ITSWC201520150803.pdf智能交通世界大会ITS智慧城市社区人工智能AI物联网IT报告课件教案.pdf

《ITSWC201520150803.pdf智能交通世界大会ITS智慧城市社区人工智能AI物联网IT报告课件教案.pdf》由会员分享，可在线阅读，更多相关《ITSWC201520150803.pdf智能交通世界大会ITS智慧城市社区人工智能AI物联网IT报告课件教案.pdf（10页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、-最新资料推荐最新资料推荐-ITSWC201520150803.pdfITSWC201520150803.pdf智能交通世界大会智能交通世界大会ITSITS智慧城市社区人工智能智慧城市社区人工智能 AIAI 物联网物联网 ITIT 报告课件教案报告课件教案 1 A traffic simulation model of pedestrian and vehicle forevacuation under natural disaster Hisatomo Hanabusa GeneralManager of Technical Department,i-Transport Lab.Co.,Lt

2、d.Shin-Surugadai BLDG 9F,3-10,Kanda-Ogawamachi,Chiyoda-ku,Tokyo,101-0052,Japan TEL:+81-3-5283-8527,FAX:+81-3-5283-8528,E-mail:hanabusai-transportlab.jpTadashiKomiya Researcher,i-Transport Lab.Co.,Ltd.Shin-SurugadaiBLDG 9F,3-10,Kanda-Ogawamachi,Chiyoda-ku,Tokyo,101-0052,JapanTEL:+81-3-5283-8527,FAX:+

3、81-3-5283-8528,E-mail:komiyai-transportlab.jpRyotaHoriguchiCEO,i-Transport Lab.Co.,Ltd.Shin-Surugadai BLDG 9F,3-10,Kanda-Ogawamachi,Chiyoda-ku,Tokyo,101-0052,Japan TEL:+81-3-5283-8527,FAX:+81-3-5283-8528,E-mail:rhoriguchii-transportlab.jp Masao Kuwahara Professor,Department of Human-Social Informati

4、on Sciences,GraduateSchoolofInformationSciences,TohokuUniversityTEL:E-mail:Aoba6-6-06,Aramaki,Aoba-ku,Sendai,980-8579,Japan+81-22-795-7492,FAX:+81-22-795-7494,kuwaharaplan.civil.tohoku.ac.jp ABSTRACT In this study,we1 1/1010developed a traffic simulation framework of pedestrian andvehicletoevaluatet

5、heevacuationplansundernaturaldisasters.In this paper,we introduce the concept of ourframework and explain the traffic model and the road crossingbehavior model of pedestrian and vehicle in the simulation.This framework has a function to be attached the disaster eventmodel(link closing and lane closi

6、ng).And the evacuationbehavior model can be installed depending on the purpose of theevaluation of evacuation measure.As the future work,we havea plan to apply this framework to urban city and describe theresultsofsomeevacuationplans.Keywords:TrafficSimulation,Pedestrian,Vehicle,Evacuation,Natural d

7、isaster1.INTRODUCTION In Japan,many researches of evacuation planunder natural disaster are studied on the background of ourexperiences such as the Great East Japan Earthquake and theGreat 2 Hanshin Earthquake.Oshima et al.1 evaluated thetraffic measures to reduce the congestion in Tokyo on the dayo

8、f the Great East Japan Earthquake.Erick et al.2 and Innamiet al.3 installed Tsunami event model to their trafficsimulation model and simulated the evacuation of vehicles andpedestrians.In the case of traffic simulation study,it is saidthat the behavior models of vehicle and pedestrian under-最新资料推荐最新

9、资料推荐-disaster have been implemented in a traffic simulation modelin recent years.However,the knowledge about the understandingofthetrafficsituationbydisasterandtheeffectiveevacuation plan for large area(the whole of city)are far fromaccumulated.For example,the latescent bottleneck pointwhenthe peopl

10、e evacuate from natural disaster by vehicle or by walkshould be understand.In this case,the traffic demands forevacuation from natural disaster should be included not onlythe vehicles on the road network but also the people inbuildings,homes,traffic facilities and so on.On the otherhand,the impassab

11、le roads by fired and collapsed buildings andthe heavy congestion of vehicles and pedestrians by theconcentrated inbound traffic should be considered when we makethe evacuate plan to guide the people to the safety places.Tounderstand these issues,the traffic simulation model which candynamically rep

12、roduce the traffic condition(especially queuespillback,capacity reduction andgridlock)bynaturaldisaster is needed.And it is important that the conflictingof vehicles and pedestrians is one ofthe needs to express thedisruption of the traffic flow.Therefore,we develop thetraffic simulation framework w

13、ith the conflicting model of3 3/1010pedestrians and vehicles to evaluate the evacuate plan fromnatural disaster.2.THE CONCEPT OF SIMULATION FRAMEWORK Ourtraffic simulation framework is developed to reproduce varioustraffic situations on road traffic under natural disaster andto evaluate the effects

14、of evacuation measures.The frameworkionmodel for the region-wide road network(10 or 20 square km,uppedestrian and vehicle(link capacity reduction by the crossingbehaviobehavior with many-to-Disaster event model(link closure event,lane closure event,destination change event,etc.)Figure 1 shows the tr

15、afficsimulation framework.The framework consists of the trafficsimulation model integrated pedestrian modeland traffic flowmodel,the disaster event model and the evacuation behaviormodel.The traffic simulation model includes the crossingbehavior model and the interfaces to accept the event requestsf

16、rom the disaster event model and the evacuation behavior model.3 Figure 1 1 :Simulation Framework 3.TRAFFIC FLOW MODELAND PEDESTRIAN MODEL In this section,we introduce the traffic-最新资料推荐最新资料推荐-flow model and the pedestrian model.For the traffic flow model,we prepared the traffic simulation model SOU

17、ND(a Simulationmodel On Urban road Networks with Dynamic route choice)4.Thevehicles in SOUND is discretely moved based on the Newellssimplified kinematic wave model56 and can be controlled bytraffic signal and the link parameters(link capacity,jamdensity,free travel speed,saturation flow rate,number

18、 oflane,etc.).SOUND has two types of route choice model(minimumcost choice model and logit choice model based on Dialsalgorithm).On the other hand,the pedestrian model has a simplemoving model by the constant speed controlled by the density(number of person)of link referred to Fundamental Diagram(K-

19、Vrelationship).The pedestrian model is also installed the sameroute choice mode l as SOUND.Figure 2 2:Basic Idea of thepedestrian model 4 Figure 3 and Figure 4 shows the interactivepoints between vehicle road network and pedestrian network.Thetraffic flow model and the pedestrian model share some cr

20、ossingpoints on each road network.And the vehicle on the vehiclesroad network performs the stop-and-go behaviour when thepedestrian on the crossing point.Figure 3 3:Interactivepoints on road network Figure 4 4:Network structure and5 5/1010crossing behaviour in an intersection 5 4.INPUT DATA FORSIMUL

21、ATION The basic input data for the simulation frameworkconsist of the road network data,zone data and OD table data(time series for each vehicle/person type).In addition,thedisaster event data such as link closure event(the set of eventtime,closing link id)and the evacuation behavior data(newOD tabl

22、e and Destination data for evacuation)are needed forthedisastersimulation.Thenewdestinationdataforevacuation can be used the location data of the evacuation placesuch as school and park.Figure 5 shows the exampleof networkdata and zone data,and Figure 6 shows the example of POI mapdata.Figure 5 5:Ro

23、ad network data and zone data Figure6 6:Example of POI data(evacuation place)6 5.CONFLICTIONBEHAVIOR MODEL In this study,two basic models to express theconfliction behaviours between vehicles and pedestrians areimplemented.Figure 7 shows the concept of the conflictionbehaviours model.The crossing mo

24、del has a function thatvehicle is stopped when pedestrian is existed(walking)on thecrossing point of vehicles road and pedestrians road.On theother hand,the capacity reduction model has a function thatthe capacity of vehicles road(link)is reduced when the densityofpedestriansroad(link)ishigh(oneofsi

25、mulation-最新资料推荐最新资料推荐-parameters).This situation by the capacity reduction model isassumed the flooding of pedestrian by the concentrated to thepublic space such as train station.Figure 7 7:ConflictionBehaviour Model 6.EVACUATION MODEL The traffic impact by theevacuation behavior can be reproduced b

26、y the destination changebehaviormodelandthedepartureselectionmodel(determination of start or stop trip)for the vehicle and theperson which is planned to departure before the disaster eventoccur.If the vehicles or the persons running on the roadnetwork is caught in the disaster(link closure event)suc

27、h astsunami,they are stopped the trip and counted up as the resultof victims by the disaster.Figure 8 shows the image of thedisaster event model and the evacuation behaviors in oursimulation framework.7 Figure 8 8:Example of DisasterEvent and Evacuation behaviours 7.EFFECT OF CONFLICTION OFVEHICLES

28、AND PEDESTRIANS Figure 9 shows the calculation resultof the effect to vehicle traffic by pedestrian traffic in anintersection(Jimbocho-intersection)in Tokyo.In this casestudy,we verified the reduction pattern of the capacity ofvehicle traffic by the traffic of pedestrian.The trafficvolumes of vehicl

29、e from each direction and the signal timing7 7/1010data are surveyed at peak time from 9:00 to 10:00.Thesimulation is calculated with shifting the traffic volume ofpedestrian through the intersection.The conflicting behavior(crossing of vehicles and pedestrians)model and the capacityreduction model(

30、vehicles road)by the pedestrians flooding tothe vehicles road are implemented.The traffic volume and theaverage speed of vehicle are declined drastically 8 in case ofover 2019(pedestrian traffic).The models will be validatedand improved comparing tothe real data as afuture work.Figure 9 9:Effect to

31、Vehicle Traffic by Pedestrian in anIntersection 8.FUTURE WORK For the next step,our simulationframework will be verified using the verification manual fortraffic simulation model with disaster and evacuation model byJapan Association for Earthquake Engineering7(JAEE)beforethe application to case stu

32、dy.JAEE offers the five outputs(vehicle/pedestrian generation,moving speed,route choice,impact of traffic flow by escape route and congested situation)for verification and three outputs(network data and od volumedata of validation area,simulation result)for validation todescribe.And we have a plan t

33、o evaluate evacuation plans forthe 23 words in Tokyo using our framework.Figure 10 showsthetarget area.The traffic situation in disaster is calibrated-最新资料推荐最新资料推荐-by using the detector data and the probe data in the day of theGreat East Japan Earthquake with some disaster event modelbased on the ha

34、zard map in Tokyo(linkclosure events and laneclosureevents).9Figure10:Targetnetworkforevaluation(Tokyo 23 words)9.ACKNOWLEDGEMENT This trafficsimulationframeworkwasdevelopedinaResearchandDevelopment Project coordinated by National Institute ofInformation and Communications Technology(NICT).The autho

35、rsthank for the DOMINGO(Data Oriented Mobility Information Group)especially for Yosuke Kawasaki from Tohoku University,KazuakiEtoh and Takeshi Ohata from Oriental Consultants Co,Ltd.,Norihiro Sudo from Japan Weather Association to promote thisresearch and development.10.REFFERENCES 1DaisukeOshima,Sh

36、inji Tanaka,Takashi Oguchi,Evaluation of trafficcontrol policy in disaster case by using traffic simulationmodel,19th ITS World Congress,Vienna,Austria,2019 2 ErickMas,Anawat Suppasri,Fumihiko Imamura,Shunichi Koshimura,Agent-basedSimulationofthe2019GreatEastJapanEarthquake/Tsunami Evacuation:An Int

37、egrated Model of TsunamiInundation and Evacuation,41 Journal of Natural DisasterScience,Volume 34,Number 1,2019 3Junji Innami,Yozo Goto,9 9/1010Takashi Suematsu,Muzailin Affan,Yudha Nurdin:Development ofTsunami Evacuation Simulation anditsusing as DisasterEducational Program,International Symposium

38、on EarthquakeEngineering,JAEE,Vol.1,2019.4 Tohiso Yoshii,SOUND:ATraffic Simulation Model for Oversaturated Traffic Flow on 10Urban Expressways,WCTR,1995.5 G.F.Newell,A simplifiedtheory of kinematic waves in highway traffic,part I:Generaltheory,Transportation Research Part B:Methodological,Volume27,I

39、ssue 4,August 1993 ,Pages 281287 6 G.F.Newell,Asimplified theory of kinematic waves in highway traffic,partI:A simplified theory of kinematic waves in highway traffic,part II:Queueing at freeway bottlenecks,TransportationResearch Part B:Methodological,Volume 27,Issue 4,August1993,Pages 289303 7 http:/www.jaee.gr.jp/jp