CN106205156B - A kind of intersection self-healing combination control method for the mutation of part lane flow - Google Patents
A kind of intersection self-healing combination control method for the mutation of part lane flow Download PDFInfo
- Publication number
- CN106205156B CN106205156B CN201610664684.4A CN201610664684A CN106205156B CN 106205156 B CN106205156 B CN 106205156B CN 201610664684 A CN201610664684 A CN 201610664684A CN 106205156 B CN106205156 B CN 106205156B
- Authority
- CN
- China
- Prior art keywords
- intersection
- operation conditions
- saturation degree
- traffic
- queue length
- 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
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
Abstract
The invention discloses a kind of intersection self-healing combination control methods for the mutation of part lane flow, by simultaneously using the operation conditions of multiple evaluation index evaluation analysis intersection traffics, based on minimal type evaluation index and BP neural network, more comprehensively accurately judge the intersection operation conditions grade under different situations, control is then optimized to the traffic signals of intersection as foundation using obtained intersection operation conditions grade.By monitoring intersection traffic operation conditions in real time, integrative design intersection system is enabled to carry out the adjustment of traffic signals according to real-time traffic noise prediction, so that intersection is more unimpeded, the whistle control system that method of the invention can be directed to the urban road intersection that part lane flow varies widely realizes self-healing control.
Description
Technical field
It is more particularly to a kind of the present invention relates to a kind of intersection self-healing combination control method for the mutation of part lane flow
Intersection traffic control system based on minimal type index Evaluation Method, BP neural network and particle swarm optimization algorithm from
Heal control method.
Background technology
Section and intersection constitute urban road system, and wherein intersection is the key that urban transportation.Due to not Tongfang
To traffic flow a series of traffic behaviors occur at the intersection, therefore intersection limits the traffic capacity of road, and
Still the place of traffic accident most easily occurs.So the key of solving road traffic problems is exactly to realize to the effective of intersection
Management.Traditional traffic signal control system has fixed controlling cycle using fixed timing, general its, it is impossible to according to reality
The situation of traffic flow changes timing of all directions etc. in real time, results in the waste of a large amount of time and resource.Now in road,
Vehicle increases suddenly especially on intersection, certain tracks causes intersection crowded, and operational efficiency is not high, can only generally wait
Treat that traffic police reaches floor manager and coordinates, not only the stand-by period is longer but also a large amount of manpower of consumption.So by detecting each side
Upward car flow information simultaneously optimizes the control of traffic signals, adjusts the control parameters such as the long green light time of each phase in real time,
Vehicle more reasonably can be controlled and guided, improves the traffic capacity of road.It, can especially in the case where lane flow is mutated
To realize the self-recovery of traffic control system by adjusting timing etc..
The signal control of urban road intersection refers to utilize the friendship for including beacon signal and traffic-police's hand signal
Messenger is detached conflicting traffic flow from the time, reaches guiding vehicle and pedestrian successfully by intersection
Target.By the control to urban road intersection traffic signals, vehicle etc. can be made by crossing, to reduce vehicle orderlyly
Queue waiting time improves the operational efficiency of intersection.Common urban road intersection control signal is light signal, is led to
Cross red, yellow, green three colour lamp light realizes the distribution weighed to vehicles and pedestrians.Green light is allows to pass through;Amber light is believed for warning
Number;Red light is no through traffic.It will be counted within green time yellow time in this method.
One of common theoretical method of research urban road intersection system is has model method.It is traditional based on statistics
Delay Model is proposed by Webster and Akcelik, and should be used extensively in the fixation timing strategy of Single Intersection.
Webster delay formula are the formula of a certain phase track delay of a very classical calculating, since it calculates the preferable institute of effect
To be widely used by people.The index of intersection traffic is evaluated other than mean delay, saturation degree and average queue length
The operation conditions of intersection can more directly be reacted.But when intersection is crowded, saturation degree no longer shows transport need
Feature, delay is also not suitable for optimization aim as traffic signalization.At this time in order to avoid queue length is excessive caused
Overflow and the influence to upstream intersection, it should control queue length, evacuate intersection as early as possible and be detained queuing.
The evaluation method of minimal type index:When in multiple indexs of selection all indexs all level off to 0 when, the operation of system
It is considered as to make overall plans and coordinate, so comprehensive evaluation result is superior.When all indexs all level off to 1 when, the operation of system is considered
It is to make overall plans and coordinate, but comprehensive evaluation result is poor.When some in index are relatively large, and other are relatively small, then it is assumed that this
One system of sample is uncoordinated, and comprehensive evaluation result is general.
Artificial neural network is complicated nonlinear network system, it can solve the problems, such as that traditional algorithm cannot.Nerve net
Network has the ability of quick processing information and the ability of very strong processing unascertained information.When selecting suitable parameter, it
Relatively small square, realization complexity and a nonlinear can be converged to.The research of artificial neural network starts from
1943, by the development of more than 60 years, it has been widely used in engineering research field.At present, the development of neural network mainly collects
In applying.Multilayer feedforward neural network is one of most widely used neural network.It is a kind of multitiered network, is had good
None-linear approximation ability and it is simple in structure the advantages of.BP neural network is in the side such as pattern-recognition, intelligent control, classification, prediction
Face achieves significant achievement.
At present, the main research direction of scholar of China's research traffic control optimizes for vehicle detection and signal.Detection side
Based on identification in mode, using image processing techniques, vehicle flowrate size at the same level recognizes situation violating the regulations etc..In signal optimization, state
Interior scholar tends to, using biological heuritic approach, optimizing be carried out to delay time at stop of intersection etc..Common algorithm has:The shoal of fish
Algorithm, ant group algorithm, genetic algorithm, particle cluster algorithm etc..Particle swarm optimization algorithm (PSO) is one of typical case of swarm intelligence,
Referred to as PSO algorithms or particle cluster algorithm.Particle cluster algorithm is complete by one kind of doctor Eberhart and doctor's Kennedy invention
Office's optimization evolution algorithm.System initialization is one group of RANDOM SOLUTION at the beginning, and iteration searches for optimal value in multiple times later.And for
Determining Delay Model, needs to adjust timing cause delay minimum, that is, global optimum to be found, so particle cluster algorithm can be with
Used in the resolving to Delay Model.Compared with ant group algorithm, particle group optimizing can effectively optimization system parameter, so as to
It is enough quickly to approach optimal solution.It is that individual can make full use of itself and population that particle cluster algorithm, which has the key of excellent specific property,
Experience adjustments itself state, carry out the iteration of next step.So PSO algorithms are suitable for solving the optimization of some continuous functions
Problem.Compared to genetic algorithm, in most instances, PSO can quickly converge on optimal solution.Therefore, by particle cluster algorithm
More rapidly simple for Real-time solution signal timing dial, control effect is more preferable.
Self-healing was a biological concept originally, and self-healing refers to the self-recovery mechanism of a kind of stabilization and balance, institute
With self-healing, that is, self-recovery balance or stablize.Traffic signalization is introduced into, just refers to have occurred when intersection
Some accidents cause in the case that its traffic capacity is affected, only to adjust its traffic signals by certain control program
Configuration so as to restore the normal operation of intersection.
Invention content
The technical problems to be solved by the invention are to provide a kind of intersection self-healing for the mutation of part lane flow
Control method.
The present invention uses following technical scheme to solve above-mentioned technical problem:
The present invention provides a kind of intersection self-healing combination control method for the mutation of part lane flow, is as follows
Step 1, while consider three intersection saturation degree, mean delay and average queue length indexs, utilize minimum
Type metrics evaluation method builds intersection overall target;
Step 2, according to intersection saturation degree, intersection traffic operation conditions is divided into five grades, wherein, divide according to
According to for:If saturation degree (0,0.3] in the range of intersection traffic operation conditions grade be 1, if saturation degree (0.3,0.6] model
Then intersection traffic operation conditions grade is 2 in enclosing, if saturation degree (0.6,0.8] in the range of intersection traffic operation conditions
Grade is 3, if saturation degree (0.8,1] in the range of intersection traffic operation conditions grade be 4, if saturation degree be more than 1 if hand over
Prong traffic noise prediction grade is 5;
Step 3, from several groups of intersection saturation degrees of intersection traffic extracting data, mean delay and average queue length
Value, obtains corresponding intersection comprehensive index value and intersection traffic operation conditions grade;
Step 4, the intersection saturation degree obtained in step 3, mean delay, average queue length, intersection synthesis are referred to
The training sample of scale value and intersection traffic operation conditions grade as BP neural network, the training for carrying out BP neural network are learned
It practises, wherein, intersection saturation degree, mean delay, the input that average queue length is neural network, intersection traffic operation conditions
Grade is the output of neural network;
Step 5, the traffic data of current intersection is acquired in real time, extracts current intersection saturation degree, mean delay peace
The BP neural network that training is completed in input step 4 after equal queue length, obtains current intersection traffic operation conditions grade;
Step 6, if the current intersection traffic operation conditions grade that step 5 obtains is less than or equal to 3, current demand signal is kept
Timing;If current intersection traffic operation conditions grade is 4, with the minimum optimization aim of intersection overall target, to intersecting
Mouth signal timing dial optimizes;It is minimum with the intersection queue length that be averaged if current intersection traffic operation conditions grade is 5
For optimization aim, intersection signal timing is optimized.
Scheme is advanced optimized as the present invention, the expression formula of intersection overall target is:
In formula, r is intersection overall target, and x, Q, d is respectively intersection saturation degree, average queue length, mean delay;
The expression formula of intersection saturation degree is:
In formula, xiFor the saturation degree of i-th of phase,ωiFor the weight of i-th of phase,X
The sum of saturation degree for four phases;qiFor the vehicle arriving rate of i-th of phase, ciFor the traffic capacity of i-th of phase, siFor
The saturation volume rate of i-th of phase, CiFor the period of i-th of phase, giGreen time for i-th of phase;
The be averaged expression formula of queue length of intersection is:
In formula, QiFor the queue length of i-th of phase,
The expression formula of intersection mean delay is:
In formula, diFor the delay in i-th of track,C is Intersections
Cycle duration, giFor the green time of i-th of phase, qiFor the vehicle arriving rate in i-th of track, SiFor the full of i-th track
And flow rate.
Scheme is advanced optimized as the present invention, with the minimum optimization aim of intersection overall target, to intersecting message
Number timing optimizes, specially:According to the expression formula of intersection overall targetUtilize population
Optimization algorithm Real-time solution causes the signal timing dial of intersection overall target minimum, and current demand signal timing is optimized with this.
Scheme is advanced optimized as the present invention, be averaged the minimum optimization aim of queue length with intersection, to intersection
Mouth signal timing dial optimizes, specially:Be averaged the expression formula of queue length according to intersectionUtilize particle
Colony optimization algorithm Real-time solution so that intersection is averaged the signal timing dial of queue length minimum, carries out current demand signal timing with this
Optimization.
As the scheme that advanced optimizes of the present invention, the intersection traffic data in step 3 pass through VISSIM analog crossovers
Acquisition is led in oral sex.
The present invention compared with prior art, has following technique effect using above technical scheme:The present invention passes through simultaneously
The traffic circulation state of three Traffic Evaluation metrics evaluation intersections is considered, compared with existing considers single evaluation traffic indicators
More comprehensively, the practical operation situation of intersection has accurately been reacted, and has obtained the expression formula of comprehensive evaluation index;Utilize training
Good BP neural network can be used for intersection traffic according to the operating condition of real time traffic data Fast Evaluation intersection
Monitoring in real time;It is excellent using population when vehicle mutations in track certain in intersection cause intersection operation conditions poor
Change algorithm rapid solving optimization object function and carry out timing designing;For the intersection of different operation conditions grades, choose respectively
Different optimization object functions, more corresponds to actual needs, and is mutated by the adjustment part lane flow to traffic signals
Intersection rapidly can voluntarily restore unimpeded.
Description of the drawings
Fig. 1 is the block diagram of control method in the present invention.
Fig. 2 is the control object i.e. VISSIM simulation model figures of four phase intersections in the present invention.
Fig. 3 is the training error curve graph of BP neural network in the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings and specific embodiment is described in further detail technical scheme of the present invention:
As shown in Figure 1, the present invention is devised based on minimal type index Evaluation Method, BP neural network and Particle Swarm Optimization
The self-healing combination control method of the intersection traffic control system of method, the VISSIM emulation moulds for the four phase intersections studied
Type is as shown in Figure 2.
First, for the method for solving of four phase intersection operation conditions comprehensive evaluation indexs
1) selection of three Traffic Evaluation indexs:
Saturation degree is one of index of most important Assessment of Serviceability of Roads, and calculation formula q/c, wherein q are that vehicle reaches
Rate, c are the traffic capacitys.Saturation value is bigger, and the service level of road is poorer.In the present invention, the U.S. is used for reference《Current handbook》
Classification to Assessment of Serviceability of Roads, with reference to Chinese practice situation and existing research, by the congestion levels of Chinese road and service water
Following five class is bisected into, as shown in table 1:
Grade 1 is that first service is horizontal:The coast is clear, service level is fine, road saturation degree (0,0.3] in the range of;
Grade 2 is that second service is horizontal:The coast is clear, service level is fine, road saturation degree (0.3,0.6] range
It is interior;
Grade 3 is third service level:Road has a little crowded, and service level is relatively preferable, road saturation degree (0.6,
0.8] in the range of;
Class 4 i.e. the 4th service level:Road is crowded, and service level is poor, road saturation degree (0.8,1] in the range of;
Class 5 i.e. the 5th service level:Road is seriously crowded, and service level is very poor, and road saturation degree is more than 1.
1 intersection saturation degree of table and the intersection operation conditions grade table of comparisons
In the present invention, for four phase intersections, a crucial track is respectively selected from four phases, by the key track
Saturation degree of the saturation degree as respective phase, then using the weighted sum of the saturation degree of four phases as the saturation degree of intersection.
The expression formula of intersection saturation degree is:
In formula, xiFor the saturation degree of i-th of phase, unit veh/s,ωiPower for i-th of phase
Weight,X is the sum of saturation degree of four phases;qiFor the vehicle arriving rate of i-th of phase, unit veh/s;ciIt is i-th
The traffic capacity of a phase, unit veh/s;siFor the saturation volume rate of i-th of phase, unit veh/s;CiWeek for i-th of phase
Phase, unit s;giFor the green time of i-th of phase, unit s.
In the present invention, for four phase intersections, a crucial track is respectively selected from four phases, by the key track
Queue length of the queue length as respective phase, then queue length weighted sum being averaged as intersection using four phases
Queue length.If the average queue length of intersection is bigger, traffic efficiency is lower, and traffic is poorer.Pass through distribution one
The green time of four phases in a crosspoint, to reduce the average queue length of entire intersection to the greatest extent.
The be averaged expression formula of queue length Q of intersection is:
In formula, QiFor the queue length of i-th of phase, unit veh,
It is averaged queue length Q according to intersection, intersection can be obtained and be averaged queue length and intersection operation conditions etc.
Grade control case, as shown in table 2.
2 intersection of table is averaged queue length and the intersection operation conditions grade table of comparisons
Delay refers to the loss of time of the vehicle by intersection, it can reflect fuel consumption in vehicles, in operational process
The loss of time and driver level of comfort.Therefore, delay is the index of most common evaluation signalized intersections.Mean delay is
The important parameter of traffic efficiency is evaluated, mean delay is bigger, and road traffic operating status is poorer.
In the present invention, for four phase intersections, a crucial track is respectively selected from four phases, by the key track
Mean delay of the mean delay as respective phase, then mean delay weighted sum being averaged as intersection using four phases
Delay.
The expression formula of intersection mean delay d is:
In formula, diFor the delay in i-th of track,C is Intersections
Cycle duration, giFor the green time of i-th of phase, qiFor the vehicle arriving rate in i-th of track, SiFor the full of i-th track
And flow rate.
According to the expression formula of intersection mean delay d, intersection mean delay and intersection operation conditions etc. can be obtained
Grade control case, as shown in table 2.
3 intersection mean delay of table and the intersection operation conditions grade table of comparisons
2) using minimum type metrics evaluation method, by three intersection saturation degree, mean delay, average queue length evaluations
Index processing is an intersection overall target, and the expression formula of intersection overall target isWherein,
R is intersection overall target.Intersection traffic operation conditions grade corresponding to three indexs is f (x, Q, d).
Intersection saturation degree, mean delay, average three evaluation indexes of queue length are transported corresponding to different intersection traffics
The threshold value of row grade, as shown in table 4.
4 three evaluation indexes of table correspond to the threshold value of different traffic circulation grades
After three evaluation indexes are normalized, three evaluation indexes correspond to different intersection traffics operation etc.
The threshold value of grade, as shown in table 5.
5 three evaluation indexes of table correspond to the threshold value of different intersection traffic Operation class
According to the expression formula of intersection overall targetCan obtain intersection overall target with
The intersection operation conditions grade table of comparisons, as shown in table 6.
6 intersection overall target of table and the intersection operation conditions grade table of comparisons
Intersection saturation degree, mean delay, average input of three evaluation indexes of queue length as BP neural network, are handed over
Outputs of the prong operation conditions grade f (x, Q, d) as BP neural network.Led to using VISSIM analog crossover oral sexes, if obtaining
Dry group intersection saturation degree, mean delay, average queue length and corresponding intersection operation conditions grade are as BP nerve nets
BP neural network after training is used for quickly determining intersection traffic operation conditions etc. according to traffic data by the training sample of network
Grade, avoids numerous and diverse calculating, rapidly and accurately judges intersection traffic operation conditions.
In the present invention, run by the vehicle of VISSIM simulation softwares emulation intersection, obtain 80 groups of traffic datas.Select it
In 60 groups as learning sample, be in addition used as test sample for 20 groups, the accuracy of the BP neural network after training can reach
89%.If hidden layer has 10 layers, BP discriminations can reach more than 89%.The training error curve graph of BP neural network is such as
Shown in Fig. 3.
The BP neural network completed using training, by the intersection saturation degree obtained in real time, mean delay and average queuing
Length quickly obtains intersection traffic operation conditions grade, carries out the real-time monitoring of intersection operation conditions as input.Work as prison
When measuring intersection operation conditions grade less than or equal to 3, current demand signal timing is kept.When intersection operation conditions monitored etc.
When grade is 4, by intersection overall target minimum target as an optimization, using particle swarm optimization algorithm to intersection overall target mould
Type is solved, and obtains the traffic signal timing so that comprehensive evaluation index minimum, current demand signal timing is optimized.Work as intersection
When mouthful operation conditions grade is 5, intersection is averaged queue length minimum target as an optimization, using particle cluster algorithm to intersecting
The average queue length model of mouth is solved, and is obtained so that the traffic signal timing of average queue length minimum, matches current demand signal
When optimize.For the intersection of different traffic conditions, different majorized functions is selected more to correspond to actual needs, can also obtained
Better control effect.
Particle swarm optimization algorithm (PSO) is one of typical case of swarm intelligence, referred to as PSO algorithms or particle cluster algorithm.PSO
The basic thought of algorithm is as follows:Each particle is treated as to the solution of required majorization of solutions problem.
First, these particles of system random initializtion, and a fitness function is set in advance as evaluation particle
Standard.Then, each particle is allowed to fly in potential solution space, and this transformable amount updates the side of its movement using speed
To and distance.Usual particle tracks current best particle, stops after certain condition is reached always by multiple search
Iteration updates to obtain optimum solution.Particle follows two extreme values in the search process in each stage, is particle up to now respectively
Oneself obtained optimum solution of search, there are one be optimum solution that group entire so far searches.
The realization step of particle swarm optimization algorithm based on matlab language is as follows:
(1) it initializes
First, the parameters in PSO algorithms are set:The upper and lower limit U of search spacedAnd Ld, Studying factors k1, k2, it is used to
Sex factor ω, the maximum iteration T of algorithmmax, the velocity interval [v of particlemin,vmax];Then, random initializtion particle
Position piAnd its speed vi, current location is the personal best particle P as each particlei, global pole is found out from individual extreme value
Value, records the particle serial number g of the value and its position Pg(being global optimum position).
(2) each particle is evaluated
According to previously selected fitness function, the corresponding numerical value of each particle is obtained.If this particle of this numeric ratio
Current individual extreme value is more excellent, then replaces personal best particle P with the position of this particlei, while refresh individual extreme value.Such as
In the individual extreme value of all particles of fruit it is best than current global extremum also than get well, then replaced with the position of the best particle
Global optimum position Pg, while change global extremum and its serial number g.
(3) the state update of particle
According to speed formula and location formula, each particle corresponding parametric values is refreshed respectively.If vi> vmaxBy its
It is set as vmaxIf vi<vminIt is set to vmin。
(4) it checks whether to meet termination condition
If the number of iteration has reached preset maximum times value Tmax, then it is no longer iterated and provides
Otherwise optimal solution goes to step (2).
In emulation, the parameters setting in PSO algorithms is as follows:The upper and lower limit U of search spacedAnd LdIt is set to
120, -120, Studying factors k1、k22 are all set as, inertial factor ω is set as 0.729, the maximum iteration T of algorithmmaxIt is set as
2000, the velocity interval [v of particlemin,vmax] be set as [- 20,20].
Work as intersection it can be seen from optimization timing effect comparison table of the table 7 for different operation conditions level crossings mouths
In the case that part lane flow mutation causes intersection operation conditions poor, the control effect of fixed timing scheme is very
Difference, the numerical value of comprehensive evaluation index is very big, and corresponding intersection operation conditions grade is just very big, and optimizes timing scheme and pass through choosing
Optimization object function is taken, the timing of optimization is provided using particle swarm optimization algorithm.Control of the timing for intersection, reason will be optimized
By the operation conditions that can above improve intersection, include reducing the saturation degree of intersection under normal circumstances, when reducing mean delay
Between and average queue length, can be fast and then with the minimum main target of optimization of average queue length in the case of especially crowded
Fast dissipation queuing vehicle improves traffic efficiency.
Table 7 is directed to the optimization timing effect comparison table of different operation conditions level crossings mouths
The above, the only specific embodiment in the present invention, but protection scope of the present invention is not limited thereto are appointed
What be familiar with the people of the technology disclosed herein technical scope in, it will be appreciated that the transformation or replacement expected should all be covered
Within the scope of the present invention, therefore, protection scope of the present invention should be subject to the protection domain of claims.
Claims (4)
1. a kind of intersection self-healing combination control method for the mutation of part lane flow, which is characterized in that be as follows
Step 1, while consider three intersection saturation degree, mean delay and average queue length indexs, referred to using minimum type
Mark evaluation assessment structure intersection overall target;Wherein, the expression formula of intersection overall target is:
In formula, r is intersection overall target, and x, Q, d is respectively intersection saturation degree, average queue length, mean delay;
The expression formula of intersection saturation degree is:
In formula, xiFor the saturation degree in i-th of track,ωiFor the weight of i-th of phase,X is four
The sum of saturation degree of a phase;qiFor the vehicle arriving rate in i-th of track, ciFor the traffic capacity in i-th of track, SiIt is i-th
The saturation volume rate in track, CiFor the period of i-th of phase, giGreen time for i-th of phase;
The be averaged expression formula of queue length of intersection is:
In formula, QiFor the queue length in i-th of track,
The expression formula of intersection mean delay is:
In formula, diFor the delay in i-th of track,C is the week of Intersections
Phase duration, giFor the green time of i-th of phase, qiFor the vehicle arriving rate in i-th of track, SiSaturated flow for i-th of track
Rate;
Step 2, according to intersection saturation degree, intersection traffic operation conditions is divided into five grades, wherein, partitioning standards
For:If saturation degree (0,0.3] in the range of intersection traffic operation conditions grade be 1, if saturation degree (0.3,0.6] range
Inside then intersection traffic operation conditions grade is 2, if saturation degree (0.6,0.8] in the range of intersection traffic operation conditions etc.
Grade for 3, if saturation degree (0.8,1] in the range of intersection traffic operation conditions grade be 4, intersect if saturation degree is more than 1
Mouth traffic noise prediction grade is 5;
Step 3, from several groups of intersection saturation degrees of intersection traffic extracting data, mean delay and average queue length value,
Obtain corresponding intersection comprehensive index value and intersection traffic operation conditions grade;
Step 4, by the intersection saturation degree obtained in step 3, mean delay, average queue length, intersection comprehensive index value
And training sample of the intersection traffic operation conditions grade as BP neural network, the training study of BP neural network is carried out,
Wherein, intersection saturation degree, mean delay, the input that average queue length is neural network, intersection traffic operation conditions etc.
Grade is the output of neural network;
Step 5, the traffic data of current intersection is acquired in real time, extracts current intersection saturation degree, mean delay and average row
The BP neural network that training is completed in input step 4 after team leader's degree, obtains current intersection traffic operation conditions grade;
Step 6, if the current intersection traffic operation conditions grade that step 5 obtains is less than or equal to 3, current demand signal is kept to match
When;If current intersection traffic operation conditions grade is 4, with the minimum optimization aim of intersection overall target, to intersection
Signal timing dial optimizes;It is minimum with the intersection queue length that is averaged if current intersection traffic operation conditions grade is 5
Optimization aim optimizes intersection signal timing.
2. a kind of intersection self-healing combination control method for the mutation of part lane flow according to claim 1, special
Sign is, with the minimum optimization aim of intersection overall target, intersection signal timing is optimized, specially:According to friendship
The expression formula of prong overall targetSo that intersection is comprehensive using particle swarm optimization algorithm Real-time solution
The signal timing dial of index minimum is closed, current demand signal timing is optimized with this.
3. a kind of intersection self-healing combination control method for the mutation of part lane flow according to claim 1, special
Sign is, is averaged the minimum optimization aim of queue length, intersection signal timing is optimized, specially with intersection:Root
Be averaged the expression formula of queue length according to intersectionCause intersection using particle swarm optimization algorithm Real-time solution
The signal timing dial of average queue length minimum, optimizes current demand signal timing with this.
4. a kind of intersection self-healing combination control method for the mutation of part lane flow according to claim 1, special
Sign is that the intersection traffic data in step 3 lead to acquisition by VISSIM analog crossover oral sexes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610664684.4A CN106205156B (en) | 2016-08-12 | 2016-08-12 | A kind of intersection self-healing combination control method for the mutation of part lane flow |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610664684.4A CN106205156B (en) | 2016-08-12 | 2016-08-12 | A kind of intersection self-healing combination control method for the mutation of part lane flow |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106205156A CN106205156A (en) | 2016-12-07 |
CN106205156B true CN106205156B (en) | 2018-06-26 |
Family
ID=57514458
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610664684.4A Active CN106205156B (en) | 2016-08-12 | 2016-08-12 | A kind of intersection self-healing combination control method for the mutation of part lane flow |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106205156B (en) |
Families Citing this family (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106683442A (en) * | 2016-12-28 | 2017-05-17 | 安徽科力信息产业有限责任公司 | Multi-index based intersection signal timing plan evaluation method |
CN108399763B (en) * | 2018-03-07 | 2021-04-27 | 辽宁工业大学 | Intersection traffic signal lamp control algorithm based on neural network |
CN108320516B (en) * | 2018-04-08 | 2020-06-09 | 华中师范大学 | Road traffic capacity evaluation method based on sharp point mutation and quantum particle swarm optimization |
CN108417057B (en) * | 2018-05-15 | 2023-06-23 | 哈尔滨工业大学 | Intelligent signal lamp timing system |
CN109191875B (en) * | 2018-09-17 | 2020-12-29 | 杭州中奥科技有限公司 | Signal control scheme generation method and device |
CN110060489B (en) * | 2019-03-27 | 2020-10-27 | 浙江工业大学 | Signal lamp timing scheme recommendation method based on neural network |
CN110136455B (en) * | 2019-05-08 | 2021-06-25 | 济南大学 | Traffic signal lamp timing method |
CN110335478B (en) * | 2019-07-10 | 2021-01-12 | 江苏航天大为科技股份有限公司 | Cross-subarea linkage control method based on deep learning |
CN112668815A (en) * | 2019-09-30 | 2021-04-16 | 北京国双科技有限公司 | Automobile data processing method and device |
CN111127892A (en) * | 2019-12-27 | 2020-05-08 | 北京易华录信息技术股份有限公司 | Intersection timing parameter optimization model construction and intersection signal optimization method |
CN111899537B (en) * | 2020-07-01 | 2023-03-28 | 山东摩西网络科技有限公司 | Intersection signal control mobile tuning device and method based on edge calculation |
CN112632127B (en) * | 2020-12-29 | 2022-07-15 | 国华卫星数据科技有限公司 | Data processing method for real-time data acquisition and time sequence of equipment operation |
CN113112827B (en) * | 2021-04-14 | 2022-03-25 | 深圳市旗扬特种装备技术工程有限公司 | Intelligent traffic control method and intelligent traffic control system |
CN113936454B (en) * | 2021-09-22 | 2023-02-21 | 上海宝康电子控制工程有限公司 | Information processing method and system based on track fusion |
CN114898555B (en) * | 2022-05-09 | 2024-01-05 | 华路易云科技有限公司 | Comprehensive evaluation method for road intersection |
CN115083165B (en) * | 2022-06-29 | 2023-11-28 | 北京智城交建科技有限公司 | Smart city intersection traffic command control method, device and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3844207B2 (en) * | 2001-12-28 | 2006-11-08 | オムロン株式会社 | Information processing apparatus and method, and program |
CN101042805A (en) * | 2007-03-09 | 2007-09-26 | 吉林大学 | Control method for single crossing mixed traffic signal |
CN101727746A (en) * | 2009-09-18 | 2010-06-09 | 苏州大学 | Method for dynamically estimating travel time of motor vehicle on urban road under control of signal lamps |
CN102169634A (en) * | 2011-04-01 | 2011-08-31 | 大连理工大学 | A priority evacuation control method for traffic congestion |
CN103077615A (en) * | 2012-12-20 | 2013-05-01 | 长沙理工大学 | Online learning method for optimizing signalized intersection queuing length |
CN103761883A (en) * | 2014-01-29 | 2014-04-30 | 中国科学技术大学 | Self-learning method and system for traffic signal control |
-
2016
- 2016-08-12 CN CN201610664684.4A patent/CN106205156B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3844207B2 (en) * | 2001-12-28 | 2006-11-08 | オムロン株式会社 | Information processing apparatus and method, and program |
CN101042805A (en) * | 2007-03-09 | 2007-09-26 | 吉林大学 | Control method for single crossing mixed traffic signal |
CN101727746A (en) * | 2009-09-18 | 2010-06-09 | 苏州大学 | Method for dynamically estimating travel time of motor vehicle on urban road under control of signal lamps |
CN102169634A (en) * | 2011-04-01 | 2011-08-31 | 大连理工大学 | A priority evacuation control method for traffic congestion |
CN103077615A (en) * | 2012-12-20 | 2013-05-01 | 长沙理工大学 | Online learning method for optimizing signalized intersection queuing length |
CN103761883A (en) * | 2014-01-29 | 2014-04-30 | 中国科学技术大学 | Self-learning method and system for traffic signal control |
Non-Patent Citations (1)
Title |
---|
过饱和交叉口交通信号控制动态规划优化模型;李瑞敏等;《交通运输工程学报》;20151231;第15卷(第6期);第101-109页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106205156A (en) | 2016-12-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106205156B (en) | A kind of intersection self-healing combination control method for the mutation of part lane flow | |
Celtek et al. | Real-time traffic signal control with swarm optimization methods | |
CN104778834B (en) | Urban road traffic jam judging method based on vehicle GPS data | |
CN111696348B (en) | Multifunctional intelligent signal control system and method | |
CN103530704B (en) | A kind of air dynamic traffic volume in terminal airspace prognoses system and method thereof | |
WO2021218235A1 (en) | Road traffic jam early warning method and system | |
WO2023056696A1 (en) | Urban rail transit short-term passenger flow forecasting method based on recurrent neural network | |
CN109493599A (en) | A kind of Short-time Traffic Flow Forecasting Methods based on production confrontation network | |
CN105654744B (en) | A kind of improvement traffic signal control method based on Q study | |
CN104318773B (en) | A kind of traffic congestion assay method based on traffic congestion space-time total amount | |
CN110570672B (en) | Regional traffic signal lamp control method based on graph neural network | |
CN109360429A (en) | A kind of urban highway traffic dispatching method and system based on simulative optimization | |
Yao et al. | Emergency vehicle route oriented signal coordinated control model with two-level programming | |
CN109272745A (en) | A kind of track of vehicle prediction technique based on deep neural network | |
CN111739284A (en) | Traffic signal lamp intelligent timing method based on genetic algorithm optimization fuzzy control | |
CN108171998A (en) | A kind of crossing self-adapting traffic signal control system and its method of work based on the alert data of electricity | |
CN103106793A (en) | Traffic state discriminated method based on real-time driving direction and transit time quantum information | |
CN111028504A (en) | Urban expressway intelligent traffic control method and system | |
CN109118787A (en) | A kind of car speed prediction technique based on deep neural network | |
CN105551250A (en) | Method for discriminating urban road intersection operation state on the basis of interval clustering | |
CN110111573A (en) | A kind of tailback integrated dispatch method based on Internet of Things | |
CN115662113A (en) | Signalized intersection people-vehicle game conflict risk assessment and early warning method | |
Chen et al. | Dynamic traffic light optimization and Control System using model-predictive control method | |
Alam et al. | Intellegent traffic light control system for isolated intersection using fuzzy logic | |
Fang et al. | Multi-Objective Traffic Signal Control Using Network-Wide Agent Coordinated Reinforcement Learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |