CN117742373A - Unmanned aerial vehicle path planning method based on improved genetic algorithm - Google Patents
Unmanned aerial vehicle path planning method based on improved genetic algorithm Download PDFInfo
- Publication number
- CN117742373A CN117742373A CN202311512310.7A CN202311512310A CN117742373A CN 117742373 A CN117742373 A CN 117742373A CN 202311512310 A CN202311512310 A CN 202311512310A CN 117742373 A CN117742373 A CN 117742373A
- Authority
- CN
- China
- Prior art keywords
- population
- individuals
- individual
- aerial vehicle
- unmanned aerial
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002068 genetic effect Effects 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000035772 mutation Effects 0.000 claims description 29
- 230000010076 replication Effects 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 8
- 230000002028 premature Effects 0.000 abstract description 5
- 238000005070 sampling Methods 0.000 abstract description 4
- 108090000623 proteins and genes Proteins 0.000 description 11
- 230000006870 function Effects 0.000 description 8
- 238000010606 normalization Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of path planning, and particularly relates to an unmanned aerial vehicle path planning method based on an improved genetic algorithm. The invention provides the improved genetic algorithm reference diversity operator to control genetic parameters so as to monitor population diversity, and uses a tabu table to monitor sub-generation quality, thereby preventing the problem of premature generation, optimizing the quality of the sub-generation and improving the algorithm solving efficiency; meanwhile, the optimized genetic algorithm also utilizes system information feedback to adjust indexes and parameters, so that the algorithm has strong adaptability in different environments; the new crossover operator is used, so that the exploration capacity of the algorithm to the space is improved; and a large-scale sampling population initialization mode is adopted, the population initialization robustness is improved, and finally, the efficiency of the unmanned plane path planning process is optimized.
Description
Technical Field
The invention belongs to the technical field of path planning, and particularly relates to an unmanned aerial vehicle path planning method based on an improved genetic algorithm.
Background
Unmanned aerial vehicle path planning technology is widely applied in various fields, including aviation, agriculture, logistics, public security and the like. The algorithms commonly used for path normalization include an a-algorithm, a DWA algorithm, an artificial potential field method, a genetic algorithm and the like, wherein the genetic algorithm has the advantages of high robustness and intelligence for the problem of finding an optimal solution, but the existing genetic algorithm cannot be well applied in the unmanned aerial vehicle realizing autonomous flight due to the low solving efficiency of the genetic algorithm and the problem of easy generation of 'premature convergence', and specific reasons include:
1. the performance and the result of the existing genetic algorithm are affected by parameter selection and adjustment, but parameters often need to be debugged by experienced people to achieve good effects, and the parameters of the genetic algorithm are different according to the characteristics of different problems;
2. the existing genetic algorithm has poor utilization of feedback information in the system, when the flight environment of the unmanned aerial vehicle is complex, the existing genetic algorithm cannot sense the external environment, so that a large number of redundant iterations can be performed on the algorithm, and the solving efficiency is quite low;
3. the existing genetic algorithm has dependence on population initialization, the randomness in the iterative solution process is too high, so that the final result may be trapped in a local optimal solution, and a plurality of iterations are usually required to search for a better path, which greatly increases the time consumed by algorithm solution.
Disclosure of Invention
The invention aims to solve the problems of low solving efficiency and high randomness of the existing unmanned aerial vehicle path planning method based on the genetic algorithm.
In order to solve the technical problems, the invention provides an unmanned aerial vehicle path planning method based on an improved genetic algorithm, which comprises the following steps:
s1, forming an individual used in a genetic algorithm by taking a path randomly formed by coordinate points in a three-dimensional map used by unmanned aerial vehicle flight as a data source;
s2, initializing a population according to the individuals;
s3, setting parameters of the genetic algorithm according to the population;
s4, calculating the fitness of each individual in the population by using a preset fitness function;
s5, updating a tabu table according to the fitness of each individual in the population;
s6, judging whether the genetic algorithm meets a preset termination condition, wherein:
if yes: s61, determining optimal individuals in the population, and outputting paths corresponding to the optimal individuals as optimal paths;
if no: s62, detecting diversity of the population, and carrying out cross mutation treatment on the population according to a preset cross mutation algorithm so as to update the population;
screening out similar individuals in the population according to the tabu list;
and carrying out population cross mutation processing on the similar individuals by using the preset cross mutation algorithm so as to update the population, and returning to the step S3.
Further, the step S2 specifically includes:
defining the population to comprise m individuals, randomly selecting 10m individuals from all the individuals, sequencing the randomly selected individuals according to fitness by utilizing the preset fitness function, and selecting m individuals in the sequence according to proportion to form the population.
Still further, the parameters of the genetic algorithm include at least one of a replication probability, a crossover probability, a mutation probability, an information entropy at k iterations, a crossover probability at k iterations, and an influence factor of the mutation probability.
Still further, the predetermined fitness function is:
wherein F is i Represents the fitness of the individual i, m represents the total number of individuals in the population, N represents the total iteration number of the population, N k Representing the current iteration number, D (i, j) represents the distance between the jth coordinate point and the (j+1) th coordinate point in the individual i, and theta (i, j) represents the angle value between the jth coordinate point and the front and rear coordinate points in the individual i, D max Represents the maximum distance value between the coordinate points in the individual, and C (i, j) represents the collision coefficient.
Further, step S5 specifically includes:
sorting the individuals according to the fitness value of each individual in the population;
the x individuals with the lowest fitness values in the sequence are stored in the tabu table.
Still further, the preset cross variation algorithm includes:
the individuals replicate with a replication probability, cross with a crossover probability, or mutate with a mutation probability, wherein:
definition of Pf i Is the replication probability for individual i, which satisfies:
wherein m is k Is a diversity operator, F best Is the maximum fitness value in the current population, F i The fitness value of the current individual i is represented, and e is a natural constant;
definition of Pc i Is the crossover probability for individual i, which satisfies:
definition Pm i Is the probability of variation for individual i, which satisfies:
further, the diversity operator m k The method meets the following conditions:
wherein e k And (5) representing the entropy value of the population of the kth generation, and q represents a preset diversity influence factor.
Further, in step S62, the step of screening out similar individuals in the population according to the tabu list is specifically:
and calculating the minimum replacement times between the individuals in the population and the individuals in the tabu list by adopting a hamming distance mode, and judging the individuals with the minimum replacement times smaller than a threshold t as the similar individuals.
Further, in step S1, coordinate points in the three-dimensional map are encoded using binary encoding.
The invention has the beneficial effects that an unmanned aerial vehicle path planning method based on an improved genetic algorithm is provided, the algorithm refers to a diversity operator to control genetic parameters so as to monitor population diversity, and a tabu table is used for monitoring sub-generation quality, so that the problem of premature offspring is prevented, the offspring quality is optimized, and the algorithm solving efficiency is improved; meanwhile, the optimized genetic algorithm also utilizes system information feedback to adjust indexes and parameters, so that the algorithm has strong adaptability in different environments; the new crossover operator is used, so that the exploration capacity of the algorithm to the space is improved; and a large-scale sampling population initialization mode is adopted, the population initialization robustness is improved, and finally, the efficiency of the unmanned plane path planning process is optimized.
Drawings
Fig. 1 is a schematic flow chart of steps of an unmanned aerial vehicle path planning method based on an improved genetic algorithm according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a crossover operator provided by an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Referring to fig. 1, fig. 1 is a schematic step flow diagram of an unmanned aerial vehicle path planning method based on an improved genetic algorithm according to an embodiment of the present invention, where the unmanned aerial vehicle path planning method includes the following steps:
s1, forming an individual used in a genetic algorithm by taking a path randomly formed by coordinate points in a three-dimensional map used by unmanned aerial vehicle flight as a data source.
In step S1, coordinate points in the three-dimensional map are encoded using binary encoding. Specifically, the codes in the genetic algorithm can map the solution in the problem space to the individual expression form in the genetic algorithm, and the codes not only determine the representation method of the search space, but also determine how operations such as crossing, mutation and the like are applied to the individual, so that the selection of a proper coding mode can directly influence the solving efficiency of the subsequent implementation modes of the genetic algorithm. The embodiment of the invention adopts binary coding, introduces a grid method when constructing a three-dimensional map, and takes the three-dimensional coordinate X of the central point of each grid i 、Y i 、Z i It is converted to binary, respectively, and the string of binary numbers is regarded as a "gene". A plurality of "genes" are combined to form a "chromosome", i.e., a complete path is obtained.
When describing the size of a three-dimensional map, it is assumed that the size of the map is n x n, i.e., the width, length, and height are all n units. To represent the coordinates of each point in three-dimensional space, a binary encoding of p bits is required to represent the position of each dimension. This means that the value of each coordinate axis ranges from 0 to 2 p (i.e. 2 p -1) from 0 to 2 can be represented by binary coding of p bits p -1 total 2 p And different integer values. For example, if a point representing the maximum in a three-dimensional space is (100,100,100), the corresponding binary code is (1100100,1100100,1100100).
S2, initializing the population according to the individuals.
Aiming at the problem of premature convergence of the population, the problem is caused to a certain extent by poor population initialization quality, low population diversity and a large number of individuals concentrated near the local optimal solution of the genetic algorithm, and if the number of the initialized population is increased, the whole algorithm consumes longer time, and the convergence speed is slowed down. The embodiment of the invention carries out initialization processing on the population in a mode of sampling in a large scale, and the step S2 is specifically as follows:
defining the population to comprise m individuals, randomly selecting 10m individuals from all the individuals, sequencing the randomly selected individuals according to fitness by utilizing the preset fitness function, and selecting m individuals in the sequence according to proportion to form the population.
The robustness of the initialized population can be improved by using a large number of samples as candidate individuals, the diversity of the population is improved, and the population initialization quality is effectively improved.
S3, setting parameters of the genetic algorithm according to the population.
The parameters of the genetic algorithm comprise at least one of replication probability, crossover probability, mutation probability, information entropy at k iterations, crossover probability at k iterations and influence factor of mutation probability.
S4, calculating the fitness of each individual in the population by using a preset fitness function.
The fitness function can calculate the fitness of each individual in the population, and the fitness of the individual is the only index for judging the quality of the solution and is also a direct factor for determining the replication, crossing or variation of the individual. Based on the fact that the unmanned aerial vehicle climbs, dives or turns greatly in the real flight process, requirements on the attitude control algorithm of the unmanned aerial vehicle and the hardware performance of the unmanned aerial vehicle are high, therefore, when the genetic algorithm is used for path planning, the embodiment of the invention avoids large-angle inflection points as much as possible, and the flight path planned by the algorithm is smoother.
The preset fitness function is as follows:
wherein F is i Represents the fitness of the individual i, m represents the total number of individuals in the population, N represents the total iteration number of the population, N k Representing the current iteration number, D (i, j) represents the Euclidean distance between the jth coordinate point and the (j+1) th coordinate point in the individual i, and theta (i, j) represents the angle value between the jth coordinate point and the front and rear coordinate points in the individual i, D max Represents the maximum distance value between the coordinate points in the individual, and C (i, j) represents the collision coefficient.
Among the above parameters, the euclidean distance is specifically:
the angle value is specifically:
the angle value may be regarded as the corner size of the path, which is an important indicator of the path smoothness. The greater the angle value of each coordinate point in the path, the higher the smoothness of the final path.
Weighting of path lengthGradually decreasing, while the weight of the path smoothness is +.>Gradually increasing, and when the population gradually approaches the shortest distance at the end of algorithm iteration, further optimizing the path smoothness of the population, improving the optimal path quality and enabling the unmanned aerial vehicle to fly more stably.
D max Is each coordinate point in the individualThe maximum distance value between the coordinate points, D (i, j)/Dmax can normalize the coordinate point distances and eliminate the dimension in order to eliminate the dimension influence of the distance and the angle; (θ (i, j) -90)/180 is the same.
As shown in the following formula, the larger the inflection point turning amplitude is, the lower the fitness value is; the smaller the inflection point turning amplitude or the no-turning, the higher the fitness thereof. And finally, the individual with the highest overall smoothness of the path obtains a higher fitness value.
The collision coefficient C (i, j) indicates whether there is an obstacle in the line between the j-th coordinate point and the j+1-th coordinate point in the individual i in the population, and the value is 0 if there is an obstacle, so that the fitness of the individual is indicated as 0, and the value is 1 if there is no obstacle.
S5, updating a tabu table according to the fitness of each individual in the population.
Aiming at the problems that the existing genetic algorithm is poor in solving speed and searching efficiency, and a large number of new individuals generated in an iterating way are extremely low in fitness or quite similar to individuals with low fitness in an original population, in order to improve the searching efficiency of the genetic algorithm, the embodiment of the invention refers to a tabu list to optimize the low-efficiency evolution mode. And comparing the individuals with low fitness in the tabu list with newly generated individuals in the population, and re-mutating or crossing the individuals with high similarity in the new population, so that the diversity of the new population is ensured, and the exploration capability and the searching efficiency of the genetic algorithm are improved.
The step S5 specifically comprises the following steps:
sorting the individuals according to the fitness value of each individual in the population;
the x individuals with the lowest fitness values in the sequence are stored in the tabu table.
S6, judging whether the genetic algorithm meets a preset termination condition, wherein:
if yes: s61, determining the optimal individuals in the population, and outputting paths corresponding to the optimal individuals as optimal paths.
If no: s62, detecting diversity of the population, and carrying out cross mutation treatment on the population according to a preset cross mutation algorithm so as to update the population;
screening out similar individuals in the population according to the tabu list;
and carrying out population cross mutation processing on the similar individuals by using the preset cross mutation algorithm so as to update the population, and returning to the step S3.
The preset cross mutation algorithm comprises the following steps:
the individuals replicate with a replication probability, cross with a crossover probability, or mutate with a mutation probability.
Individuals in the population are copied by adopting an elite retention mechanism, namely, individuals with high fitness are easier to copy, more optimal solutions are added to the population, and enough samples are added to find the optimal solutions; while the cross operation of individuals in the population can increase the space exploration capability, and individuals with high fitness value have better opportunity to find the global optimal solution, the existing cross operator is random operation, the offspring simply inherits a part of genes of father and a part of genes of mother, the performance of the offspring cannot be guaranteed to be better than that of the parents, and the problem that better points in the neighborhood can be ignored and the result is sub-converged to the local optimal is possibly caused based on the limited search range. The mutation operation of individuals in the population is helpful to jump out of the local optimal solution, the lower mutation probability can lead to search to fall into the local optimal solution, and the higher mutation probability can destroy the superiority of the individuals.
Wherein: definition of Pf i Is the replication probability for individual i, which satisfies:
wherein m is k Is a diversity operator, F best Is the maximum fitness value in the current population, F i Indicating adaptation of the current individual iA degree value, e, is a natural constant; when the population diversity is reduced and the fitness value of the individuals is low, the probability of generating the copying operation is low; when the population diversity is increased and the fitness value of the individuals is high, the probability of generating the copying operation is high; the introduction of the diversity operator can also ensure that when the diversity of the population is reduced, the probability of generating the copying operation of the excellent individuals is relatively reduced, and the algorithm is prevented from generating a large number of copying operations to further sink into the local optimal solution when the algorithm is sunk into the local optimal solution.
The embodiment of the invention designs three new crossover operators, so that the filial generation keeps the common gene segments of father and mother, and the genes of the crossover segments are determined by the first crossover, the second crossover and the third crossover. As shown in fig. 2, where a is the "parent" common gene segment, b is the "father" and "mother" gene segments, respectively, and c is the first crossover segment; the green section is a second cross section; the yellow segment is the third crossover segment.
The first crossing is to directly connect the two crossing points, and a straight line segment is generated as a crossing segment gene; the second intersection is to select the segment connected with any coordinate point of the two intersection points and the parent as the intersection gene segment; the third intersection is to select the segment connected with any two coordinate points of the two intersection points and the "parent" as the "intersection gene segment". The three new crossover operators of the patent make the algorithm embody the ideas of 'inheritance' and 'evolution', so that the algorithm has stronger exploration capacity on space.
Definition of Pc i Is the crossover probability for individual i, which satisfies:
definition Pm i Is the probability of variation for individual i, which satisfies:
the method aims at the situation that the genetic algorithm falls into a local optimal value due to the fact that the similarity of individuals in the population is too high. The embodiment of the invention introduces a diversity operator, which monitors the fitness value of the population individuals, so that more mutation or crossover operation is carried out when the population diversity is reduced, thereby ensuring the population diversity and avoiding sinking into a local optimal value.
The diversity operator represents entropy values of individual fitness in the population, namely the degree of confusion of the individual fitness in the population, and the higher the entropy values are, the more diversity is; the lower the entropy value, the more single the diversity. The diversity operator m k The method meets the following conditions:
wherein e k And (5) representing the entropy value of the population of the kth generation, and q represents a preset diversity influence factor.
Specifically, the steps of calculating the diversity operator are as follows:
1. firstly, carrying out normalization treatment on individual fitness in a population, and unifying dimensions:
Z i representing the fitness value of the ith individual after normalization.
2. Calculating the weight of each individual in the whole body:
P i is the ithThe individual fitness value after normalization treatment accounts for the ratio of the sum of the fitness values of the whole population.
3. Calculating entropy value of the k generation population:
e k is the entropy value of the kth generation population, and represents the confusion degree of the population fitness value of the kth generation, and indirectly represents the diversity of the population.
4. Finally, entropy value of the population is converted into a diversity operator m k ：
Final diversity operator m k Will directly affect individual replication, crossover and mutation probabilities. The diversity influence factor q represents that the population entropy value of q generations is selected as a reference value, and the population diversity overall shows a descending trend due to the convergence of an algorithm (each individual can evolve towards an optimal solution or a local optimal solution), so that the higher the value of the q value is, the larger the influence of a diversity operator in the individual replication, crossing and mutation probability is; the lower the value, the opposite is true. The q factor has a value range of [2,10 ]]Has better functionality.
In step S62, a step of screening similar individuals in the population according to the tabu list is specifically:
and calculating the minimum replacement times between the individuals in the population and the individuals in the tabu list by adopting a hamming distance mode, and judging the individuals with the minimum replacement times smaller than a threshold t as the similar individuals.
The similarity between binary character strings of different individuals can be calculated rapidly by means of the characteristic of the computing equipment such as a computer and the like for computing the Hamming distance, so that the algorithm has a good screening function on population offspring, the population can effectively evolve and find the optimal solution faster, and the exploration capacity and the searching efficiency of the genetic algorithm are improved.
The invention has the beneficial effects that an unmanned aerial vehicle path planning method based on an improved genetic algorithm is provided, the algorithm refers to a diversity operator to control genetic parameters so as to monitor population diversity, and a tabu table is used for monitoring sub-generation quality, so that the problem of premature offspring is prevented, the offspring quality is optimized, and the algorithm solving efficiency is improved; meanwhile, the optimized genetic algorithm also utilizes system information feedback to adjust indexes and parameters, so that the algorithm has strong adaptability in different environments; the new crossover operator is used, so that the exploration capacity of the algorithm to the space is improved; and a large-scale sampling population initialization mode is adopted, the population initialization robustness is improved, and finally, the efficiency of the unmanned plane path planning process is optimized.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM) or the like.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.
While the embodiments of the present invention have been illustrated and described in connection with the drawings, what is presently considered to be the most practical and preferred embodiments of the invention, it is to be understood that the invention is not limited to the disclosed embodiments, but on the contrary, is intended to cover various equivalent modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (9)
1. The unmanned aerial vehicle path planning method based on the improved genetic algorithm is characterized by comprising the following steps of:
s1, forming an individual used in a genetic algorithm by taking a path randomly formed by coordinate points in a three-dimensional map used by unmanned aerial vehicle flight as a data source;
s2, initializing a population according to the individuals;
s3, setting parameters of the genetic algorithm according to the population;
s4, calculating the fitness of each individual in the population by using a preset fitness function;
s5, updating a tabu table according to the fitness of each individual in the population;
s6, judging whether the genetic algorithm meets a preset termination condition, wherein:
if yes: s61, determining optimal individuals in the population, and outputting paths corresponding to the optimal individuals as optimal paths;
if no: s62, detecting diversity of the population, and carrying out cross mutation treatment on the population according to a preset cross mutation algorithm so as to update the population;
screening out similar individuals in the population according to the tabu list;
and carrying out population cross mutation processing on the similar individuals by using the preset cross mutation algorithm so as to update the population, and returning to the step S3.
2. The unmanned aerial vehicle path planning method according to claim 1, wherein step S2 is specifically:
defining the population to comprise m individuals, randomly selecting 10m individuals from all the individuals, sequencing the randomly selected individuals according to fitness by utilizing the preset fitness function, and selecting m individuals in the sequence according to proportion to form the population.
3. The unmanned aerial vehicle path planning method of claim 1, wherein the parameters of the genetic algorithm comprise at least one of a replication probability, a crossover probability, a mutation probability, an information entropy at k iterations, a crossover probability at k iterations, and an influence factor of a mutation probability.
4. The unmanned aerial vehicle path planning method of claim 1, wherein the predetermined fitness function is:
wherein F is i Represents the fitness of the individual i, m represents the total number of individuals in the population, N represents the total iteration number of the population, N k Representing the current iteration number, D (i, j) representing the distance between the j-th coordinate point and the j+1-th coordinate point in the individual i, θ (i, j) representing the angle value of the j-th coordinate point and the front and rear coordinate points in the individual i,D max represents the maximum distance value between the coordinate points in the individual, and C (i, j) represents the collision coefficient.
5. The unmanned aerial vehicle path planning method according to claim 1, wherein step S5 is specifically:
sorting the individuals according to the fitness value of each individual in the population;
the x individuals with the lowest fitness values in the sequence are stored in the tabu table.
6. The unmanned aerial vehicle path planning method of claim 1, wherein the preset crossover variation algorithm comprises:
the individuals replicate with a replication probability, cross with a crossover probability, or mutate with a mutation probability, wherein:
definition of Pf i Is the replication probability for individual i, which satisfies:
wherein m is k Is a diversity operator, F best Is the maximum fitness value in the current population, F i The fitness value of the current individual i is represented, and e is a natural constant;
definition of Pc i Is the crossover probability for individual i, which satisfies:
definition Pm i Is the probability of variation for individual i, which satisfies:
7. the unmanned aerial vehicle path planning method of claim 6, wherein the diversity operator m k The method meets the following conditions:
wherein e k And (5) representing the entropy value of the population of the kth generation, and q represents a preset diversity influence factor.
8. The unmanned aerial vehicle path planning method according to claim 1, wherein in step S62, the step of screening out similar individuals in the population according to the tabu list is specifically:
and calculating the minimum replacement times between the individuals in the population and the individuals in the tabu list by adopting a hamming distance mode, and judging the individuals with the minimum replacement times smaller than a threshold t as the similar individuals.
9. The unmanned aerial vehicle path planning method of claim 1, wherein in step S1, the coordinate points in the three-dimensional map are encoded using binary encoding.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311512310.7A CN117742373A (en) | 2023-11-13 | 2023-11-13 | Unmanned aerial vehicle path planning method based on improved genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311512310.7A CN117742373A (en) | 2023-11-13 | 2023-11-13 | Unmanned aerial vehicle path planning method based on improved genetic algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117742373A true CN117742373A (en) | 2024-03-22 |
Family
ID=90251551
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311512310.7A Pending CN117742373A (en) | 2023-11-13 | 2023-11-13 | Unmanned aerial vehicle path planning method based on improved genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117742373A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021022637A1 (en) * | 2019-08-06 | 2021-02-11 | 南京赛沃夫海洋科技有限公司 | Unmanned surface vehicle path planning method and system based on improved genetic algorithm |
| CN115202394A (en) * | 2022-07-13 | 2022-10-18 | 安徽成方智能科技有限公司 | Unmanned aerial vehicle full-coverage path planning method based on improved genetic algorithm |
-
2023
- 2023-11-13 CN CN202311512310.7A patent/CN117742373A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021022637A1 (en) * | 2019-08-06 | 2021-02-11 | 南京赛沃夫海洋科技有限公司 | Unmanned surface vehicle path planning method and system based on improved genetic algorithm |
| CN115202394A (en) * | 2022-07-13 | 2022-10-18 | 安徽成方智能科技有限公司 | Unmanned aerial vehicle full-coverage path planning method based on improved genetic algorithm |
Non-Patent Citations (3)
| Title |
|---|
| 吴俊;方锦明;: "采用排挤机制小生境技术改进禁忌搜索算法", 化工自动化及仪表, vol. 37, no. 08, 31 December 2010 (2010-12-31), pages 34 - 36 * |
| 梁金泉: "基于改进遗传算法的航迹规划研究", 中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑, no. 01, 15 January 2013 (2013-01-15), pages 1 - 43 * |
| 欧阳浩;陈波;: "求解最小生成树的一种小生境遗传禁忌算法", 计算机工程与应用, vol. 49, no. 01, 31 December 2013 (2013-12-31), pages 59 - 61 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11775833B2 (en) | Accelerated TR-L-BFGS algorithm for neural network | |
| Babatunde et al. | A genetic algorithm-based feature selection | |
| CN105488528B (en) | Neural network image classification method based on improving expert inquiry method | |
| Qu et al. | Chicken swarm optimization based on elite opposition-based learning | |
| Ishibuchi et al. | Diversity improvement by non-geometric binary crossover in evolutionary multiobjective optimization | |
| CN102013037A (en) | Method and device for searching path based on particle swarm optimization (PSO) | |
| Dong et al. | An Adaptive Multiobjective Genetic Algorithm with Fuzzy c‐Means for Automatic Data Clustering | |
| CN112214031B (en) | Multi-node collaborative landing position planning method based on genetic particle swarm optimization | |
| CN110287985B (en) | Depth neural network image identification method based on variable topology structure with variation particle swarm optimization | |
| Wang et al. | Optimal path planning based on hybrid genetic-cuckoo search algorithm | |
| Kadlec et al. | Particle swarm optimization for problems with variable number of dimensions | |
| CN113282756A (en) | Text clustering intelligent evaluation method based on hybrid clustering | |
| CN112381157A (en) | Fuzzy C-means clustering method based on goblet sea squirt group algorithm optimization | |
| Bandyopadhyay et al. | VGA-classifier: design and applications | |
| CN115099133A (en) | TLMPA-BP-based cluster system reliability evaluation method | |
| CN117742373A (en) | Unmanned aerial vehicle path planning method based on improved genetic algorithm | |
| Zhao et al. | An improved differential evolution algorithm for maritime collision avoidance route planning | |
| CN111614609B (en) | GA-PSO-DBN-based intrusion detection method | |
| CN114004326B (en) | ELM neural network optimization method based on improved suburban wolf algorithm | |
| Saha et al. | A fuzzy genetic clustering technique using a new symmetry based distance for automatic evolution of clusters | |
| Petrowski et al. | Evolutionary algorithms | |
| Fofanah et al. | Experimental Exploration of Evolutionary Algorithms and their Applications in Complex Problems: Genetic Algorithm and Particle Swarm Optimization Algorithm | |
| Tareq et al. | A new density-based method for clustering data stream using genetic algorithm | |
| CN113869380B (en) | Kmeans clustering-based data classification method and device and electronic equipment | |
| Nascimento et al. | Ensembling heterogeneous learning models with boosting |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |