CN113858200B - Group robot control method for improving multi-universe inspired by foraging behavior of slime mold - Google Patents
Group robot control method for improving multi-universe inspired by foraging behavior of slime mold Download PDFInfo
- Publication number
- CN113858200B CN113858200B CN202111151161.7A CN202111151161A CN113858200B CN 113858200 B CN113858200 B CN 113858200B CN 202111151161 A CN202111151161 A CN 202111151161A CN 113858200 B CN113858200 B CN 113858200B
- Authority
- CN
- China
- Prior art keywords
- universe
- robot
- position vector
- representing
- robots
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 230000019637 foraging behavior Effects 0.000 title claims abstract description 27
- 239000013598 vector Substances 0.000 claims abstract description 127
- 238000005457 optimization Methods 0.000 claims abstract description 87
- 238000012546 transfer Methods 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 10
- 230000003247 decreasing effect Effects 0.000 claims description 5
- 230000007246 mechanism Effects 0.000 claims description 4
- 230000033001 locomotion Effects 0.000 abstract description 11
- 241001467460 Myxogastria Species 0.000 abstract description 6
- JJWKPURADFRFRB-UHFFFAOYSA-N carbonyl sulfide Chemical compound O=C=S JJWKPURADFRFRB-UHFFFAOYSA-N 0.000 description 26
- 230000006870 function Effects 0.000 description 25
- 238000001135 Friedman test Methods 0.000 description 4
- 238000001793 Wilcoxon signed-rank test Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 241000254158 Lampyridae Species 0.000 description 2
- 230000000052 comparative effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000251468 Actinopterygii Species 0.000 description 1
- 241000271566 Aves Species 0.000 description 1
- 244000025254 Cannabis sativa Species 0.000 description 1
- 235000012766 Cannabis sativa ssp. sativa var. sativa Nutrition 0.000 description 1
- 235000012765 Cannabis sativa ssp. sativa var. spontanea Nutrition 0.000 description 1
- 241000255581 Drosophila <fruit fly, genus> Species 0.000 description 1
- 241000238631 Hexapoda Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 241000251555 Tunicata Species 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 235000009120 camo Nutrition 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 235000005607 chanvre indien Nutrition 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 230000002431 foraging effect Effects 0.000 description 1
- 239000011487 hemp Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Mechanical Engineering (AREA)
- Robotics (AREA)
- Artificial Intelligence (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Fuzzy Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Feedback Control In General (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a method for controlling a swarm robot for improving a multi-universe inspired by foraging behaviors of slime mold, which comprises the following steps: acquiring initial positions of all robots in a group of robots, and determining a multi-element universe population of the robots; determining a primary updating position vector according to the multi-universe population and the optimization parameters of the robot based on the multi-universe algorithm; determining a secondary update position vector according to the primary update position vector of the robot in each universe based on the slime foraging behavior model of the slime mold; and when the secondary update position vector of the robot meets the preset condition, determining the output position vector of the robot according to the secondary update position vector. The multi-universe optimization algorithm is improved in a mode of the myxomycete foraging behavior, after the movement is completed according to the movement rule of the multi-universe optimization algorithm, the optimal solution is further searched between the local optimal solution and the global optimal solution according to the myxomycete foraging behavior, and the convergence speed is ensured to be high, so that the control efficiency of the swarm robots is improved.
Description
Technical Field
The invention relates to the technical field of robot control, in particular to a swarm robot control method for improving a multi-universe inspired by slime foraging behavior.
Background
In recent years, as optimization problems have become more complex in real life, it has become increasingly difficult to solve such problems using traditional gradient-based methods. Therefore, the requirements for stability and reliability of the algorithm for solving the problem are also increasing. Therefore, swarm intelligence algorithms, which are algorithms that simulate the group behavior of insects, animals, birds, and fish, are widely used to solve such problems. Due to the simplicity, high efficiency and low computational complexity of swarm intelligence algorithms, many researchers have proposed many advanced swarm intelligence algorithms, such as continuous ant colony optimization algorithm, bat optimization algorithm, differential evolution algorithm, firefly optimization algorithm, moth optimization algorithm, particle swarm optimization algorithm, sine and cosine optimization algorithm, goblet sea squirt swarm optimization algorithm, drosophila optimization algorithm, and the like. In addition, a group intelligent algorithm, which is called a multi-element universe optimization algorithm, has strong capability of solving the optimal scheme. Therefore, many researchers have proposed and applied many improved versions of the multivariate cosmic optimization algorithm to various fields.
In the prior art, in the technical field of group robot (including unmanned aerial vehicles and the like) control, due to the limited searching capability and the slow convergence speed of a multi-universe optimization algorithm, the solution is easy to fall into a local minimum value, and the group robot control efficiency is low.
Accordingly, there is a need for improvements and developments in the art.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a swarm robot control method for improving the multivariate universe inspired by the foraging behavior of slime mold aiming at the defects in the prior art, and the method is used for solving the problem of low control efficiency of the swarm robot in the prior art.
The technical scheme adopted by the invention for solving the technical problem is as follows:
a method for controlling a swarm robot for improving a multi-universe inspired by myxomycete foraging behaviors comprises the following steps:
acquiring initial positions of all robots in group robots, and determining a multi-universe group of the robots according to the initial positions of the robots; wherein the multi-universe population includes a plurality of universes;
determining a primary updating position vector of the robot in each universe according to the multi-universe population and the optimization parameters of the robot based on a multi-universe algorithm;
determining a secondary update position vector of the robot in each universe according to the primary update position vector of the robot in each universe on the basis of the slime foraging behavior model;
and when the secondary updating position vector of the robot meets a preset condition, determining the output position vector of the robot according to the secondary updating position vector of the robot in each universe.
The control method, wherein the control method further comprises:
and when the secondary updating position vector of the robot does not meet the preset condition, forming a multi-universe population of the robot through the secondary updating position vectors of the robot in all universes, and continuously executing a step of determining the primary updating position vector of the robot in each universe according to the multi-universe population and the optimized parameters of the robot based on a multi-universe algorithm until the secondary updating position vector of the robot meets the preset condition.
The control method described above, wherein the universe includes a position vector of each robot in the group of robots, and the optimization parameters include: probability of existence of wormholes and travel distance values; the probability of the wormholes is determined according to the current iteration times and the maximum iteration times, and the travel distance value is determined according to the current iteration times, the maximum iteration times and the utilization degree;
the method for determining the one-time updating position vector of the robot in each universe according to the multi-universe population and the optimization parameters of the robot based on the multi-universe algorithm comprises the following steps:
determining the fitness value of each universe in the multi-universe population;
determining an optimal universe according to the fitness value of each universe;
obtaining transfer position vectors of the robots in each universe through white hole/black hole orbital transfer according to the position vectors of the robots in each universe;
and obtaining a primary updating position vector of the robot in each universe according to the initial position of the robot in each universe, the transfer position vector of the robot, the probability of the existence of the wormholes, the travel distance value and the optimal universe.
The control method described above, wherein the transfer position vector of the robot is:
wherein,
a transfer position vector of the jth robot representing the ith universe,
position vector, r, of jth robot representing ith universe 1 Represents [0,1]Random number in range, NI (U) i ) Represents the normalized fitness value of the ith universe,
a position vector representing the jth robot of the kth universe selected by the roulette mechanism.
The control method described above, wherein the once-updated position vector of the robot is:
wherein,
one-time update position vector, x, of jth robot representing ith universe j Represents the position vector of the j-th robot in the optimal universe, TDR represents the travel distance value, ub j To represent
Upper limit of (lb) j To represent
WEP represents the probability of wormholes, r 2 ，r 3 ，r 4 All represent [0,1]Random number in range, WEP min Minimum probability of wormhole presence, WEP max The maximum probability of the existence of the wormholes is represented, T represents the current iteration number, T represents the maximum iteration number, and p represents the utilization degree.
The control method described above, wherein the secondary update position vector of the robot is:
q＝tanh|S(i)-DF|
wherein,
the quadratic update position vectors, rand, r, of the jth robot representing the ith universe all represent [0,1 ]]Random number in the range, LB
UB represents the lower limit of
Z represents a set of integers,
the position vector of the corresponding robot in the universe representing the highest fitness value in the past iteration,
representing a linear decreasing value from 0 to 1,
the weight of the ith universe is represented,
respectively representing the position vectors of two robots randomly chosen from the population of robots,
representing a linear decreasing value from 1 to 0, q representing an intermediate variable, S (i) representing the fitness value of the ith universe, DF representing the highest fitness value of the universe in the previous iteration, a representing a parameter, bF representing the highest fitness value of the universe in the current iteration, and wF representing the lowest fitness value of the universe in the current iteration.
The control method, wherein the preset condition includes: the current iteration times reach the maximum iteration times, or the iteration time reaches the termination time.
The control method, wherein when the secondary updated position vector of the robot satisfies a preset condition, determining an output position vector of the robot according to the secondary updated position vector of the robot in each universe, includes:
when the secondary updating position vector of the robot meets a preset condition, determining the fitness value of each universe;
and taking the secondary updating position vector of the robot in the universe with the maximum fitness value as the output position vector of the robot.
A computer device comprising a memory storing a computer program and a processor, wherein the processor implements the steps of any of the methods described above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored, wherein the computer program, when being executed by a processor, carries out the steps of the method of any of the above.
Has the advantages that: the movement modes of the white hole and the black hole in the multi-universe optimization algorithm are improved by adopting the mode of the myxomycete foraging behavior, after the white hole and the black hole finish own movement according to the movement rule of the multi-universe optimization algorithm, the optimal solution is further searched between the local optimal solution and the global optimal solution according to the myxomycete foraging behavior, and the convergence speed is ensured to be high, so that the control efficiency of the swarm robots is improved.
Drawings
Fig. 1 is a first flowchart of a method for controlling a swarm robot for improving multivariate universe inspired by foraging behavior of slime mold.
Fig. 2 is a second flowchart of the swarm robot control method for improving the multi-universe inspired by the foraging behavior of slime mold.
Fig. 3 is a diagram showing the results of friedman's test of the improved multivariate cosmic optimization algorithm of the invention and some similar algorithms.
FIG. 4 is a convergence curve of the improved multivariate cosmic optimization algorithm of the invention with some of the same algorithms.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring also to fig. 1-4, the present invention provides some embodiments of a swarm robot control method for improving the multi-universe inspired by slime foraging behavior.
In order to solve the problem of low robot control efficiency caused by limited search capacity and slow convergence speed, the invention provides an improved multivariate universe optimization algorithm. The algorithm mainly adopts a mode of the slime foraging behavior of the slime to improve the motion modes of the white hole and the black hole in the multi-universe optimization algorithm, wherein the slime foraging behavior mainly acts on the white hole and the black hole to complete own motion according to the motion rules of the multi-universe optimization algorithm, then an optimal solution is further searched between local optimal and global optimal according to the slime foraging behavior, and the convergence speed is ensured to be high, so that the control efficiency of the swarm robots is improved.
As shown in fig. 1 and fig. 2, a method for controlling a swarm robot for improving a multi-universe inspired by myxomycete foraging behavior according to an embodiment of the present invention includes the following steps:
s100, acquiring initial positions of all robots in group robots, and determining a multi-element universe population of the robots according to the initial positions of the robots; wherein the multi-universe population includes a plurality of universes.
Specifically, in the multivariate cosmic optimization algorithm, one universe is considered as a solution to the optimization problem, each object in a single universe is considered as a position vector of the robot of the corresponding solution, and the expansion rate of the universe is proportional to the degree of adaptation of the corresponding solution. The initialization of the solution is used to start the optimization process, with each set of updates to the solution being performed by a progressive iteration rule. The random initialization U of the multi-cosmic population is defined as equation (1).
Where n represents the number of universes, corresponding to the universe population size. d represents the number of robots, corresponding to the dimensions of the search space.
The (i) th universe is represented,
a position vector of the jth robot representing the ith universe. Position vector of jth robot in ith universe
Determined according to the initial position of the jth robot, specifically, the moving directions are multiple because the robot moves from the initial position, and the initial position of the robot is determined according to the moving directionsAnd obtaining a position vector of the robot from the initial position and the moving direction. The moving directions of the same robot in different universes can be the same or different.
And S200, determining a primary updating position vector of the robot in each universe according to the multi-universe population and the optimized parameters of the robot based on the multi-universe algorithm.
Specifically, the universe includes a position vector of each of the group of robots. Based on the multivariate universe algorithm, determining a primary update position vector of each robot in each universe population according to the multivariate universe population and the optimization parameters, namely, based on the multivariate universe algorithm, optimizing the multivariate universe population to obtain an optimized multivariate universe population, namely, updating the position vector of each robot in the multivariate universe population for the first time to obtain a primary update position vector.
Specifically, the optimization parameters refer to parameters used in optimization of the multivariate cosmic algorithm, and the optimization parameters include: probability of wormhole existence and travel distance value; the probability of the wormholes is determined according to the current iteration times and the maximum iteration times, and the travel distance value is determined according to the current iteration times, the maximum iteration times and the utilization degree. The probability of the existence of the wormholes refers to the probability of the existence of the wormholes in the multi-element universe population, and the travel distance value refers to the step length of the movement of the position vector of the robot towards the optimal universe.
Step S200 specifically includes:
and step S210, determining the fitness value of each universe in the multi-universe population.
And S220, determining the optimal universe according to the adaptability value of each universe.
Specifically, for each universe, the fitness value of the universe is determined, and the universe with the highest fitness value in the universe population is used as the optimal universe.
And step S230, obtaining the transfer position vector of the robot in each universe through white hole/black hole orbital transfer according to the position vector of the robot in each universe.
In particular, since each individual universe has a different fitness value (inflation rate), the update process follows a roulette wheel mechanism, with the position vectors of the robots in a single universe being transferred through the white/black hole orbits.
The transfer position vector of the robot is as follows:
wherein,
a transfer position vector of the jth robot representing the ith universe,
position vector, r, of jth robot representing ith universe 1 Represents [0,1]Random number in range, NI (U) i ) Represents the normalized fitness value of the ith universe,
a position vector representing the jth robot of the kth universe selected by the roulette mechanism.
And S240, obtaining a primary updating position vector of the robot in each universe according to the initial position of the robot in each universe, the transfer position vector of the robot, the probability of the existence of the wormholes, the travel distance value and the optimal universe.
Specifically, when the magnitude of the expansion ratio is not considered, the self-adaptive value (expansion ratio) is increased to realize local change, and the position vector of the internal robot excited by the universe moves to the current optimal universe.
The once updated position vector of the robot is as follows:
wherein,
one-time update position vector, x, of jth robot representing ith universe j Representing the position vector of the jth robot in the optimal universe, TDR representing the travel distance value, ub j To represent
Upper limit of (lb) j To represent
WEP represents the probability of wormholes, r 2 ，r 3 ，r 4 All represent [0,1]Random number in range, WEP min Minimum probability of wormhole presence, WEP min Can be set as desired, e.g., WEP min ＝0.2，WEP max Indicating the maximum probability of the presence of wormholes, WEP max Can be set as desired, e.g., WEP max =1,t denotes the current number of iterations, T denotes the maximum number of iterations, p denotes the degree of utilization, and p can be set as needed, for example, p =6.
And S300, determining secondary updating position vectors of the robots in each universe according to the primary updating position vectors of the robots in each universe based on the slime foraging behavior model of the slime mold.
Specifically, after the primary optimized multi-element universe population is obtained, the primary optimized multi-element universe population is further optimized based on the slime foraging behavior model to obtain a secondary optimized multi-element universe population, namely, the primary updated position vectors of all robots in the primary optimized multi-element universe population are updated for the second time to obtain a secondary updated position vector.
The secondary update position vector of the robot is as follows:
q＝tanh|S(i)-DF|
wherein,
the quadratic update position vectors, rand, r, of the jth robot representing the ith universe all represent [0,1 ]]Random number in the range, LB
Lower limit of (1), UB represents
Z represents a set of integers,
the position vector of the corresponding robot in the universe representing the highest fitness value in the past iteration,
representing a linear decreasing value from 0 to 1,
represents the ithThe weight of the universe is such that,
respectively representing the position vectors of two robots randomly selected from a group of robots,
representing a linear decreasing value from 1 to 0, q representing an intermediate variable, S (i) representing the fitness value of the ith universe, DF representing the highest fitness value of the universe in the previous iteration, a representing a parameter, bF representing the highest fitness value of the universe in the current iteration, and wF representing the lowest fitness value of the universe in the current iteration.
And S400, when the secondary updating position vector of the robot meets a preset condition, determining the output position vector of the robot according to the secondary updating position vector of the robot in each universe.
Specifically, the preset conditions include: and when the current iteration times reach the maximum iteration times or the iteration time reaches the end time, the current iteration times of the secondary updating position vector of the robot reach the maximum iteration times or the iteration time of the secondary updating position vector of the robot reaches the end time, determining the output position vector of the robot according to the secondary updating position vector of the robot in each universe. And then, after the secondary updating position vector of the robot meets the preset condition, determining the optimal universe from the universe population, and taking the secondary updating position vector of the robot in the optimal universe as the output vector of the robot.
Step S400 specifically includes:
and S410, when the secondary updating position vector of the robot meets a preset condition, determining the fitness value of each universe.
And step S420, taking the secondary updating position vector of the robot in the universe with the maximum fitness value as an output position vector of the robot.
And when the secondary update position vector of the robot meets the preset condition, determining the fitness value of the universe, taking the universe with the maximum fitness value in the universe population as the optimal universe, and taking the secondary update position vector of the robot in the optimal universe as the output position vector of the robot so as to control the movement of the robot.
And S500, when the secondary updated position vectors of the robots do not meet preset conditions, forming multiple universe populations of the robots through the secondary updated position vectors of the robots in all universes, and continuously executing a step of determining the primary updated position vectors of the robots in all universes according to the multiple universe populations and the optimized parameters of the robots based on a multiple universe algorithm until the secondary updated position vectors of the robots meet the preset conditions.
Specifically, when the secondary update position vector of the robot does not meet the preset condition, iteration needs to be continued, and the position vector of the robot is continuously updated.
In the improved multi-universe optimization algorithm, firstly, all robots finish the first updating of position vectors mainly by means of the movement modes of white holes, black holes and wormholes of the multi-universe optimization algorithm; after the position vector is updated for the first time, in order to improve the capability of the robot for avoiding falling into the local optimum, the robot searches for a better position by simulating the foraging mode of the slime mold, so that a better fitness value is obtained.
In order to verify the effectiveness of the method, the method and other algorithms are tested on a plurality of reference functions, and the result shows that the method provided by the invention not only can obtain a better solution, but also has a faster convergence rate, and improves the control efficiency of the robot.
When the benchmark functions are used for proving the performance of the improved multivariate cosmic optimization algorithm, 12 benchmark functions are selected as test functions, and detailed information of F1-F12 is shown in Table 1. In order to ensure the relative fairness of the experiments. All algorithms performed comparative experiments under the same conditions. The robot population size is set to 30, the maximum evaluation number is uniformly set to 30 ten thousand times, and all algorithms are tested on the basis function for 30 times to reduce the influence of random conditions. For the comparison results, the mean AVG, variance STD, wilcoxon signed rank test and friedman test were used for analysis.
The improved multi-element universe optimization algorithm is compared with a plurality of similar algorithms, so that the core advantages of the improved multi-element universe optimization algorithm are further proved. In the experiment, the basic algorithms involved in the comparison include MVO (Multi-wise Optimizer, multivariate universe Optimization Algorithm), ACOR (Ant Colony Optimization For Continuous domain Domains), BA (Bat Optimization, bat Algorithm), DE (Differential Evolution Algorithm), FA (Firefly Algorithm), MFO (move-Flame Optimization, moth Flame Optimization Algorithm), PSO (Particle Swarm Optimization, particle Swarm Algorithm), SCA (Sine Algorithm, sine Optimization Algorithm), SSA (spark Search Optimization, hemp Search Algorithm), FOA (free Fly Optimization Algorithm), in which the values of the relevant parameters are all the parameter values set by the Algorithm itself. The results of comparing the improved multivariate cosmic optimization algorithm with 10 homogeneous algorithms are shown in tables 2 and 3, wherein the results include the analysis results of mean and variance, AVG represents the mean obtained by the algorithm on the basis function, and STD represents the variance obtained by the algorithm. Further, tables 2 and 3 also include the related results obtained by Wilcoxon signed Rank test, that is, ' + ' indicates the number of functions that performed the improved multivariate cosmic optimization algorithm better than the other methods in the test functions 30, ' - ' indicates the number of functions that performed the improved multivariate cosmic optimization algorithm worse than the other methods, ' = ' indicates that the improved multivariate cosmic optimization algorithm and the other methods performed equally, mean ' indicates the comprehensive ranking average obtained by each method, and ' Rank ' is the final ranking result obtained by each algorithm according to Mean.
Regarding the obtained function optimal values, the improved multi-element universe optimization algorithm obtains optimal values on 9 function problems, namely the minimum mean value, and the DE obtains optimal values on 2 function problems, which shows that the improved multi-element universe optimization algorithm has better processing capability than the DE when processing different optimization problems, and simultaneously shows that the improved multi-element universe optimization algorithm has greater advantages when compared with the similar algorithm on a reference function. The result that the improved multi-element universe optimization algorithm obtains the minimum value on 8 function problems, the DE obtains the minimum value on 3 function problems, and the FOA obtains the minimum value on 1 function problem effectively proves that the improved multi-element universe optimization algorithm has better stability than the DE and the FOA, and further shows that the improved multi-element universe optimization algorithm has better stability than other similar methods. In view of the mean value and the variance obtained by the improved multivariate cosmic optimization algorithm on the 13 reference functions, the improved multivariate cosmic optimization algorithm is not only a group intelligent optimization algorithm with strong capability of obtaining an optimal solution in a solution space, but also has high stability in the process of obtaining the optimal solution.
Further, with respect to the results of the Wilcoxon signed rank test in tables 2 and 3, the improved multivariate cosmic optimization algorithm obtained No.1 with the ranked integrated mean of 1.33 and DE obtained No.2 with the ranked integrated mean of 2.83. In a comparison of the improved multivariate cosmic optimization algorithm with other similar methods, it can also be seen that the improved multivariate cosmic optimization algorithm performs better on 8 function problems than on DE ranked No.2, 3 function problems are worse than on DE,1 function problem performs equally. In addition, the improved multi-element universe optimization algorithm is superior to the SSA in 11 function problems and inferior to the SSA in 1 function problem. It is also worth mentioning that the improved multivariate cosmic optimization algorithm performs better than the same kind of algorithms except for DE and SSA over 12 reference functions. The results of using friedman test are presented in fig. 3 to further evaluate and demonstrate the performance of the improved multivariate cosmic optimization algorithm, where 'friedman ranking' means ranking results obtained according to friedman test. After analysis, the improved multi-element universe optimization algorithm and 9 basic methods thereof have the advantages that the improved multi-element universe optimization algorithm is ranked with the value of 1.41 obtained and the DE is ranked with the value of 2.89 obtained and the value is ranked with the value of 2.1, and the improved multi-element universe optimization algorithm and the like are compared. Therefore, the improved multivariate universe optimization algorithm is an excellent group intelligent optimization algorithm and is further proved to be sufficient by combining the Wilcoxon signed rank test result and the Friedman test result.
In fig. 4, it shows the convergence curves of the improved multivariate cosmic optimization algorithm and its 10 homogeneous methods at F1, F2, F3, F5, F7, F9, where 'bestfittness' means that each method obtains the optimal fitness value during the evaluation process. The convergence curves of the given 6 reference functions show that the convergence accuracy of the improved multivariate cosmic optimization algorithm is better than that of other algorithms. Further, through analysis of the convergence curves of F1, F2 and F3, the improved multivariate cosmic optimization algorithm is enhanced in convergence speed to a certain extent; through analysis of the convergence curves of F5, F7 and F9, the result shows that the improved multivariate cosmic optimization algorithm has stronger capability of avoiding falling into local optimization than other algorithms. Therefore, by combining the analysis of all given convergence curves, it can be obviously found that the core advantages of the improved multivariate cosmic optimization algorithm are further fully proved when the improved multivariate cosmic optimization algorithm is compared with other similar algorithms.
In conclusion, after comparison with similar algorithms, the improved multivariate cosmic optimization algorithm is found to be a group intelligent optimization algorithm which can obtain a high-quality solution, has stronger stability, faster convergence speed, higher convergence precision and stronger capability of avoiding falling into local optimum.
TABLE 1 reference function description
TABLE 2 comparison of the improved multivariate cosmic optimization algorithm with 10 homogeneous algorithms
TABLE 3 (TABLE 2) comparative results of the improved multivariate cosmic optimization algorithm and 10 homogeneous algorithms
Based on the method for controlling the population robot for improving the multivariate universe inspired by the foraging behavior of the slime mold in any embodiment, the invention also provides a preferred embodiment of computer equipment, which comprises the following steps:
the computer device comprises a memory and a processor, the memory storing a computer program, the processor implementing the steps when executing the computer program:
acquiring initial positions of all robots in group robots, and determining a multi-universe group of the robots according to the initial positions of the robots;
determining a primary updating position of the robot in each universe according to the multi-universe population and the optimization parameters of the robot based on a multi-universe algorithm;
determining a secondary updating position of the robot in each universe according to the primary updating position of the robot in each universe population based on the slime foraging behavior model of the slime mold;
and when the secondary update position of the robot meets a preset condition, determining the output position of the robot according to the secondary update position of the robot in each universe.
Based on the method for controlling the population robot for improving the multivariate universe inspired by the foraging behavior of the slime mold in any embodiment, the invention also provides a preferred embodiment of a computer-readable storage medium:
a computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:
acquiring initial positions of all robots in group robots, and determining a multi-universe group of the robots according to the initial positions of the robots;
determining a primary updating position of the robot in each universe according to the multi-universe population and the optimization parameters of the robot based on a multi-universe algorithm;
determining a secondary updating position of the robot in each universe according to the primary updating position of the robot in each universe population based on the slime foraging behavior model of the slime mold;
and when the secondary update position of the robot meets a preset condition, determining the output position of the robot according to the secondary update position of the robot in each universe.
It will be understood that the invention is not limited to the examples described above, but that modifications and variations will occur to those skilled in the art in light of the above teachings, and that all such modifications and variations are considered to be within the scope of the invention as defined by the appended claims.
Claims (10)
1. A method for controlling a colony robot for improving a multi-universe inspired by the foraging behavior of slime mold is characterized by comprising the following steps:
acquiring initial positions of all robots in group robots, and determining a multi-universe group of the robots according to the initial positions of the robots; wherein the multi-universe population includes a plurality of universes;
determining a primary updating position vector of the robot in each universe according to the multi-universe population and the optimization parameters of the robot based on a multi-universe algorithm;
determining a secondary update position vector of the robot in each universe according to the primary update position vector of the robot in each universe based on the slime foraging behavior model of the slime mold;
and when the secondary updating position vector of the robot meets a preset condition, determining the output position vector of the robot according to the secondary updating position vector of the robot in each universe.
2. The control method according to claim 1, characterized by further comprising:
and when the secondary updating position vector of the robot does not meet the preset condition, forming a multi-universe population of the robot through the secondary updating position vectors of the robot in all universes, and continuously executing a step of determining the primary updating position vector of the robot in each universe according to the multi-universe population and the optimized parameters of the robot based on a multi-universe algorithm until the secondary updating position vector of the robot meets the preset condition.
3. The method of claim 1, wherein the universe includes a position vector for each of the plurality of robots, and wherein the optimization parameters include: probability of existence of wormholes and travel distance values; the probability of the wormholes is determined according to the current iteration times and the maximum iteration times, and the travel distance value is determined according to the current iteration times, the maximum iteration times and the utilization degree;
the method for determining the one-time updating position vector of the robot in each universe according to the multi-universe population and the optimization parameters of the robot based on the multi-universe algorithm comprises the following steps:
determining the fitness value of each universe in the multi-universe population;
determining an optimal universe according to the fitness value of each universe;
obtaining transfer position vectors of the robots in each universe through white hole/black hole orbital transfer according to the position vectors of the robots in each universe;
and obtaining a primary updating position vector of the robot in each universe according to the initial position of the robot in each universe, the transfer position vector of the robot, the probability of the existence of the wormholes, the travel distance value and the optimal universe.
4. The control method according to claim 3, wherein the transfer position vector of the robot is:
wherein,
a transfer position vector of the jth robot representing the ith universe,
position vector, r, of jth robot representing ith universe 1 Represents [0,1]Random number in range, NI (U) i ) Represents the normalized fitness value of the ith universe,
a position vector representing the jth robot of the kth universe selected by the roulette mechanism.
5. The control method according to claim 4, wherein the once-updated position vector of the robot is:
wherein,
one-time update position vector, x, of jth robot representing ith universe j Representing the position vector of the jth robot in the optimal universe, TDR representing the travel distance value, ub j To represent
Upper limit of (lb) j To represent
WEP represents the probability of wormholes, r 2 ,r 3 ,r 4 All represent [0,1]Random number in range, WEP min Minimum probability of wormhole presence, WEP max The maximum probability of the existence of the wormholes is represented, T represents the current iteration number, T represents the maximum iteration number, and p represents the utilization degree.
6. The control method of claim 5, wherein the second updated position vector of the robot is:
q＝tanh|S(i)-DF|
wherein,
the quadratic update position vectors, rand, r, of the jth robot representing the ith universe all represent [0,1 ]]Random numbers within the range, LB denotes
UB represents the lower limit of
Z represents a set of integers,
the position vector of the corresponding robot in the universe representing the highest fitness value in the past iteration,
the weight of the ith universe is represented,
respectively representing the position vectors of two robots randomly selected from a group of robots,
representing a linear decreasing value from 1 to 0, q representing an intermediate variable, S (i) representing the fitness value of the ith universe, DF representing the highest fitness value of the universe in the previous iteration, a representing a parameter, bF representing the highest fitness value of the universe in the current iteration, and wF representing the lowest fitness value of the universe in the current iteration.
7. The control method according to any one of claims 1 to 6, characterized in that the preset conditions include: the current iteration number reaches the maximum iteration number, or the iteration time reaches the termination time.
8. The method according to any one of claims 1 to 6, wherein determining the output position vector of the robot according to the secondarily updated position vector of the robot in each universe when the secondarily updated position vector of the robot satisfies a preset condition comprises:
when the secondary updating position vector of the robot meets a preset condition, determining the fitness value of each universe;
and taking the secondary updating position vector of the robot in the universe with the maximum fitness value as the output position vector of the robot.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 8 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111151161.7A CN113858200B (en) | 2021-09-29 | 2021-09-29 | Group robot control method for improving multi-universe inspired by foraging behavior of slime mold |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111151161.7A CN113858200B (en) | 2021-09-29 | 2021-09-29 | Group robot control method for improving multi-universe inspired by foraging behavior of slime mold |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113858200A CN113858200A (en) | 2021-12-31 |
| CN113858200B true CN113858200B (en) | 2022-10-28 |
Family
ID=78992631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111151161.7A Active CN113858200B (en) | 2021-09-29 | 2021-09-29 | Group robot control method for improving multi-universe inspired by foraging behavior of slime mold |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113858200B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115543005A (en) * | 2022-09-01 | 2022-12-30 | 三峡大学 | Photovoltaic maximum power tracking control method based on differential evolution slime mold algorithm |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005539296A (en) * | 2002-08-21 | 2005-12-22 | ニール，ソロモン | System, method and apparatus for an organization of self-configurable mobile robot agents within a multi-robot system |
| CN107066705B (en) * | 2017-03-27 | 2020-11-03 | 东莞理工学院 | Group robot search algorithm and verification method based on multi-head folliculus foraging |
| US10850395B2 (en) * | 2017-05-19 | 2020-12-01 | Stc.Unm | System and methods for multiple-place swarm foraging with dynamic depots |
| KR20190141576A (en) * | 2018-06-14 | 2019-12-24 | 삼성전자주식회사 | Apparatus for controlling swarm devices using dynamic rule based blockchain and method thereof |
| CN110288131B (en) * | 2019-06-04 | 2021-10-12 | 三峡大学 | Bionic optimization method for traffic network vulnerability problem based on myxomycete foraging behavior |
| CN111339635B (en) * | 2020-01-17 | 2023-06-30 | 大连大学 | DNA storage coding optimization method of multi-element universe algorithm based on damping factors |
| CN111600303A (en) * | 2020-06-15 | 2020-08-28 | 湘潭大学 | Grid-connected type wind/light/storage micro-grid optimization configuration method based on improved multivariate universe algorithm |
| CN112132280A (en) * | 2020-09-28 | 2020-12-25 | 大连大学 | Method for optimizing BP neural network based on improved bacterial foraging algorithm |
| CN113110520B (en) * | 2021-05-26 | 2022-08-09 | 重庆邮智机器人研究院有限公司 | Robot path planning method based on multiple intelligent optimization parallel algorithms |
-
2021
- 2021-09-29 CN CN202111151161.7A patent/CN113858200B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113858200A (en) | 2021-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pedersen | Good parameters for particle swarm optimization | |
| CN110062390B (en) | Wireless sensor network node optimization deployment method based on improved wolf colony algorithm | |
| CN113858200B (en) | Group robot control method for improving multi-universe inspired by foraging behavior of slime mold | |
| CN105005820B (en) | Target assignment optimizing method based on particle swarm algorithm of population explosion | |
| Muhammed et al. | Improved fitness-dependent optimizer algorithm | |
| WO2023071592A1 (en) | Network structure search method for ultra-large search space, system and medium | |
| CN111404731A (en) | SDN network controller deployment method and system based on double heuristic algorithm | |
| CN107784353A (en) | A kind of function optimization method based on cuckoo searching algorithm | |
| CN111797966A (en) | Multi-machine cooperative global target distribution method based on improved sheep swarm algorithm | |
| CN114995460B (en) | Robot path planning method based on self-adaptive improved ant colony algorithm | |
| CN113467481B (en) | Path planning method based on improved Sarsa algorithm | |
| CN114662638A (en) | Mobile robot path planning method based on improved artificial bee colony algorithm | |
| CN117273057A (en) | Multi-agent collaborative countermeasure decision-making method and device based on reinforcement learning | |
| CN115373400A (en) | Robot path planning method and system based on dynamic update mechanism ant colony algorithm | |
| Wang et al. | Cooperative velocity updating model based particle swarm optimization | |
| CN117523359A (en) | Image comparison and identification method and device based on reinforcement learning | |
| Iwamatsu | Locating all the global minima using multi-species particle swarm optimizer: The inertia weight and the constriction factor variants | |
| CN109902796A (en) | A kind of improved wolf pack algorithm based on fireworks explosion | |
| Schoene et al. | Step-optimized particle swarm optimization | |
| Masuda et al. | Radius parallel self-organizing map (RPSOM) | |
| Wang et al. | Empirical study of hybrid particle swarm optimizers with the simplex method operator | |
| CN117113216B (en) | Heterogeneous multi-unmanned aerial vehicle hit decision method and device based on dynamic Bayesian network | |
| CN114997617B (en) | Multi-unmanned platform multi-target combined detection task allocation method and system | |
| Qu et al. | Variational policy propagation for multi-agent reinforcement learning | |
| Snáel et al. | Particle swarm optimization with protozoic behaviour |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |