CN115994666A - Unmanned aerial vehicle task allocation method based on adaptive genetic learning particle swarm algorithm - Google Patents
Unmanned aerial vehicle task allocation method based on adaptive genetic learning particle swarm algorithm Download PDFInfo
- Publication number
- CN115994666A CN115994666A CN202310070727.6A CN202310070727A CN115994666A CN 115994666 A CN115994666 A CN 115994666A CN 202310070727 A CN202310070727 A CN 202310070727A CN 115994666 A CN115994666 A CN 115994666A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- particles
- aerial vehicle
- particle swarm
- elite
- 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
Images
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
- Feedback Control In General (AREA)
Abstract
The invention discloses an unmanned aerial vehicle task allocation method based on a self-adaptive genetic learning particle swarm algorithm, which belongs to the technical field of unmanned aerial vehicle intelligent inspection control, and aims at detecting and supplementing materials for m pieces of equipment distributed at different positions in a scene by n pieces of unmanned aerial vehicles of different types scattered in a plurality of machine nests, wherein each piece of equipment needs different quantity and different kinds of material supplementation, each unmanned aerial vehicle can carry different quantities of food and material to complete the inspection task, an inspection relation non-decision variable between the unmanned aerial vehicle and the equipment is defined, and a model of multiple unmanned aerial vehicle multi-task types is constructed; the self-adaptive genetic learning particle swarm algorithm is provided for solving the model in the step S1, and unmanned aerial vehicle cooperative task allocation is obtained; according to the invention, the combination optimization model which is more fit with the actual is established by considering the different types of assistance required by the equipment, the algorithm searching capability is improved through the two-level connection structure, and a reasonable patrol allocation scheme can be quickly and effectively found.
Description
Technical field:
the invention relates to the technical field of intelligent inspection control of unmanned aerial vehicles, in particular to an unmanned aerial vehicle task allocation method based on a self-adaptive genetic learning particle swarm algorithm.
The background technology is as follows:
in order to meet the increasingly complex diversified task demands, the single unmanned aerial vehicle has limited resources and the capability of executing tasks, and is difficult to independently complete. The multi-unmanned aerial vehicle system has the advantages of being more flexible, higher in fault tolerance and the like, and can effectively and efficiently execute various complex tasks. However, if multiple unmanned aerial vehicles simply form a cluster, and a good cooperative mechanism is not available, the efficiency of the unmanned aerial vehicles for executing tasks cannot be improved, and serious conflict and resource waste may be caused due to the increase of the system scale. Therefore, the reasonable allocation of tasks in a multi-constraint complex environment is a key for improving the overall performance of the system and realizing multi-machine cooperation.
The problem of multi-unmanned aerial vehicle collaborative task allocation refers to that one or a group of ordered tasks are allocated to unmanned aerial vehicles, and the overall income of a system is guaranteed to be optimal. The problem has the characteristics of high complexity, strong constraint, high calculation difficulty and the like, and reasonable problem modeling and an effective solving algorithm are vital to the processing of the problem. In modeling, the number, the isomerism, the task execution capacity and other factors of the unmanned aerial vehicle are considered; in addition, the category of the task, the distributed environment and various cooperative constraint conditions are considered. There are many particle swarm-based methods for dealing with the problem of multi-unmanned aerial vehicle cooperative task allocation. Nevertheless, the following difficulties still exist in handling the multi-unmanned aerial vehicle cooperative task allocation in different task scenarios: 1) How to build an optimization model that meets the practical problem. 2) How to quickly and effectively obtain the optimal scheme of the model.
The invention comprises the following steps:
in order to overcome the defects of the prior art, the invention aims to provide an unmanned aerial vehicle task allocation method based on a self-adaptive genetic learning particle swarm algorithm, which considers the difference of the types of assistance required by equipment, defines an undetermined variable of the patrol relation between an unmanned aerial vehicle and the equipment, takes the average waiting time and the range distance as evaluation indexes of a task allocation scheme, adopts a penalty function method to process various constraints existing in the problem, and establishes a combined optimization model which is more fit with reality; the decision variable constraint is processed by adopting a real vector coding mechanism to simplify model solving, the algorithm searching capability is improved through a two-level joint structure, the particle swarm optimizing capability is effectively enhanced, the convergence speed of the particle swarm under different conditions is flexibly controlled, and the optimal scheme of the model is rapidly and effectively obtained.
The technical scheme of the invention is as follows:
the unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm comprises the following steps:
s1: aiming at n unmanned aerial vehicles of different types dispersed in a plurality of machine nests, detecting and supplementing material for m devices distributed at different positions in a scene, wherein each device needs different amounts and different kinds of material supplementation, each unmanned aerial vehicle can carry different amounts of food and material to complete a patrol task, a real vector coding mechanism is adopted to represent the relationship between the unmanned aerial vehicle and the task, and a model of multiple unmanned aerial vehicle multitasking types is constructed;
s2: the self-adaptive genetic learning particle swarm algorithm is used for solving the model in the step S1, and unmanned aerial vehicle cooperative task allocation is obtained.
The step S1 specifically includes the following steps:
s11: the problem is specifically described as: n different types of unmanned aerial vehicles distributed in a plurality of machine nests detect and supplement materials for m devices distributed at different positions in a scene, wherein each device needs to be supplemented with different amounts and different types of materials;
s12: a global equipment set T is set, wherein each equipment needs a certain amount of food and material assistance, the corresponding task set K= {1,2}, and an unmanned aerial vehicle set U is set, wherein each unmanned aerial vehicle can carry different amounts of food and material to complete the inspection task, and for each equipment, the material assistance can be completed by different unmanned aerial vehicles;
s13: defining an inspection relation non-decision variable between the unmanned aerial vehicle and the equipment;
s14: and taking the average waiting time and the range distance as evaluation indexes of the task allocation scheme, and adopting a penalty function method to treat various constraints existing in the problem.
The step S2 specifically includes the following steps:
s21: generating a high-quality elite population by adopting a genetic learning strategy to replace a historical optimal position to update particle positions, wherein the method specifically comprises three operations of crossing, mutation and selection;
s22: in the iterative evolution process, if the fitness value of elite particles generated by continuous sg iterations is unchanged, the evolution of the particles is considered to be stopped, and at the moment, for each dimension of the elite particles, two particles with better fitness values (the first 20%) in two pbest are selected for modification through a tournament selection mechanism, and the particles have small probability of variation;
s23: guiding particle speed and position update through elite population;
s24: and solving the multi-unmanned aerial vehicle cooperative task allocation by adopting an AGLPSO algorithm.
The step S21 specifically includes the following steps:
firstly, comparing historical optimal fitness values of current particles i and random particles k based on a ring topology structure to determine a value corresponding to a certain dimension in a crossed individual, and further realizing crossed operation;
then, judging whether each dimension of the crossed individuals is smaller than a given probability value, if so, resetting the value corresponding to the dimension as a random value in a search space, and further realizing mutation operation;
and finally, comparing the candidate particles generated by variation with the elite particles corresponding to the previous iteration according to a greedy method, and selecting a better individual as the elite particles of the current iteration to realize the selection operation.
The step S23 specifically includes the following steps:
an adaptive updating strategy is adopted to change the value of omega, the convergence speed of an algorithm is flexibly controlled according to the evolution speed of a particle swarm and the aggregation degree of particles, an evolution speed threshold value is set, and when the evolution speed of the particle swarm is greater than the evolution speed threshold value, the algorithm should keep global optimization in a large range; when the evolution speed is smaller than the threshold value of the evolution speed, the algorithm should perform a local search in a small range to find the optimal solution more quickly; the degree of aggregation of particles depends on the proximity of the fitness value of the particles in the population.
The step S24 specifically includes the following steps:
firstly, inputting initial information of an unmanned plane and survivors according to a patrol scene, setting the maximum iteration number Gmax of an algorithm, the number N of particles, initial values and final values of an inertia weight reference value, the upper limit and the lower limit of a learning factor, a mutation rate, the maximum stagnation algebra and a scale factor; initializing the iteration times G=0 and initializing the particle population;
secondly, calculating individual fitness of the particles, and updating historical optimal positions and global optimal positions of the particles; then, generating elite particles by adopting a genetic learning strategy, and updating the global optimal position;
then, updating stagnant elite particles by adopting an elite learning strategy and calculating inertia weight and learning factors; updating particle speed and position, calculating individual fitness of particles, and updating historical optimal position and global optimal position of particles;
and finally, the obtained global optimal position is the optimal solution, and the optimal solution is decoded to obtain an optimal inspection scheme marked by a calculation rule.
The invention has the advantages that:
1. according to the invention, the difference of the types of assistance required by the equipment is considered, the non-decision variable of the patrol relation between the unmanned aerial vehicle and the equipment is defined, the average waiting time and the range distance are used as evaluation indexes of a task allocation scheme, and a penalty function method is adopted to treat various constraints existing in the problem, so that a combined optimization model which is more fit with the reality is established.
2. According to different tasks of the unmanned aerial vehicle, the decision variable constraint is processed by adopting a real vector coding mechanism, so that model solving can be simplified, and the complexity of the problem is reduced.
3. The invention improves the algorithm searching capability through a two-level linkage structure: generating high-quality elite particles through a genetic learning strategy, and updating the particles with evolution stagnation by adopting the elite learning strategy to jump out of local optimum; and the second layer guides the searching direction of the population by elite particles, and adopts a self-adaptive evolutionary strategy to improve the optimizing capability of an algorithm in different evolutionary periods according to the evolutionary speed of the particle swarm and the aggregation degree of the particles, so that the optimizing capability of the particle swarm is effectively enhanced by a two-layer cascade structure, and the convergence speed of the particle swarm under different conditions is flexibly controlled.
Description of the drawings:
FIG. 1 is a flow chart of the method of the present invention.
The specific embodiment is as follows:
the following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.
As shown in fig. 1, the unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm disclosed by the invention comprises the following steps:
s1: aiming at n unmanned aerial vehicles of different types dispersed in a plurality of machine nests, detecting and supplementing material for m devices distributed at different positions in a scene, wherein each device needs different amounts and different kinds of material supplementation, each unmanned aerial vehicle can carry different amounts of food and material to complete a patrol task, a real vector coding mechanism is adopted to represent the relationship between the unmanned aerial vehicle and the task, and a model of multiple unmanned aerial vehicle multitasking types is constructed; the machine nest is arranged on the trunk pole tower;
s2: the self-adaptive genetic learning particle swarm algorithm is used for solving the model in the step S1, and unmanned aerial vehicle cooperative task allocation is obtained.
The step S1 is specifically as follows:
s11: the problem is specifically described as: n different types of unmanned aerial vehicles distributed in a plurality of machine nests detect and supplement materials for m devices distributed at different positions in a scene, wherein each device needs to be supplemented with different amounts and different types of materials;
s12: a global equipment set T is provided, wherein each equipment needs a certain amount of food and material assistance, the corresponding task sets K= {1, 2) and an unmanned aerial vehicle set U are respectively provided, each unmanned aerial vehicle can carry different amounts of food and material to complete the inspection task, and for each equipment, the material assistance can be completed by different unmanned aerial vehicles;
s13: defining an inspection relation non-decision variable between the unmanned aerial vehicle and the equipment;
s14: and taking the average waiting time and the range distance as evaluation indexes of the task allocation scheme, and adopting a penalty function method to treat various constraints existing in the problem.
The step S2 is specifically as follows:
s21: generating a high-quality elite population by adopting a genetic learning strategy to replace a historical optimal position to update particle positions, wherein the method specifically comprises three operations of crossing, mutation and selection;
s22: in the iterative evolution process, if the fitness value of elite particles generated by continuous sg iterations is unchanged, the evolution of the particles is considered to be stopped, and at the moment, for each dimension of the elite particles, two particles with better fitness values (the first 20%) in two pbest are selected for modification through a tournament selection mechanism, and the particles have small probability of variation;
s23: guiding particle speed and position update through elite population;
s24: and solving the multi-unmanned aerial vehicle cooperative task allocation by adopting an AGLPSO algorithm.
The step S21 is specifically as follows:
firstly, comparing historical optimal fitness values of current particles i and random particles k based on a ring topology structure to determine a value corresponding to a certain dimension in a crossed individual, and further realizing crossed operation;
then, judging whether each dimension of the crossed individuals is smaller than a given probability value, if so, resetting the value corresponding to the dimension as a random value in a search space, and further realizing mutation operation;
and finally, comparing the candidate particles generated by variation with the elite particles corresponding to the previous iteration according to a greedy method, and selecting a better individual as the elite particles of the current iteration to realize the selection operation.
The step S23 is specifically as follows:
an adaptive updating strategy is adopted to change the value of omega, the convergence speed of an algorithm is flexibly controlled according to the evolution speed of a particle swarm and the aggregation degree of particles, an evolution speed threshold value is set, and when the evolution speed of the particle swarm is greater than the evolution speed threshold value, the algorithm should keep global optimization in a large range; when the evolution speed is smaller than the threshold value of the evolution speed, the algorithm should perform a local search in a small range to find the optimal solution more quickly; the degree of aggregation of particles depends on the proximity of the fitness value of the particles in the population.
The step S24 is specifically as follows:
firstly, inputting initial information of an unmanned plane and survivors according to a patrol scene, setting the maximum iteration number Gmax of an algorithm, the number N of particles, initial values and final values of an inertia weight reference value, the upper limit and the lower limit of a learning factor, a mutation rate, the maximum stagnation algebra and a scale factor; initializing the iteration times G=0 and initializing the particle population;
secondly, calculating individual fitness of the particles, and updating historical optimal positions and global optimal positions of the particles; then, generating elite particles by adopting a genetic learning strategy, and updating the global optimal position;
then, updating stagnant elite particles by adopting an elite learning strategy and calculating inertia weight and learning factors; updating particle speed and position, calculating individual fitness of particles, and updating historical optimal position and global optimal position of particles;
and finally, the obtained global optimal position is the optimal solution, and the optimal solution is decoded to obtain an optimal inspection scheme marked by a calculation rule.
The invention adopts a two-level connection structure to enhance the optimizing capability of an algorithm, and is concretely implemented as follows:
step 2.1, generating elite population according to genetic learning strategy, guiding searching direction of particle swarm, and updating the elite particles with evolution stagnation according to elite learning strategy
And generating a high-quality elite population by adopting a genetic learning strategy to replace a historical optimal position to update the particle position. The method specifically comprises three operations of crossing, mutation and selection, and is described in detail as follows:
firstly, comparing historical optimal fitness values of current particles i and random particles k based on a ring topology structure to determine a value corresponding to a certain dimension in a crossed individual, and further realizing a crossed operation:
wherein O is i Represents the ith crossed individual; pbest (p best) i Representing a historical optimal position of the ith particle; d is the dimension of the search space; n is n i1 And n i2 An index representing two particles adjacent to the i-th particle; phi is a random number between (0, 1); f (·) represents the fitness function corresponding to the task allocation problem.
Then, crossing each dimension of the individual with a small probability p ml Judging whether the variation exists, if so, resetting the value corresponding to the dimension as a random value in the search space, and further realizing the variation operation:
wherein lb d And ub d Representing the lower and upper limits of the d-th dimension in the search space, respectively, rand (0, 1) represents the random number between (0, 1).
Finally, comparing the candidate particles generated by variation with the elite particles corresponding to the previous iteration according to a greedy method, and selecting a better individual as the elite particle of the current iteration to realize the selection operation:
wherein e i Indicating the ith elite particle.
In the iterative evolution process, if the fitness value of elite particles generated by continuous sg iterations is unchanged, the evolution of the particles is considered to be stagnant. At this time, for each dimension of elite particles, the better of the two pbest particles with better fitness value (the first 20%) is selected for modification by a tournament selection mechanism, and there is a small probability p m2 Variation occurs.
At this time, each dimension of elite particles can learn from different preferred particles and has a certain probability of variation, so that the diversity of the population is further enhanced, the particles are assisted to jump out of a local optimal region, and the particles are searched for a potential preferred region.
Step 2.2, dynamically adjusting inertia weight and learning factor according to the self-adaptive evolution strategy by the particle swarm, so that the particle swarm can flexibly control convergence rate according to actual search conditions
Particle velocity and position update are guided by elite population, and the formula is as follows:
wherein x is i And v i Respectively representing the position and velocity of the ith particle, i=1, 2, …, N; n is the total number of particles; gbest= (gbest) 1 ,gbest 2 ,...,gbest D ) Representing the global optimal position searched by the particle swarm; omega is the inertial weight; c 1 And c 2 Is a learning factor.
The value of the inertia weight omega has an important influence on the algorithm performance. When omega is larger, the algorithm has stronger global searching capability; when ω is smaller, the algorithm is more prone to local searches. Many documents use linear changes to update the value of ω, but they are not flexible enough to handle the complex, nonlinear characteristics that are present in the problem. Thus, an adaptive update strategy is employed herein to change the value of ω, flexibly controlling the convergence rate of the algorithm based on the evolutionary speed of the particle population and the degree of aggregation of the particles. The rate of evolution of the population of particles depends on the variation of the global optimum. Since the task allocation problem described herein is a process of minimizing the objective function, the evolution speed factor h of the particle swarm is defined as follows:
wherein F is mean The average fitness value of the particle swarm is represented, and N is the total number of particles in the swarm. It can be seen that 0<s is less than or equal to 1, and the smaller s is, the more dispersed the particles are, the less the algorithm is easy to fall into local optimum; when particles are aggregated, the algorithm should increase the search scope to improve the global optimization capability.
And 3, solving the cooperative task allocation of the multiple unmanned aerial vehicles by adopting a self-adaptive genetic learning particle swarm algorithm.
The process for solving the multi-unmanned aerial vehicle cooperative task allocation problem by adopting the AGLPSO algorithm is as follows:
step1: inputting initial information of the unmanned aerial vehicle and survivors according to the inspection scene;
step2: setting the initial value and the final value omega of the maximum iteration number Gmax, the number N of particles and the reference value of the inertia weight omega of the algorithm i ，ω f Learning factor c 1 ，c 2 Upper and lower limit c of (2) 1max ，c 1min ，c 2max ，c 2min Mutation rate p m1 ，p m2 Maximum stagnation algebra sg and scaling factor omega h ，ω s . Initializing the iteration times G=0 and initializing the particle population;
step3: calculating individual fitness of particles, and updating the historical optimal position and the global optimal position of the particles;
step4: generating elite particles by adopting a genetic learning strategy, and updating the global optimal position;
step5: updating stagnant elite particles by adopting an elite learning strategy;
step6: calculating inertia weight and learning factor;
step7: updating particle velocity and position;
step8: calculating individual fitness of particles, and updating the historical optimal position and the global optimal position of the particles;
step9: updating iteration times G=G+1, judging whether G is less than or equal to Gmax, if so, turning to step4, otherwise, ending the algorithm, and obtaining a global optimal position in step8 as an optimal solution;
step10: and decoding the optimal solution to obtain an optimal inspection scheme marked by a calculation rule.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (6)
1. The unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm is characterized by comprising the following steps of: the method comprises the following steps:
s1: aiming at n unmanned aerial vehicles of different types dispersed in a plurality of machine nests, detecting and supplementing material for m devices distributed at different positions in a scene, wherein each device needs different amounts and different kinds of material supplementation, each unmanned aerial vehicle can carry different amounts of food and material to complete a patrol task, a real vector coding mechanism is adopted to represent the relationship between the unmanned aerial vehicle and the task, and a model of multiple unmanned aerial vehicle multitasking types is constructed;
s2: the self-adaptive genetic learning particle swarm algorithm is used for solving the model in the step S1, and unmanned aerial vehicle cooperative task allocation is obtained.
2. The unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm according to claim 1, wherein the step S1 is specifically as follows:
s11: the problem is specifically described as: n different types of unmanned aerial vehicles distributed in a plurality of machine nests detect and supplement materials for m devices distributed at different positions in a scene, wherein each device needs to be supplemented with different amounts and different types of materials;
s12: a global equipment set T is set, wherein each equipment needs a certain amount of food and material assistance, the corresponding task set K= {1,2}, and an unmanned aerial vehicle set U is set, wherein each unmanned aerial vehicle can carry different amounts of food and material to complete the inspection task, and for each equipment, the material assistance can be completed by different unmanned aerial vehicles;
s13: defining an inspection relation non-decision variable between the unmanned aerial vehicle and the equipment;
s14: and taking the average waiting time and the range distance as evaluation indexes of the task allocation scheme, and adopting a penalty function method to treat various constraints existing in the problem.
3. The unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm according to claim 1, wherein the step S2 is specifically as follows:
s21: generating a high-quality elite population by adopting a genetic learning strategy to replace a historical optimal position to update particle positions, wherein the method specifically comprises three operations of crossing, mutation and selection;
s22: in the iterative evolution process, if the fitness value of elite particles generated by continuous sg iterations is unchanged, the evolution of the particles is considered to be stopped, and at the moment, for each dimension of elite particles, the better particles in two pbest with better fitness values are selected for modification through a tournament selection mechanism, and the variation occurs with small probability;
s23: guiding particle speed and position update through elite population;
s24: and solving the multi-unmanned aerial vehicle cooperative task allocation by adopting an AGLPSO algorithm.
4. The unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm according to claim 3, wherein the step S21 is specifically as follows:
firstly, comparing historical optimal fitness values of current particles i and random particles k based on a ring topology structure to determine a value corresponding to a certain dimension in a crossed individual, and further realizing crossed operation;
then, judging whether each dimension of the crossed individuals is smaller than a given probability value, if so, resetting the value corresponding to the dimension as a random value in a search space, and further realizing mutation operation;
and finally, comparing the candidate particles generated by variation with the elite particles corresponding to the previous iteration according to a greedy method, and selecting a better individual as the elite particles of the current iteration to realize the selection operation.
5. The unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm according to claim 3, wherein the step S23 is specifically as follows:
an adaptive updating strategy is adopted to change the value of omega, the convergence speed of an algorithm is flexibly controlled according to the evolution speed of a particle swarm and the aggregation degree of particles, an evolution speed threshold value is set, and when the evolution speed of the particle swarm is greater than the evolution speed threshold value, the algorithm should keep global optimization in a large range; when the evolution speed is smaller than the threshold value of the evolution speed, the algorithm should perform a local search in a small range to find the optimal solution more quickly; the degree of aggregation of particles depends on the proximity of the fitness value of the particles in the population.
6. The unmanned aerial vehicle task allocation method based on the adaptive genetic learning particle swarm algorithm according to claim 3, wherein the step S24 is specifically as follows:
firstly, inputting initial information of an unmanned plane and survivors according to a patrol scene, setting the maximum iteration number Gmax of an algorithm, the number N of particles, initial values and final values of an inertia weight reference value, the upper limit and the lower limit of a learning factor, a mutation rate, the maximum stagnation algebra and a scale factor; initializing the iteration times G=0 and initializing the particle population;
secondly, calculating individual fitness of the particles, and updating historical optimal positions and global optimal positions of the particles; then, generating elite particles by adopting a genetic learning strategy, and updating the global optimal position;
then, updating stagnant elite particles by adopting an elite learning strategy and calculating inertia weight and learning factors; updating particle speed and position, calculating individual fitness of particles, and updating historical optimal position and global optimal position of particles;
and finally, the obtained global optimal position is the optimal solution, and the optimal solution is decoded to obtain an optimal inspection scheme marked by a calculation rule.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310070727.6A CN115994666A (en) | 2023-01-13 | 2023-01-13 | Unmanned aerial vehicle task allocation method based on adaptive genetic learning particle swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310070727.6A CN115994666A (en) | 2023-01-13 | 2023-01-13 | Unmanned aerial vehicle task allocation method based on adaptive genetic learning particle swarm algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115994666A true CN115994666A (en) | 2023-04-21 |
Family
ID=85990072
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310070727.6A Pending CN115994666A (en) | 2023-01-13 | 2023-01-13 | Unmanned aerial vehicle task allocation method based on adaptive genetic learning particle swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115994666A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116245346A (en) * | 2023-05-12 | 2023-06-09 | 深圳大学 | Multi-unmanned aerial vehicle task allocation method based on multiple swarm genetic algorithms and local search |
| CN117631615A (en) * | 2023-10-12 | 2024-03-01 | 中国电建集团山东电力管道工程有限公司 | Production workshop data acquisition and processing method and system based on Internet of things equipment |
-
2023
- 2023-01-13 CN CN202310070727.6A patent/CN115994666A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116245346A (en) * | 2023-05-12 | 2023-06-09 | 深圳大学 | Multi-unmanned aerial vehicle task allocation method based on multiple swarm genetic algorithms and local search |
| CN116245346B (en) * | 2023-05-12 | 2023-07-28 | 深圳大学 | Multi-unmanned aerial vehicle task allocation method based on multiple swarm genetic algorithms and local search |
| CN117631615A (en) * | 2023-10-12 | 2024-03-01 | 中国电建集团山东电力管道工程有限公司 | Production workshop data acquisition and processing method and system based on Internet of things equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115994666A (en) | Unmanned aerial vehicle task allocation method based on adaptive genetic learning particle swarm algorithm | |
| Luo et al. | Short-term traffic flow prediction based on least square support vector machine with hybrid optimization algorithm | |
| CN108399451A (en) | A kind of Hybrid Particle Swarm Optimization of combination genetic algorithm | |
| Zhao et al. | Evolved neural network ensemble by multiple heterogeneous swarm intelligence | |
| CN103164742A (en) | Server performance prediction method based on particle swarm optimization nerve network | |
| CN109102124B (en) | Dynamic multi-target multi-path induction method and system based on decomposition and storage medium | |
| CN113435025B (en) | Robot high-performance track automatic generation method combined with multistage optimization model | |
| CN112926825B (en) | Multi-unmanned aerial vehicle task allocation method based on multi-target quantum krill group mechanism | |
| CN113205172A (en) | Multitask evolution algorithm based on self-adaptive knowledge migration | |
| CN115146866A (en) | Multi-equivalent optimal path planning method considering actual multi-constraints | |
| CN114580763A (en) | Power load prediction method based on improved dragonfly and lightweight gradient lifting tree model | |
| CN117519244B (en) | Unmanned plane cluster collaborative detection multi-target path planning method and system | |
| CN118195471A (en) | Optimal management method and system for transferring and moving robot transportation tasks | |
| Deng et al. | Solving the Food-Energy-Water Nexus Problem via Intelligent Optimization Algorithms | |
| CN117474036A (en) | Advancing method based on fusion strategy intelligent algorithm | |
| Bi et al. | Self-adaptive Teaching-learning-based Optimizer with Improved RBF and Sparse Autoencoder for Complex Optimization Problems | |
| CN116523158A (en) | Multi-unmanned aerial vehicle track planning method, device, equipment and storage medium | |
| CN114650321A (en) | Task scheduling method for edge computing and edge computing terminal | |
| Mutingi | System reliability optimization: A fuzzy multi-objective genetic algorithm approach | |
| Li et al. | A knowledge-based initialization technique of genetic algorithm for the travelling salesman problem | |
| CN116405498A (en) | Container scheduling method and system based on entropy weight method and multi-strategy particle swarm algorithm | |
| Ebadi et al. | Resource Allocation in The Cloud Environment with Supervised Machine learning for Effective Data Transmission | |
| CN115879361A (en) | Intelligent determination method for operation, maintenance, overhaul and debugging cost of primary equipment | |
| Ye et al. | A self-organizing assisted multi-task algorithm for constrained multi-objective optimization problems | |
| Salim et al. | Discrete bee algorithms and their application in multivariable function optimization |
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 |