A method and system for allocating unmanned aerial vehicle tasks

Abstract

The present invention relates to a method for allocating unmanned aerial vehicle tasks based on an improved chaotic adaptive genetic algorithm, and the method belongs to the field of information technology. The specific steps are described as follows: ① Based on the chaotic mapping, a random population is generated as the initial population of the genetic algorithm; ② Based on the fitness of the initial population of the genetic algorithm, the definition of the interval concentration adjustment parameter is given; ③ An improved adaptive crossover probability and mutation probability calculation method is proposed; ④ Based on the improved adaptive crossover probability and mutation probability calculation method, a method for allocating unmanned aerial vehicle tasks based on an improved chaotic adaptive genetic algorithm is designed. The present invention fully considers the respective advantages of the adaptive genetic algorithm and the chaotic genetic algorithm, and proposes a method for allocating unmanned aerial vehicle tasks based on an improved chaotic adaptive genetic algorithm, which provides a certain reference for the task allocation of unmanned aerial vehicle clusters.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method and system for allocating tasks for unmanned aerial vehicles (UAVs), in particular to a method for allocating tasks for UAVs using a chaotic adaptive genetic algorithm, and specifically to a method for allocating tasks for UAVs based on an improved chaotic adaptive genetic algorithm. Background Art

[0002] Unmanned aerial vehicles (UAVs) are widely used in battlefield reconnaissance, joint attacks, emergency rescue and other operations due to their low cost, high flexibility and strong concealment. With the rapid development and widespread application of UAV technology, the task allocation problem of UAV swarms has become one of the research hotspots. UAV task allocation is to assign one or a group of ordered tasks to UAVs and ensure that their overall benefits are optimal. Genetic algorithm is a classic algorithm for UAV task allocation swarm intelligence. It has the advantages of strong global search ability and good adaptability. However, there are still problems of premature maturity and falling into local optimality in genetic algorithms. Chaos algorithm has randomness and determinism, which expands the search space, enhances the diversity of initial solutions, and avoids falling into local optimal solutions.

[0003] Based on this, the present invention analyzes the advantages and disadvantages of genetic algorithm and chaos algorithm, and provides a UAV task allocation method based on improved chaos adaptive genetic algorithm: Figure 1 As shown in the figure, the randomness and determinism of chaotic mapping are used to generate the initial population of genetic algorithm to prevent premature falling into the local optimal solution; based on the fitness of the initial population of genetic algorithm, the interval concentration adjustment parameter definition is given; the calculation method of the crossover probability and mutation probability of the improved adaptive genetic algorithm is designed, and the parameters are adaptively adjusted according to the fitness results; based on the calculation method of the improved adaptive crossover probability and mutation probability, a UAV task allocation method based on the improved chaotic adaptive genetic algorithm is designed. This method fully considers the randomness and determinism of the chaotic algorithm, can enhance the global search ability of the algorithm, and the adaptive crossover rate and mutation rate can automatically adjust the parameters according to the fitness and diversity of the current population to avoid premature problems. This method enables UAVs to respond to various tasks efficiently, especially in emergency situations, to quickly adjust task priorities and resource allocation, significantly improving the efficiency and success rate of task execution. Therefore, the UAV allocation method of chaotic adaptive genetic algorithm has become an optimization tool with great potential and application value in UAV task allocation.

[0004] In addition, on the one hand, there are differences in understanding among those skilled in the art; on the other hand, the applicant studied a large number of documents and patents when making the present invention, but due to space limitations, not all details and contents are listed in detail. However, this does not mean that the present invention does not have the characteristics of these prior arts. On the contrary, the present invention already has all the characteristics of the prior art, and the applicant reserves the right to add relevant prior art to the background technology. Summary of the invention

[0005] The purpose of the present invention is to provide a chaotic adaptive UAV task allocation method, considering that the existing methods still have a lot of room for improvement in the problems of premature maturity and falling into local optimality in genetic algorithms. The present invention can expand the search space and enhance the diversity of initial solutions based on the randomness and determinism of the chaotic algorithm, avoiding falling into the local optimal solution, while the adaptive crossover rate and mutation rate improve the convergence speed of the algorithm and avoid premature maturity and other problems. The present invention designs a UAV task allocation method based on an improved chaotic adaptive genetic algorithm through the advantages of the chaotic algorithm and the advantages of the adaptive genetic algorithm.

[0006] To achieve the above purpose, the present invention proposes a method for allocating tasks of unmanned aerial vehicles. The present invention proposes a method for allocating tasks of unmanned aerial vehicles based on an improved chaotic adaptive genetic algorithm. Figure 1 As shown, the method comprises the following steps:

[0007] (1) Based on chaotic mapping, the initial population definition of chaotic genetic algorithm is given;

[0008] (2) Based on the initial population fitness, the interval concentration adjustment parameter definition is given;

[0009] (3) Propose an improved method for calculating adaptive crossover probability and mutation probability;

[0010] (4) Based on the improved crossover probability and mutation probability, an improved chaotic adaptive genetic algorithm method for UAV task allocation is designed.

[0011] According to a preferred embodiment, the initial population of the genetic algorithm is generated based on the chaotic Logistic map. The population randomly generated by the existing genetic algorithm may contain a large number of low-quality individuals, which will lead to low population quality and easy to fall into the local optimal solution. The present invention proposes to use the ergodicity of the Logistic chaotic map to perform a global search to improve the quality of the initial population individuals. The expression of the Logistic chaotic map is:

[0012] s r+1 =μs r (1-s r ), r=0,1,2,… (1),

[0013] Wherein, μ is a control parameter, usually 3.57<μ≤4; different μ values ​​will result in different states of the Logistic curve. Preferably, the μ value selected in the present invention is 4.0.

[0014] According to a preferred embodiment, based on the initial population of the genetic algorithm, the definition of the interval concentration adjustment parameter is given. The definition is as follows:

[0015] (1) To adaptively adjust the crossover probability and mutation probability, define γ and λ as the concentration of two intervals.

[0016] The adjustment parameters are:

[0017]

[0018] (2) Assume that f' is the larger fitness value of the individual to be crossed, f avg is the average fitness value of the population, f max is the maximum fitness value of the population, f min is the minimum fitness value of the population. To characterize the distribution of the population in the upper and lower intervals of the average fitness value, define parameters a (upper interval) and b (lower interval), whose ranges are 0.5 and <a<1、0.5<b<1。

[0019] (3) The specific adjustment judgment method is as follows: when λ>a, it indicates that the individuals are concentrated in the upper half, and the closer λ is to 1, the higher the concentration in the upper half; when γ>b, it indicates that the individuals are concentrated in the lower half, and the closer γ is to 1, the higher the concentration in the lower half.

[0020] By analyzing the costs of different parameters under the same conditions, the optimal concentration adjustment parameters are selected to avoid excessive adjustments in the adaptive crossover rate and mutation rate and increase smoothness.

[0021] According to a preferred embodiment, based on the interval concentration adjustment parameter definition, an improved adaptive crossover probability and mutation probability calculation method is proposed. Considering that the adaptive crossover probability and mutation probability can improve the convergence speed and global search ability of the algorithm and avoid premature convergence to the local optimal solution, the present invention will use the improved adaptive crossover probability and mutation probability to perform UAV task allocation.

[0022] In the process of UAV task allocation, when the fitness value of the task allocation scheme is higher than the average fitness value, the adaptive crossover rate P is reduced. c and adaptive mutation rate P m ; When the fitness value of the task allocation scheme is not higher than the average fitness value, increase P c and P m .

[0023] When calculating the crossover probability and mutation probability, the effect of the Sigmoid function on the parameters a and b can adapt the size of the adjustment item to different fitness levels. In this way, at high fitness, the adjustment of the crossover probability is more sensitive, which helps to protect the excellent solution; at low fitness, the adjustment of the crossover probability is more moderate, which helps to avoid the algorithm from falling into the local optimal solution. Sigmoid is a smooth adjustment function, which is defined as:

[0024]

[0025] Based on the above principles and judgment methods, a new definition of crossover probability and mutation probability is proposed:

[0026] (1) When f'>f avg When P c and P m They are defined as follows:

[0027]

[0028] (2) When f'<=f avg When P c and P m They are defined as follows:

[0029]

[0030] Among them, P c1 , P c2 is the initial crossover probability of the individual to be crossed; P m1 , P m2 is the initial mutation probability of the individual to be mutated. 0.08 and 0.008 are the adjustment coefficients of the concentration adjustment parameter.

[0031] According to a preferred embodiment, a chaotic adaptive genetic algorithm UAV task allocation method is designed based on chaotic mapping and improved adaptive crossover probability and mutation probability.

[0032] The detailed task allocation process is as follows:

[0033] Step 1: Input: 5-tuple<B,U,M,T,R> .

[0034] Step 2: Initialize the population using chaotic mapping.

[0035] Step 3: Use formula (1)(3)(4) to calculate the UAV task allocation process.

[0036] Step 4: Record the optimal solution of the current iteration.

[0037] Step 5: Use roulette probability to select individuals with higher fitness.

[0038] Step 6: Use the improved adaptive crossover rate P c Perform crossover operation.

[0039] Step 7: Use the improved adaptive mutation rate P m Perform mutation operation.

[0040] Step8: output: optimal fitness value.

[0041] Another aspect of the present invention relates to a UAV task allocation system based on an improved chaotic adaptive genetic algorithm. The system is configured as follows:

[0042] (1) Based on chaotic mapping, the initial population definition of chaotic genetic algorithm is given;

[0043] (2) Based on the initial population fitness, the interval concentration adjustment parameter definition is given;

[0044] (3) Propose an improved method for calculating adaptive crossover probability and mutation probability;

[0045] (4) Based on the improved crossover probability and mutation probability, an improved chaotic adaptive genetic algorithm method for UAV task allocation is designed.

[0046] The beneficial effects of the present invention relative to the prior art are:

[0047] Based on the chaotic mapping, the initial population of the genetic algorithm is generated; secondly, the generated initial population is used to give the definition of the interval concentration adjustment parameter; then, according to the interval concentration adjustment parameter, the adaptive crossover probability and mutation probability are improved; finally, the chaotic algorithm and the improved adaptive genetic algorithm are combined to design a chaotic adaptive genetic algorithm. This invention effectively describes the algorithm's consideration of the payload capacity, flight endurance, functionality and other characteristics of the drone during the task allocation process. At the same time, factors such as the value benefit and time required for each task are also evaluated to ensure that the allocation plan can not only meet the task requirements but also maximize the efficiency of drone resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A flow chart of the method involved in this application;

[0049] Figure 2 This is a comparison chart of different parameters;

[0050] Figure 3 The objective function graph for task completion;

[0051] Figure 4 Provide a cost map for the UAV to perform missions;

[0052] Figure 5Develop a time cost map for UAV mission execution;

[0053] Figure 6 This is the target benefit diagram for drones. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. For those skilled in the art, the specific meanings of the terms in the present invention can be understood in specific circumstances.

[0055] This embodiment provides a method for coupling a dependent command and control network. In order to verify the feasibility, applicability and effectiveness of the present invention, simulation analysis is designed from five aspects. The analysis includes different adaptive crossover probability and mutation probability adjustment parameters, different objective functions for task completion, different range costs, comparison of the time cost of UAV task execution, and comparison of the target benefits of UAV task execution. The initial parameters are set as follows: the number of multi-functional UAVs is 9 as shown in Table 1, the initial tasks are 30 as shown in Table 2; the tasks are classified into: reconnaissance tasks, attack tasks, and damage assessment tasks.

[0056] (1) The influence of adaptive crossover probability and mutation probability adjustment parameters on the objective function of UAV mission execution

[0057] In order to analyze the selection of adaptive adjustment coefficient, the task allocation cost under the same conditions and different parameters is simulated and analyzed; Figure 2 It can be seen that all parameter combinations quickly reduce the cost value in the initial stage, indicating that the algorithm has good initial convergence performance. In particular, the blue line (0.08 & 0.008) and the red line (0.09 & 0.009) drop most rapidly in the early stage, indicating that this parameter combination is more efficient in the early stage of the algorithm. The cost values ​​of the adjustment parameters 0.08 and 0.008 selected by the present invention are the lowest, indicating that the adjustment effect of coefficients 0.08 and 0.008 is the best, followed by 0.09 and 0.009; based on the above analysis, the optimal values ​​of the adaptive adjustment coefficients are 0.08 and 0.008.

[0058] (2) Comparison of objective functions for task completion

[0059] like Figure 3 As shown in Figure 2, after 500 iterations, the objective function values ​​obtained by the four algorithms when completing task allocation. The smaller the objective function value, the better the solution obtained. Figure 3 It can be seen that the algorithm CAGA proposed in this paper has the lowest objective function value compared with other algorithms, which proves that CAGA can obtain a better allocation solution.

[0060] (3) Comparison of the range costs of UAV missions

[0061] Depend on Figure 4 It can be seen that as the number of iterations increases, the total flight distance is constantly decreasing. Compared with other algorithms, the CAGA algorithm proposed in this paper has the smallest flight distance when the task is completed and the fastest convergence speed; the solution result of the SAGA algorithm is second, followed by AGA, and the GA solution result is the worst; therefore, it shows that the method of the present invention has a better choice in UAV route planning.

[0062] (4) Comparison of time costs of UAV mission execution

[0063] from Figure 5 It can be observed that as the number of iterations increases, the time cost of the UAV to perform the task is constantly decreasing; the CAGA algorithm has the fastest convergence speed and the lowest task time cost. This shows that the algorithm of the present invention still has significant advantages in performance.

[0064] (5) Comparison of target benefits of UAV missions

[0065] like Figure 6 As shown in Figure 1, the target total revenue of the four algorithms after task allocation is given in 500 iterations. Figure 6 The performance of different algorithms for the target benefit in the UAV task allocation problem is shown. The figure compares the CAGA algorithm proposed in this paper with the existing genetic algorithm and two different adaptive genetic algorithms.

[0066] From the simulation results, it can be seen that the target benefits of the four algorithms rise rapidly in the initial stage (the first 50 iterations), indicating that each algorithm can quickly find a good solution in the initial stage. However, as the number of iterations increases, the performance of each algorithm begins to differ. As shown in Table 3, the simulation results show that the use of the adaptive genetic algorithm (CAGA) can significantly improve the target benefit of UAV task allocation compared with other versions of the adaptive genetic algorithm (AGA and SAGA), especially CAGA, which finally achieved the highest target benefit by introducing mixed mutation operations.

[0067] In summary, the UAV task allocation method based on the improved chaotic adaptive genetic algorithm proposed in this invention improves the performance indicators such as the objective function, range cost, time cost, and target benefit of task allocation, and effectively solves the problems of low convergence accuracy and slow convergence speed of the existing adaptive genetic algorithm. In the future, we will further combine the actual application of UAV clusters to explore parameter selection and task types that are more in line with application scenarios to improve the adaptability of the algorithm.

[0068] Table 1 UAV parameter settings

[0069]

[0070] Table 2 Task point parameter settings

[0071]

[0072]

[0073] Table 3 Simulation parameters of four algorithms

[0074]

[0075]

[0076] It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the disclosure scope of the present invention and fall within the protection scope of the present invention. Those skilled in the art should understand that the present invention specification and its drawings are illustrative and do not constitute a limitation on the claims. The scope of protection of the present invention is defined by the claims and their equivalents. The present invention specification contains multiple inventive concepts, such as "preferably", "according to a preferred embodiment" or "optionally", all of which indicate that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application based on each inventive concept. Throughout the text, the features guided by "preferably" are only an optional method and should not be understood as being required. Therefore, the applicant reserves the right to abandon or delete the relevant preferred features at any time.

Claims

1. A method for assigning unmanned aerial vehicle tasks based on an improved chaotic adaptive genetic algorithm, characterized in that: The steps include: (1) Based on chaotic mapping, the initial population definition of chaotic genetic algorithm is given; (2) Based on the initial population fitness, the interval concentration adjustment parameter definition is given, and γ and λ are defined as two interval concentration adjustment parameters: Among them, f' is the maximum fitness value of the individual to be crossed, f avg is the average fitness value of the population, f max is the maximum fitness value of the population, f min is the minimum fitness value of the population, In order to characterize the distribution of the population in the upper and lower intervals of the average fitness value, the parameter a is defined as the upper half interval and b as the lower half interval; (3) An improved method for calculating the adaptive crossover probability and mutation probability is proposed. When calculating the crossover probability and mutation probability, the effect of the Sigmoid function on the parameters a and b makes the size of the adjustment item adapt to different fitness levels. The Sigmoid is a smooth adjustment function, which is defined as: The new crossover probability and mutation probability are defined as: (i) When f'>f avg When P c and P m They are defined as follows: (ii) When f'<=f avg When P c and P m They are defined as follows: Among them, P c1 , P c2 is the initial crossover probability of the individual to be crossed; P m1 , P m2 is the initial mutation probability of the individual to be mutated; (4) Based on the improved crossover probability and mutation probability, an improved chaotic adaptive genetic algorithm method for UAV task allocation is designed.

2. The method for allocating unmanned aerial vehicle tasks according to claim 1, characterized in that: Based on chaotic Logistic mapping, the definition of initial population of chaotic genetic algorithm is given, in which global search is carried out based on the ergodic advantage of Logistic chaotic mapping, thereby improving the individual quality of initial population.

3. The method for allocating unmanned aerial vehicle tasks according to claim 2, characterized in that: The expression of the Logistic chaotic map is: s r+1 =μs r (1-s r ),r=0,1,2,L (1), Where μ is the control parameter; with different μ values, the Logistic curve presents different states.

4. The method for allocating unmanned aerial vehicle tasks according to claim 3, characterized in that: The μ value is greater than 3.57 and not greater than 4.

5. The method for allocating unmanned aerial vehicle tasks according to claim 4, characterized in that: The value of μ is 4.

6. The method for allocating unmanned aerial vehicle tasks according to claim 3, characterized in that: Based on the initial population fitness of the genetic algorithm, the definition of the interval concentration adjustment parameter is given, and the definition includes the following: 0.5 <a<1、0.5<b<1; The specific adjustment judgment method is as follows: when λ>α, it indicates that the individuals are concentrated in the upper half, and the closer λ is to 1, the higher the concentration in the upper half; when γ>b, it indicates that the individuals are concentrated in the lower half, and the closer γ is to 1, the higher the concentration in the lower half.

7. The method for allocating unmanned aerial vehicle tasks according to claim 6, characterized in that: In the process of UAV task allocation, when the fitness value of the task allocation scheme is higher than the average fitness value, the adaptive crossover probability P is reduced. c and adaptive mutation probability P m ; When the fitness value of the task allocation scheme is not higher than the average fitness value, increase P c and P m .

8. The method for allocating unmanned aerial vehicle tasks according to claim 7, characterized in that: Based on the improved crossover probability and mutation probability, an improved chaotic adaptive genetic algorithm UAV task allocation method is designed. The detailed task allocation process is as follows: Step 1: Input-quintuple<B,U,M,T,R> ; Step 2: Initialize the population using chaotic mapping; Step 3: Use formulas (1), (3), and (4) to calculate the UAV task allocation process; Step 4: Record the optimal solution of the current iteration; Step 5: Use roulette probability to select individuals with higher fitness; Step 6: Use the improved adaptive crossover probability P c Perform crossover operations; Step 7: Use the improved adaptive mutation probability P m Perform mutation operations; Step8: output-optimal fitness value.

9. A UAV task allocation system based on improved chaotic adaptive genetic algorithm, characterized in that: The system is capable of executing the drone task allocation method according to any one of claims 1 to 8, and is configured as follows: (1) Based on chaotic mapping, the initial population definition of chaotic genetic algorithm is given; (2) Based on the initial population fitness, the interval concentration adjustment parameter definition is given; (3) Propose an improved method for calculating adaptive crossover probability and mutation probability; (4) Based on the improved crossover probability and mutation probability, an improved chaotic adaptive genetic algorithm method for UAV task allocation is designed.

Images