CN107329473B - Robot inspection path planning method based on mean value and sound search - Google Patents

Robot inspection path planning method based on mean value and sound search Download PDF

Info

Publication number
CN107329473B
CN107329473B CN201710553903.6A CN201710553903A CN107329473B CN 107329473 B CN107329473 B CN 107329473B CN 201710553903 A CN201710553903 A CN 201710553903A CN 107329473 B CN107329473 B CN 107329473B
Authority
CN
China
Prior art keywords
individual
mean value
fes
robot
harmony
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
Application number
CN201710553903.6A
Other languages
Chinese (zh)
Other versions
CN107329473A (en
Inventor
郭肇禄
李大海
王洋
杨火根
刘小生
余法红
李康顺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangxi University of Technology
Original Assignee
Jiangxi University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangxi University of Technology filed Critical Jiangxi University of Technology
Priority to CN201710553903.6A priority Critical patent/CN107329473B/en
Publication of CN107329473A publication Critical patent/CN107329473A/en
Application granted granted Critical
Publication of CN107329473B publication Critical patent/CN107329473B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
  • Manipulator (AREA)

Abstract

The invention discloses a robot inspection path planning method based on mean value and sound search. The invention adopts mean value and sound search to plan the inspection path of the robot. In the mean value and sound search, firstly, the mean value of all individuals in the harmony library is calculated, then, mean value information is used as a base point for generating a new individual, and the mean value information of the harmony library is fused with the optimal individual in the harmony library to be used as a guide direction of a Gaussian mutation operator, so that the search performance of the algorithm is enhanced, and the routing efficiency of the robot patrol route planning is improved. The invention can improve the routing planning efficiency of the robot inspection path to a certain extent.

Description

Robot inspection path planning method based on mean value and sound search
Technical Field
The invention relates to the field of robot path planning, in particular to a robot inspection path planning method based on mean value and sound search.
Background
The inspection robot plays a crucial role in modern industrial production. The inspection robot is widely applied to the fields of power inspection, warehouse inspection, forest fire prevention inspection and the like. In the design and development of the inspection robot, the inspection path planning of the inspection robot is a supporting technology in the base. In routing inspection path planning of a robot, people often encounter the following routing inspection path planning problems: the coordinate positions of a plurality of inspection points are given, a path is required to be planned for an inspection robot, the inspection robot is enabled to pass through each inspection point once and only once from a source inspection point and then returns to the source inspection point, and the cost of the path passing by the inspection robot is required to be minimized. The routing inspection path planning problem is an NP completeness problem, and when the scale of the problem is large, a routing inspection path which is acceptable in engineering is difficult to plan for the routing inspection robot effectively by a traditional path planning algorithm. For the routing problem, researchers often design an evolutionary algorithm simulating nature to solve. The harmony search algorithm is a newly proposed evolutionary algorithm, and the basic principle of the harmony search algorithm is an optimization algorithm designed by simulating the creation process of musicians.
The routing planning problem of the robot inspection is essentially an optimization problem, and the harmony search algorithm has superior performance in solving the optimization problem. To this end, many scholars have proposed various improved harmony search algorithms to solve various optimization problems, such as that the tomalli and the lynsen use the harmony search algorithm to optimize the training parameters of the support vector machine and use the trained support vector machine to predict the network traffic, and experimental results show that the proposed method can obtain higher network traffic prediction accuracy than some conventional methods (the tomalli, the lynsen harmony search algorithm optimizes the network traffic prediction of the support vector machine [ J ] microcomputer application, 2017,33(01): 67-70); the information gain is fused into the harmony search algorithm to realize the emotion feature selection of voice data, and the experimental result shows that the method can effectively select proper emotion features and can achieve high emotion recognition rate (the information gain and harmony search is fused [ J ] the small-sized microcomputer system, 2017,38(05): 1164-.
It can be known from the existing research results that the harmony search algorithm has the advantages of simple operation operator, strong global search capability and the like, has been successfully applied to various engineering optimization problems, and obtains satisfactory results in solving a plurality of engineering optimization problems, but the traditional harmony search algorithm has the defects of low convergence rate and low planning efficiency when solving some complex robot routing inspection path planning problems.
Disclosure of Invention
The invention aims to provide a robot routing inspection path planning method based on mean value and acoustic search, which can overcome the defects that the convergence speed is low and the planning efficiency is low easily when the traditional acoustic search is used for solving some complex robot routing inspection path planning problems to a certain extent.
The technical scheme of the invention is as follows: a robot inspection path planning method based on mean value and acoustic search comprises the following steps:
step 1, inputting coordinates of each inspection point, and determining the number D of the inspection points;
step 2, initializing a sound library size HMS by a user, and selecting a probability HMCR, a disturbance probability PAR and a maximum evaluation time MAX _ FEs;
step 3, setting the current evolution algebra t as 0 and setting the current evaluation times FEs as 0;
step 4, randomly generating an initial harmony library
Figure BDA0001345205140000021
Wherein the individual subscript i ═ 1, 2., HMS; and is
Figure BDA0001345205140000022
Is a harmony library MtThe ith individual of (1); individuals
Figure BDA0001345205140000023
The sequential weight of the D inspection points is stored;
Figure BDA0001345205140000024
is an individual
Figure BDA0001345205140000025
The sequential weight of the jth routing inspection point in the table, and a dimension subscript j is 1, 2.
Step 5, countingCaculation sound library MtThe fitness value of each individual;
step 6, making the current evaluation times FEs equal to FEs + HMS;
step 7, storing the harmony database MtBest individual Best in (1)t；
Step 8, executing search operation based on the mean value information to generate a new individual UtThe method comprises the following specific steps:
step 8.1, making a counter kj equal to 1;
step 8.2, if the counter kj is larger than D, turning to step 9, otherwise, turning to step 8.3;
step 8.3, calculating the sum sound library M according to the formula (1)tMean of the kth dimension
Figure BDA0001345205140000026
Step 8.4, generating a random real number PR between [0,1 ];
step 8.5, if PR is less than HMCR, go to step 8.6, otherwise go to step 8.19;
step 8.6, randomly generating a positive integer NR1 between [1, HMS ];
step 8.7, order
Figure BDA0001345205140000031
Step 8.8, randomly generating two real numbers TAR and TG between [0,1 ];
step 8.9, if TAR is less than PAR, go to step 8.10, otherwise go to step 8.21;
step 8.10, if TG is less than 0.5, go to step 8.11, otherwise go to step 8.16;
step 8.11, randomly generating a positive integer NR2 which is not equal to NR1 between [1, HMS ];
step 8.12, order the elite valueWherein rand is a random real number generating function subject to uniform distribution;
step 8.13, randomly generating a real number MW between [0,1 ];
step 8.14, order
Figure BDA0001345205140000033
Step 8.15, go to step 8.21;
step 8.16, let Gaussian mean value
Figure BDA0001345205140000034
And standard deviation of Gauss
Figure BDA0001345205140000035
Wherein abs is an absolute value function;
step 8.17, order
Figure BDA0001345205140000036
Wherein NormRand represents a Gaussian random number generating function;
step 8.18, go to step 8.21;
step 8.19, at [1, HMS]Randomly generating a positive integer NR3, making the search factor SLK ═ rand (0,1) × 2-1, and then making the mean term
Figure BDA0001345205140000037
Wherein rand is a random real number generating function subject to uniform distribution;
step 8.20, orderWherein the combination factor SW is [0,1]Random real numbers in between;
step 8.21, let the counter kj be kj +1, and then go to step 8.2;
step 9, calculating the individual UtAn adaptation value of;
step 10, the harmony library MtThe worst individual among them was designated as MWorstt；
Step 11, if the individual UtIs superior to MWorsttThen use the individual UtAlternative MWorsttOtherwise, keep MWorsttThe change is not changed;
step 12, setting the current evaluation frequency FEs to FEs + 1;
step 13, making the current evolution algebra t equal to t + 1;
step 14, save the harmony database MtBest individual Best in (1)t；
Step 15, repeating the steps 8 to 14 until the current evaluation times FEs reaches MAX _ FEs, and finishing the process, wherein the optimal individual Best obtained in the execution processtThe decoding is the routing inspection path of the robot, and routing inspection path planning of the robot can be realized.
The invention adopts mean value and sound search to plan the inspection path of the robot. In the mean value and sound search, firstly, the mean value of all individuals in the harmony library is calculated, then, mean value information is used as a base point for generating a new individual, and the mean value information of the harmony library is fused with the optimal individual in the harmony library to be used as a guide direction of a Gaussian mutation operator, so that the search performance of the algorithm is enhanced, and the routing efficiency of the robot patrol route planning is improved. The invention can improve the routing planning efficiency of the robot inspection path to a certain extent.
Drawings
FIG. 1 is a flow chart of the present invention.
Fig. 2 is a patrol point used for planning a patrol path of the robot in the embodiment, and each small black point in the diagram represents a patrol point.
Detailed Description
The technical scheme of the invention is further specifically described by the following embodiments and the accompanying drawings.
Example (b):
step 1, inputting coordinates of each inspection point shown in fig. 2, and determining the number D of the inspection points to be 50;
step 2, initializing the size HMS of the sound library by the user to be 50, selecting the probability HMCR to be 0.95, the disturbance probability PAR to be 0.6 and the maximum evaluation time MAX _ FEs to be 300000;
step 3, setting the current evolution algebra t as 0 and setting the current evaluation times FEs as 0;
step 4, randomly generating an initial harmony library
Figure BDA0001345205140000041
Wherein the individual subscript i ═ 1, 2., HMS; and isIs a harmony library MtThe ith individual of (1); individuals
Figure BDA0001345205140000043
The sequential weight of the D inspection points is stored;
Figure BDA0001345205140000051
is an individual
Figure BDA0001345205140000052
The sequential weight of the jth routing inspection point in the table, and a dimension subscript j is 1, 2.
Step 5, calculating a harmony database MtThe fitness value of each individual;
step 6, making the current evaluation times FEs equal to FEs + HMS;
step 7, storing the harmony database MtBest individual Best in (1)t；
Step 8, executing search operation based on the mean value information to generate a new individual UtThe method comprises the following specific steps:
step 8.1, making a counter kj equal to 1;
step 8.2, if the counter kj is larger than D, turning to step 9, otherwise, turning to step 8.3;
step 8.3, calculating the sum sound library M according to the formula (1)tMean of the kth dimension
Figure BDA0001345205140000053
Figure BDA0001345205140000054
Step 8.4, generating a random real number PR between [0,1 ];
step 8.5, if PR is less than HMCR, go to step 8.6, otherwise go to step 8.19;
step 8.6, randomly generating a positive integer NR1 between [1, HMS ];
step 8.7, order
Figure BDA0001345205140000055
Step 8.8, randomly generating two real numbers TAR and TG between [0,1 ];
step 8.9, if TAR is less than PAR, go to step 8.10, otherwise go to step 8.21;
step 8.10, if TG is less than 0.5, go to step 8.11, otherwise go to step 8.16;
step 8.11, randomly generating a positive integer NR2 which is not equal to NR1 between [1, HMS ];
step 8.12, order the elite value
Figure BDA0001345205140000056
Wherein rand is a random real number generating function subject to uniform distribution;
step 8.13, randomly generating a real number MW between [0,1 ];
step 8.14, order
Figure BDA0001345205140000057
Step 8.15, go to step 8.21;
step 8.16, let Gaussian mean value
Figure BDA0001345205140000061
And standard deviation of Gauss
Figure BDA0001345205140000062
Wherein abs is an absolute value function;
step 8.17, order
Figure BDA0001345205140000063
Wherein NormRand represents a Gaussian random number generating function;
step 8.18, go to step 8.21;
step 8.19, at [1, HMS]Randomly generating a positive integer NR3, making the search factor SLK ═ rand (0,1) × 2-1, and then making the mean term
Figure BDA0001345205140000064
Wherein rand is a random real number generating function subject to uniform distribution;
step 8.20, orderWherein the combination factor SW is [0,1]Random real numbers in between;
step 8.21, let the counter kj be kj +1, and then go to step 8.2;
step 9, calculating the individual UtAn adaptation value of;
step 10, the harmony library MtThe worst individual among them was designated as MWorstt；
Step 11, if the individual UtIs superior to MWorsttThen use the individual UtAlternative MWorsttOtherwise, keep MWorsttThe change is not changed;
step 12, setting the current evaluation frequency FEs to FEs + 1;
step 13, making the current evolution algebra t equal to t + 1;
step 14, save the harmony database MtBest individual Best in (1)t；
Step 15, repeating the steps 8 to 14 until the current evaluation times FEs reaches MAX _ FEs, and finishing the process, wherein the optimal individual Best obtained in the execution processtDecoding the route to the routing inspection path of the robot, and obtaining the planned route of the routing inspection robot.
The specific embodiments described herein are merely illustrative of the spirit of the invention. Various modifications or additions may be made to the described embodiments or alternatives may be employed by those skilled in the art without departing from the spirit or ambit of the invention as defined in the appended claims.

Claims (1)

1. A robot inspection path planning method based on mean value and acoustic search is characterized by comprising the following steps:
step 1, inputting coordinates of each inspection point, and determining the number D of the inspection points;
step 2, initializing a sound library size HMS by a user, and selecting a probability HMCR, a disturbance probability PAR and a maximum evaluation time MAX _ FEs;
step 3, setting the current evolution algebra t as 0 and setting the current evaluation times FEs as 0;
step 4, randomly generating an initial harmony library
Figure FDA0001345205130000011
Wherein the individual subscript i ═ 1, 2., HMS; and is
Figure FDA0001345205130000012
Is a harmony library MtThe ith individual of (1); individuals
Figure FDA0001345205130000013
The sequential weight of the D inspection points is stored;
Figure FDA0001345205130000014
is an individual
Figure FDA0001345205130000015
The sequential weight of the jth routing inspection point in the table, and a dimension subscript j is 1, 2.
Step 5, calculating a harmony database MtThe fitness value of each individual;
step 6, making the current evaluation times FEs equal to FEs + HMS;
step 7, storing the harmony database MtBest individual Best in (1)t；
Step 8, executing search operation based on the mean value information to generate a new individual UtIn particularThe method comprises the following steps:
step 8.1, making a counter kj equal to 1;
step 8.2, if the counter kj is larger than D, turning to step 9, otherwise, turning to step 8.3;
step 8.3, calculating the sum sound library M according to the formula (1)tMean of the kth dimension
Figure FDA0001345205130000016
Figure FDA0001345205130000017
Step 8.4, generating a random real number PR between [0,1 ];
step 8.5, if PR is less than HMCR, go to step 8.6, otherwise go to step 8.19;
step 8.6, randomly generating a positive integer NR1 between [1, HMS ];
step 8.7, order
Figure FDA0001345205130000018
Step 8.8, randomly generating two real numbers TAR and TG between [0,1 ];
step 8.9, if TAR is less than PAR, go to step 8.10, otherwise go to step 8.21;
step 8.10, if TG is less than 0.5, go to step 8.11, otherwise go to step 8.16;
step 8.11, randomly generating a positive integer NR2 which is not equal to NR1 between [1, HMS ];
step 8.12, order the elite value
Figure FDA0001345205130000021
Wherein rand is a random real number generating function subject to uniform distribution;
step 8.13, randomly generating a real number MW between [0,1 ];
step 8.14, order
Figure FDA0001345205130000022
Step 8.15, go to step 8.21;
step 8.16, let Gaussian mean value
Figure FDA0001345205130000023
And standard deviation of Gauss
Figure FDA0001345205130000024
Wherein abs is an absolute value function;
step 8.17, order
Figure FDA0001345205130000025
Wherein NormRand represents a Gaussian random number generating function;
step 8.18, go to step 8.21;
step 8.19, at [1, HMS]Randomly generating a positive integer NR3, making the search factor SLK ═ rand (0,1) × 2-1, and then making the mean term
Figure FDA0001345205130000026
Wherein rand is a random real number generating function subject to uniform distribution;
step 8.20, order
Figure FDA0001345205130000027
Wherein the combination factor SW is [0,1]Random real numbers in between;
step 8.21, let the counter kj be kj +1, and then go to step 8.2;
step 9, calculating the individual UtAn adaptation value of;
step 10, the harmony library MtThe worst individual among them was designated as MWorstt；
Step 11, if the individual UtIs superior to MWorsttThen use the individual UtAlternative MWorsttOtherwise, keep MWorsttThe change is not changed;
step 12, setting the current evaluation frequency FEs to FEs + 1;
step 13, making the current evolution algebra t equal to t + 1;
step 14, save the harmony database MtBest individual Best in (1)t；
Step 15, repeating the steps 8 to 14 until the current evaluation times FEs reaches MAX _ FEs, and finishing the process, wherein the optimal individual Best obtained in the execution processtThe decoding is the routing inspection path of the robot, and routing inspection path planning of the robot can be realized.
CN201710553903.6A 2017-07-09 2017-07-09 Robot inspection path planning method based on mean value and sound search Active CN107329473B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710553903.6A CN107329473B (en) 2017-07-09 2017-07-09 Robot inspection path planning method based on mean value and sound search

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710553903.6A CN107329473B (en) 2017-07-09 2017-07-09 Robot inspection path planning method based on mean value and sound search

Publications (2)

Publication Number Publication Date
CN107329473A CN107329473A (en) 2017-11-07
CN107329473B true CN107329473B (en) 2020-02-07

Family

ID=60197185

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710553903.6A Active CN107329473B (en) 2017-07-09 2017-07-09 Robot inspection path planning method based on mean value and sound search

Country Status (1)

Country Link
CN (1) CN107329473B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108267138B (en) * 2018-01-27 2021-05-11 江西理工大学 Welding robot path planning method applying adaptive elite-oriented search
CN110689205B (en) * 2019-10-09 2023-02-07 甘肃农业大学 Vehicle path optimization method and device, readable storage medium and electronic equipment
CN115357044A (en) * 2022-08-16 2022-11-18 国网浙江省电力有限公司台州供电公司 Unmanned aerial vehicle cluster distribution network line inspection path planning method, equipment and medium
CN115796421B (en) * 2023-01-30 2023-05-16 北京智芯微电子科技有限公司 Automatic inspection method and system for operation monitoring of distribution network equipment and terminal equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102789493A (en) * 2012-07-06 2012-11-21 葛彦强 Self-adaptive dual-harmony optimization method
CN104197941A (en) * 2014-09-16 2014-12-10 哈尔滨恒誉名翔科技有限公司 Defense penetration path planning method of underwater vehicle based on harmony search algorithm
CN104317721A (en) * 2014-11-12 2015-01-28 大连交通大学 Regression test case selection method based on improved harmony search algorithm
CN105631540A (en) * 2015-12-27 2016-06-01 哈尔滨米米米业科技有限公司 Underwater vehicle path planning system based on harmony search algorithm
CN106447026A (en) * 2016-09-13 2017-02-22 哈尔滨工程大学 Cuckoo harmony search mechanism-based IIR (infinite impulse response) digital filter generating method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102789493A (en) * 2012-07-06 2012-11-21 葛彦强 Self-adaptive dual-harmony optimization method
CN104197941A (en) * 2014-09-16 2014-12-10 哈尔滨恒誉名翔科技有限公司 Defense penetration path planning method of underwater vehicle based on harmony search algorithm
CN104317721A (en) * 2014-11-12 2015-01-28 大连交通大学 Regression test case selection method based on improved harmony search algorithm
CN105631540A (en) * 2015-12-27 2016-06-01 哈尔滨米米米业科技有限公司 Underwater vehicle path planning system based on harmony search algorithm
CN106447026A (en) * 2016-09-13 2017-02-22 哈尔滨工程大学 Cuckoo harmony search mechanism-based IIR (infinite impulse response) digital filter generating method

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
An improved global-best harmony search algorithm;Mohammed El-Abd;《Applied Mathematics and Computation》;20131001;第222卷;第94-106页 *
全局竞争和声搜索算法;夏红刚等;《控制与决策》;20160228;第31卷(第2期);第310-316页 *
和声搜索算法探索能力研究及其修正;欧阳海滨等;《控制理论与应用》;20140131;第31卷(第1期);第57-65页 *
基于改进和声搜索算法的车辆路径规划问题;赵再兴;《中国优秀硕士学位论文全文数据库-经济与管理科学辑》;20110715(第07期);第J145-260页 *

Also Published As

Publication number Publication date
CN107329473A (en) 2017-11-07

Similar Documents

Publication Publication Date Title
CN107329473B (en) Robot inspection path planning method based on mean value and sound search
Sak et al. Fast and accurate recurrent neural network acoustic models for speech recognition
CN108446290B (en) Streaming real-time conversation management
CN106441316B (en) Historical data-based single-point road network matching method
CN109346056B (en) Speech synthesis method and device based on depth measurement network
Ge et al. Partially-connected differentiable architecture search for deepfake and spoofing detection
CN113253050B (en) Traveling wave fault location method based on whale optimization Kalman filtering algorithm
CN110967594A (en) Method and device for positioning faults of power distribution network containing inverter type distributed power supply
Comarela et al. Robot routing in sparse wireless sensor networks with continuous ant colony optimization
US20240005942A1 (en) Frame-level permutation invariant training for source separation
Xu et al. Language independent query-by-example spoken term detection using n-best phone sequences and partial matching
Wang et al. Using DTW to measure trajectory distance in grid space
Xu et al. Multi-Scale Approaches to the MediaEval 2015" Emotion in Music" Task.
Zhang et al. Distance-based weight transfer for fine-tuning from near-field to far-field speaker verification
KR101727306B1 (en) Languange model clustering based speech recognition apparatus and method
CN109979422A (en) Fundamental frequency processing method, device, equipment and computer readable storage medium
Kala et al. Very fast unit selection using Viterbi search with zero-concatenation-cost chains
Du et al. Combining quantum-behaved PSO and K2 algorithm for enhancing gene network construction
Meirong et al. Query-by-example on-device keyword spotting using convolutional recurrent neural network and connectionist temporal classification
Ramos et al. An evaluation of different evolutionary approaches applied in the process of automatic transcription of music scores into tablatures
Choi et al. Short-utterance embedding enhancement method based on time series forecasting technique for text-independent speaker verification
Холєв et al. Comparative analysis of neural network models for the problem of speaker recognition
CN108267138B (en) Welding robot path planning method applying adaptive elite-oriented search
Souza et al. FPGA implementation of genetic algorithms for frequency estimation in power systems
Alhamada et al. Deep learning for environmentally robust speech recognition

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