CN103593535B - Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration - Google Patents
Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration Download PDFInfo
- Publication number
- CN103593535B CN103593535B CN201310597678.8A CN201310597678A CN103593535B CN 103593535 B CN103593535 B CN 103593535B CN 201310597678 A CN201310597678 A CN 201310597678A CN 103593535 B CN103593535 B CN 103593535B
- Authority
- CN
- China
- Prior art keywords
- data
- traffic
- vehicle
- road
- network
- 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
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to an urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration. The system comprises a parallel simulation evaluation module, a traffic demand module, a road network description module, a traffic evolution module, a scheme description module and a road network analysis module. The method includes the parallel simulation evaluation step, the traffic demand step, the road network description step, the traffic evolution step, the scheme description step and the road network analysis step. A macroscopic scale, a mid-scale and a micro-scale are organically integrated, a current traffic condition can be quickly evaluated, and a scientific basis for optimizing a traffic signal scheme is provided for an urban intelligent traffic control system.
Description
Technical field
The present invention relates to a kind of urban transportation complicated self-adapting network parallel simulation system based on Multiscale Fusion and side
Method, belongs to Urban Traffic Simulation field.
Background technology
According to the applicant understood, microcosmic integrated wagon flow loading technique in existing in prior art, based on parallel simulation skill
The traffic signalization technology of art, based on the urban microscopic emulation technology of cellular automata, Complex Networks Analysis technology, Hang Renwei
See emulation technology, based on the macro-traffic emulation technology of GIS, see traffic simulation technology based in statistics, based on multiple agent
Microscopic traffic simulation technology, and based on grand, in, the system integration technology of microscopic simulation platform, etc..
The above technology is all the technology that forward position is compared in current city traffic simulation field.But due to urban transportation system
System is to macroscopical high integrity on space scale from microcosmic, in time scale from second level to grade highly continuous system,
Its complexity determines any single yardstick and is all difficult to traffic behavior is more objectively emulated, and is embodied in：
In the macroscopic artificial of urban transportation is analyzed, the network traffics stress model mainly history according to a city macroscopic view
OD demand（OD refers to the traffic flow between default Origin And Destination）, combine related geography information knowledge with the method for statistics and satisfy the need
Net flow carries out producing, attracts, iteration equalizing distribution loading is carrying out macroscopic artificial.However, in actual road network, most of feelings
Being in a kind of non-balanced state condition lower network flow more, even if changing, also having the transient process of a gradual change, this is with people's
Driving habits and intersection signal control have certain relation, and just cannot consider this in macroscopical network simulation analysis
The small yardstick of sample.
In the middle sight simulation analysis of urban transportation, employ the functional relationship between parameter in a large number, by mathematical statisticss
Or the method for mathematical modeling describes the variation relation between traffic parameter, but the process of modeling and statistics will be through certain
Idealization assume or process, such as in the emulation relation of section vehicle flowrate and speed, both cannot describe for example section stop,
In weather and section, the impact of the random material elements such as vehicle in and out port, also cannot describe the change of other road section traffic volume situations
Change the impact to this section car speed, this makes middle sight emulation can only simulate a kind of Utopian variation tendency, and needs
Long-term probability demarcation is carried out to parameter, workload is huge so that its practical application effect is not fine.
In the microsimulation analysis of urban transportation, individual behavior often only changes according to the change of environment at one's side,
Because yardstick is excessively small, various small changes in true traffic system all can produce impact, such as one pedestrian to result
Make a dash across the red light, a Public Transit Bus Stopping roadside, these microcosmic influence factors all can occur randomly at any time truly in bad border so that microcosmic
Emulation is finally difficult to the simulated effect accomplishing to match with truth.
Based on grand, in, in the traffic simulation integrated system that is used interchangeably of microscopic simulation platform although various emulation platform
Between parameter can transmit, and data can only from macroscopic artificial pass to middle see emulation, then again by middle sight emulation pass to micro-
See emulation.But the principle of simulation each within emulation platform or constant, the Parameters variation simply emulating.Therefore, this
It is actually a kind of multiple dimensioned alternately simulation process, because macroscopical, middle sight, the respective phantom of microcosmic be not real
Realize merging, the problem in above single simulation process still exists；And, the data fusion of different scale traffic simulation result
Process is also a very big problem.
In current urban transportation intelligent control technology, various analogue systems are excessively poorly efficient and huge, and its main cause exists
More single in current analogue system scale ratio, need just can be given through parameter calibration repeatedly, lengthy and tedious more correctly to emulate
As a result, and macroscopical and middle see emulation and be all based on the statistical simulation of a period of time, the signal timing dial being difficult to crossing provides essence
Thin emulation is supported；And the emulation technology based on the system integration, its integrated analogue system excessively huge it is also difficult to as city
The kernel that intelligent traffic signal controls, being difficult to municipal intelligent traffic control system provides real-time simulation evaluation and support.?
In this case, urban transportation intelligence control system is difficult to know whether the timing scheme oneself providing is rationally effective, Zhi Nengcong
The timing behavior of oneself is little by little adjusted in substantial amounts of history timing.
Content of the invention
The technical problem to be solved is：The problem existing for prior art, proposes one kind based on multiple dimensioned
The urban transportation complicated self-adapting network parallel simulation system and method merging, macroscopical, middle sight, micro-scale are organically blended
Together, can rapidly provide the evaluation of current traffic condition, provide for municipal intelligent traffic control system and optimize traffic signal side
The scientific basis of case.
For reaching above-mentioned purpose, technical scheme is as follows：
A kind of urban transportation complicated self-adapting network parallel simulation system based on Multiscale Fusion, is characterized in that, including
With lower module：
Parallel simulation evaluation module, for linkage simulation evaluation is carried out to target traffic network and its signal time distributing conception,
And provide Optimal Signals timing scheme；
Traffic demand module, for transferring the traffic flow data of target traffic network, and according to history macroscopic view OD data,
The microcosmic measured data of the middle sight statistical data at target road section and crossing and wagon detector is generating Vehicle Agent；Car
Intelligent body has the ability of making decisions on one's own, and specifically includes Path selection, speed modification, observes traffic lights and react, with speeding
Ability with lane change；
Road network describing module, for setting up the geographical description data of target traffic network, is given by analyzing road net data
The traffic information data of Vehicle Agent；
Traffic genetic module, makes a response according to micro environment factor for simulating vehicle intelligent body, according to macro environment
Factor and middle sight environmental factorss make motion decision-making, and the behavior of all Vehicle Agents is incorporated formation emulation in traffic network
Traffic scene；
Scheme describing module, for crossing signals timing protocol each in real-time update analogue system or traffic guidance side
Case data, and provide desired data to analogue system；Traffic guidance protocol for vehicle supervision department pass through Traffic Announcement or
The road conditions real time information that person crossing variable message board provides to vehicle driver, including：Congestion, restricted driving, green channel or traffic
Accident；
Road network analysis module, for road network gradual change cascade in traffic network in analysis following a period of time from currently
Tender spots, line impedance and crossing congestion coefficient；In the time serieses macroscopically analyzing each traffic key element in road network, in
Statistical vehicle flowrate and pedestrian's quantity in sight；And in the way of sequential update, traffic key element each in road network is carried out in temporal sequence
Line emulates.
Present system technical scheme perfect further is as follows：
Preferably, in traffic demand module, history macroscopic view OD data is：It has been stored in the traffic flow data of historical record,
Specifically include distributed data on room and time for the wagon flow starting point and ending point in target traffic network, wagon flow starting point and
Traffic flow distributed data between path data between terminating point, and wagon flow starting point and ending point；
The middle sight statistical data at target road section and crossing is：For target road section and crossing, in its single traffic signal control
In cycle processed, according to drive towards from a crossing volume of traffic at another crossing, each flow direction of each crossing roll the volume of traffic, link travel away from
The statistical data analysis that time, the arrival rate of intersection vehicle flux and steering rate draw；Specifically include：The gathering around of target road section and crossing
Crowded degrees of data, impedance data, arrival rate data, steering rate data, fluctuation data, non-machine mix row data, and wherein fluctuation data is
The crossing vehicle that analogue system is given rolls cycle flow diagram away from and true crossing vehicle rolls the flat vehicle between cycle flow diagram away from
Pcu average error value, non-machine mix row data be crossing wagon flow in terms of unit vehicle pcu through stop line when in different pedestrians
Or non-maneuver interference under by the time, unit vehicle pcu is the car equivalent being converted by various types of automobiles；
The microcosmic measured data of wagon detector is：The real-time traffic amount of 3 seconds to the 9 seconds ranks being uploaded by wagon detector
Data.
Preferably, in road network describing module, geographical description data include the topological structure of road network, section, crossing, track,
Parking lot, bus platform；Road net data includes node degree, degree of communication, betweenness, shortest path；Traffic information data is i.e. from currently
The crowded degrees of data of traffic network in the following a period of time risen；In traffic genetic module, micro environment factor include before and after's car,
Track, crossing, signal lighties；Macro environment factor includes the traffic network congestion status being obtained by macroscopic artificial analysis；Middle sight ring
Border factor includes seeing, by middle, link proportion and the flowed fluctuation parameter that simulation analysis obtain；Motion decision-making includes changing path, repaiies
Change average speed；In road network analysis module, in road network, each traffic key element includes network, section, crossing, signal lighties, wagon flow, car
, pedestrian.
The present invention also provides：A kind of urban transportation complicated self-adapting network parallel simulation side using aforementioned analogue system
Method, is characterized in that, including parallel simulation evaluation procedure：Parallel simulation evaluation module is according to the real-time traffic number of target traffic network
Carry out simulation evaluation according to historical statistical data, provide operating index PI under Setting signal timing scheme for the target traffic network
Value, according to this PI value repeatedly adjustment signal time distributing conception parameter, and provides the PI value after adjustment respectively, after obtain PI through comparing
The optimum signal time distributing conception of value；Wherein, operating index PI value is run in advance by current demand signal timing scheme for target traffic network
Overall delay level after fixing time, overall delay level is to stop vehicle number accumulation on a timeline in target traffic network
Degree.
The inventive method technical scheme perfect further is as follows：
Preferably, also include traffic demand step：Traffic demand module is according to history macroscopic view OD data, target road section and road
The middle sight statistical data of mouth provides the following vehicle flowrate of target traffic network, in conjunction with the microcosmic measured data of wagon detector,
The Vehicle Agent of continuous service within continuous time is simulated under micro environment；Meanwhile, according to history macroscopic view OD data, mesh
Mark section and the middle sight statistical data at crossing, provide the operational factor of each Vehicle Agent；Wherein, the operation ginseng of Vehicle Agent
Number includes driving trace, average speed, security reaction time.
Preferably, also include road network and describe step：The running of road network describing module is as follows：
L1. the nodal community matrix of target traffic network, road-net node connection matrix, road-net node connection attribute square are set up
Battle array, and artificial urban traffic network is built according to the data of above matrix；
L2. calculate and provide node degree matrix and node degree distribution matrix, betweenness matrix and the betweenness of target traffic network
Distribution matrix, each OD between path matrix；OD is to i.e. wagon flow starting point and ending point pair；
L3. calculated in real time and from currently not according to macroscopical OD data, middle sight statistical data and microcosmic measured data
Carry out the road network congestion data in a period of time；Road network congestion data includes vehicle density, journey time；
L4. the road conditions providing Vehicle Agent are calculated according to macroscopical OD data, middle sight statistical data and microcosmic measured data
Information data, and this data is loaded onto in the traffic network of artificial urban.
Preferably, also include traffic evolutionary step：By the concrete behavior of traffic genetic module evolution Vehicle Agent, wrap
Include：Vehicle Agent determines whether to change path according to the road network vehicle flowrate under macro-scale, crowded degrees of data and impedance data；
Whether Vehicle Agent changes average speed according to the current road segment impedance data under medium measure and vehicle flow fluctuating data judging；
Vehicle Agent mixes row data judging according to the non-machine in the crossing under medium measure and sails the average speed rolling crossing away from into；Vehicular intelligent
Body determines whether to accelerate, slows down, stops or lane change according to the car in front and back under micro-scale and signal lighties situation.
It is highly preferred that the detailed process of traffic evolutionary step is as follows：
Y1. crowded degrees of data J is calculated according to road network vehicle flowrate data；Vehicle Agent calculates according to logit model and is directed to
The Path selection probability P in different paths, and judge whether to change road with reference to this Vehicle Agent default individual psychology threshold value R
Footpath, concrete judge process is as follows：
If J is less than R, and this Vehicle Agent to be presently in the Path selection probability P in path be maximum in all paths
Value, then this Vehicle Agent does not change path；If J is less than R, and Path selection probability P >=0.5 in all paths, then this vehicle
Intelligent body does not change path；If J is less than R, and the Path selection probability P in all paths<0.5, then this Vehicle Agent change road
Footpath, and select the maximum path of Path selection probability P；If J is more than R, this Vehicle Agent changes path, and selects path to select
Select the maximum path of probability P；
Wherein, crowded degrees of data J represents the overall congestion level of road network, is calculated as follows：
In formula, xaRepresent the stroke vehicle flowrate of path a, taRepresent the journey time of path a, t0aRepresent zero stream of path a
Impedance, that is, when path a wagon flow is amount, one car drives through the time required for this path；
Logit model is multinomial logit model M NL, Muti-nomial Logit, and formula is as follows:
I.e. Path selection probability PjRatio equal to path index function sums all in path j exponential function and road network；
In formula：PjFor the Path selection probability to path j for the Vehicle Agent, θ represents Vehicle Agent to traffic network
Familiarity, J represents the degree of crowding of path j,Represent the average degree of crowding in all paths in road network, Jm represents path m
The degree of crowding, path m is one of all paths in road network；
When in the road network information that scheme describing module provides containing traffic guidance protocol, θ value improves；
Y2. Vehicle Agent determines the average speed of itself according to link proportion data and vehicle flow fluctuating data；Wherein, car
Stream fluctuation data is the deviation being occurred in real time by vehicle checker between data and simulation and prediction data；
Link proportion data is calculated as follows：
In formula:xaRepresent the stroke vehicle flowrate of path a, taRepresent the running time of path a, t0aRepresent zero stream of path a
Impedance；UaRepresent the traffic capacity of path a, α, β are constant.
Y3. Vehicle Agent according to the state of in front and back's car, signal lighties, the non-machine in crossing mix row data determine whether to accelerate,
Slow down, stop or lane change；Wherein, the non-machine in crossing mixes row data and is obtained by carrying out cluster analyses to crossing wagon detector data
, signal lamp data is provided by scheme descriptive model；
Vehicle Agent controls speed according to Nasch model：
Step1:Accelerate, vn=min(vn+1,vmax)；
Step2:Slow down, vn=min(vn,dn)；
dn=xn+1-xn-lveh, dnRepresent the safe distance between the n-th car and front truck n+1；lvehRepresent length of wagon；
Step3:Random slowing down, with predetermined probabilities P0Random generation, vn=max(vn- 1,0), in order to simulate by uncertain because
The random deceleration that element causes；
Step4:Motion, xn=xn+vn, xn、vnRepresent position and the speed of the n-th car respectively；
The lane change model of Vehicle Agent is as follows：
Track right-to-left is designated as 1,2 successively ..., N, Vehicle Agent lane-change according to the following rules：
R1. consider right road：IfAndThen new_ln(t)=lanen(t-1)-1；Otherwise,
new_ln(t)=lanen(t-1)；
R2. consider left road：IfAnd [Or vn(t)=0] andAndThen new_ln(t)=lanen(t-1)+1；
R3. carry out lane-change：In probability 1-pignoreUnder, make lanen(t)=new_ln(t)；
During process, successively each track is processed to avoid Vehicle Agent to touch by dextrosinistral order
Hit；
The distance between Adjacent vehicles before and after expression t vehicle n and Zuo Dao, right road and Ben Daonei；X is l, r
Or default, l is left road, r is right road, defaults to this road；Y be+or-,+for vehicle n front truck ,-for vehicle n rear car；lanen
T () represents the mark in t place track for the vehicle n, that is, 1,2 ..., one of N；new_lnWhen () represents lane-change t, vehicle n may
The track entering；
Y4. Vehicle Agent passes through crossing according to signal lighties, front vehicle position and vehicle speed condition and sails next section into, with
When traffic genetic module count certain section and sail the vehicle flowrate rolling away from into；
Signal lighties carry into execution a plan describing module offer signal time distributing conception；
When a certain phase place red light lights, this phase place does not cross the first car stagnation of movement of stop line, and subsequent vehicle is according to front
Car situation is slowly slowed down and is entered parking queueing condition；
When a certain phase place green light lights, first car starts, and mixes row data and downstream road section according to the non-machine in crossing
Queuing vehicle number and the degree of crowding, determine the average speed of vehicle；Wherein, the queuing vehicle number of downstream road section and the degree of crowding
There is provided by road network analysis module.
Preferably, also include scheme and describe step：The running of scheme describing module is as follows：
M1：The real time data of input section wagon detector, crossing signals timing protocol and traffic guidance scheme number
According to；And provide data to road network describing module；Wherein, the real time data of section wagon detector is in every 3 seconds to 60 second time
Unit vehicle pcu quantity, crossing signals timing protocol includes signal phase duration and the phase contrast at each crossing in road network
Data；
M2：Read the traffic information data of the Vehicle Agent that road network describing module provides, and be supplied to traffic evolution mould
Block and road network analysis module are analyzed, and then read analysis the data obtained in case M1 step input is used, export each crossing simultaneously and hand over
Logical delay and the evaluation index of whole road grid traffic delay；Wherein, road network describing module and traffic genetic module are to run parallel
, traffic genetic module and road network analysis module are that order is run；
M3：Repeat M1, M2 to be iterated computing.
Preferably, also include road network analytical procedure：The running of road network analysis module is as follows：
S1. the actual tolerance limit by updating section analyzes the cascading failure behavior in transportation network, is shown below：
Wherein, Cpij (0) represents from node i to the actual tolerance limit in the section of node j, and li (t) expression node i is in t
Flow, Cni represents the traffic capacity or the maximum queue length of node i；
S2. calculate degree of crowding FC and the road network performance E of traffic network, be shown below：
Wherein,Represent the shortest route time between j for the node i.
The Multiscale Fusion emulation technology means that the present invention provides, can make macroscopical and middle sight emulation more have real
Evolution Feature, also solves microscopic simulation due to being excessively finely difficult to complete the simulation problems under long-time and big region simultaneously,
And do not need to carry out the huge system integration to macroscopical, middle sight and microscopic simulation system, realize Multiscale Fusion in-circuit emulation
The miniaturization of system and kernelised.
The emulation technology each independent with existing macroscopical, middle sight, microcosmic and grand, in, the system integration skill of micro- three kinds of platforms
Art is compared, instant invention overcomes the single problem of each phantom yardstick in current analogue system, using Vehicle Agent skill
Art has organically merged macroscopical, middle sight, microcosmic, real and virtual emulation yardstick on space scale, melts in time scale
Close history, the real and following emulation yardstick, in same model system, city can have been handed over the multiple dimensioned method portrayed
Lead to and carry out continuous comprehensive simulating so that simulation result is more true and efficient, and compare for the offer of urban transportation Based Intelligent Control
Reliable traffic simulating system kernel.
Brief description
Fig. 1 is the operation schematic diagram of embodiment of the present invention analogue system.
Fig. 2 is the composition schematic diagram of Fig. 1 embodiment analogue system.
Fig. 3 is the simulation result figure one of application experiment case in Fig. 1 embodiment.
Fig. 4 is the simulation result figure two of application experiment case in Fig. 1 embodiment.
Specific embodiment
It is described in further detail with reference to the accompanying drawings and in conjunction with the embodiments to the present invention.But the invention is not restricted to
The example going out.
Embodiment
As shown in Figure 1 to Figure 2, the urban transportation complicated self-adapting network parallel simulation based on Multiscale Fusion for the present embodiment
System, including with lower module：
Parallel simulation evaluation module, for linkage simulation evaluation is carried out to target traffic network and its signal time distributing conception,
And provide Optimal Signals timing scheme；
Traffic demand module, for transferring the traffic flow data of target traffic network, and according to history macroscopic view OD data,
The microcosmic measured data of the middle sight statistical data at target road section and crossing and wagon detector is generating Vehicle Agent；Car
Intelligent body has the ability of making decisions on one's own, and specifically includes Path selection, speed modification, observes traffic lights and react, with speeding
Ability with lane change；
Wherein, history macroscopic view OD data is：It has been stored in the traffic flow data of historical record, specifically included target traffic road
Distributed data on room and time for the wagon flow starting point and ending point in net, the number of path between wagon flow starting point and ending point
According to, and the traffic flow distributed data between wagon flow starting point and ending point；
The middle sight statistical data at target road section and crossing is：For target road section and crossing, in its single traffic signal control
In cycle processed, according to drive towards from a crossing volume of traffic at another crossing, each flow direction of each crossing roll the volume of traffic, link travel away from
The statistical data analysis that time, the arrival rate of intersection vehicle flux and steering rate draw；Specifically include：The gathering around of target road section and crossing
Crowded degrees of data, impedance data, arrival rate data, steering rate data, fluctuation data, non-machine mix row data, and wherein fluctuation data is
The crossing vehicle that analogue system is given rolls cycle flow diagram away from and true crossing vehicle rolls the flat vehicle between cycle flow diagram away from
Pcu average error value, non-machine mix row data be crossing wagon flow in terms of unit vehicle pcu through stop line when in different pedestrians
Or non-maneuver interference under by the time, unit vehicle pcu is the car equivalent being converted by various types of automobiles；
The microcosmic measured data of wagon detector is：The real-time traffic amount of 3 seconds to the 9 seconds ranks being uploaded by wagon detector
Data；
Road network describing module, for setting up the geographical description data of target traffic network, is given by analyzing road net data
The traffic information data of Vehicle Agent；
Wherein, geographical description data includes the topological structure of road network, section, crossing, track, parking lot, bus platform；Road
Network data includes node degree, degree of communication, betweenness, shortest path；Traffic information data is i.e. in the following a period of time from currently
The crowded degrees of data of traffic network；
Traffic genetic module, makes a response according to micro environment factor for simulating vehicle intelligent body, according to macro environment
Factor and middle sight environmental factorss make motion decision-making, and the behavior of all Vehicle Agents is incorporated formation emulation in traffic network
Traffic scene；
Wherein, micro environment factor includes before and after's car, track, crossing, signal lighties；Macro environment factor includes imitative by macroscopic view
The traffic network congestion status that true analysis obtains；Middle environmental factorss of seeing include seeing, by middle, link proportion and the stream that simulation analysis obtain
Amount fluctuation parameters；Motion decision-making includes changing path, modification average speed；
Scheme describing module, for crossing signals timing protocol each in real-time update analogue system or traffic guidance side
Case data, and provide desired data to analogue system；Traffic guidance protocol for vehicle supervision department pass through Traffic Announcement or
The road conditions real time information that person crossing variable message board provides to vehicle driver, including：Congestion, restricted driving, green channel or traffic
Accident；
Road network analysis module, for road network gradual change cascade in traffic network in analysis following a period of time from currently
Tender spots, line impedance and crossing congestion coefficient；In the time serieses macroscopically analyzing each traffic key element in road network, in
Statistical vehicle flowrate and pedestrian's quantity in sight；And in the way of sequential update, traffic key element each in road network is carried out in temporal sequence
Line emulates；
Wherein, in road network, each traffic key element includes network, section, crossing, signal lighties, wagon flow, vehicle, pedestrian.
Using the urban transportation complicated self-adapting network parallel simulation method of system above, comprise the following steps：
Parallel simulation evaluation procedure：Parallel simulation evaluation module is according to the real time traffic data of target traffic network and history
Statistical data carries out simulation evaluation, provides operating index PI value under Setting signal timing scheme for the target traffic network, according to
This PI value repeatedly adjusts signal time distributing conception parameter, and provides the PI value after adjustment respectively, after obtain PI value optimum through comparing
Signal time distributing conception；
Wherein, operating index PI value for target traffic network by current demand signal timing scheme run the scheduled time after overall
Delay level, overall delay level is to stop vehicle number accumulation degree on a timeline in target traffic network；
Using climbing method, genetic algorithm or fuzzy logic algorithm during adjustment signal time distributing conception parameter.
Traffic demand step：Traffic demand module is according to the middle sight statistics at history macroscopic view OD data, target road section and crossing
Data provides the following vehicle flowrate of target traffic network, in conjunction with the microcosmic measured data of wagon detector, under micro environment
Simulate the Vehicle Agent of continuous service within continuous time；Meanwhile, according to history macroscopic view OD data, target road section and crossing
Middle sight statistical data, provide the operational factor of each Vehicle Agent；
Wherein, the operational factor of Vehicle Agent includes driving trace, average speed, security reaction time.
Road network describes step：The running of road network describing module is as follows：
L1. the nodal community matrix of target traffic network, road-net node connection matrix, road-net node connection attribute square are set up
Battle array, and artificial urban traffic network is built according to the data of above matrix；
L2. calculate and provide node degree matrix and node degree distribution matrix, betweenness matrix and the betweenness of target traffic network
Distribution matrix, each OD between path matrix；OD is to i.e. wagon flow starting point and ending point pair；
L3. calculated in real time and from currently not according to macroscopical OD data, middle sight statistical data and microcosmic measured data
Carry out the road network congestion data in a period of time；Road network congestion data includes vehicle density, journey time；
L4. the road conditions providing Vehicle Agent are calculated according to macroscopical OD data, middle sight statistical data and microcosmic measured data
Information data, and this data is loaded onto in the traffic network of artificial urban.
Traffic evolutionary step：By the concrete behavior of traffic genetic module evolution Vehicle Agent, including：Vehicle Agent root
Determine whether to change path according to the road network vehicle flowrate under macro-scale, crowded degrees of data and impedance data；Vehicle Agent according to
Whether the current road segment impedance data under medium measure and vehicle flow fluctuating data judging change average speed；Vehicle Agent according to
The non-machine in crossing under medium measure mixes row data judging and sails the average speed rolling crossing away from into；Vehicle Agent is according to micro-scale
Under before and after car and signal lighties situation determine whether to accelerate, slow down, stop or lane change.
The detailed process of traffic evolutionary step is as follows：
Y1. crowded degrees of data J is calculated according to road network vehicle flowrate data；Vehicle Agent calculates according to logit model and is directed to
The Path selection probability P in different paths, and judge whether to change road with reference to this Vehicle Agent default individual psychology threshold value R
Footpath, concrete judge process is as follows：
If J is less than R, and this Vehicle Agent to be presently in the Path selection probability P in path be maximum in all paths
Value, then this Vehicle Agent does not change path；If J is less than R, and Path selection probability P >=0.5 in all paths, then this vehicle
Intelligent body does not change path；If J is less than R, and the Path selection probability P in all paths<0.5, then this Vehicle Agent change road
Footpath, and select the maximum path of Path selection probability P；If J is more than R, this Vehicle Agent changes path, and selects path to select
Select the maximum path of probability P；
Wherein, crowded degrees of data J represents the overall congestion level of road network, is calculated as follows：
In formula, xaRepresent the stroke vehicle flowrate of path a, taRepresent the journey time of path a, t0aRepresent zero stream of path a
Impedance, that is, when path a wagon flow is amount, one car drives through the time required for this path；
Logit model is multinomial logit model M NL, Muti-nomial Logit, and formula is as follows:
I.e. Path selection probability PjRatio equal to path index function sums all in path j exponential function and road network；
In formula：PjFor the Path selection probability to path j for the Vehicle Agent, θ represents Vehicle Agent to traffic network
Familiarity, J represents the degree of crowding of path j,Represent the average degree of crowding in all paths in road network, Jm represents path m
The degree of crowding, path m is one of all paths in road network；
When in the road network information that scheme describing module provides containing traffic guidance protocol, θ value improves；
Y2. Vehicle Agent determines the average speed of itself according to link proportion data and vehicle flow fluctuating data；
Wherein, vehicle flow fluctuating data is the deviation being occurred in real time by vehicle checker between data and simulation and prediction data；
Link proportion data is calculated as follows：
In formula:xaRepresent the stroke vehicle flowrate of path a, taRepresent the running time of path a, t0aRepresent zero stream of path a
Impedance；UaRepresent the traffic capacity of path a, α, β are constant（Preferably α=0.15, β=4）.
Y3. Vehicle Agent according to the state of in front and back's car, signal lighties, the non-machine in crossing mix row data determine whether to accelerate,
Slow down, stop or lane change；
Wherein, the non-machine in crossing mixes row data by carrying out cluster analyses acquisition, signal lighties to crossing wagon detector data
Data is provided by scheme descriptive model；
Vehicle Agent controls speed according to Nasch model：
Step1:Accelerate, vn=min(vn+1,vmax)；
Step2:Slow down, vn=min(vn,dn)；
dn=xn+1-xn-lveh, dnRepresent the safe distance between the n-th car and front truck n+1；lvehRepresent length of wagon；
Step3:Random slowing down, with predetermined probabilities P0Random generation, vn=max(vn- 1,0), in order to simulate by uncertain because
Element（As road conditions, driver's phychology）The random deceleration causing；
Step4:Motion, xn=xn+vn, xn、vnRepresent position and the speed of the n-th car respectively；
The lane change model of Vehicle Agent is as follows：
Track right-to-left is designated as 1,2 successively ..., N, Vehicle Agent lane-change according to the following rules：
R1. consider right road：IfAndThen new_ln(t)=lanen(t-1)-1；Otherwise,
new_ln(t)=lanen(t-1)；
R2. consider left road：IfAnd [Or vn(t)=0] andAndThen new_ln(t)=lanen(t-1)+1；
R3. carry out lane-change：In probability 1-pignoreUnder, make lanen(t)=new_ln(t)；
During concrete process, successively each track is processed by dextrosinistral order to avoid Vehicle Agent to occur
Collision；
The distance between Adjacent vehicles before and after expression t vehicle n and Zuo Dao, right road and Ben Daonei；X be l, r or
Default, l is left road, r is right road, defaults to this road；Y be+or-,+for vehicle n front truck ,-for vehicle n rear car；lanen
T () represents the mark in t place track for the vehicle n, that is, 1,2 ..., one of N；new_lnWhen () represents lane-change t, vehicle n may
The track entering；
Y4. Vehicle Agent passes through crossing according to signal lighties, front vehicle position and vehicle speed condition and sails next section into, with
When traffic genetic module count certain section and sail the vehicle flowrate rolling away from into；
Signal lighties carry into execution a plan describing module offer signal time distributing conception；
When a certain phase place red light lights, this phase place does not cross the first car stagnation of movement of stop line, and subsequent vehicle is according to front
Car situation is slowly slowed down and is entered parking queueing condition；
When a certain phase place green light lights, first car starts, and mixes row data and downstream road section according to the non-machine in crossing
Queuing vehicle number and the degree of crowding, determine the average speed of vehicle；Wherein, the queuing vehicle number of downstream road section and the degree of crowding
There is provided by road network analysis module.
Scheme describes step：The running of scheme describing module is as follows：
M1：The real time data of input section wagon detector, crossing signals timing protocol and traffic guidance scheme number
According to；And provide data to road network describing module；Wherein, the real time data of section wagon detector is in every 3 seconds to 60 second time
Unit vehicle pcu quantity, crossing signals timing protocol includes signal phase duration and the phase contrast at each crossing in road network
Data；
M2：Read the traffic information data of the Vehicle Agent that road network describing module provides, and be supplied to traffic evolution mould
Block and road network analysis module are analyzed, and then read analysis the data obtained in case M1 step input is used, export each crossing simultaneously and hand over
Logical delay and the evaluation index of whole road grid traffic delay；Wherein, road network describing module and traffic genetic module are to run parallel
, traffic genetic module and road network analysis module are that order is run；
M3：Repeat M1, M2 to be iterated computing.
So can access real-time vehicle flowrate data, signal time distributing conception data, traffic guidance data, by section
Roll vehicle flowrate away from and carry out iteratively faster simulation calculating with sailing into of crossing, provide the traffic of urban road network in future time section
Road condition change situation, and feed back to outside coherent signal control system in time associated control parameters are modified, simultaneously
Amended parameter is input in the present embodiment analogue system to reappraise following traffic conditions again, repeatedly implements this
Process, till finding the control parameter of optimum.
Road network analytical procedure：The running of road network analysis module is as follows：
S1. the actual tolerance limit by updating section analyzes the cascading failure behavior in transportation network, is shown below：
Wherein, Cpij (0) represents from node i to the actual tolerance limit in the section of node j, and li (t) expression node i is in t
Flow, Cni represents the traffic capacity or the maximum queue length of node i；
Specifically, the value of li (t) is relevant with the arrival vehicle fleet size at crossing and signal time distributing conception, in real time data
Part, the vehicle flowrate data that vehicle flowrate digital independent vehicle checker uploads, in emulation evolution part, reads vehicle flowrate emulation between section
Propagation data.
From Cni, when node i is more than its ability in the flow of t, this node is then in congestion state, thus
The traffic capacity in section about is made to reduce.
S2. calculate degree of crowding FC and the road network performance E of traffic network, be shown below：
Wherein,Represent the shortest route time between j for the node i.
Additionally, degree of crowding FC and road network performance E can also be described as simultaneously left and right OD path the degree of crowding and list
Traffic efficiency between individual OD.
Present embodiments provide the urban transportation complicated self-adapting network parallel simulation technique means of Multiscale Fusion, can be
Play an important role in traffic intelligent control and substantially beneficial.The core as intellectual traffic control is evaluated in traffic control, for handing over
The logical Optimization about control parameter controlling provides effect assessment.Urban transportation is a complicated adaptive network, the friendship at each crossing
Logical control parameter and flow parameter all can produce different degrees of impact to its junction perimeter, circuit or even region, will do at present
To the Traffic Simulation Evaluation of urban road network, generally require very big server and support, and the disunity due to model, need
Be compared complexity multisystem Data Integration after just enable, effective parallel simulation of traffic control effect difficult to realize and
Feedback.And the urban transportation complicated self-adapting network parallel simulation technique means of the present embodiment Multiscale Fusion are then urban transportation
The miniaturization of road network emulation provides new method, its role is to, it can be various auto-adaptive control scheme real-time onlines
Single intersection Control effect appraisement in offer future time period, circuit Control effect appraisement and Region control effect assessment, and will
Evaluation result feed back to correlation control system so that control system the crossing parameter in urban road network can be made whole
The optimization of body.
Application experiment case：
Experimental subject：Nanjing Fujian road and three decorated archway crossings
（1）Set up simulated environment, describe relational structure relation, each section of road network by way of parameter inputs in detail
Shape facility, crossing shape facility, canal draw the geographical features such as type.Analyze the statistics experience number of current road network by preceding method
According to, such as different OD nodes between shortest path, second shortest path etc..
（2）By the macroscopical and middle current road segment resistance seeing section near simulation analysis Nanjing Fujian road and three decorated archway crossings
View the car in anti-, arrival rate, steering rate etc. flow data, the vehicle in conjunction with the section upstream crossing of Real-time Collection imports data, generates
Vehicle Agent.
（3）Vehicle Agent passes through macroscopic artificial, cognitive current road network is removed in middle sight emulation, section, the traffic shape on crossing
Condition is simultaneously reacted, and changes the driving habits of oneself according to certain statistical probability, changes expection and sails path and expected travel speed
Deng；And carry out in microscopic simulation environment between Vehicle Agent with speeding, lane change, the microcosmic evolution simulation such as overtake other vehicles, stop；Again
Using microscopic simulation as vehicle stress model, macroscopic artificial and middle sight emulation are loaded into wagon flow statistical data or vehicle form
Among.By Vehicle Agent, multiple yardsticks such as microscopic simulation, middle sight emulation, macroscopic artificial, real scale, virtual scale are melted
It is combined together.
（4）The time serieses of comprehensive each traffic parameter of history, the actual value of Current traffic parameter, right after Multiscale Fusion
The circumstances of future transportation carries out the fine emulation of microcosmic, and by signal timing dial interface, in real time or timing scheme is changed in timing, leads to
Cross optimized algorithm to iterate following traffic evolution under various signal timing dials of computing.
（5）Provide crossing delay and the stop frequency of current timing scheme by aforementioned emulation mode, and through aforementioned imitative
True method optimize after crossing delay and stop frequency correlation curve, thus being preferably provided at line for the timing scheme of road network
Multiscale Fusion emulation support.
Simulation result is as shown in Figure 3 and Figure 4.
Fig. 3 illustrates the 8 minutes downstream road junction in Nanjing Fujian road by being evolved using aforementioned analogue system and method
Correlation curve between simulation value and vehicle checker actual value, the 3rd signal period because the stop of buses causes emulation
The fluctuation of data, sees emulation and judges that such fluctuation is not enough to produce lasting emulation impact, therefore this subwave in then using
Move by Vehicle Agent record, see vehicle number in section and revised, but Vehicle Agent does not change itself lasting driving
Behavior.
Fig. 4 is the change using queue length after aforementioned emulation mode auxiliary, and result shows that stability substantially has improvement.
In addition to the implementation, the present invention can also have other embodiment.All employing equivalents or equivalent transformation shape
The technical scheme becoming, all falls within the protection domain of application claims.
Claims (9)
1. a kind of urban transportation complicated self-adapting network parallel simulation system based on Multiscale Fusion, is characterized in that, including with
Lower module：
Parallel simulation evaluation module, for carrying out linkage simulation evaluation to target traffic network and its signal time distributing conception, and gives
Go out Optimal Signals timing scheme；
Traffic demand module, for transferring the traffic flow data of target traffic network, and according to history macroscopic view OD data, target
The microcosmic measured data of the middle sight statistical data at section and crossing and wagon detector is generating Vehicle Agent；Vehicle intelligence
Can have the ability of making decisions on one's own by body, specifically include Path selection, speed modification, observe traffic lights and react, with speeding and becoming
The ability in road；
Road network describing module, for setting up the geographical description data of target traffic network, provides vehicle by analyzing road net data
The traffic information data of intelligent body；
In road network describing module, geographical description data includes the topological structure of road network, section, crossing, track, parking lot, public transport
Platform；Road net data includes node degree, degree of communication, betweenness, shortest path；Traffic information data i.e. one section of future from currently
The crowded degrees of data of traffic network in time；
The running of road network describing module is：
L1. the nodal community matrix of target traffic network, road-net node connection matrix, road-net node connection attribute matrix are set up,
And artificial urban traffic network is built according to the data of above matrix；
L2. calculate and provide node degree matrix and node degree distribution matrix, betweenness matrix and the betweenness distribution of target traffic network
Matrix, each OD between path matrix；OD is to i.e. wagon flow starting point and ending point pair；
L3. real-time and the future one from currently are calculated according to macroscopical OD data, middle sight statistical data and microcosmic measured data
Road network congestion data in the section time；Road network congestion data includes vehicle density, journey time；
L4. the traffic information providing Vehicle Agent is calculated according to macroscopical OD data, middle sight statistical data and microcosmic measured data
Data, and this data is loaded onto in the traffic network of artificial urban；
Traffic genetic module, makes a response according to micro environment factor for simulating vehicle intelligent body, according to macro environment factor
Make motion decision-making with middle sight environmental factorss, and the behavior of all Vehicle Agents is incorporated formation emulation traffic in traffic network
Scene；
Scheme describing module, for crossing signals timing protocol each in real-time update analogue system or traffic guidance scheme number
According to, and provide desired data to analogue system；Traffic guidance protocol passes through Traffic Announcement or road for vehicle supervision department
The road conditions real time information that mouth variable message board provides to vehicle driver, including：Congestion, restricted driving, green channel or traffic thing
Therefore；
Road network analysis module, for the fragility of road network gradual change cascade in traffic network in analysis following a period of time from currently
Point, line impedance and crossing congestion coefficient；In the time serieses macroscopically analyzing each traffic key element in road network, on middle sight
Statistical vehicle flowrate and pedestrian's quantity；And in the way of sequential update, traffic key element each in road network is imitated online in temporal sequence
Very.
2. the urban transportation complicated self-adapting network parallel simulation system based on Multiscale Fusion according to claim 1, its
Feature is, in traffic demand module, history macroscopic view OD data is：It has been stored in the traffic flow data of historical record, specifically included
Distributed data on room and time for the wagon flow starting point and ending point in target traffic network, wagon flow starting point and ending point it
Between path data, and the traffic flow distributed data between wagon flow starting point and ending point；
The middle sight statistical data at target road section and crossing is：For target road section and crossing, in its single traffic signalization week
In phase, according to the volume of traffic, when the rolling the volume of traffic, link travel away from of each flow direction of each crossing of driving towards another crossing from a crossing
Between, the statistical data analysis that draw of the arrival rate of intersection vehicle flux and steering rate；Specifically include：Target road section and crossing crowded
Degrees of data, impedance data, arrival rate data, steering rate data, fluctuation data, non-machine mix row data, and wherein fluctuation data is imitative
The crossing vehicle that true system is given rolls cycle flow diagram away from and true crossing vehicle rolls the unit vehicle between cycle flow diagram away from
Pcu average error value, non-machine mix row data be crossing wagon flow in terms of unit vehicle pcu through stop line when in different pedestrians or
Person's non-maneuver interference under by the time, unit vehicle pcu is the car equivalent being converted by various types of automobiles；
The microcosmic measured data of wagon detector is：The real-time traffic amount number of 3 seconds to the 9 seconds ranks being uploaded by wagon detector
According to.
3. the urban transportation complicated self-adapting network parallel simulation system based on Multiscale Fusion according to claim 1, its
Feature is, in traffic genetic module, micro environment factor includes before and after's car, track, crossing, signal lighties；Macro environment factor bag
Include the traffic network congestion status being obtained by macroscopic artificial analysis；Middle environmental factorss of seeing include seeing, by middle, the road that simulation analysis obtain
Section impedance and flowed fluctuation parameter；Motion decision-making includes changing path, modification average speed；In road network analysis module, in road network
Each traffic key element includes network, section, crossing, signal lighties, wagon flow, vehicle, pedestrian.
4. the urban transportation complicated self-adapting network parallel simulation side of a kind of employing claim 1 or analogue system described in 2 or 3
Method, is characterized in that, including parallel simulation evaluation procedure：Parallel simulation evaluation module is according to the real-time traffic number of target traffic network
Carry out simulation evaluation according to historical statistical data, provide operating index PI under Setting signal timing scheme for the target traffic network
Value, according to this PI value repeatedly adjustment signal time distributing conception parameter, and provides the PI value after adjustment respectively, after obtain PI through comparing
The optimum signal time distributing conception of value；Wherein, operating index PI value is run in advance by current demand signal timing scheme for target traffic network
Overall delay level after fixing time, overall delay level is to stop vehicle number accumulation on a timeline in target traffic network
Degree.
5. urban transportation complicated self-adapting network parallel simulation method according to claim 4, is characterized in that, also include traffic
Demand step：Traffic demand module provides target according to the middle sight statistical data at history macroscopic view OD data, target road section and crossing
The following vehicle flowrate of traffic network, in conjunction with the microcosmic measured data of wagon detector, simulates continuous under micro environment
The Vehicle Agent of continuous service in time；Meanwhile, the middle sight statistical number according to history macroscopic view OD data, target road section and crossing
According to providing the operational factor of each Vehicle Agent；Wherein, the operational factor of Vehicle Agent include driving trace, average speed,
The security reaction time.
6. urban transportation complicated self-adapting network parallel simulation method according to claim 4, is characterized in that, also include traffic
Evolutionary step：By the concrete behavior of traffic genetic module evolution Vehicle Agent, including：Vehicle Agent is according under macro-scale
Road network vehicle flowrate, crowded degrees of data and impedance data determine whether to change path；Vehicle Agent is according under medium measure
Whether current road segment impedance data and vehicle flow fluctuating data judging change average speed；Vehicle Agent is according under medium measure
The non-machine in crossing mixes row data judging and sails the average speed rolling crossing away from into；Vehicle Agent according to car before and after under micro-scale with
And signal lighties situation determines whether to accelerate, slows down, stops or lane change.
7. urban transportation complicated self-adapting network parallel simulation method according to claim 6, is characterized in that, traffic is developed step
Rapid detailed process is：
Y1. crowded degrees of data J is calculated according to road network vehicle flowrate data；Vehicle Agent calculates for difference according to logit model
The Path selection probability P in path, and judge whether to change path, tool with reference to this Vehicle Agent default individual psychology threshold value R
Body judge process is：
If J is less than R, and this Vehicle Agent to be presently in the Path selection probability P in path be maximum in all paths, then
This Vehicle Agent does not change path；If J is less than R, and Path selection probability P >=0.5 in all paths, then this Vehicle Agent
Do not change path；If J is less than R, and the Path selection probability P in all paths<0.5, then this Vehicle Agent change path, and select
The maximum path of routing footpath select probability P；If J is more than R, this Vehicle Agent changes path, and selects Path selection probability P
Maximum path；
Wherein, crowded degrees of data J represents the overall congestion level of road network, is calculated as follows：
In formula, xaRepresent the stroke vehicle flowrate of path a, taRepresent the journey time of path a, t0aRepresent zero flow impedance of path a,
I.e. when path a wagon flow is amount, one car drives through the time required for this path；
Logit model is multinomial logit model M NL, Muti-nomial Logit, and formula is:
I.e. Path selection probability PjRatio equal to path index function sums all in path j exponential function and road network；
In formula：PjFor the Path selection probability to path j for the Vehicle Agent, θ represents that Vehicle Agent is familiar with journey to traffic network
Degree, J represents the degree of crowding of path j,Represent the average degree of crowding in all paths in road network, Jm represents that path m's is crowded
Degree, path m is one of all paths in road network；
When in the road network information that scheme describing module provides containing traffic guidance protocol, θ value improves；
Y2. Vehicle Agent determines the average speed of itself according to link proportion data and vehicle flow fluctuating data；Wherein, traffic flow wave
Dynamic data is the deviation being occurred in real time by vehicle checker between data and simulation and prediction data；
Link proportion data is calculated as follows：
In formula:xaRepresent the stroke vehicle flowrate of path a, taRepresent the running time of path a, t0aRepresent zero flow impedance of path a；
UaRepresent the traffic capacity of path a, α, β are constant；
Y3. Vehicle Agent according to the state of in front and back's car, signal lighties, the non-machine in crossing mix row data determine whether to accelerate, slow down,
Stop or lane change；Wherein, the non-machine in crossing mixes row data by carrying out cluster analyses acquisition to crossing wagon detector data, letter
Signal lamp data is provided by scheme descriptive model；
Vehicle Agent controls speed according to Nasch model：
Step1:Accelerate, vn=min (vn+1,vmax)；
Step2:Slow down, vn=min (vn,dn)；
dn=xn+1-xn-lveh, dnRepresent the safe distance between the n-th car and front truck n+1；lvehRepresent length of wagon；
Step3:Random slowing down, with predetermined probabilities P0Random generation, vn=max (vn- 1,0), made by uncertain factor in order to simulate
The random deceleration becoming；
Step4:Motion, xn=xn+vn, xn、vnRepresent position and the speed of the n-th car respectively；
The lane change model of Vehicle Agent is：
Track right-to-left is designated as 1,2 successively ..., N, Vehicle Agent presses following rule lane-change：
R1. consider right road：IfAndThen new_ln(t)=lanen(t-1)-1；Otherwise, new_
ln(t)=lanen(t-1)；
R2. consider left road：IfAndOr vn(t)=0] and
AndThen new_ln(t)=lanen(t-1)+1；
R3. carry out lane-change：In probability 1-pignoreUnder, make lanen(t)=new_ln(t)；
During process, successively each track is processed to avoid Vehicle Agent to collide by dextrosinistral order；
The distance between Adjacent vehicles before and after expression t vehicle n and Zuo Dao, right road and Ben Daonei；X is l, r or default,
L is left road, r is right road, defaults to this road；Y be+or-,+for vehicle n front truck ,-for vehicle n rear car；lanenT () represents
Vehicle n in the mark in t place track, that is, 1,2 ..., one of N；new_ln(t) represent lane-change when vehicle n possibly into
Track；
Y4. Vehicle Agent passes through crossing according to signal lighties, front vehicle position and vehicle speed condition and sails next section into, hands over simultaneously
Logical genetic module counts certain section and sails the vehicle flowrate rolling away from into；
Signal lighties carry into execution a plan describing module offer signal time distributing conception；
When a certain phase place red light lights, this phase place does not cross the first car stagnation of movement of stop line, and subsequent vehicle is according to front truck feelings
Condition is slowly slowed down and is entered parking queueing condition；
When a certain phase place green light lights, first car starts, and mixes the row of row data and downstream road section according to the non-machine in crossing
Team's vehicle number and the degree of crowding, determine the average speed of vehicle；Wherein, the queuing vehicle number of downstream road section and the degree of crowding are by road
Net analysis module provides.
8. urban transportation complicated self-adapting network parallel simulation method according to claim 4, is characterized in that, also include scheme
Description step：The running of scheme describing module is：
M1：The real time data of input section wagon detector, crossing signals timing protocol and traffic guidance protocol；And
There is provided data to road network describing module；Wherein, the real time data of section wagon detector is the unit in every 3 seconds to 60 second time
Vehicle pcu quantity, crossing signals timing protocol includes signal phase duration and the phase data at each crossing in road network；
M2：Read the traffic information data of Vehicle Agent that road network describing module provides, and be supplied to traffic genetic module and
Road network analysis module is analyzed, and then reads analysis the data obtained in case M1 step input is used, exports each crossing traffic simultaneously and prolong
Evaluation index with whole road grid traffic delay by mistake；Wherein, road network describing module and traffic genetic module run parallel, and hand over
Logical genetic module and road network analysis module are that order is run；
M3：Repeat M1, M2 to be iterated computing.
9. urban transportation complicated self-adapting network parallel simulation method according to claim 4, is characterized in that, also include road network
Analytical procedure：The running of road network analysis module is：
S1. the actual tolerance limit by updating section analyzes the cascading failure behavior in transportation network, and its formula is：
Wherein, Cpij (0) represents from node i to the actual tolerance limit in the section of node j, and li (t) expression node i is in the stream of t
Amount, Cni represents the traffic capacity or the maximum queue length of node i；
S2. degree of crowding FC and the road network performance E of traffic network are calculated, its formula is：
Wherein,Represent the shortest route time between j for the node i.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310597678.8A CN103593535B (en) | 2013-11-22 | 2013-11-22 | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310597678.8A CN103593535B (en) | 2013-11-22 | 2013-11-22 | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103593535A CN103593535A (en) | 2014-02-19 |
| CN103593535B true CN103593535B (en) | 2017-02-22 |
Family
ID=50083673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310597678.8A Active CN103593535B (en) | 2013-11-22 | 2013-11-22 | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103593535B (en) |
Families Citing this family (62)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504170A (en) * | 2014-11-20 | 2015-04-08 | 中国科学院计算技术研究所 | Animated simulation method and system of vehicle |
| CN104464320B (en) * | 2014-12-15 | 2016-09-07 | 东南大学 | Based on true road network characteristic and the shortest path abductive approach of dynamic travel time |
| JP6398759B2 (en) * | 2015-02-03 | 2018-10-03 | 株式会社デンソー | Vehicle communication equipment |
| CN104866654B (en) * | 2015-05-06 | 2017-10-13 | 广州市交通规划研究院 | A kind of construction method of integrated urban dynamic traffic emulation platform |
| CN105046943B (en) * | 2015-06-12 | 2017-08-11 | 郑州轻工业学院 | The modeling method of traffic faults in a kind of vehicle movement simulation |
| CN105574243A (en) * | 2015-12-14 | 2016-05-11 | 北京航空航天大学 | High-performance mesoscopic traffic simulation method based on lane entrance time |
| TWI618647B (en) * | 2016-02-02 | 2018-03-21 | 財團法人資訊工業策進會 | System and method of detection, tracking and identification of evolutionary adaptation of vehicle lamp |
| CN105679037B (en) * | 2016-03-29 | 2019-03-08 | 招商局重庆交通科研设计院有限公司 | A kind of dynamic path planning method based on user's trip habit |
| CN105956333A (en) * | 2016-06-02 | 2016-09-21 | 苏州科技学院 | Method for modeling object-oriented three-level page table data structure of multi-resolution traffic system |
| CN106023629B (en) * | 2016-06-06 | 2018-12-25 | 西安电子科技大学昆山创新研究院 | A kind of path recommended method and device |
| CN106202606A (en) | 2016-06-22 | 2016-12-07 | 百度在线网络技术（北京）有限公司 | A kind of characteristic information acquisition methods simulating intelligent body and device |
| CN106355877B (en) * | 2016-08-23 | 2020-01-14 | 重庆大学 | Particle filter algorithm-based simulation estimation method for influence range of highway traffic incident |
| CN106529064A (en) * | 2016-11-15 | 2017-03-22 | 北京航空航天大学 | Multi-agent based route selection simulation system in vehicle online environment |
| CN106803148B (en) * | 2017-01-11 | 2019-09-06 | 成都通园科技有限公司 | A kind of data safety management system applied to smart city |
| CN106657381B (en) * | 2017-01-11 | 2019-10-18 | 深圳汇生通科技股份有限公司 | A kind of wisdom public traffic network augmented reality application system |
| CN106934107A (en) * | 2017-02-20 | 2017-07-07 | 北京百度网讯科技有限公司 | Traffic trip scenario building method, device, equipment and storage medium |
| CN106981204B (en) * | 2017-04-01 | 2018-08-24 | 三峡大学 | A kind of information processing method and device |
| CN107038862B (en) * | 2017-05-04 | 2020-02-14 | 东南大学 | Traffic distribution correction method under tail number restriction condition |
| CN107103142B (en) * | 2017-07-11 | 2019-12-03 | 交通运输部公路科学研究所 | Emulation technology is deduced towards the comprehensive traffic network operation situation of highway and the railway network |
| CN107293132A (en) * | 2017-08-14 | 2017-10-24 | 常州市规划设计院 | Intersections timing design method based on traffic delay |
| CN107908876A (en) * | 2017-11-16 | 2018-04-13 | 宁波工程学院 | Motor vehicle driven by mixed power operating mode Forecasting Methodology based on multiple dimensioned convolutional neural networks |
| CN108182812B (en) * | 2018-01-03 | 2020-01-21 | 浙江师范大学 | Urban road intersection ecological driving behavior optimization method based on reinforcement learning |
| CN108681806B (en) * | 2018-04-09 | 2021-07-27 | 东南大学 | Public transport network connectivity evaluation method based on traveler activities |
| CN109086910B (en) * | 2018-06-11 | 2021-10-15 | 北京工商大学 | Urban rail transit operation road network topological structure modeling method |
| CN109664884B (en) * | 2018-11-19 | 2020-06-09 | 江苏大学 | Extension self-adaptive lane keeping control method under variable vehicle speed |
| CN109326131A (en) * | 2018-11-28 | 2019-02-12 | 南京莱斯信息技术股份有限公司 | A kind of coordinating control of traffic signals method in adjust automatically period and split |
| CN109360429B (en) * | 2018-12-13 | 2021-03-23 | 武汉泰坦智慧科技有限公司 | Urban road traffic scheduling method and system based on simulation optimization |
| CN109637158B (en) * | 2018-12-27 | 2020-04-21 | 东南大学 | Intelligent vehicle confluence early warning method for expressway entrance ramp area |
| CN110196912B (en) * | 2019-04-15 | 2022-09-23 | 贵州电网有限责任公司 | Power grid archive parallel model construction method based on trust rule network |
| CN110211447A (en) * | 2019-05-28 | 2019-09-06 | 北京建筑大学 | A kind of batch microscopic traffic simulation laboratory automation realization method and system |
| CN110276541B (en) * | 2019-06-14 | 2023-04-28 | 上海理工大学 | Road organization evaluation method considering traveler adaptation process |
| US11682295B2 (en) | 2019-06-20 | 2023-06-20 | Here Global B.V. | Adjusting historical probe data for new road geometry |
| CN110472271A (en) * | 2019-07-01 | 2019-11-19 | 电子科技大学 | A kind of non-motorized lane Mixed contact construction method of microscopic traffic simulation |
| KR102309714B1 (en) * | 2019-12-30 | 2021-10-06 | 건국대학교 산학협력단 | Traffic simulation method based on parallel processing and appratus therefor |
| CN111275965B (en) * | 2020-01-20 | 2021-02-05 | 交通运输部科学研究院 | Real-time traffic simulation analysis system and method based on internet big data |
| CN111383453B (en) * | 2020-02-18 | 2022-05-20 | 山东摩西网络科技有限公司 | Traffic signal control on-line simulation and real-time tracking feedback system and operation method |
| CN111292535B (en) * | 2020-02-19 | 2020-09-22 | 河海大学 | Road network traffic state evaluation method for passenger travel in vehicle and road cooperation environment |
| CN111309845B (en) * | 2020-05-15 | 2020-09-08 | 北京全路通信信号研究设计院集团有限公司 | Rail transit road network macroscopic situation data display method and system |
| CN111680377B (en) * | 2020-06-11 | 2024-02-09 | 杭州海康威视数字技术股份有限公司 | Traffic situation simulation method, system and electronic equipment |
| CN111753424B (en) * | 2020-06-26 | 2023-08-01 | 北京百度网讯科技有限公司 | Simulation method, simulation device, electronic equipment and storage medium |
| CN111508240B (en) * | 2020-06-29 | 2020-09-22 | 之江实验室 | Traffic flow prediction method based on mixed feature mining |
| CN111881557B (en) * | 2020-07-01 | 2023-09-29 | 浙江中控信息产业股份有限公司 | Traffic flow simulation method based on average speed of road |
| CN112288272B (en) * | 2020-10-29 | 2021-05-07 | 北京交通大学 | Subway passenger flow regulation and control plan compilation method based on demand evolution and flow propagation |
| US11760379B2 (en) * | 2021-01-11 | 2023-09-19 | Toyota Research Institute, Inc. | Navigating an autonomous vehicle through an intersection |
| CN113268857B (en) * | 2021-04-21 | 2024-04-09 | 东南大学 | Urban expressway interweaving area microscopic traffic simulation method and device based on multiple intelligent agents |
| CN113192330B (en) * | 2021-04-26 | 2022-05-31 | 上海德衡数据科技有限公司 | Multi-agent-based vehicle management method, management system, device and medium |
| CN113297794A (en) * | 2021-05-28 | 2021-08-24 | 佳都科技集团股份有限公司 | Training method of simulator, prediction method of traffic road conditions and related device |
| CN113392106A (en) * | 2021-05-31 | 2021-09-14 | 东南大学 | Multi-agent traffic simulation data storage and updating method |
| CN113532449B (en) * | 2021-06-21 | 2023-11-21 | 阿波罗智联(北京)科技有限公司 | Intelligent traffic network acquisition method and device, electronic equipment and storage medium |
| CN113538905A (en) * | 2021-06-25 | 2021-10-22 | 华录智达科技股份有限公司 | Urban traffic operation simulation system |
| CN113312737B (en) * | 2021-07-28 | 2022-01-07 | 深圳市城市交通规划设计研究中心股份有限公司 | Multilayer multi-mode traffic simulation road network construction method |
| CN113724492A (en) * | 2021-07-29 | 2021-11-30 | 东南大学 | Method for predicting red light running behavior of rider by considering vehicle type and group |
| CN113792424B (en) * | 2021-09-06 | 2024-01-19 | 清华大学深圳国际研究生院 | Multi-lane changing method and system under heterogeneous traffic flow of automatic driving vehicle |
| CN113936461B (en) * | 2021-10-15 | 2024-04-19 | 吉林大学 | Simulation method and system for signal control intersection vehicle mixed running |
| CN114004077A (en) * | 2021-10-28 | 2022-02-01 | 腾讯科技（深圳）有限公司 | Traffic simulation conversion method, device, computer equipment and storage medium |
| CN114677837B (en) * | 2022-03-16 | 2023-08-25 | 杭州海康威视数字技术股份有限公司 | Traffic simulation method and device based on radar data and electronic equipment |
| CN114611337B (en) * | 2022-05-11 | 2022-09-23 | 深圳市城市交通规划设计研究中心股份有限公司 | Distributed traffic simulation method and device and storage medium |
| CN114937366B (en) * | 2022-07-22 | 2022-11-25 | 深圳市城市交通规划设计研究中心股份有限公司 | Traffic flow calculation method based on multi-scale traffic demand and supply conversion |
| CN116127786B (en) * | 2023-04-07 | 2023-06-30 | 南京大学 | System and method for measuring and calculating security state of slow traffic group based on city slow traffic simulation |
| CN116153079B (en) * | 2023-04-17 | 2023-06-23 | 湖南工商大学 | Road traffic flow prediction method and device based on vehicle track analysis |
| CN117556570A (en) * | 2023-09-28 | 2024-02-13 | 北京建筑大学 | Bus lane optimization design method, device, equipment and storage medium thereof |
| CN117709811B (en) * | 2024-02-05 | 2024-04-19 | 河北省交通规划设计研究院有限公司 | Urban planning system and method based on urban simulation |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101789182A (en) * | 2010-02-05 | 2010-07-28 | 北京工业大学 | Traffic signal control system and method based on parallel simulation technique |
| CN102393928A (en) * | 2011-11-22 | 2012-03-28 | 广州市交通规划研究所 | Traffic simulation integration system based on interaction usages of macroscopic, mid-scopic and microscopic traffic simulation platforms |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7321555B2 (en) * | 2003-04-16 | 2008-01-22 | International Business Machines Corporation | Multilevel analysis of self-similar network traffic |
-
2013
- 2013-11-22 CN CN201310597678.8A patent/CN103593535B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101789182A (en) * | 2010-02-05 | 2010-07-28 | 北京工业大学 | Traffic signal control system and method based on parallel simulation technique |
| CN102393928A (en) * | 2011-11-22 | 2012-03-28 | 广州市交通规划研究所 | Traffic simulation integration system based on interaction usages of macroscopic, mid-scopic and microscopic traffic simulation platforms |
Non-Patent Citations (1)
| Title |
|---|
| 基于多尺度融合的城市交通路网仿真系统研究;李晓峰等;《计算机科学》;20130228;第40卷(第2期);第108-111页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103593535A (en) | 2014-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103593535B (en) | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration | |
| Khamis et al. | Adaptive multi-objective reinforcement learning with hybrid exploration for traffic signal control based on cooperative multi-agent framework | |
| CN114495527B (en) | Internet-connected intersection vehicle road collaborative optimization method and system in mixed traffic environment | |
| Vasic et al. | Cellular automata simulation of traffic including cars and bicycles | |
| Ren et al. | Heterogeneous cellular automata model for straight-through bicycle traffic at signalized intersection | |
| CN105047057B (en) | A kind of high through street Macro-traffic Flow emulation mode of multi-user's multilane | |
| Alekszejenkó et al. | SUMO Based Platform for Cooperative Intelligent Automotive Agents. | |
| Burghout | Mesoscopic simulation models for short-term prediction | |
| Hou et al. | Large-scale vehicle platooning: Advances and challenges in scheduling and planning techniques | |
| Lansdowne | Traffic simulation using agent-based modelling | |
| Du et al. | Impacts of vehicle-to-everything enabled applications: literature review of existing studies | |
| Ioannou et al. | Integrated roadway/adaptive cruise control system: Safety, performance, environmental and near term deployment considerations | |
| Shang et al. | A new cellular automaton model for traffic flow considering realistic turn signal effect | |
| Fekete et al. | Methods for improving the flow of traffic | |
| Beza et al. | How PTV Vissim Has Been Calibrated for the Simulation of Automated Vehicles in Literature? | |
| Nguyen | Evaluation of SCATSIM–RTA Adaptive Traffic Network Simulation Model | |
| Zhang et al. | Cellular automaton traffic model considering driver’s reaction to velocity and headway distance with variable possibility of randomization | |
| Alkhatib et al. | A New System for Road Traffic Optimisation Using the Virtual Traffic Light Technology. | |
| Jin et al. | Simulation-driven multi-segment variable speed limits on freeway | |
| Othman et al. | Performance analysis to evaluate overtaking behavior on highways | |
| Wang et al. | Modeling driver behavior on urban streets | |
| Jamshidnejad et al. | Traffic simulation of an urban network system using agent-based modeling | |
| Alanazi et al. | Improving the performance of unsignalized t-intersections within CAVs mixed traffic | |
| CN114360290B (en) | Reinforced learning-based method for selecting vehicle group lanes in front of intersection | |
| CN113643528B (en) | Signal lamp control method, model training method, system, device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Parallel simulation system and method of urban traffic complex adaptive network based on multi scale fusion Effective date of registration: 20210702 Granted publication date: 20170222 Pledgee: The 14th Research Institute of China Electronics Technology Group Corporation Pledgor: NANJING LOPU Co.,Ltd. Registration number: Y2021320000112 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20220914 Granted publication date: 20170222 Pledgee: The 14th Research Institute of China Electronics Technology Group Corporation Pledgor: NANJING LOPU Co.,Ltd. Registration number: Y2021320000112 |