基于蚁群算法的移动机器人全局路径规划方法研究-史恩秀

20146456doi10．6041/j．issn．1000-1298．2014．06．009*710048。m、α、βρ。。。。TP242.6A1000-1298201406-0053-052013-07-072013-07-21*[e-mail]。。1－2。。A*3、4、5－6、7、89。、、10。。m、α、βρ。11.1。。11。12。1。1Fig．1AntsforagingsimulationabT=0cT=11AEBCDH。ACHEECHABCD1BHD21a。AE301。1。BC、CD、BH、HDAE。BC、CD、BH、HD1b。BCDBHD。1c20B、CDE。BCDBCD。1.2。ＲS。。ＲS。。。。。123…n。。。。S=123…N。g001g102gXYn。、S。1.32。2Fig．2Antcolonyalgorithmflowchart10101。D=01…n－1m、α、β、ρNc。Bkkk=12…m。SES。2ijpkijt=ταijtηβijt∑s∈Cταistηβistj∈C0j瓝C{1ηijt=1dij2C=D－Bk3dij———ijηij———ijτij———ij123BkkjBk。42～3。5τijτijt+1=ρτijt+Δτijtt+14Δτijtt+1=∑mk=1Δτkijtt+15Δτkijt=Q/Lkkij0kij{6Q———Lk———k64～5。26。α、β、ρ、m、Q。3。4520143Fig．3Twopathplanningenvironmenta1b22.1。13。7。α=1β=5ρ=0.5Q=100m10、30、50、80、100、150200。4180250。mmm。4mFig．4Effectofantnumbermonalgorithmperformance2.2。ρ14。ρρ。m=100α=1β=5Q=100ρ=0.1、0.2、0.3、0.4、0.5、0.70.95。51ρ=0.72ρ=0.5。5ρFig．5Effectofpheromoneevaporationcoefficientρonalgorithmperformance2.3α。β。α=0。β=0。α=1β6。6βFig．6Effectofβonalgorithmperformance6αβββ=7α。75567αFig．7Effectofαonalgorithmperformance。7αα=1、β=7。3。Matlab。1400。m=80Nc=100α=1β=71ρ=0.72ρ=0.5。89。8。98Fig．8Theoptimaltrajectorya1b2。。2832。9Fig．9Convergencecurvesa1b241。m、α、β、ρ。2mNcmm。3ρρ。4β。5α。1．J．2012342202－206．LiQingWangLijunChenBoetal．AnimprovedartificialpotentialfieldmethodwithparametersoptimizationbasedongeneticalgorithmsJ．JournalofUniversityofScienceandTechnologyBeijing2012342202－206．inChinese2．J．201128127－31．LianXiaofengLiuZaiwenZuoMin．ＲesearchondynamicfuzzyartificialpotentialfieldmethodformobilerobotpathplannlingJ．ComputerSimulation201128127－31．inChinese3．A*J．20125281085－1089．652014WangDianjun．Indoormobile-robotpathplanningbasedonanimprovedA*algorithmJ．JournalofTsinghuaUniversityScienceandTechnology20125281085－1089．inChinese4．J．2011284193－195303．ShiTiefeng．ＲesearchonpathplanningformobilerobotbasedonimprovedgeneticalgorithmJ．ComputerSimulation2011284193－195303．inChinese5．J．2011286231－234．ZhangYinlingNiuXiaomei．SimulationresearchonmobilerobotpathplanningbasedonantcolonyoptimizationJ．ComputerSimulation2011286231－234．inChinese6．α、β、ρ-TSPJ．2004287597－601．YeZhiweiZhengZhaobao．ConfigurationofparametersαβρinantalgorithmJ．JournalofWuhanUniversityGeomaticsandInformationScience2004287597－601．inChinese7．J．20123171085－1089．ZhouLikunLiuHongzhao．Anadaptiveartificialfishschoolalgorithmforpathplanningofmobiletank-clearingrobotJ．MechanicalScienceandTechnology20123171085－1089．inChinese8．J．2010323397－402．LiQingXuYinmeiZhangDezhengetal．GlobalpathplanningmethodformobilerobotsbasedontheparticleswarmalgorithmJ．JournalofUniversityofScienceandTechnologyBeijing2010323397－402．inChinese9．J．2013303323－326375．YeZhaoliYuanMingxinChengShuaietal．NewfireworksexplosiveimmuneplanningalgorithmformobilerobotsJ．ComputerSimulation2013303323－326375．inChinese10．J．2012411108－113．11．J．2011355637－641．ZhaoJuanpingGaoXianwenFuXiuhui．ImprovedantcolonyoptimizationalgorithmforsolvingpathplanningproblemofmobilerobotJ．JournalofNanjingUniversityofScienceandTechnology2011355637－641．inChinese12ErinBAbiyevＲIbrahimD．TeachingrobotnavigationinthepresenceofobstaclesusingacomputersimulationprogramJ．Procedia-SocialandBehavioralSciences201022565－571．13ArdiyantoIMiuraJ．Ｒeal-timenavigationusingrandomizedkinodynamicplanningwitharrivaltimefieldJ．ＲoboticsandAutonomousSystems201260121579－1591．14．J．2011333274－286．YeWeiyaoWangChunxiangYangMingetal．VirtualobstaclesbasedpathplanningformobilerobotsJ．Ｒobot2011333274－286．inChinese15BradenEStenningTimothyDBarfoot．Pathplanningwithvariable-fidelityterrainassessmentJ．ＲoboticsandAutonomousSystems20126091135－1148．16VolosChKKyprianidisIMStouboulosIN．AchaoticpathplanninggeneratorforautonomousmobilerobotsJ．ＲoboticsandAutonomousSystems2012604651－656．17KoutsonikolasaDimitriosDasaSMHuCY．PathplanningofmobilelandmarksforlocalizationinwirelesssensornetworksJ．ComputerCommunications200730132577－2592．ＲesearchonMethodofGlobalPath-planningforMobileＲobotBasedonAnt-colonyAlgorithmShiEnxiuChenMinminLiJunHuangYumeiSchoolofMechanicalandInstrumentalXi'anUniversityofTechnologyXi'an710048ChinaAbstractTheglobalpath-planningmethodforMＲisstudiedbasedonthecharacteristicsofantcolonyalgorithm．TheenvironmentinformationofaplaneonwhichmobilerobotMＲworksisexpressedbythegridmethod．Themainparametersusedbyantcolonyalgorithmsuchasthenumberofantmarousefactorαexpectedarousefactorβandinformationelementhangovercoefficientsρwhichaffecttheresultofpathplannedincludingthelengthofthepathplannedandtheefficiencyplanningpatharesimulated．Itisfoundfromthesimulationresultthatthebetterglobalpathcanbegotten．Thelengthofthepathplannedisshorterandtheefficiencyofplanningpathishigherwhentheparametersmαβandρarematchedbetter．BasedontheresultsofabovesimulationthepathisplannedforMＲwhichworksontwodifferentwork-planeusingantcolonyalgorithmwiththebestmatchingparametersobtainedfromthesimulation．TheaccuracyoftheoreticalresearchisverifiedbytheglobalpathplannedforMＲworkingunderthedifferentwork-plane．KeywordsMobilerobotAnt-colonyalgorithmWorkpathplanningMatchingparameters756