Heterogeneous unmanned aerial vehicle path planning method and system

Abstract

The invention provides a heterogeneous unmanned aerial vehicle path planning method and system, and the method comprises the steps: carrying out the intersection processing and variation processing of gene segments of chromosomes in an unmanned aerial vehicle path planning chromosome population in a current iteration period based on an improved genetic algorithm, and obtaining a new unmanned aerial vehicle path planning chromosome population; and when it is determined that the individual fitness of chromosomes in the new unmanned aerial vehicle path planning chromosome population meets preset fitness, obtaining a target chromosome from the new unmanned aerial vehicle path planning chromosome population, and performing unmanned aerial vehicle path planning according to unmanned aerial vehicle path planning corresponding to the target chromosome. And constructing an optimal cooperative flight path of the first unmanned aerial vehicle and the second unmanned aerial vehicle in the target area. According to the method, the problem of multi-target optimization path planning of the multi-heterogeneous unmanned aerial vehicle can be effectively solved, and the efficiency and adaptability of path planning of the unmanned aerial vehicle are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) path planning, and particularly to a method and system for heterogeneous UAV path planning. Background Art

[0002] Modern UAVs are equipped with high-resolution cameras and thermal imaging sensors. By combining with satellite positioning systems, they can operate flexibly under adverse weather and complex sea conditions, accurately locate the positions of missing persons or specific vessels, provide real-time data for rescue teams, and thus quickly cover vast sea areas and monitor and capture possible distress signals and floating objects in real time.

[0003] Existing UAV path planning algorithms mainly include the A-star algorithm, particle swarm optimization algorithm, artificial potential field algorithm, and genetic algorithm (Genetic Algorithms, abbreviated as GA), etc. Among them, the genetic algorithm has the advantages of robustness and stability in solving multi-UAV planning. However, due to the respective advantages of different types of UAVs in terms of their coverage areas, task execution speeds, and economic efficiencies, the existing genetic algorithm cannot well solve the collaborative planning problem among heterogeneous UAVs.

[0004] Therefore, there is an urgent need for a method and system for heterogeneous UAV path planning to solve the above problems. Summary of the Invention

[0005] Aiming at the problems existing in the prior art, the present invention provides a method and system for heterogeneous UAV path planning.

[0006] The present invention provides a method for heterogeneous UAV path planning, including: Based on an improved genetic algorithm, perform crossover processing and mutation processing on gene segments of chromosomes in the UAV path planning chromosome population of the current iteration cycle to obtain a new UAV path planning chromosome population; Wherein, the improved genetic algorithm includes a first crossover operator, a second crossover operator, and a mutation operator. The first crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model within the same chromosome. The second crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model between different chromosomes. The mutation operator is used to add or delete gene segments within a chromosome. The UAV path planning chromosome of the current iteration cycle includes the quantity information and model attributes of the first UAV and the second UAV, and the survey area path information of the first UAV and the second UAV in the target area. The maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV; When it is determined that the individual fitness of the chromosomes in the new UAV path planning chromosome population meets the preset fitness, obtain the target chromosome from the new UAV path planning chromosome population, so as to construct the optimal cooperative flight path of the first UAV and the second UAV in the target area according to the UAV path planning corresponding to the target chromosome.

[0007] According to a heterogeneous UAV path planning method provided by the present invention, the method further includes: Divide the area to be surveyed into the target area composed of a plurality of preset survey square grids, wherein the target area is a square area, and the preset survey square grid is constructed based on the maximum survey radius of the second UAV; Number each of the preset survey square grids in the target area in sequence to obtain the survey area number information corresponding to each of the preset survey square grids; According to the preset survey square grids corresponding to the initial survey area points of the first UAV and the second UAV in the target area respectively, obtain the survey area number information corresponding to different initial path plans; Perform chromosome encoding according to the model attribute corresponding to the first UAV and the survey area number information to obtain the first UAV gene segments corresponding to different initial path plans, and splice all the first UAV gene segments in the same initial path plan based on the quantity information of the first UAV to obtain the first gene sequence; Perform chromosome encoding according to the model attribute corresponding to the second UAV and the survey area number information to obtain the second UAV gene segments corresponding to different initial path plans, and splice all the second UAV gene segments in the same initial path plan based on the quantity information of the second UAV to obtain the second gene sequence; wherein each gene in the first UAV gene segment and the second UAV gene segment corresponds to a piece of survey area number information; Construct the initial UAV path planning chromosome corresponding to the same initial path plan according to the first gene sequence and the second gene sequence, and construct the initial UAV path planning chromosome population according to all the initial UAV path planning chromosomes; Based on the improved genetic algorithm, perform the crossover process and the mutation process on the gene segments in the initial UAV path planning chromosome to obtain a new initial UAV path planning chromosome; Construct the UAV path planning chromosome population for the next iteration cycle according to the initial UAV path planning chromosome population and the new initial UAV path planning chromosome.

[0008] According to a heterogeneous UAV path planning method provided by the present invention, constructing a UAV path planning chromosome population for the next iteration cycle based on the initial UAV path planning chromosome population and the new initial UAV path planning chromosome includes: Based on the roulette wheel algorithm, determining new individuals from multiple new initial UAV path planning chromosomes; Adding the new individuals to the initial UAV path planning chromosome population to obtain a new initial UAV path planning chromosome population; Based on a preset population size and the preset fitness, obtaining parental individuals for the next iteration cycle from the new initial UAV path planning chromosome population; Constructing the UAV path planning chromosome population for the next iteration cycle according to the parental individuals for the next iteration cycle.

[0009] According to a heterogeneous UAV path planning method provided by the present invention, the first crossover operator is specifically used for: Randomly obtaining a target self-crossover chromosome in the UAV path planning chromosome population of the current iteration cycle, where the target self-crossover chromosome is a chromosome in the UAV path planning chromosome population of the current iteration cycle to which the first crossover operator is to be applied; Performing a reverse operation, an exchange operation, and a sliding operation on gene segments in at least one first UAV gene segment and / or gene segments in at least one second UAV gene segment within the target self-crossover chromosome to obtain a new UAV path planning chromosome; Wherein, the gene segments correspond to one or more of the survey area number information; The reverse operation is used to reverse the order of genes within the gene segment; The exchange operation is used to exchange the gene positions of any one or more genes within the gene segment with other genes within the gene segment; The exchange operation is also used to exchange the gene positions of any one or more genes within the gene segment between different gene segments under the same UAV model in the same chromosome; The sliding operation is used to move any one or more genes within the gene segment from the current gene position to a gene position between other genes within the gene segment.

[0010] According to a heterogeneous UAV path planning method provided by the present invention, the second crossover operator is specifically used for: Randomly obtain a first target mutually crossed chromosome and a second target mutually crossed chromosome in the drone path planning chromosome population of the current iteration cycle, wherein the first target mutually crossed chromosome and the second target mutually crossed chromosome are chromosomes to be executed by the second crossover operator in the drone path planning chromosome population of the current iteration cycle; Exchanging gene positions of at least one gene segment of the first drone gene segment in the first target mutually crossed chromosome with a corresponding number of gene segments of the first drone gene segment in the second target mutually crossed chromosome, and / or exchanging gene positions of at least one gene segment of the second drone gene segment in the first target mutually crossed chromosome with a corresponding number of gene segments of the second drone gene segment in the second target mutually crossed chromosome, to obtain a mutually crossed first target mutually crossed chromosome and a mutually crossed second target mutually crossed chromosome, wherein the gene segments correspond to one or more of the survey area number information; The survey area number information in the gene segments of the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing is judged, if there is no repeated survey area number information in all gene segments of the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing, then the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing are both determined as the new drone path planning chromosome; If the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing each have repeated survey area number information in the first drone gene segment, the first target mutually crossed chromosome after the mutual crossing is restored to the first target mutually crossed chromosome, and the second target mutually crossed chromosome after the mutual crossing is restored to the second target mutually crossed chromosome; If there is no repeated survey area number information in the first drone gene segment of the first target mutually crossed chromosome after mutual crossing and the second target mutually crossed chromosome after mutual crossing, and there is repeated survey area number information in the second drone gene segment of the first target mutually crossed chromosome after mutual crossing and the second target mutually crossed chromosome after mutual crossing, the repeated genes in the second drone gene segment of the first target mutually crossed chromosome after mutual crossing and the second target mutually crossed chromosome after mutual crossing are removed to obtain the new drone path planning chromosome, wherein the repeated genes are determined by the following formula: ; ; Wherein, represents the duplicate gene in the first target cross chromosome after the mutual cross; represents the duplicate gene in the second target cross chromosome after the mutual cross; represents the th gene in the second UAV gene segment in the first target cross chromosome after the mutual cross, represents the th gene in the second UAV gene segment in the first target cross chromosome after the mutual cross, represents the th gene in the second UAV gene segment in the second target cross chromosome after the mutual cross, represents the th gene in the second UAV gene segment in the second target cross chromosome after the mutual cross; represents the first UAV survey area number information set, and the first UAV survey area number information set is the survey area number information set covered by the genes corresponding to the first UAV gene segment in the first target cross chromosome after the mutual cross; represents the second UAV survey area number information set, and the second UAV survey area number information set is the survey area number information set covered by the genes corresponding to the first UAV gene segment in the second target cross chromosome after the mutual cross.

[0011] According to a heterogeneous UAV path planning method provided by the present invention, the second crossover operator is further used for: Obtain a first survey area number information set and a second survey area number information set, wherein the first survey area number information set represents the survey area number information set covered by the genes corresponding to all gene segments in the first target cross chromosome after the mutual cross; the second survey area number information set represents the survey area number information set covered by the genes corresponding to all gene segments in the second target cross chromosome after the mutual cross; Calculate the difference set between the first survey area number information set and the first UAV survey area number information set to obtain a first difference set; Calculate the difference set between the second survey area number information set and the second UAV survey area number information set to obtain a second difference set; Based on the survey area number information in the first difference set, re-encode to obtain the second UAV gene segment in the first new individual, and obtain the first remaining difference set, where the first new individual is the new UAV path planning chromosome obtained based on the first target crossover chromosome after the mutual crossover; the first remaining difference set is the set of the survey area number information in the first difference set that is not encoded as a gene during the process of re-encoding to obtain the second UAV gene segment in the first new individual; Based on the survey area number information in the second difference set, re-encode to obtain the second UAV gene segment in the second new individual, and obtain the second remaining difference set, where the second new individual is the new UAV path planning chromosome obtained based on the second target crossover chromosome after the mutual crossover; the second remaining difference set is the set of the survey area number information in the second difference set that is not encoded as a gene during the process of re-encoding to obtain the second UAV gene segment in the second new individual; Obtain the first shortest gene segment and the second shortest gene segment, where the first shortest gene segment is the shortest second UAV gene segment in the first new individual, and the second shortest gene segment is the shortest second UAV gene segment in the second new individual; Encode the survey area number information in the first remaining difference set into corresponding genes, and add them to the first shortest gene segment to perform gene expansion on the first new individual to obtain the first new individual after gene expansion; Encode the survey area number information in the second remaining difference set into corresponding genes, and add them to the second shortest gene segment to perform gene expansion on the second new individual to obtain the second new individual after gene expansion.

[0012] According to a heterogeneous UAV path planning method provided by the present invention, the mutation operator is specifically used for: Randomly obtain the target mutation chromosome in the UAV path planning chromosome population of the current iteration cycle, where the target mutation chromosome is the chromosome to be executed with the mutation operator in the UAV path planning chromosome population of the current iteration cycle; When determining that the first UAV gene segment in the target mutant chromosome is the gene segment to be mutated, add the first mutant gene to the first target mutant gene segment in the target mutant chromosome, and remove the genes identical to the first mutant gene in other gene segments of the target mutant chromosome except the first target mutant gene segment; and regenerate the second UAV gene segment in the target mutant chromosome based on the survey area numbering information in the target area other than the survey area numbering information covered by the first UAV gene segment in the target mutant chromosome, to obtain the mutated target mutant chromosome, where the first mutant gene is encoded based on the randomly obtained survey area numbering information; the first target mutant gene segment is the shortest first UAV gene segment in the target mutant chromosome; When determining that the second UAV gene segment in the target mutant chromosome is the gene segment to be mutated, add the second mutant gene to the second target mutant gene segment in the target mutant chromosome, and remove the genes identical to the second mutant gene in the third target mutant gene segment in the target mutant chromosome, to obtain the mutated target mutant chromosome; where the second target mutant gene segment is the shortest second UAV gene segment in the target mutant chromosome; the second mutant gene is randomly determined from other genes except the genes in the second target mutant gene segment and the first UAV gene segment in the target mutant chromosome; the third target mutant gene segment is the second UAV gene segment in the target mutant chromosome that has genes identical to the second mutant gene.

[0013] According to a heterogeneous UAV path planning method provided by the present invention, the mutation operator is further used for: When determining that the first UAV gene segment in the target mutant chromosome is the gene segment to be mutated, remove the first removal gene from the first target removal gene segment in the target mutant chromosome, and regenerate the second UAV gene segment in the target mutant chromosome based on the survey area numbering information in the target area other than the survey area numbering information covered by the first UAV gene segment in the target mutant chromosome, to obtain the mutated target mutant chromosome, where the first removal gene is encoded based on the randomly obtained survey area numbering information; the first target removal gene segment is the longest first UAV gene segment in the target mutant chromosome; When determining that the second UAV gene segment in the target mutant chromosome is the gene segment to be mutated, remove the second removal gene from the second target removal gene segment in the target mutant chromosome, and add the same gene as the second removal gene to the third target removal gene segment in the target mutant chromosome to obtain the mutated target mutant chromosome; wherein, the second target removal gene segment is the longest second UAV gene segment in the target mutant chromosome; the second removal gene is randomly determined based on the genes in the second target removal gene segment in the target mutant chromosome; the third target removal gene segment is the shortest second UAV gene segment in the target mutant chromosome.

[0014] According to a heterogeneous UAV path planning method provided by the present invention, the calculation formula of the individual fitness is: ; Wherein, represents the individual fitness, represents the total distance of all UAV flight paths, represents the number of redundant survey area number information in the target area, represents the side length of the target area, represents the flight time of the UAV corresponding to the maximum flight path among all UAVs, represents the distance weight, represents the redundant coverage penalty weight, represents the time weight.

[0015] The present invention also provides a heterogeneous UAV path planning system, including: A processing module, configured to perform crossover processing and mutation processing on the gene segments of the chromosomes in the UAV path planning chromosome population in the current iteration cycle based on an improved genetic algorithm to obtain a new UAV path planning chromosome population; Wherein, the improved genetic algorithm includes a first crossover operator, a second crossover operator and a mutation operator. The first crossover operator is used to perform crossover processing on the gene segments corresponding to the same UAV model in the same chromosome, and the second crossover operator is used to perform crossover processing on the gene segments corresponding to the same UAV model between different chromosomes. The mutation operator is used to add or remove gene segments in the chromosome; the UAV path planning chromosome in the current iteration cycle includes the quantity information and model attributes of the first UAV and the second UAV, and the survey area path information of the first UAV and the second UAV in the target area. The maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV; A path planning module, configured to obtain a target chromosome from the new UAV path planning chromosome population when it is determined that the individual fitness of the chromosomes in the new UAV path planning chromosome population meets a preset fitness, so as to construct an optimal cooperative flight path of the first UAV and the second UAV in the target area according to the UAV path planning corresponding to the target chromosome.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the heterogeneous UAV path planning method as described in any one of the above is implemented.

[0017] The heterogeneous UAV path planning method and system provided by the present invention perform crossover processing and mutation processing on the gene segments of the chromosomes in the UAV path planning chromosome population of the current iteration cycle through a genetic algorithm in a specific coding form, improving the iteration update efficiency of the UAV path planning chromosome population, effectively solving the multi-objective optimization path planning problem of multiple heterogeneous UAVs, and enhancing the efficiency and adaptability of UAV path planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a schematic flowchart of the heterogeneous UAV path planning method provided by the present invention; Figure 2 It is a schematic diagram of the relationship between the target area and the UAV survey radius provided by the present invention; Figure 3 It is a schematic diagram of the coding method of the improved genetic algorithm provided by the present invention; Figure 4 It is a schematic diagram of the self-crossover process of the same type of UAV single gene segment of the improved genetic algorithm provided by the present invention; Figure 5 It is a schematic diagram of the self-crossover process of multiple gene segments of the same type of UAV of the improved genetic algorithm provided by the present invention; Figure 6 It is a schematic diagram of the mutual crossover process of the improved genetic algorithm provided by the present invention; Figure 7 It is a schematic diagram of the mutation process of the improved genetic algorithm provided by the present invention; Figure 8The figure showing the collaborative planning simulation experiment results of UAV A provided by the present invention; Figure 9 The individual case figure of the collaborative planning simulation experiment results of UAV B provided by the present invention; Figure 10 The schematic diagram of the simulation results of the optimal adaptive value iteration graph of the improved genetic algorithm provided by the present invention; Figure 11 The structural schematic diagram of the heterogeneous UAV path planning system provided by the present invention; Figure 12 The structural schematic diagram of the electronic device provided by the present invention. Detailed implementation manners

[0020] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the scope of protection of the present invention.

[0021] With its excellent high maneuverability and rapid deployment characteristics, UAVs can quickly respond in emergency search and rescue operations, thus significantly improving the search and rescue efficiency and success rate at critical moments. In the operation of UAVs, path planning is one of the core links, and common algorithms include A* algorithm, particle swarm optimization algorithm, artificial potential field algorithm, genetic algorithm and other methods. Among them, the genetic algorithm shows excellent robustness and stability when dealing with multi-UAV planning tasks. However, in the face of the different advantages of different types of UAVs in terms of coverage area, task execution speed, and economic efficiency in practical applications, the traditional genetic algorithm encounters challenges in coordinating the collaborative planning of heterogeneous UAVs and is difficult to fully utilize the unique effectiveness of various UAVs. The main reasons include: First, the existing genetic algorithm is usually used to solve the collaborative planning among UAVs of the same type, but there are significant differences in coverage area, task execution speed, and economic benefits among different types of UAVs. Since the existing genetic algorithm fails to effectively consider these differences, when multi-UAV collaborative operations are carried out, the advantages of each type of UAV cannot be fully utilized, affecting the efficiency and effect of the overall task. In addition, when the existing genetic algorithm deals with the collaborative path planning of heterogeneous UAVs, it often lacks pertinence and flexibility and is difficult to adapt to real-time changes in complex environments. This limitation makes it impossible to achieve efficient collaboration and resource optimization allocation among different types of UAVs in practical applications.

[0022] Second, in the process of continuing the population evolution of the existing genetic algorithm, the currently mature crossover and mutation operators are adopted. Since there is little change in gene diversity, the effect is relatively single, resulting in insufficient solution space and thus prone to falling into the "premature" state. To address this problem, the present invention provides custom genetic operators, including a first crossover operator, a second crossover operator, and sudden increase gene points and missing gene points introduced during the mutation process, thereby effectively solving the path optimization problem of heterogeneous unmanned aerial vehicles (UAVs) in collaborative planning. The application of these custom operators enables the algorithm to more flexibly adapt to the characteristics of different types of UAVs, thereby optimizing specific challenges in the solution space, expanding the solutions of collaborative planning, enabling the algorithm to better select the optimal solution from multiple solutions, thus improving the efficiency and adaptability of path planning and meeting the actual needs in complex environments.

[0023] Third, the collaborative path planning between heterogeneous UAVs usually involves multiple optimization objectives, such as time, distance, and resources. In the process of multi-objective optimization of multiple UAVs by the existing genetic algorithm, when weighing multiple optimization objectives such as time, distance, and resources, the algorithm will face the problem that it is difficult to simultaneously meet multiple optimization objectives, resulting in the inability to achieve the ideal comprehensive optimal effect. In addition, the existing genetic methods lack an effective trade-off mechanism when dealing with the conflicts between different optimization objectives, resulting in the inability to accurately reflect the priorities among the objectives, thereby affecting the quality and efficiency of the overall path planning.

[0024] To address the above problems of the existing technologies, the present invention provides a path planning method for heterogeneous UAVs based on an improved genetic algorithm. According to the flight capabilities and mission requirements of heterogeneous UAVs, by designing a genetic algorithm with specific individual encoding, the efficiency can be effectively increased, the randomness of the algorithm is increased, and multi-objective factors such as time, distance, and resource utilization rate are integrated, and finally the optimal path planning is achieved.

[0025] Figure 1 For the flow schematic diagram of the path planning method for heterogeneous UAVs provided by the present invention, as Figure 1 shown, the present invention provides a path planning method for heterogeneous UAVs, including: Step 101, based on the improved genetic algorithm, perform crossover processing and mutation processing on the gene segments of the chromosomes in the chromosome population of the UAV path planning in the current iteration cycle to obtain a new chromosome population of the UAV path planning; Among them, the improved genetic algorithm includes a first crossover operator, a second crossover operator, and a mutation operator. The first crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model within the same chromosome. The second crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model between different chromosomes. The mutation operator is used to add or remove gene segments within a chromosome. The UAV path planning chromosome in the current iteration period includes the quantity information and model attributes of the first UAV and the second UAV, as well as the survey area path information of the first UAV and the second UAV within the target area. The maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV.

[0026] In the UAV path planning problem, the goal is to determine how a group of UAVs fly within a specific target area to complete the survey task, which involves the selection of UAV models (different models may have different performances, such as the maximum survey radius), quantity allocation, and specific flight path planning.

[0027] In the present invention, for the path planning of UAVs of different models, the existing genetic algorithm is improved, which can realize the path planning of heterogeneous UAVs and improve the performance and efficiency of the genetic algorithm. In the genetic algorithm, each possible solution is encoded as a chromosome.

[0028] In the present invention, a chromosome represents a specific UAV path planning scheme. A chromosome is composed of multiple gene segments, and each gene segment corresponds to the information of a specific UAV, including quantity, model, and survey path. The chromosome contains the quantity information and model attributes of the first UAV and the second UAV. Among them, the model attribute determines the maximum survey radius of the UAV, which is an important factor affecting path planning. Specifically, it is reflected in the maximum survey radius between the first UAV and the second UAV. That is, when planning the path, the ability limitations of different UAVs need to be considered to ensure that the planned path is within the capabilities of the UAVs.

[0029] In the present invention, the specific operation of the improved genetic algorithm is implemented based on the first crossover operator, the second crossover operator, and the mutation operator. Specifically, the first crossover operator acts within the same chromosome and can perform crossover processing on gene segments corresponding to the same UAV model, indicating that in the same planning scheme, UAVs of the same model can exchange partial path information, thereby exploring new path combinations. The second crossover operator acts between different chromosomes and performs crossover processing on gene segments corresponding to the same UAV model, allowing UAVs of the same model in different planning schemes to share path information, increasing the diversity of the population, and contributing to finding better solutions. The mutation operator introduces randomness by adding or removing gene segments within the chromosome, which helps the algorithm jump out of the local optimal solution and explore a wider solution space.

[0030] Step 102, when it is determined that the individual fitness of the chromosomes in the new UAV path planning chromosome population meets the preset fitness, obtain the target chromosome from the new UAV path planning chromosome population, so as to construct the optimal cooperative flight path of the first UAV and the second UAV in the target area according to the UAV path planning corresponding to the target chromosome.

[0031] In the present invention, each chromosome (i.e., each possible UAV path planning scheme) has a corresponding fitness, which reflects the quality or superiority of the chromosome. In the UAV path planning problem, the fitness may be calculated based on multiple factors, such as the total length of the flight path, the cooperative efficiency between UAVs, the completion degree of the survey task, etc. The preset fitness is a threshold set in advance, used to determine whether the optimal chromosome in the current iteration cycle is good enough to stop the iteration process.

[0032] Furthermore, after each iteration ends, the genetic algorithm calculates the individual fitness of each chromosome in the new UAV path planning chromosome population and compares it with the preset fitness. If there is at least one chromosome in the population whose fitness value reaches or exceeds the preset fitness, then the genetic algorithm considers that a good enough solution has been found and the iteration can be stopped.

[0033] In the present invention, when it is determined that the individual fitness meets the preset fitness, the genetic algorithm selects the chromosome with the highest fitness from the new UAV path planning chromosome population as the target chromosome. This chromosome represents the optimal UAV path planning scheme found in the current iteration cycle. Once the target chromosome is determined, the genetic algorithm constructs the optimal cooperative flight path of the UAVs in the target area according to the information encoded in the chromosome, including determining the take-off point, flight path, survey area of each UAV, and the cooperative method with other UAVs, etc.

[0034] In the present invention, the optimal cooperative flight path refers to a flight path for a group of heterogeneous unmanned aerial vehicles (UAVs) to complete a survey task in the target area in the most efficient and cooperative manner. It is required that not only collisions between UAVs be avoided, but also information be shared as much as possible and they work cooperatively to improve the overall survey efficiency and accuracy.

[0035] The heterogeneous UAV path planning method provided by the present invention performs crossover processing and mutation processing on the gene segments of the chromosomes in the UAV path planning chromosome population of the current iteration cycle through a genetic algorithm in a specific coding form, improving the iteration update efficiency of the UAV path planning chromosome population, effectively solving the multi-objective optimization path planning problem of multiple heterogeneous UAVs, and enhancing the efficiency and adaptability of UAV path planning.

[0036] Based on the above embodiments, the method further includes: Dividing the area to be surveyed into the target area composed of a plurality of preset survey square grids, wherein the target area is a square area, and the preset survey square grids are constructed based on the maximum survey radius of the second UAV; Sequentially numbering each of the preset survey square grids in the target area to obtain the survey area numbering information corresponding to each of the preset survey square grids; According to the preset survey square grids corresponding to the initial survey area points of the first UAV and the second UAV in the target area respectively, obtaining the survey area numbering information corresponding to different initial path plans; Performing chromosome coding according to the model attribute of the first UAV and the survey area numbering information to obtain the gene segments of the first UAV corresponding to different initial path plans, and based on the quantity information of the first UAV, splicing all the gene segments of the first UAV in the same initial path plan to obtain a first gene sequence; Performing chromosome coding according to the model attribute of the second UAV and the survey area numbering information to obtain the gene segments of the second UAV corresponding to different initial path plans, and based on the quantity information of the second UAV, splicing all the gene segments of the second UAV in the same initial path plan to obtain a second gene sequence; wherein each gene in the gene segments of the first UAV and the gene segments of the second UAV corresponds to a survey area numbering information; Constructing an initial UAV path planning chromosome corresponding to the same initial path plan according to the first gene sequence and the second gene sequence, and constructing an initial UAV path planning chromosome population according to all the initial UAV path planning chromosomes; Based on the improved genetic algorithm, perform the crossover process and the mutation process on the gene segments in the initial UAV path planning chromosome to obtain a new initial UAV path planning chromosome; Construct a UAV path planning chromosome population for the next iteration cycle according to the initial UAV path planning chromosome population and the new initial UAV path planning chromosome.

[0037] In the present invention, in the initial stage of the improved genetic algorithm, after determining the model and quantity of UAVs for a certain survey area, first, create corresponding resource sets for the UAVs according to the model attributes and quantity, where the model attributes are composed of the UAV flight speed and the UAV survey radius.

[0038] Specifically, the present invention defines 2 UAV models, namely the first UAV and the second UAV. Among them, the UAV model with a larger survey radius is UAV A (i.e., the first UAV), and the UAV model with a smaller survey ability is UAV B (i.e., the second UAV). The quantity of UAV A is n , and the model attribute Q A is: , and its corresponding resource set ; the quantity of UAV B is m , and the model attribute Q B is , and its corresponding resource set .

[0039] Figure 2 For the schematic diagram of the relationship between the target area and the UAV survey radius provided by the present invention, reference can be made to Figure 2 As shown, in the present invention, the actual scenario is set in a specific accident - prone sea area for survey among multiple heterogeneous UAVs. Taking the distance traveled by a ship in 1 hour as the side length, the obtained target area is a square area. At the same time, the square area is divided into grids, and the side length of the smallest unit square (i.e., the preset survey square grid) needs to satisfy the maximum survey radius of the UAV B with a smaller survey range, that is, the UAV with a smaller survey radius surveys one preset survey square grid at a time.

[0040] Reference can be made to Figure 2 As shown, after dividing the area to be surveyed into multiple preset survey square grids, the size of each preset survey square grid is the same, and the side length of the preset survey square grid is equal to the maximum survey radius of the UAV B with a weaker survey ability, so as to satisfy the UAV B to complete the path area and achieve the full - area coverage effect. For the UAV A with a stronger survey ability, it is defined that UAV A moves from the center point of the current preset survey square in the target area to the center point of a certain preset survey square grid in the target area The distance is , when is satisfied ( is the maximum survey radius of UAV A), the center point is located outside the coverage radius of UAV A. At this time, the preset survey square grid corresponding to the center point is not fully observed by UAV A.

[0041] If the center point of a preset survey square grid has a distance that satisfies , then the preset survey square grid is considered to have been surveyed by UAV A. It should be noted that in the present invention, only by determining , it can be determined that the preset survey square grids covered by UAV A at this time have been fully surveyed by UAV A. For some partial areas in the covered area that are not fully covered by the survey radius of UAV A in certain preset survey square grids, since their areas are smaller than the size of the missing ship, they can be ignored and regarded as the preset survey square grids having been fully covered.

[0042] Further, a population is created with specific coding according to the model of the UAV and the task requirements, and initialization processing is performed on each individual in the population. Figure 3 is a schematic diagram of the coding method of the improved genetic algorithm provided by the present invention, which can be referred to Figure 3 as shown. The chromosome consists of , , and the survey area number information . Among them, is the number of segments of the preset survey square grids divided in the target area. The preset survey square grids that have been divided are encoded from 1 (it can be based on the bottommost part of the target area and encoded in sequence from left to right), which is used to represent the area positions surveyed by the UAV; while , represent the models between heterogeneous UAVs. Among them, represents UAV A, represents UAV B. A single UAV path consists of the UAV model and the area points it surveys. For example, it can be referred to Figure 3 as shown. The sequence of UAV A is , forming a corresponding first UAV gene segment, and then multiple first UAV gene segments are combined to construct a first gene sequence; the sequence of UAV B , a corresponding second UAV gene segment is formed, and then multiple second UAV gene segments are combined to construct a second gene sequence. Then, the first gene sequence and the second gene sequence are spliced together, so that all UAV genes are jointly stored on the same gene sequence to form a chromosome, and a chromosome forms an individual, that is, a path planning for multi-heterogeneous UAV cooperation at one time.

[0043] In the present invention, based on the improved genetic algorithm, the gene segments in the initial UAV path planning chromosome are subjected to the above-mentioned crossover processing and mutation processing to obtain a batch of new UAV path planning chromosomes. These new chromosomes, together with the chromosomes in the initial population, constitute the chromosome population of the next iteration cycle, so as to serve as the input for the next round of genetic algorithm operations, and continue to perform operations such as selection, crossover, and mutation until a predetermined number of iterations is reached or a chromosome that meets the preset fitness is obtained.

[0044] In the chromosome structure of the improved genetic algorithm constructed by the present invention, by embedding UAV identifiers of different models in the gene strip, such as model "a" and "b", which respectively represent UAVs with large coverage radii and small coverage radii (such as UAV A with strong surveying capabilities and UAV B with weak capabilities). Each path point not only identifies the flight position of the UAV, but also contains the coverage radius information of the UAV of this model, thus implicitly indicating the survey range of the UAV for this path point. For example, when the gene segment of UAV A contains a certain specific path point, it means that the point and the surrounding area within the coverage radius of A have been surveyed, while UAV B only surveys this path point due to its smaller reconnaissance radius.

[0045] Furthermore, by adding these model and coverage radius information to the chromosome gene strip in the present invention, the algorithm can automatically judge the survey coverage range of each UAV, and preferentially allocate UAVs with large coverage ranges to survey large areas when planning the path, reducing the redundant workload of other UAVs in the same area. This design enables heterogeneous UAVs to give full play to their different survey advantages during collaborative work, ensuring the full coverage of the entire mission area, while reducing the repeated coverage rate, improving the efficiency of the mission and the resource utilization rate. Compared with the existing path planning of homogeneous UAVs, this gene strip structure design not only significantly enhances the collaborative efficiency, but also optimizes the use of mission resources and has stronger adaptability.

[0046] On the basis of the above embodiment, constructing the UAV path planning chromosome population of the next iteration cycle according to the initial UAV path planning chromosome population and the new initial UAV path planning chromosome includes: Based on the roulette wheel algorithm, new individuals are determined from multiple new initial UAV path planning chromosomes; Add the new individual to the initial UAV path planning chromosome population to obtain a new initial UAV path planning chromosome population; Based on the preset population size and the preset fitness, obtain the parental individuals for the next iteration cycle from the new initial UAV path planning chromosome population; Construct the UAV path planning chromosome population for the next iteration cycle according to the parental individuals for the next iteration cycle.

[0047] In the present invention, through the roulette wheel algorithm, new individuals are selected from the new initial UAV path planning chromosome or the UAV path planning chromosome population obtained in a certain iteration cycle, and the selected new individuals and the parental individuals (i.e., the UAV path planning chromosome in the previous iteration cycle) form the total population. Then, individuals with the top individual fitness and a size of pop_size (i.e., the preset population size) are selected from the obtained total population to form the parental population in the next iteration cycle.

[0048] In the present invention, it is judged whether the improved genetic algorithm satisfies the iteration termination condition. When the number of iterations reaches the maximum number of iterations t max or the individual fitness of a certain individual reaches a predetermined threshold f threshold , the optimal individual in the population is determined, and the output of the optimal individual is used as the optimal path, and the optimal information of each UAV is visually displayed; If the above iteration termination condition is not satisfied, the newly generated total population is used as the UAV path planning chromosome population in the next iteration cycle to generate a new population again until the iteration termination condition is satisfied.

[0049] On the basis of the above embodiments, the calculation formula of the individual fitness is: ; wherein, represents the individual fitness, represents the total distance of the travel paths of all UAVs, represents the number of redundant survey area number information in the target area, represents the side length of the target area, represents the flight time of the UAV corresponding to the maximum flight path among all UAVs, represents the distance weight, represents the redundant coverage penalty weight, represents the time weight.

[0050] In the present invention, a preset fitness function is used to calculate the adaptive value of each individual in the population, that is, the individual fitness. Considering the task requirements, the flight task in the present invention is regional full-coverage survey, and the total task time is set as the time consumed by the drone with the longest flight duration among all drones. Among them, the fuel cost of the task is determined by the total distance traveled by the drone. Set as the sum of the distances of all drone flight paths, and take drone A as the basis of resource utilization rate, and use whether to repeatedly observe regional points as the evaluation criterion to calculate the redundant coverage penalty. The redundant coverage penalty satisfies: ; wherein, represents the number of redundant survey area number information in the target area, that is, the number of survey area number information covered multiple times. is the size of the entire grid, that is, the side length of the target area. Therefore, represents the area of the entire target area.

[0051] Finally, construct the calculation formula of individual fitness: .

[0052] By introducing a redundant coverage penalty mechanism, the present invention is used to reduce the repeated coverage of the same area by drones of different models. By calculating the number of points of redundant coverage and using the total grid area to generate a penalty term, the fitness function can dynamically balance the influence of repeated coverage. Through this penalty mechanism, the algorithm preferentially selects non-repeated areas for survey during path planning, effectively reducing resource waste and ensuring a more accurate and economical path selection. Compared with the existing genetic algorithm, the present invention significantly improves the task efficiency in reducing redundant coverage, makes the collaborative operation of drones more efficient and energy-saving, and greatly enhances its applicability in complex task scenarios.

[0053] Based on the above embodiment, the first crossover operator is specifically used for: Randomly obtain the target self-crossover chromosome in the drone path planning chromosome population of the current iteration cycle, where the target self-crossover chromosome is the chromosome to be executed by the first crossover operator in the drone path planning chromosome population of the current iteration cycle; Perform reverse order operation, exchange operation and sliding operation on the gene segments in at least one of the first drone gene segments and / or the gene segments in at least one of the second drone gene segments in the target self-crossover chromosome to obtain a new drone path planning chromosome; wherein, the gene segment corresponds to one or more of the survey area number information; The reverse operation is used to reverse the order of the genes within the gene segment; The exchange operation is used to exchange the positions of any one or more genes within the gene segment with other genes within the gene segment; The exchange operation is also used to exchange the positions of any one or more genes within the gene segment between different gene segments of the same UAV model on the same chromosome; The sliding operation is used to move any one or more genes within the gene segment from their current gene positions to gene positions between other genes within the gene segment.

[0054] In the present invention, in each iteration process of the genetic algorithm, the UAV path planning chromosome population of the current iteration cycle (if it is the initial stage of iteration, it is the initial UAV path planning chromosome population) is processed by various self-set crossover operators such as self-crossover and mutual crossover, and then the individuals obtained by crossover are subjected to mutation operations to increase the diversity change of the population so as to increase the solution space.

[0055] Specifically, in the present invention, the self-crossover method is executed based on the first crossover operator, and is divided into self-crossover of single gene segments of the same type of UAV under one chromosome, and self-crossover of multiple gene segments of the same type of UAV. Figure 4 For the schematic diagram of the self-crossover process of single gene segments of the same type of UAV in the improved genetic algorithm provided by the present invention, as Figure 4 shown, the main crossover methods involved are three methods: reverse operation, exchange operation, and sliding operation. Specifically, the reverse operation is expressed as: ; Among them, represents reversing the part of the genes within the gene segment of the target self-crossover chromosome from position to . For reference, as Figure 4 shown, the in the gene segment of a certain UAV B has its order reversed and becomes , which is stored in the chromosome as the new gene obtained by crossover.

[0056] Furthermore, assuming the chromosome is , the exchange operation is performed at positions to , and can be expressed as: ; The exchange operation is used to exchange the genes at positions to within the gene segment of the target self-crossover chromosome. For reference, as Figure 4As shown, the genes within the gene segment of a certain drone B and undergo gene exchange, and the new genome is stored in the chromosome.

[0057] Furthermore, the sliding operation can be expressed as: ; The sliding operation is used to move the gene to the position , and move the genes between the original position and the new position forward or backward. As Figure 4 shown, in the sliding operation, the genes within the gene segment of the drone B sequence slide and move between the gene and , while both slide forward by 1 unit.

[0058] Figure 5 is a schematic diagram of the self-crossing process of multiple gene segments of the same model of drones provided by the improved genetic algorithm of the present invention. As Figure 5 shown, the self-crossing process of multiple gene segments of the same model of drones satisfies: ; ; Among them, and are respectively different gene segments of drones of the same model in the chromosome , that is, the path point exchange between drones of the same model under a single collaborative planning task.

[0059] The specific self-crossing process of multiple gene segments of the same model of drones is as follows: randomly select the fragment from index to to in the gene segment, and the fragment from index to to in the gene segment are exchanged to obtain a new individual. During the crossing process, the length of the selected gene fragment must satisfy being greater than 0, that is, the length from index to to satisfies the distance to avoid the exchange of empty gene segments and ensure that its length is sufficient to complete the crossing operation. As Figure 5 shown, under one gene segment of a drone B and An exchange has occurred. To avoid an exception where the path length is zero, it is necessary to judge whether the gene fragment length is greater than 0 before crossover to ensure the effectiveness of the crossover operation and the integrity of the path.

[0060] Based on the above embodiments, the second crossover operator is specifically used for: Randomly obtain the first target crossover chromosome and the second target crossover chromosome in the UAV path planning chromosome population of the current iteration cycle, where the first target crossover chromosome and the second target crossover chromosome are the chromosomes in the UAV path planning chromosome population of the current iteration cycle to which the second crossover operator is to be applied; Exchange the gene segments of at least one of the first UAV gene segments in the first target crossover chromosome with the corresponding number of gene segments in the first UAV gene segment in the second target crossover chromosome, and / or exchange the gene segments of at least one of the second UAV gene segments in the first target crossover chromosome with the corresponding number of gene segments in the second UAV gene segment in the second target crossover chromosome to obtain the mutually crossed first target crossover chromosome and the mutually crossed second target crossover chromosome, where the gene segment corresponds to one or more of the survey area numbering information; Judge the survey area numbering information in the gene segments of the mutually crossed first target crossover chromosome and the mutually crossed second target crossover chromosome. If there is no duplicate survey area numbering information in all the gene segments of the mutually crossed first target crossover chromosome and the mutually crossed second target crossover chromosome, then determine both the mutually crossed first target crossover chromosome and the mutually crossed second target crossover chromosome as the new UAV path planning chromosomes; If there is duplicate survey area numbering information in the first UAV gene segments of the mutually crossed first target crossover chromosome and the mutually crossed second target crossover chromosome respectively, then restore the mutually crossed first target crossover chromosome to the first target crossover chromosome and restore the mutually crossed second target crossover chromosome to the second target crossover chromosome; If there is no duplicate survey area number information in the first UAV gene segment of the first target crossover chromosome and the second target crossover chromosome after the crossover, and there is duplicate survey area number information in the second UAV gene segment of the first target crossover chromosome and the second target crossover chromosome after the crossover, the duplicate genes in the second UAV gene segment of the first target crossover chromosome and the second target crossover chromosome after the crossover are removed to obtain the new UAV path planning chromosome, where the duplicate genes are determined by the following formula: ; ; Wherein, represents the duplicate gene in the first target crossover chromosome after the crossover, represents the duplicate gene in the second target crossover chromosome after the crossover; represents the th gene in the second UAV gene segment of the first target crossover chromosome after the crossover, represents the th gene in the second UAV gene segment of the first target crossover chromosome after the crossover, represents the th gene in the second UAV gene segment of the second target crossover chromosome after the crossover, represents the th gene in the second UAV gene segment of the second target crossover chromosome after the crossover; represents the first UAV survey area number information set, and the first UAV survey area number information set is the survey area number information set covered by the genes corresponding to the first UAV gene segment in the first target crossover chromosome after the crossover; represents the second UAV survey area number information set, and the second UAV survey area number information set is the survey area number information set covered by the genes corresponding to the first UAV gene segment in the second target crossover chromosome after the crossover.

[0061] Figure 6 is a schematic diagram of the crossover process of the improved genetic algorithm provided by the present invention, which can be referred to Figure 6As shown, in the present invention, the second crossover operator is used to perform the crossover process between two chromosomes, and the crossover process is only for the crossover between gene segments of the same type. To ensure the full coverage of the path and the integrity of the crossover, the present invention sets two chromosomes as and (i.e., the first target crossover chromosome and the second target crossover chromosome). After the crossover is completed, the new chromosomes are and . The crossover process is described as follows. Among them, the second crossover operator mainly performs the exchange operation between genes. The exchange operation process can refer to the exchange operation between genes in the embodiment.

[0062] Specifically, after the gene exchange occurs between the segments of the same type of unmanned aerial vehicle between chromosome and chromosome , if duplicate points (i.e., duplicate survey area number information) appear in the gene segment of unmanned aerial vehicle A of the new chromosomes and , considering that the unmanned aerial vehicle needs to calculate the coverage radius, removing the duplicate points will affect the path points of the unmanned aerial vehicle B segment. Therefore, this crossover process is abandoned, and the new chromosomes and remain unchanged, and they satisfy: ; ; If there are no duplicate points, the gene segment of unmanned aerial vehicle A of the new chromosome is the result after the crossover.

[0063] Furthermore, if duplicate points appear in the gene segment of unmanned aerial vehicle B of the new chromosomes and , the duplicate points need to be removed. Since the newly obtained path points may appear within the coverage range of unmanned aerial vehicle A, the path points (i.e., duplicate genes) need to be removed. Among them, the path points p to be removed satisfy: ； ; The gene segments of unmanned aerial vehicle A of the newly obtained chromosomes and remain unchanged: ; ; The gene segments of unmanned aerial vehicle B of the newly obtained chromosomes and need to satisfy: ; 。

[0064] Based on the above embodiments, the second crossover operator is further configured to: Obtain a first survey area number information set and a second survey area number information set, where the first survey area number information set represents the set of survey area number information covered by the genes corresponding to all gene segments in the first target mutually crossed chromosome after mutual crossing; the second survey area number information set represents the set of survey area number information covered by the genes corresponding to all gene segments in the second target mutually crossed chromosome after mutual crossing; Calculate the difference set between the first survey area number information set and the first UAV survey area number information set to obtain a first difference set; Calculate the difference set between the second survey area number information set and the second UAV survey area number information set to obtain a second difference set; Based on the survey area number information in the first difference set, re-encode to obtain the second UAV gene segment in the first new individual, and obtain a first remaining difference set, where the first new individual is the new UAV path planning chromosome obtained based on the first target mutually crossed chromosome after mutual crossing; the first remaining difference set is the set of survey area number information in the first difference set that is not encoded as a gene during the process of re-encoding to obtain the second UAV gene segment in the first new individual; Based on the survey area number information in the second difference set, re-encode to obtain the second UAV gene segment in the second new individual, and obtain a second remaining difference set, where the second new individual is the new UAV path planning chromosome obtained based on the second target mutually crossed chromosome after mutual crossing; the second remaining difference set is the set of survey area number information in the second difference set that is not encoded as a gene during the process of re-encoding to obtain the second UAV gene segment in the second new individual; Obtain a first shortest gene segment and a second shortest gene segment, where the first shortest gene segment is the shortest second UAV gene segment in the first new individual, and the second shortest gene segment is the shortest second UAV gene segment in the second new individual; Encode the survey area number information in the first remaining difference set into corresponding genes and add them to the first shortest gene segment to expand the genes of the first new individual to obtain a first new individual with expanded genes; Encode the survey area number information in the second remaining difference set into corresponding genes and add them to the second shortest gene segment to expand the genes of the second new individual to obtain a second new individual with expanded genes.

[0065] In the present invention, for the UAV A in the newly generated chromosome, the changed genome is obtained through the mutual crossing process in the above embodiments. Since the survey radius of UAV A is relatively large and there are multiple path points covered, it is necessary to regenerate the path points in the gene segment of UAV B. First, subtract the coverage points of UAV A in the corresponding chromosome, and the remaining point sets (i.e., the first difference set) and (i.e., the second difference set) are: ; ; Among them, represents the set of survey area number information covered by the genes corresponding to all gene segments in the first target mutually crossed chromosome after mutual crossing, represents the set of survey area number information covered by the genes corresponding to all gene segments in the second target mutually crossed chromosome after mutual crossing, the set of survey area number information covered by the gene corresponding to the first UAV gene segment in the first target mutually crossed chromosome after mutual crossing, represents the set of survey area number information covered by the gene corresponding to the first UAV gene segment in the second target mutually crossed chromosome after mutual crossing.

[0066] The UAV B gene segment parts in the newly generated two chromosomes and are respectively: ; ; Among them, represents the path points in the UAV B gene fragment in the regenerated chromosome.

[0067] Furthermore, after removing the duplicate points and the points that appear within the coverage range of UAV A, there will be a situation of non-full coverage in the entire chromosome, so it is necessary to further complete the full coverage of the path points. First, obtain the remaining path point sets (i.e., the first remaining difference set) and (i.e., the second remaining difference set) in the total point set after removing the points covered by UAV A and the points already encoded as UAV B path points. Then, in order to satisfy the balance of the UAV B path segment, find the shortest UAV B path, which satisfies: ; ; Among them, let the UAV B under chromosome have gene points, each gene segment is represented by indicating that the chromosome UAV B has gene points, each gene segment is represented by indicating that the gene segment with the minimum length has an index of .

[0068] Furthermore, the path points in the remaining difference sets (i.e., the first remaining difference set and the second remaining difference set) are successively where is the encoding of the survey area number information in the first remaining difference set as the corresponding gene, is the encoding of the survey area number information in the second remaining difference set as the corresponding gene) are respectively added to the corresponding shortest gene segments: ; ; The gene segments of UAV B in the newly generated chromosome and satisfy: ; ; Finally, the newly generated chromosome is obtained by adding the gene segment 2 parts of UAV A and UAV B, that is: ; .

[0069] For reference, Figure 6 as shown, the in the gene segment of UAV B in the chromosome and the in the gene segment of UAV B in the chromosome have completed the interaction, that is, the cross process.

[0070] The present invention realizes operations such as reverse order, exchange, and sliding within the gene segments of the same model UAVs on the same chromosome. These operations locally adjust the gene sequence within the UAV path, continuously changing the arrangement of path points inside the chromosome, enabling the same model UAVs to flexibly adjust the path when performing tasks to adapt to the requirements of different regions and effectively avoiding the "premature" phenomenon.

[0071] For the mutual cross operation on the same model gene segments between different chromosomes, it ensures the gene diversity in the multi-UAV task. This mutual cross method only alternates with each other at specific path points. By eliminating duplicate path points, it ensures full path coverage while reducing duplicate coverage, further improving the rationality of the planning scheme and the resource utilization efficiency.

[0072] Based on the above embodiment, the mutation operator is specifically used for: Randomly obtain a target mutation chromosome in the drone path planning chromosome population of the current iteration cycle, wherein the target mutation chromosome is a chromosome to be executed by the mutation operator in the drone path planning chromosome population of the current iteration cycle; When it is determined that the first drone gene segment in the target mutation chromosome is the gene segment to be mutated, the first mutation gene is added to the first target mutation gene segment in the target mutation chromosome, and the genes identical to the first mutation gene in the gene segments other than the first target mutation gene segment in the target mutation chromosome are eliminated; and based on the survey area number information other than the survey area number information covered by the first drone gene segment in the target mutation chromosome in the target region, the second drone gene segment in the target mutation chromosome is regenerated to obtain the mutated target mutation chromosome, wherein the first mutation gene is obtained by encoding based on the randomly obtained survey area number information; the first target mutation gene segment is the shortest first drone gene segment in the target mutation chromosome; When it is determined that the second drone gene segment in the target mutation chromosome is the gene segment to be mutated, the second mutation gene is added to the second target mutation gene segment in the target mutation chromosome, and the gene identical to the second mutation gene in the third target mutation gene segment in the target mutation chromosome is removed to obtain the mutated target mutation chromosome; wherein the second target mutation gene segment is the shortest second drone gene segment in the target mutation chromosome; the second mutation gene is randomly determined by other genes except the second target mutation gene segment and the genes in the first drone gene segment in the target mutation chromosome; the third target mutation gene segment is the second drone gene segment in the target mutation chromosome that contains the same gene as the second mutation gene.

[0073] Figure 7 The schematic diagram of the mutation process of the improved genetic algorithm provided by the present invention can be referred to Figure 7 As shown, the calculation of the mutation factor in the genetic algorithm is also divided into mutations occurring on the gene segment A or B of the drone. In the present invention, the mutation mode is Operation, that is, gene addition operation.

[0074] Specifically, for the gene addition operation on the gene segment of drone A, let the original chromosome be , randomly select the shortest gene segment in drone A (i.e. the first target mutation gene segment) to add a path point , to avoid repeated surveys of the same area, the selected points should exclude the points on the mutated gene segments, so as to obtain the gene segment part of UAV A in the new chromosome . At this time, for the gene segment of UAV B, new path points are regenerated to obtain the gene segment part of UAV B , and then a new chromosome is obtained , and the specific process is as follows: Randomly select the shortest gene segment of UAV A , and add path points : ; Among them, the selection of path points should meet the following requirements: ; Among them, the added path points need to remove the same path points on the gene segment of UAV A.

[0075] Since the gene segment of UAV A does not change due to other gene segments, it is expressed as: ; The gene segment of UAV B is the point set obtained by subtracting the path points covered by the gene segment of UAV A from all path points and regenerated: ; Finally, the new chromosome is the combination of the gene segments of UAV A and UAV B: ; Among them, is the point set obtained by subtracting the path points covered by the gene segment of UAV A from all path points.

[0076] Furthermore, for the gene addition operation on the gene segment of UAV B, let the original chromosome be , randomly select the shortest gene segment of UAV B, and add a path point from it. To ensure the full coverage requirement of the path points, a balancing operation is required, that is, the just-added path point is removed on the gene segment of UAV B (this gene segment is the third target mutated gene segment) .

[0077] To avoid repeated surveys of the same area, the selected points should exclude the points on the mutated gene segments and the points covered by UAV A, so as to obtain the gene part of UAV B in the new chromosome as , while the gene segment part of UAV A remains unchanged and is equal to , because the path of UAV A is not affected in this mutation operation. Therefore, the new chromosome changes as follows: Randomly select the shortest gene segment on UAV B , and add path points : ; Among them, the selection of path points should meet the following conditions: ; Among them, represents all the path point sets covered by the points on the gene segment of UAV A.

[0078] Then, remove the path point from the gene segment of UAV B (i.e., the third target mutated gene segment): .

[0079] At this time, the other gene segments of UAV B remain unchanged, and the gene segment of the new UAV B is represented as: .

[0080] Finally, the new chromosome is a combination of the gene segments of UAV A and UAV B: .

[0081] For reference, Figure 7 as shown, in the gene segment of the shortest UAV B under chromosome , is randomly added under the condition of ensuring non-repeated paths. To ensure path integrity, this path point is removed from a corresponding gene segment of UAV B.

[0082] Based on the above embodiments, the mutation operator is further used for: When determining that the first UAV gene segment in the target mutated chromosome is the gene segment to be mutated, remove the first removal gene from the first target removal gene segment in the target mutated chromosome, and regenerate the second UAV gene segment in the target mutated chromosome based on other survey area number information in the target area except the survey area number information covered by the first UAV gene segment in the target mutated chromosome, to obtain the mutated target mutated chromosome, where the first removal gene is encoded based on the randomly obtained survey area number information; the first target removal gene segment is the longest first UAV gene segment in the target mutated chromosome; When determining that the second UAV gene segment in the target mutant chromosome is the gene segment to be mutated, the second removed gene is removed from the second target removed gene segment in the target mutant chromosome, and a gene identical to the second removed gene is added to the third target removed gene segment in the target mutant chromosome to obtain the mutated target mutant chromosome; wherein, the second target removed gene segment is the longest second UAV gene segment in the target mutant chromosome; the second removed gene is randomly determined based on the genes in the second target removed gene segment in the target mutant chromosome; the third target removed gene segment is the shortest second UAV gene segment in the target mutant chromosome.

[0083] In the present invention, for the operation on the UAV A gene segment, that is, the gene removal operation, let the original chromosome be , randomly select the longest gene segment on UAV A and randomly remove a path point , thereby obtaining the UAV A gene segment part of the new chromosome . At this time, for the UAV B gene segment, new path points are regenerated to obtain the UAV B gene segment part , and then the new chromosome is obtained. The specific process is as follows: Randomly select the longest gene segment of UAV A , and remove the path point : ; Since the other gene segments of the UAV A gene segment do not change, it is represented as: ; The UAV B gene segment is the point set obtained by deducting the path points covered by the UAV A gene segment from all path points , and each gene segment is regenerated specifically as: ; Finally, the new chromosome is a combination of the UAV A and UAV B gene segments: ; Among them, is the point set obtained by deducting the path points covered by the UAV A gene segment from all path points.

[0084] For the gene removal operation on the UAV B gene segment, let the original chromosome be , randomly select the longest gene segment on UAV B (i.e., the second target removed gene segment), and remove a path point from it , to ensure the full coverage requirement of waypoints, a balancing operation is needed, that is, adding the just-removed waypoint to the shortest gene segment of UAV B (i.e., the third target removal gene segment). , so that the gene part of UAV B in the new chromosome is , while the gene segment part of UAV A remains unchanged and is equal to , because the path of UAV A is not affected in this mutation operation. Therefore, the new chromosome has the following specific change process: Randomly select the longest gene segment on UAV B , and remove the waypoint : ; Add the waypoint to the shortest gene segment of UAV B : ; Since the gene segments of UAV B do not change due to other gene segments, it is expressed as: ; Finally, the new chromosome is a combination of the gene segments of UAV A and UAV B: .

[0085] As Figure 7 shown, in the longest gene segment of UAV B in chromosome , is randomly removed. To ensure path integrity, it is added to one of the shortest gene segments of UAV B.

[0086] In the mutation process of the present invention, by adding new waypoints or removing existing waypoints, it adapts to the requirements of actual tasks for coverage and path adjustment. Among them, the addition operation ensures the addition of new waypoints in the uncovered area, while the removal operation optimizes the path length and reduces unnecessary repeated paths.

[0087] In an embodiment, to verify the effectiveness of the present invention, a simulation experiment was carried out to simulate the path planning process of multi-heterogeneous UAVs. The specific algorithm parameter settings and the type and attribute information of the UAVs are shown in Table 1 and Table 2 respectively: Table 1 Parameters of the improved genetic algorithm

[0088] Table 2 Parameters of heterogeneous UAVs

[0089] In this simulation experiment, the results of the optimal cooperative path planning are shown in Table 3 and Table 4. A total of 6 UAV As and 14 UAV Bs are dispatched for the survey of regional points in this planning. In the planning results, the mission execution time of UAVA-2 is used as the total time of this planning mission, and the final total time is 6.08 hours. This result indicates that the proposed cooperative path planning scheme can effectively coordinate multiple UAVs and achieve efficient regional coverage survey.

[0090] Table 3 Information of Optimal Cooperative Path Planning I

[0091] Table 4 Information of Optimal Cooperative Path Planning II

[0092] Figure 8 This is the simulation experiment result diagram of the cooperative planning of UAV A provided by the present invention. Figure 9 This is an individual case diagram of the simulation experiment result of the cooperative planning of UAV B provided by the present invention. Among them, Figure 8 shows the planned route diagram of UAV A in the constructed two-dimensional space. Figure 9 displays the path planning of the UAV numbered UAVB-1. Figure 8 and Figure 9 The abscissa and ordinate in represent longitude and latitude information. It can be seen from the simulation diagram that under the algorithm proposed in the present invention, the path planning effect is remarkable, and it can effectively achieve the coverage of the mission area, verifying the superiority and effectiveness of the algorithm in the cooperative path planning of multiple UAVs. Figure 10 This is the schematic diagram of the simulation result of the best adaptive value iteration diagram of the improved genetic algorithm provided by the present invention. As Figure 10 shown, the improved genetic algorithm provided by the present invention shows strong exploration ability in the initial stage. Thanks to the diversity of the crossover and mutation operators, the algorithm can quickly find a path better than the initial solution. In the middle and late stages, as the algorithm gradually converges, it finally tends to be stable. This process proves that the genetic algorithm with a specific coding form adopted in the present invention has excellent robustness and convergence, and can effectively solve the multi-objective optimization path planning problem of multiple heterogeneous UAVs. This algorithm shows good balance in global search and local convergence and is suitable for the optimization of complex tasks.

[0093] Next, the heterogeneous UAV path planning system provided by the present invention will be described. The heterogeneous UAV path planning system described below can be mutually corresponding and referred to the heterogeneous UAV path planning method described above.

[0094] Figure 11 This is the structural schematic diagram of the heterogeneous UAV path planning system provided by the present invention. As Figure 11As shown in the figure, the present invention provides a heterogeneous UAV path planning system, including a processing module 1101 and a path planning module 1102. Among them, the processing module 1101 is used to perform crossover processing and mutation processing on the gene segments of the chromosomes in the UAV path planning chromosome population in the current iteration cycle based on an improved genetic algorithm, so as to obtain a new UAV path planning chromosome population. Among them, the improved genetic algorithm includes a first crossover operator, a second crossover operator, and a mutation operator. The first crossover operator is used to perform crossover processing on the gene segments corresponding to the same UAV model within the same chromosome, the second crossover operator is used to perform crossover processing on the gene segments corresponding to the same UAV model between different chromosomes, and the mutation operator is used to add or remove gene segments within the chromosome. The UAV path planning chromosome in the current iteration cycle includes the quantity information and model attributes of the first UAV and the second UAV, as well as the survey area path information of the first UAV and the second UAV in the target area. The maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV. The path planning module 1102 is used to obtain a target chromosome from the new UAV path planning chromosome population when it is determined that the individual fitness of the chromosomes in the new UAV path planning chromosome population meets the preset fitness, so as to construct an optimal cooperative flight path of the first UAV and the second UAV in the target area according to the UAV path planning corresponding to the target chromosome.

[0095] The heterogeneous UAV path planning system provided by the present invention performs crossover processing and mutation processing on the gene segments of the chromosomes in the UAV path planning chromosome population in the current iteration cycle through a genetic algorithm with a specific coding form, improving the iteration update efficiency of the UAV path planning chromosome population, effectively solving the multi-objective optimization path planning problem of multiple heterogeneous UAVs, and enhancing the efficiency and adaptability of UAV path planning.

[0096] The system provided by the embodiments of the present invention is used to execute the above method embodiments. For the specific process and detailed content, please refer to the above embodiments and will not be elaborated here.

[0097] Figure 12 It is a schematic structural diagram of the electronic device provided by the present invention, as Figure 12As shown, the electronic device may include: a processor 1201, a communications interface 1202, a memory 1203, and a communication bus 1204. Among them, the processor 1201, the communications interface 1202, and the memory 1203 complete mutual communication through the communication bus 1204. The processor 1201 may call logic instructions in the memory 1203 to execute a heterogeneous drone path planning method, which includes: based on an improved genetic algorithm, performing crossover processing and mutation processing on gene segments of chromosomes in the drone path planning chromosome population of the current iteration cycle to obtain a new drone path planning chromosome population; wherein, the improved genetic algorithm includes a first crossover operator, a second crossover operator, and a mutation operator. The first crossover operator is used to perform crossover processing on gene segments corresponding to the same drone model within the same chromosome, the second crossover operator is used to perform crossover processing on gene segments corresponding to the same drone model between different chromosomes, and the mutation operator is used to add or remove gene segments within a chromosome; the drone path planning chromosome of the current iteration cycle includes the quantity information and model attributes of a first drone and a second drone, and the survey area path information of the first drone and the second drone within the target area. The maximum survey radius corresponding to the model attribute of the first drone is greater than the maximum survey radius corresponding to the model attribute of the second drone; when it is determined that the individual fitness of the chromosomes in the new drone path planning chromosome population meets a preset fitness, a target chromosome is obtained from the new drone path planning chromosome population, so as to construct an optimal cooperative flight path of the first drone and the second drone within the target area according to the drone path planning corresponding to the target chromosome.

[0098] In addition, when the logic instructions in the above-mentioned memory 1203 can be implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0099] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the heterogeneous unmanned aerial vehicle path planning method provided by the above-mentioned various methods. The method includes: based on an improved genetic algorithm, performing crossover processing and mutation processing on gene segments of chromosomes in the chromosome population for unmanned aerial vehicle path planning in the current iteration cycle to obtain a new chromosome population for unmanned aerial vehicle path planning; wherein, the improved genetic algorithm includes a first crossover operator, a second crossover operator, and a mutation operator. The first crossover operator is used to perform crossover processing on gene segments corresponding to the same unmanned aerial vehicle model within the same chromosome. The second crossover operator is used to perform crossover processing on gene segments corresponding to the same unmanned aerial vehicle model between different chromosomes. The mutation operator is used to add or remove gene segments within a chromosome. The chromosome for unmanned aerial vehicle path planning in the current iteration cycle includes the quantity information and model attributes of the first unmanned aerial vehicle and the second unmanned aerial vehicle, and the survey area path information of the first unmanned aerial vehicle and the second unmanned aerial vehicle within the target area. The maximum survey radius corresponding to the model attribute of the first unmanned aerial vehicle is greater than the maximum survey radius corresponding to the model attribute of the second unmanned aerial vehicle. When it is determined that the individual fitness of the chromosomes in the new chromosome population for unmanned aerial vehicle path planning meets a preset fitness, a target chromosome is obtained from the new chromosome population for unmanned aerial vehicle path planning, so as to construct an optimal cooperative flight path of the first unmanned aerial vehicle and the second unmanned aerial vehicle within the target area according to the unmanned aerial vehicle path planning corresponding to the target chromosome.

[0100] In another aspect, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the heterogeneous UAV path planning method provided in the above embodiments. The method includes: based on an improved genetic algorithm, performing crossover processing and mutation processing on gene segments of chromosomes in the UAV path planning chromosome population of the current iteration cycle to obtain a new UAV path planning chromosome population; wherein, the improved genetic algorithm includes a first crossover operator, a second crossover operator and a mutation operator. The first crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model within the same chromosome, the second crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model between different chromosomes, and the mutation operator is used to add or remove gene segments within the chromosome; the UAV path planning chromosome of the current iteration cycle includes the quantity information and model attributes of the first UAV and the second UAV, as well as the survey area path information of the first UAV and the second UAV within the target area. The maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV; when it is determined that the individual fitness of the chromosomes in the new UAV path planning chromosome population meets the preset fitness, a target chromosome is obtained from the new UAV path planning chromosome population, so as to construct an optimal cooperative flight path of the first UAV and the second UAV within the target area according to the UAV path planning corresponding to the target chromosome.

[0101] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0102] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence, or the parts that contribute to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A path planning method for heterogeneous unmanned aerial vehicles, characterized in that, Including: Based on an improved genetic algorithm, perform crossover processing and mutation processing on the gene segments of the chromosomes in the chromosome population of the UAV path planning in the current iteration cycle to obtain a new chromosome population of the UAV path planning; Among them, the improved genetic algorithm includes a first crossover operator, a second crossover operator, and a mutation operator. The first crossover operator is used to perform crossover processing on the gene segments corresponding to the same UAV model within the same chromosome, and the second crossover operator is used to perform crossover processing on the gene segments corresponding to the same UAV model between different chromosomes. The mutation operator is used to add or remove gene segments within the chromosome. The chromosome of the UAV path planning in the current iteration cycle includes the quantity information and model attributes of the first UAV and the second UAV, as well as the survey area path information of the first UAV and the second UAV in the target area. The maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV; When it is determined that the individual fitness of the chromosomes in the new chromosome population of the UAV path planning meets the preset fitness, obtain the target chromosome from the new chromosome population of the UAV path planning, so as to construct the optimal cooperative flight path of the first UAV and the second UAV in the target area according to the UAV path planning corresponding to the target chromosome.

2. The heterogeneous UAV path planning method according to claim 1, wherein The method further includes: Divide the area to be surveyed into the target area composed of multiple preset survey square grids, where the target area is a square area, and the preset survey square grids are constructed based on the maximum survey radius of the second UAV; Number each of the preset survey square grids in the target area in sequence to obtain the survey area number information corresponding to each of the preset survey square grids; According to the preset survey square grids corresponding to the initial survey area points of the first UAV and the second UAV in the target area respectively, obtain the survey area number information corresponding to different initial path plans; Perform chromosome coding according to the model attribute of the first UAV and the survey area number information to obtain the gene segments of the first UAV corresponding to different initial path plans, and based on the quantity information of the first UAV, splice all the gene segments of the first UAV in the same initial path plan to obtain the first gene sequence; Perform chromosome coding according to the model attribute of the second UAV and the survey area number information to obtain the gene segments of the second UAV corresponding to different initial path plans, and based on the quantity information of the second UAV, splice all the gene segments of the second UAV in the same initial path plan to obtain the second gene sequence; where each gene in the gene segments of the first UAV and the gene segments of the second UAV corresponds to a piece of the survey area number information; Based on the first gene sequence and the second gene sequence, an initial UAV path planning chromosome corresponding to the same initial path planning is constructed, and an initial UAV path planning chromosome population is constructed based on all the initial UAV path planning chromosomes; Based on the improved genetic algorithm, the gene segments in the initial UAV path planning chromosome are subjected to the crossover process and the mutation process to obtain a new initial UAV path planning chromosome; Based on the initial UAV path planning chromosome population and the new initial UAV path planning chromosome, a UAV path planning chromosome population for the next iteration cycle is constructed.

3. The heterogeneous UAV path planning method according to claim 2, wherein The constructing a UAV path planning chromosome population for the next iteration cycle based on the initial UAV path planning chromosome population and the new initial UAV path planning chromosome includes: Based on the roulette wheel algorithm, new individuals are determined from multiple new initial UAV path planning chromosomes; The new individuals are added to the initial UAV path planning chromosome population to obtain a new initial UAV path planning chromosome population; Based on a preset population size and the preset fitness, parental individuals for the next iteration cycle are obtained from the new initial UAV path planning chromosome population; Based on the parental individuals for the next iteration cycle, a UAV path planning chromosome population for the next iteration cycle is constructed.

4. The heterogeneous UAV path planning method according to claim 2, characterized in that, The first crossover operator is specifically used for: Randomly obtaining a target self-crossover chromosome in the UAV path planning chromosome population of the current iteration cycle, where the target self-crossover chromosome is the chromosome in the UAV path planning chromosome population of the current iteration cycle to which the first crossover operator is to be applied; Performing a reverse order operation, a swap operation, and a sliding operation on the gene segments in at least one of the first UAV gene segments and / or the gene segments in at least one of the second UAV gene segments in the target self-crossover chromosome to obtain a new UAV path planning chromosome; Wherein, the gene segments correspond to one or more of the survey area number information; The reverse order operation is used to reverse the order of the genes in the gene segment; The swap operation is used to swap the positions of any one or more genes in the gene segment with other genes in the gene segment; The swap operation is also used to swap the positions of any one or more genes in the gene segments between different gene segments under the same UAV model in the same chromosome; The sliding operation is used to move any one or more genes in the gene segment from the current gene position to a gene position between other genes in the gene segment.

5. The heterogeneous UAV path planning method according to claim 2 or 4, characterized in that The second crossover operator is specifically used for: Randomly obtain a first target mutually crossed chromosome and a second target mutually crossed chromosome in the drone path planning chromosome population of the current iteration cycle, wherein the first target mutually crossed chromosome and the second target mutually crossed chromosome are chromosomes to be executed by the second crossover operator in the drone path planning chromosome population of the current iteration cycle; Exchanging gene positions of at least one gene segment of the first drone gene segment in the first target mutually crossed chromosome with a corresponding number of gene segments of the first drone gene segment in the second target mutually crossed chromosome, and / or exchanging gene positions of at least one gene segment of the second drone gene segment in the first target mutually crossed chromosome with a corresponding number of gene segments of the second drone gene segment in the second target mutually crossed chromosome, to obtain a mutually crossed first target mutually crossed chromosome and a mutually crossed second target mutually crossed chromosome, wherein the gene segments correspond to one or more of the survey area number information; The survey area number information in the gene segments of the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing is judged, if there is no repeated survey area number information in all gene segments of the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing, then the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing are both determined as the new drone path planning chromosome; If the first target mutually crossed chromosome after the mutual crossing and the second target mutually crossed chromosome after the mutual crossing each have repeated survey area number information in the first drone gene segment, the first target mutually crossed chromosome after the mutual crossing is restored to the first target mutually crossed chromosome, and the second target mutually crossed chromosome after the mutual crossing is restored to the second target mutually crossed chromosome; If there is no repeated survey area number information in the first drone gene segment of the first target mutually crossed chromosome after mutual crossing and the second target mutually crossed chromosome after mutual crossing, and there is repeated survey area number information in the second drone gene segment of the first target mutually crossed chromosome after mutual crossing and the second target mutually crossed chromosome after mutual crossing, the repeated genes in the second drone gene segment of the first target mutually crossed chromosome after mutual crossing and the second target mutually crossed chromosome after mutual crossing are removed to obtain the new drone path planning chromosome, wherein the repeated genes are determined by the following formula: ； ； Among them, represents the duplicate gene in the first target cross chromosome after the cross; represents the duplicate gene in the second target cross chromosome after the cross; represents the th gene in the second drone gene segment in the first target cross chromosome after the cross, represents the th gene in the second drone gene segment in the first target cross chromosome after the cross, represents the th gene in the second drone gene segment in the second target cross chromosome after the cross, represents the th gene in the second drone gene segment in the second target cross chromosome after the cross; represents the first drone survey area number information set, and the first drone survey area number information set is the survey area number information set covered by the genes corresponding to the first drone gene segment in the first target cross chromosome after the cross; represents the second drone survey area number information set, and the second drone survey area number information set is the survey area number information set covered by the genes corresponding to the first drone gene segment in the second target cross chromosome after the cross.

6. The heterogeneous UAV path planning method according to claim 5, characterized in that, The second crossover operator is also used for: Obtain the first set of survey area number information and the second set of survey area number information, where the first set of survey area number information represents the set of survey area number information covered by the genes corresponding to all gene segments in the first target intersected chromosome after the intersection; the second set of survey area number information represents the set of survey area number information covered by the genes corresponding to all gene segments in the second target intersected chromosome after the intersection; Calculate the difference set between the first set of survey area number information and the first set of UAV survey area number information to obtain the first difference set; Calculate the difference set between the second set of survey area number information and the second set of UAV survey area number information to obtain the second difference set; Based on the survey area number information in the first difference set, re-encode to obtain the second UAV gene segment in the first new individual, and obtain the first remaining difference set, where the first new individual is the new UAV path planning chromosome obtained based on the first target intersected chromosome after the intersection; the first remaining difference set is the set of survey area number information in the first difference set that is not encoded as a gene during the process of re-encoding to obtain the second UAV gene segment in the first new individual; Based on the survey area number information in the second difference set, re-encode to obtain the second UAV gene segment in the second new individual, and obtain the second remaining difference set, where the second new individual is the new UAV path planning chromosome obtained based on the second target intersected chromosome after the intersection; the second remaining difference set is the set of survey area number information in the second difference set that is not encoded as a gene during the process of re-encoding to obtain the second UAV gene segment in the second new individual; Obtain the first shortest gene segment and the second shortest gene segment, where the first shortest gene segment is the shortest second UAV gene segment in the first new individual, and the second shortest gene segment is the shortest second UAV gene segment in the second new individual; Encode the survey area number information in the first remaining difference set into the corresponding gene and add it to the first shortest gene segment to expand the genes of the first new individual to obtain the first new individual with expanded genes; Encode the survey area number information in the second remaining difference set into the corresponding gene and add it to the second shortest gene segment to expand the genes of the second new individual to obtain the second new individual with expanded genes.

7. The heterogeneous UAV path planning method according to claim 2, characterized in that The mutation operator is specifically used for: Randomly obtain the target mutated chromosome in the UAV path planning chromosome population of the current iteration cycle, where the target mutated chromosome is the chromosome to be executed with the mutation operator in the UAV path planning chromosome population of the current iteration cycle; When determining that the first UAV gene segment in the target mutant chromosome is the gene segment to be mutated, add the first mutant gene to the first target mutant gene segment in the target mutant chromosome, and remove the genes identical to the first mutant gene from other gene segments in the target mutant chromosome except the first target mutant gene segment; and regenerate the second UAV gene segment in the target mutant chromosome based on the survey area numbering information in the target area other than the survey area numbering information covered by the first UAV gene segment in the target mutant chromosome, to obtain the mutated target mutant chromosome, where the first mutant gene is encoded based on the randomly obtained survey area numbering information; the first target mutant gene segment is the shortest first UAV gene segment in the target mutant chromosome; When determining that the second UAV gene segment in the target mutant chromosome is the gene segment to be mutated, add the second mutant gene to the second target mutant gene segment in the target mutant chromosome, and remove the genes identical to the second mutant gene from the third target mutant gene segment in the target mutant chromosome, to obtain the mutated target mutant chromosome; where the second target mutant gene segment is the shortest second UAV gene segment in the target mutant chromosome; the second mutant gene is randomly determined from the genes other than those in the second target mutant gene segment and the first UAV gene segment in the target mutant chromosome; the third target mutant gene segment is the second UAV gene segment in the target mutant chromosome where there are genes identical to the second mutant gene.

8. The heterogeneous UAV path planning method according to claim 7, wherein The mutation operator is further used for: When determining that the first UAV gene segment in the target mutant chromosome is the gene segment to be mutated, remove the first removal gene from the first target removal gene segment in the target mutant chromosome, and regenerate the second UAV gene segment in the target mutant chromosome based on the survey area numbering information in the target area other than the survey area numbering information covered by the first UAV gene segment in the target mutant chromosome, to obtain the mutated target mutant chromosome, where the first removal gene is encoded based on the randomly obtained survey area numbering information; the first target removal gene segment is the longest first UAV gene segment in the target mutant chromosome; When determining that the second UAV gene segment in the target mutant chromosome is the gene segment to be mutated, the second removed gene is removed from the second target removed gene segment in the target mutant chromosome, and a gene identical to the second removed gene is added to the third target removed gene segment in the target mutant chromosome to obtain the mutated target mutant chromosome; wherein, the second target removed gene segment is the longest second UAV gene segment in the target mutant chromosome; the second removed gene is randomly determined based on the genes in the second target removed gene segment in the target mutant chromosome; the third target removed gene segment is the shortest second UAV gene segment in the target mutant chromosome.

9. The heterogeneous UAV path planning method according to claim 2, characterized in that The calculation formula of the individual fitness is as follows: ； Among them, represents the individual fitness, represents the total distance of all UAV flight paths, represents the number of redundant survey area number information within the target area, represents the side length of the target area, represents the flight time of the UAV corresponding to the maximum flight path among all UAVs, represents the distance weight, represents the redundant coverage penalty weight, represents the time weight.

10. A heterogeneous UAV path planning system, characterized in that: including: a processing module, configured to perform crossover processing and mutation processing on gene segments of chromosomes in the UAV path planning chromosome population in the current iteration cycle based on an improved genetic algorithm to obtain a new UAV path planning chromosome population; wherein, the improved genetic algorithm includes a first crossover operator, a second crossover operator and a mutation operator, the first crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model in the same chromosome, the second crossover operator is used to perform crossover processing on gene segments corresponding to the same UAV model between different chromosomes, and the mutation operator is used to add or remove gene segments in the chromosome; the UAV path planning chromosome in the current iteration cycle includes the quantity information and model attributes of the first UAV and the second UAV, and the survey area path information of the first UAV and the second UAV in the target area, and the maximum survey radius corresponding to the model attribute of the first UAV is greater than the maximum survey radius corresponding to the model attribute of the second UAV; a path planning module, configured to obtain a target chromosome from the new UAV path planning chromosome population when determining that the individual fitness of the chromosomes in the new UAV path planning chromosome population meets a preset fitness, so as to construct an optimal cooperative flight path of the first UAV and the second UAV in the target area according to the UAV path planning corresponding to the target chromosome.

11. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the heterogeneous UAV path planning method according to any one of claims 1 to 9.

Images