[0019] The probability of the historical similar flight data is determined according to the historical training data using kernel density estimation, and the probability of the historical similar flight data is marked as: The probability that the UAV can complete the task under the current flight characteristics is calculated by using the Bayesian formula, the probability that the UAV can complete the task is expressed by a Bayesian likelihood probability coefficient, and the calculation formula is: ; wherein, is the Bayesian likelihood probability coefficient.

[0020] In a preferred embodiment, the flight path information and the flight risk information of a single UAV are comprehensively analyzed, including:

[0021] The flight path information and the flight risk information of a single UAV are comprehensively analyzed, and a single UAV evaluation model is constructed by weighted calculation through the actual time deviation coefficient and the Bayesian likelihood probability coefficient, to generate a single UAV evaluation coefficient. The calculation formula of the single UAV evaluation coefficient is: ; wherein, is the single UAV evaluation coefficient, q = 1, 2, 3, …, Q, Q is a positive integer, q is the number of the UAV when multiple UAVs work cooperatively, , are proportional coefficients of the actual time deviation coefficient and the Bayesian likelihood probability coefficient respectively, , are both greater than 0;

[0022] A single UAV evaluation coefficient threshold is set, and the single UAV evaluation coefficient is compared with the single UAV evaluation coefficient threshold. If the single UAV evaluation coefficient is greater than the single UAV evaluation coefficient threshold, no warning signal is generated. If the single UAV evaluation coefficient is less than the single UAV evaluation coefficient threshold, the UAV with the single UAV evaluation coefficient less than the single UAV evaluation coefficient threshold generates a warning signal, and the remaining UAVs with the single UAV evaluation coefficient greater than the single UAV evaluation coefficient threshold are automatically redistributed by using the GWCA algorithm to perform tasks.

[0023] In a preferred embodiment, the matching degree information of multiple UAVs working cooperatively is collected, including:

[0024] The matching degree information of multiple UAVs working cooperatively is expressed by a cooperative matching degree coefficient;

[0025] The acquisition logic of the cooperative matching degree coefficient is as follows: the working progress of each UAV in the monitoring interval is obtained, the working progress of the UAV is determined based on the actual route obtained by the GWCA algorithm, and the ratio of the navigation distance of the UAV in the actual route to the total distance of the actual route is determined. The working progress of each UAV in the monitoring interval is marked as: , wherein, , is the navigation distance of the qth UAV in the actual route in the monitoring period, is the total distance of the actual route of the qth UAV in the monitoring period.

[0026] determining a correlation matrix of the multiple unmanned vehicles working cooperatively in the monitoring interval by a Pearson correlation coefficient, and marking the correlation matrix of the multiple unmanned vehicles working cooperatively in the monitoring interval as R, ; wherein, R is a Q*Q matrix, each element in the correlation matrix representing a Pearson correlation coefficient between unmanned vehicle j and unmanned vehicle k, m = 1, 2, 3, …, M, M is a positive integer, and m is the number of different time in the monitoring interval, is the working progress of the jth unmanned vehicle at the mth time, is the working progress of the kth unmanned vehicle at the mth time, is the average value of the working progress of the jth unmanned vehicle in the monitoring interval, is the average value of the working progress of the jth unmanned vehicle in the monitoring interval;

[0027] calculating the cooperative matching degree coefficient, and the calculation formula is: ; wherein, is the cooperative matching degree coefficient.

[0028] In a preferred embodiment, the risk information of the multiple unmanned vehicles working cooperatively is collected, including:

[0029] the risk information of the multiple unmanned vehicles working cooperatively is represented by a risk concealment coefficient;

[0030] the acquisition logic of the risk concealment coefficient is: different types of unmanned vehicles are divided by the use performance in the historical data of the unmanned vehicles, the risk events and the risk event results existing in different types of unmanned vehicles are determined, different types of unmanned vehicles are scored based on the influence degree of the risk event results on the cooperative work of the unmanned vehicles, and the risk score of the unmanned vehicles is obtained;

[0031] a regression model is constructed according to the number of unmanned vehicle types in the cooperative work of the multiple unmanned vehicles and the risk scores of different types of unmanned vehicles, and the risk concealment coefficient is obtained, and the expression of the risk concealment coefficient is: ; wherein, is the risk concealment coefficient, , , , …, is the number of different unmanned vehicle types in the cooperative work of the multiple unmanned vehicles, , , , …, is the risk score of different types of unmanned vehicles, and e is a natural number.

[0032] In a preferred embodiment, the performance of the multi-unmanned aerial vehicle cooperative work is quantified, comprising:

[0033] Through comprehensive analysis of the matching degree information and risk information of the multi-unmanned aerial vehicle cooperative work, the cooperative matching degree coefficient and the risk concealment coefficient are weighted and calculated to construct a cooperative work evaluation model to generate a cooperative work evaluation coefficient, and the calculation formula of the cooperative work evaluation coefficient is: ; wherein, is the cooperative work evaluation coefficient, , are the proportional coefficients of the cooperative matching degree coefficient and the risk concealment coefficient respectively, , are both greater than 0.

[0034] In a preferred embodiment, the alarm signal is generated, comprising:

[0035] The cooperative work evaluation coefficient threshold is set, the cooperative work evaluation coefficient is compared with the cooperative work evaluation coefficient threshold, if the cooperative work evaluation coefficient is less than the cooperative work evaluation coefficient threshold, the alarm signal is generated to inform the professional field worker to stop the multi-unmanned aerial vehicle cooperative work to avoid the risk in the working process, if the cooperative work evaluation coefficient is greater than the cooperative work evaluation coefficient threshold, the alarm signal is not generated.

[0036] The technical effects and advantages of the present application are:

[0037] The present application uses the GWCA algorithm to optimize the distribution of multi-unmanned aerial vehicle cooperative tasks, and combines real-time data and historical training data in the flight process to evaluate the path planning and flight risk of the unmanned aerial vehicle. Through the combination of the GWCA algorithm and flight path and risk analysis, efficient multi-unmanned aerial vehicle cooperative work management can be realized, and through dynamic adjustment of task distribution, quantification of cooperative work performance, potential risks can be discovered and responded in time, which helps the task scheduling of multi-unmanned aerial vehicles to become more flexible and intelligent, can cope with complex environment and task changes, and improves the robustness of the system. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to facilitate the understanding of those skilled in the art, the present application will be further described below in conjunction with the drawings;

[0039] Figure 1 is the flowchart of the present application based on the GWCA algorithm multi-unmanned aerial vehicle autonomous cooperative task distribution method. DETAILED DESCRIPTION

[0040] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0041] Embodiment 1

[0042] Figure 1 The flowchart of the method for autonomous cooperative task allocation of multiple unmanned aerial vehicles based on the GWCA algorithm of the present application specifically comprises the following steps:

[0043] S1: training unmanned aerial vehicles in a fixed scene based on the GWCA algorithm, determining the task execution sequence and route planning of the unmanned aerial vehicles at different positions in the fixed scene;

[0044] S2: determining the flight path information and flight risk information of a single unmanned aerial vehicle according to the current flight process data of the unmanned aerial vehicle and the historical training data of the unmanned aerial vehicle;

[0045] S3: comprehensively analyzing the flight path information and flight risk information of a single unmanned aerial vehicle, determining the unmanned aerial vehicle with problems, and re-distributing tasks to the remaining unmanned aerial vehicles using the GWCA algorithm;

[0046] S4: in the process of cooperative work of multiple unmanned aerial vehicles, through overall analysis of the multiple unmanned aerial vehicles, collecting the matching degree information and risk information of the cooperative work of the multiple unmanned aerial vehicles, quantifying the performance of the cooperative work of the multiple unmanned aerial vehicles, and generating an alarm signal.

[0047] The Great Wall Construction Algorithm (GWCA) is a heuristic optimization algorithm inspired by the competition and elimination mechanism among workers during the construction of the ancient Great Wall. It simulates the dynamic interaction between workers (or "individuals") to solve various constrained and unconstrained optimization problems. GWCA has the advantages of dynamic nature, simple structure, high convergence performance, and good solution quality, and has shown competitiveness and superiority in many optimization problems, especially for complex and dynamic optimization problems.

[0048] In the application of unmanned aerial vehicles, GWCA can be used for task allocation and path planning of multiple unmanned aerial vehicles. Specifically, it can solve problems such as how to allocate tasks and how to dynamically adjust task allocation strategies for unmanned aerial vehicles in multi-node tasks, including:

[0049] In GWCA, each "worker" represents a solution, and these workers update their positions according to a specific motion model during each iteration. Similarly, in multi-UAV task allocation, each "worker" can be seen as a task allocation scheme for a UAV, which is updated through algorithm iteration;

[0050] Workers improve their work efficiency through competition and elimination. In UAV task allocation, task allocation schemes are constantly optimized through competition and elimination mechanisms, ultimately selecting the optimal task allocation scheme;

[0051] Each UAV updates its "position" according to the current task status and target location, similar to the motion of workers in GWCA. In each iteration, the algorithm assigns a predefined "motion model" to each UAV (worker), which dynamically adjusts task allocation based on flight status, task priority, remaining battery life, and other information;

[0052] UAV path planning can borrow from the "labor motion model" in GWCA, simulating different flight models (such as shortest path model, efficient path model, etc.) to dynamically adjust flight paths to reduce energy consumption, avoid obstacles, and improve task completion efficiency;

[0053] Considering that UAVs may be affected by environmental, obstacle, or other unexpected situations during flight, GWCA assigns a random motion model to each worker, enabling UAVs to autonomously adjust task execution order and optimize task execution path when facing dynamic changes;

[0054] By simulating the dynamic process of workers, UAVs can automatically adjust their flight paths and task allocation under different environmental conditions. For example, when some UAVs cannot complete their tasks on time due to weather or battery issues, other UAVs can take over their tasks.

[0055] Based on the GWCA algorithm, UAVs are trained in a fixed scenario to determine the task execution order and route planning of UAVs at different locations in the fixed scenario. By using actual route distance instead of straight-line distance, task allocation and path planning are tightly coupled. That is, multiple UAVs are trained in a fixed scenario, and the optimal solution for multiple UAVs to execute tasks in a fixed scenario is gradually found through optimization. Based on the current flight process data of each UAV and the historical training data of the UAV, the flight path information and flight risk information of the individual UAV are determined. The flight path information of the individual UAV is represented by an actual time deviation coefficient, and the flight mode information of the individual UAV is represented by a Bayesian likelihood probability coefficient.

[0056] The acquisition logic of the actual time deviation coefficient is: determining a preset flight path of the unmanned aerial vehicle when performing a task by the GWCA algorithm, setting special marker points in the preset flight path, setting a monitoring time period, obtaining the special marker points passed by the unmanned aerial vehicle in the monitoring time period, determining a preset time point when the unmanned aerial vehicle arrives at the special marker points, marking the preset time point when the unmanned aerial vehicle arrives at the special marker points as: , setting a monitoring time period, obtaining the special marker points passed by the unmanned aerial vehicle in the monitoring time period, and determining a time point when the unmanned aerial vehicle actually arrives at the special marker points, marking the time point when the unmanned aerial vehicle actually arrives at the special marker points as: ;

[0057] By comparing the preset time point and the actual time point, the actual time deviation coefficient is determined, and the calculation formula of the actual time deviation coefficient is: ; wherein, is the actual time deviation coefficient.

[0058] It should be noted that the preset flight path of the unmanned aerial vehicle when performing a task is selected by the GWCA algorithm based on the different states of multiple unmanned aerial vehicles after training the scene, and different unmanned aerial vehicles should travel according to the preset flight path. The special marker points and the monitoring time period are set by professional staff, and there are usually multiple special marker points in the monitoring time period. The special marker points are usually marked according to the length of the preset flight path.

[0059] The actual time deviation coefficient can obtain the actual arrival time of the unmanned aerial vehicle in real time by monitoring the special marker points passed by the unmanned aerial vehicle on the preset flight path. By comparing the preset time point with the actual time point, deviations can be found in time and adjusted. This real-time feedback mechanism helps to quickly respond to the flight state of the unmanned aerial vehicle and ensures the smooth progress of the task.

[0060] The acquisition logic of the Bayesian likelihood probability coefficient is: determining flight characteristics of the unmanned aerial vehicle in the flight process, the flight characteristics including power, speed, height and wind speed, matching the flight characteristics with historical training data of the unmanned aerial vehicle in the scene, determining historical similar flight data by cosine similarity, and the calculation formula is: , wherein A is a flight characteristic vector of the current unmanned aerial vehicle, is a flight characteristic vector in the historical training data, and XS is a flight characteristic vector of the historical similar flight data.

[0061] It should be noted that the flight characteristics are determined by the characteristics of the unmanned aerial vehicle, and the flight characteristics determine the flight quality of the unmanned aerial vehicle. The flight quality is also affected by environmental factors such as wind speed. By matching the flight characteristics of the current unmanned aerial vehicle with the flight characteristics in the historical training data, the probability of the unmanned aerial vehicle successfully completing the current task is determined.

[0062] Based on historical similar flight data, the prior probability of the drone mission completion is determined and marked as: , determine the probability of task completion in historical similar flight data, and mark the probability of task completion in historical similar flight data as: ;

[0063] The probability of historically similar flight data is determined using kernel density estimation based on historical training data, and the probability of historically similar flight data is marked as: , the probability that the UAV can complete the mission under the current flight characteristics is calculated by the Bayesian formula, and the probability that the UAV can complete the mission is expressed by the Bayesian probability coefficient. The calculation formula is: ;in, is the Bayesian likelihood coefficient.

[0064] It should be noted that the standards for drones to complete tasks are set by professional staff. Usually, the degree of task completion is used to determine whether the drone is completed on time. The Bayesian probability coefficient is updated through prior knowledge and observed data (such as current flight characteristics). The probability estimate can be continuously adjusted as new data arrives. Therefore, when using the Bayesian formula to calculate the probability of task completion under the current flight characteristics, the relationship between the current flight status and historical data can be dynamically considered to adjust the probability of task completion in real time.

[0065] The flight path information and flight risk information of a single UAV are comprehensively analyzed, and weighted calculation is performed through the actual time deviation coefficient and the Bayesian probability coefficient to construct a single UAV evaluation model and generate a single UAV evaluation coefficient. The calculation formula for the single UAV evaluation coefficient is: ;in, is the evaluation coefficient of a single UAV, q=1, 2, 3, ..., Q, Q is a positive integer, and q is the number of the UAV when multiple UAVs work together. 、 are the proportional coefficients of the actual time deviation coefficient and the Bayesian likelihood probability coefficient, 、 Both are greater than 0.

[0066] It can be seen from the formula that the smaller the actual time deviation coefficient and the larger the Bayesian probability coefficient, the larger the single-UAV evaluation coefficient, which indicates that the UAV performs better when multiple UAVs work together, and the possibility of the UAV encountering problems and failing to complete the task is low. Conversely, the larger the actual time deviation coefficient and the smaller the Bayesian probability coefficient, the smaller the single-UAV evaluation coefficient, which indicates that the UAV performs poorly when multiple UAVs work together.

[0067] The single unmanned aerial vehicle evaluation coefficient threshold is set, the single unmanned aerial vehicle evaluation coefficient is compared with the single unmanned aerial vehicle evaluation coefficient threshold, if the single unmanned aerial vehicle evaluation coefficient is greater than the single unmanned aerial vehicle evaluation coefficient threshold, no early warning signal is generated, if the single unmanned aerial vehicle evaluation coefficient is less than the single unmanned aerial vehicle evaluation coefficient threshold, the unmanned aerial vehicle less than the single unmanned aerial vehicle evaluation coefficient threshold generates an early warning signal, and the remaining unmanned aerial vehicles greater than the single unmanned aerial vehicle evaluation coefficient threshold are automatically redistributed by using the GWCA algorithm to perform the task.

[0068] In the process of the multi-unmanned aerial vehicle cooperative work, the matching degree information and the risk information of the multi-unmanned aerial vehicle cooperative work are collected through the overall analysis of the multi-unmanned aerial vehicles, the matching degree information of the multi-unmanned aerial vehicle cooperative work is represented by a cooperative matching degree coefficient, and the risk information of the multi-unmanned aerial vehicle cooperative work is represented by a risk concealment coefficient.

[0069] The cooperative matching degree coefficient is obtained by: obtaining the work progress of each unmanned aerial vehicle in the monitoring interval, the work progress of the unmanned aerial vehicle being determined based on the actual route obtained by the GWCA algorithm, and the work progress of each unmanned aerial vehicle in the monitoring interval being marked as: , wherein, , is the navigation distance of the qth unmanned aerial vehicle in the actual route in the monitoring period, is the total distance of the actual route of the qth unmanned aerial vehicle in the monitoring period;

[0070] The correlation matrix of the multi-unmanned aerial vehicles in the monitoring interval is determined by the Pearson correlation coefficient, and the correlation matrix of the multi-unmanned aerial vehicles in the monitoring interval is marked as R, ; wherein, R is a Q*Q matrix, and each element in the correlation matrix represents the Pearson correlation coefficient between the unmanned aerial vehicle j and the unmanned aerial vehicle k, m = 1, 2, 3, …, M, M is a positive integer, and m is the number of different time points in the monitoring interval, is the work progress of the jth unmanned aerial vehicle at the mth time point, is the work progress of the kth unmanned aerial vehicle at the mth time point, is the average value of the work progress of the jth unmanned aerial vehicle in the monitoring interval, is the average value of the work progress of the jth unmanned aerial vehicle in the monitoring interval;

[0071] The cooperative matching degree coefficient is calculated, and the calculation formula is: ; wherein, is the cooperative matching degree coefficient.

[0072] It should be noted that the synergy matching degree coefficient can quickly evaluate the synergy between the unmanned vehicles, track the synergy performance of the unmanned vehicles in real time by dynamically updating the correlation matrix, and quickly reflect changes in environmental conditions (such as wind speed, air temperature, etc.) or task requirements, helping decision makers to make adaptive adjustments to the current unmanned vehicle task execution state. If the synergy matching degree coefficient of the multiple unmanned vehicles is low when they work in synergy, it indicates that the synergy performance of the multiple unmanned vehicles is poor.

[0073] The acquisition logic of the risk concealment coefficient is as follows: different types of unmanned vehicles are divided according to the use performance in the historical data of the unmanned vehicles, risk events and risk event results existing in different types of unmanned vehicles are determined, different types of unmanned vehicles are scored based on the influence degree of the risk event results on the cooperative work of the unmanned vehicles, and the risk score of the unmanned vehicles is obtained.

[0074] It should be noted that the risk events and the possible results of these risk events in the historical tasks of different types of unmanned vehicles are analyzed. Common risk events include failure to arrive on time, failure to complete tasks, etc. According to the use performance in the historical data, the unmanned vehicles can be divided into different types. Generally, the types of unmanned vehicles can be classified according to their performance, usage frequency, environmental adaptability, etc.

[0075] According to the number of unmanned vehicle types in the cooperative work of multiple unmanned vehicles and the risk scores of different types of unmanned vehicles, a regression model is constructed to obtain the risk concealment coefficient. The expression of the risk concealment coefficient is as follows: ; wherein, is the risk concealment coefficient, , , , …, is the number of different unmanned vehicle types in the cooperative work of multiple unmanned vehicles, , , , …, is the risk score of different types of unmanned vehicles, and e is a natural number.

[0076] It should be noted that the risk events of different types of unmanned vehicles are analyzed through historical data, which can more accurately identify the risks that each type of unmanned vehicle may face. According to the performance, environmental adaptability, usage frequency, etc. of the unmanned vehicles, different risk events (such as failure to arrive on time, failure to complete tasks, etc.) are scored, which can quantify the influence degree of each risk event, making the risk assessment more personalized and targeted, rather than just a unified standard.

[0077] In the scenario of multiple UAVs working cooperatively, the risk score of each UAV not only reflects its own risk, but also reflects its potential impact in the cooperative task. By evaluating the risk scores of different types of UAVs, the risk level of the overall task can be understood, and the task allocation can be optimized.

[0078] By comprehensive analysis of the matching degree information and risk information of multiple UAVs working cooperatively, the cooperative matching degree coefficient and the risk concealment coefficient are weighted and calculated to construct a cooperative working evaluation model, and a cooperative working evaluation coefficient is generated. The calculation formula of the cooperative working evaluation coefficient is: ; wherein, is the cooperative working evaluation coefficient, , the cooperative matching degree coefficient and the risk concealment coefficient are respectively proportional coefficients, , all greater than 0.

[0079] As can be seen from the formula, the greater the cooperative matching degree coefficient and the smaller the risk concealment coefficient, the greater the cooperative working evaluation coefficient, which means that the performance of multiple UAVs working cooperatively is better. Conversely, the smaller the cooperative matching degree coefficient and the greater the risk concealment coefficient, the smaller the cooperative working evaluation coefficient, which means that the performance of multiple UAVs working cooperatively is worse.

[0080] Set the cooperative working evaluation coefficient threshold, compare the cooperative working evaluation coefficient with the cooperative working evaluation coefficient threshold, if the cooperative working evaluation coefficient is less than the cooperative working evaluation coefficient threshold, generate an alarm signal to notify the professional staff to stop the cooperative work of multiple UAVs to avoid risks in the working process, if the cooperative working evaluation coefficient is greater than the cooperative working evaluation coefficient threshold, no alarm signal is generated.

[0081] The present application uses the GWCA algorithm to optimize the allocation of multiple UAVs working cooperatively, and combines real-time data and historical training data in the flight process to evaluate the path planning and flight risk of the UAV. By combining the GWCA algorithm with flight path and risk analysis, efficient multiple UAV cooperative work management can be achieved, and by dynamically adjusting task allocation, quantifying cooperative work performance, and timely discovering and responding to potential risks, it helps the task scheduling of multiple UAVs to become more flexible and intelligent, and can cope with complex environments and task changes, improving the robustness of the system.

[0082] The above formulas are dimensionless and the numerical values are calculated. The formula is obtained by software simulation of a large amount of data to obtain the most recent real situation. The preset parameters in the formula are set by the person skilled in the art according to the actual situation.

[0083] The above-described embodiments can be implemented in part or in whole through software implemented by a processor. The results of such software can be stored on storage medium, such as a computer program product. The storage medium can include persistent storage and / or removable storage depending on the implementation. Examples of persistent storage include solid state drives, versioned file systems, database systems, and other storage systems. Examples of removable storage include 'flash' (e.g., SD cards, mini and micro SD cards, etc.), 'USB' (e.g., USB thumb drives, etc.), and other storage systems. Each of these examples can be, and is, a computer-readable medium. Whether persistent storage or removable storage, the storage medium is a tangible storage that is one or both of read only and nonvolatile. Accordingly, the storage medium is distinct from mere signal per se. The computer-readable medium storing the computer program product can be included with, or peripheral to, a computer system. Such computer-readable media include, but are not limited to: volatile and non-volatile, removable and non-removable tangible storage media dictated by electronic configuration and use. Computer-readable media that store data, which is both static (e.g., recorded on a carrier) and dynamic (e.g., broadcast) are both storage media. Examples of computer-readable media include, but are not limited to: primary computer readable storage media, secondary computer readable storage media, and communication media.

[0084] It should be understood that the sequence size of the above processes does not mean the execution order, the execution order of the processes should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0085] In several embodiments provided by the present application, it should be understood that the disclosed system, device, and method can be implemented by other means. For example, the above-described device embodiments are only illustrative, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be through some interface, device or unit indirect coupling or communication connection, which can be electrical, mechanical or other forms.

[0086] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0087] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. The multi-UAV autonomous collaborative task allocation method based on GWCA algorithm is characterized by: The specific steps include: S1: Based on the GWCA algorithm, the UAV is trained in a fixed scene to determine the task execution sequence and route planning of the UAV at different locations in the fixed scene; S2: Determine the flight path information and flight risk information of each drone based on the current flight process data of each drone and the drone's historical training data; S3: Comprehensively analyze the flight path information and flight risk information of each drone, identify the drone with problems, and reassign tasks to the remaining drones using the GWCA algorithm; S4: During the multi-UAV collaborative work process, through the overall analysis of the multi-UAVs, the matching information and risk information of the multi-UAV collaborative work are collected, the performance of the multi-UAV collaborative work is quantified, and an alarm signal is generated; The flight path information of a single UAV is represented by the actual time deviation coefficient; The logic for obtaining the actual time deviation coefficient is as follows: determine the preset route path of the drone when performing the mission through the GWCA algorithm, set special marking points in the preset route path, set a monitoring time period, obtain the special marking points passed by the drone during the monitoring time period, determine the preset time point when the drone arrives at the special marking point, and mark the preset time point when the drone arrives at the special marking point as: , set the monitoring time period, obtain the special marking points that the drone passes during the monitoring time period, and determine the time point when the drone actually arrives at the special marking point, and mark the time point when the drone actually arrives at the special marking point as: ; By comparing the preset time point with the actual time point, the actual time deviation coefficient is determined. The calculation formula of the actual time deviation coefficient is: ;in, is the actual time deviation coefficient; The flight risk information of a single UAV is represented by the Bayesian probability coefficient; The logic for obtaining the Bayesian likelihood coefficient is as follows: determine the flight characteristics of the drone during flight, including power, speed, altitude, and wind speed, match them with the flight characteristics in the drone's historical training data in the scenario, and determine historical similar flight data through cosine similarity. The calculation formula is: , where A is the flight feature vector of the current UAV, is the flight feature vector in the historical training data, and XS is the flight feature vector in the historical similar flight data; Based on historical similar flight data, the prior probability of the drone mission completion is determined and marked as: , determine the probability of task completion in historical similar flight data, and mark the probability of task completion in historical similar flight data as: ; The probability of historically similar flight data is determined using kernel density estimation based on historical training data, and the probability of historically similar flight data is marked as: , the probability that the UAV can complete the mission under the current flight characteristics is calculated by the Bayesian formula, and the probability that the UAV can complete the mission is expressed by the Bayesian probability coefficient. The calculation formula is: ;in, is the Bayesian likelihood coefficient; Comprehensively analyze the flight path information and flight risk information of a single drone, including: The flight path information and flight risk information of a single UAV are comprehensively analyzed, and weighted calculation is performed through the actual time deviation coefficient and the Bayesian probability coefficient to construct a single UAV evaluation model and generate a single UAV evaluation coefficient. The calculation formula for the single UAV evaluation coefficient is: ;in, is the evaluation coefficient of a single UAV, q=1, 2, 3, ..., Q, Q is a positive integer, and q is the number of the UAV when multiple UAVs work together. 、 are the proportional coefficients of the actual time deviation coefficient and the Bayesian likelihood probability coefficient, 、 are both greater than 0; Set a single drone evaluation coefficient threshold, compare the single drone evaluation coefficient with the single drone evaluation coefficient threshold. If the single drone evaluation coefficient is greater than the single drone evaluation coefficient threshold, no warning signal is generated. If the single drone evaluation coefficient is less than the single drone evaluation coefficient threshold, a warning signal is generated for the drones less than the single drone evaluation coefficient threshold, and the remaining drones greater than the single drone evaluation coefficient threshold are automatically reassigned tasks using the GWCA algorithm. The matching degree information of multiple UAVs working together is expressed by the collaborative matching coefficient; The logic for obtaining the collaborative matching coefficient is as follows: the progress of each UAV working collaboratively within the monitoring interval is obtained. The progress of the UAV working collaboratively is determined based on the actual route obtained by the GWCA algorithm, and is determined based on the ratio of the UAV's navigation distance in the actual route to the total distance of the actual route. The progress of each UAV working collaboratively within the monitoring interval is marked as: ,in, , is the actual flight distance of the qth UAV during the monitoring period, is the total distance of the actual route of the qth UAV during the monitoring period; The correlation matrix of multiple UAVs working together in the monitoring interval is determined by the Pearson correlation coefficient, and the correlation matrix of multiple UAVs working together in the monitoring interval is marked as: R, ;in, , R is a Q*Q matrix, each element in the correlation matrix represents the Pearson correlation coefficient between UAV j and UAV k, , m=1, 2, 3, ..., M, M is a positive integer, m is the number of different moments in the monitoring interval, is the work progress of the j-th UAV at the m-th moment, is the work progress of the k-th UAV at the m-th moment, is the average working progress of the jth UAV in the monitoring interval, is the average working progress of the jth UAV in the monitoring interval; Calculate the collaborative matching coefficient, the calculation formula is: ;in, is the synergistic matching coefficient; The risk information of multi-UAV collaborative work is expressed by the risk concealment coefficient; The logic for obtaining the risk concealment coefficient is as follows: drones are classified into different types based on their usage performance in historical data, risk events and risk event outcomes for different types of drones are determined, and different types of drones are scored based on the degree of impact of risk event outcomes on drone collaboration to obtain a drone risk score. According to the number of drone types in multi-drone collaboration and the risk scores of different types of drones, a regression model is constructed to obtain the risk concealment coefficient. The expression of the risk concealment coefficient is: ;in, is the risk concealment coefficient, 、 、 、……、 is the number of different drone types in multi-drone collaboration, 、 、 、……、 Score the risk of different drone types, where e is a natural number; Quantify the performance of multi-UAV collaboration, including: Through comprehensive analysis of the matching information and risk information of multi-UAV collaborative work, the collaborative matching coefficient and risk concealment coefficient are weighted and calculated to construct a collaborative work evaluation model and generate a collaborative work evaluation coefficient. The calculation formula of the collaborative work evaluation coefficient is: ;in, is the collaborative work evaluation coefficient, 、 are the proportional coefficients of the synergy matching coefficient and the risk concealment coefficient, 、 Both are greater than 0.

2. The multi-UAV autonomous collaborative task allocation method based on the GWCA algorithm according to claim 1 is characterized in that: Generate alarm signals, including: A collaborative work evaluation coefficient threshold is set, and the collaborative work evaluation coefficient is compared with the collaborative work evaluation coefficient threshold. If the collaborative work evaluation coefficient is less than the collaborative work evaluation coefficient threshold, an alarm signal is generated to notify professional staff to stop the collaborative work of multiple drones to avoid risks during the work process. If the collaborative work evaluation coefficient is greater than the collaborative work evaluation coefficient threshold, no alarm signal is generated.

Images