CN110164147A - A kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA - Google Patents
A kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA Download PDFInfo
- Publication number
- CN110164147A CN110164147A CN201910441394.7A CN201910441394A CN110164147A CN 110164147 A CN110164147 A CN 110164147A CN 201910441394 A CN201910441394 A CN 201910441394A CN 110164147 A CN110164147 A CN 110164147A
- Authority
- CN
- China
- Prior art keywords
- node
- cluster head
- head node
- information
- cluster
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/095—Traffic lights
Abstract
The present invention relates to wireless sense network fields more particularly to a kind of intelligent traffic lamp based on improved adaptive GA-IAGA to regulate and control method, it is characterised in that: its step is to simulate crossroad environment with MATLAB, and vehicle is considered as node, random to generateNFitness function is arranged in a node, the more a height of target of coverage rate covered with ordinary node by cluster head node；Using Revised genetic algorithum, optimal cluster head is selected under environment at the parting of the ways；Cluster head node nearest in its communication range is selected according to ordinary node and the principle being added carries out sub-clustering, its information is passed to corresponding cluster head by ordinary node；Cluster head integrates information of vehicles and is sent to traffic lights, and traffic lights are regulated and controled according to information of vehicles.The present invention, which improves genetic algorithm, reaches better sub-clustering effect, and the thought of sub-clustering is introduced into traffic signal lamp system, enables traffic lights with road traffic condition by real-time monitoring.
Description
Technical field
The present invention relates to wireless sense network field more particularly to a kind of intelligent traffic lamps based on improved adaptive GA-IAGA
Regulation method.
Background technique
Currently, traffic accident has become global public transport safety problem, the real time information of adjacent vehicle, including vehicle are obtained
Speed, driving direction, position etc., the generation that can effectively avoid traffic accident, the development of car networking play one to traffic accident is reduced
Fixed effect；With the increasingly intelligence of automobile and highway, more and more automobiles and roadside infrastructure are equipped with communication and set
It is standby, entire car networking and have become inexorable trend for the relevant application development of car networking；Because of wireless sensor network
(wireless sensor network, WSN) has many advantages, such as to dispose convenient, at low cost, flexible structure and survivability is strong,
Car networking field has a wide range of applications；Cluster algorithm splits the network into cluster one by one, includes a cluster head in each cluster
With several clusters member, member node will sense by the information transmitting between vehicle and vehicle (Vehicle to Vehicle, V2V)
Know that information is sent to cluster head node, cluster head node is conscientious to data to be sent to base station with after；Genetic algorithm is a kind of based on certainly
The optimizing algorithm of right choosing principles and natural genetic mechanism, genetic algorithm is introduced into sub-clustering, makes sub-clustering more fast and accurately；
Not the problem of existing technology does not fully consider cluster head dump energy when selecting cluster head, and not by cluster head after sub-clustering
To traffic signal lamp system, the information of vehicles for causing cluster head to be collected into fails to be fully used the data transmission being collected into.
Summary of the invention
It, will the purpose of the invention is to provide a kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA
Genetic algorithm, which improves, reaches better sub-clustering effect, and the thought of sub-clustering is introduced into traffic signal lamp system, makes to hand over
Ventilating signal lamp can be with road traffic condition by real-time monitoring.
In order to solve the above technical problems, the technical solution of the present invention is as follows: a kind of intelligent transportation based on improved adaptive GA-IAGA
Signal lamp regulates and controls method, which comprises the steps of:
Step 1: establishing crossroad environment model and parameter and population is set；
Step 2: defining fitness function；
Step 3: choosing optimal cluster head node based on Revised genetic algorithum and according to fitness function；
Step 4: sub-clustering being carried out to ordinary node and its coordinate information and velocity information are passed into corresponding cluster head node, cluster
First node, which is integrated and obtains information of vehicles, is transferred to traffic lights；
Step 5: traffic lights are regulated and controled according to information of vehicles.
Further, the step 1: crossroad environment is simulated with MATLAB, vehicle is considered as node, generates N at random
The coordinate of a node, and the corresponding speed of these nodes is generated at random；To population scale NP, all node number N, cluster head node
Number M, maximum number of iterations Gm, cluster head collect the predetermined constant R of information radius0Assign initial value；Wherein, ordinary node number is N-
M。
Further, the step 2: traffic lights position is set as base station, setting cluster head collects information radius
Rc, specific radius formula is as follows
Wherein, R0For predetermined constant, the maximum communication radius of cluster head node, d are indicatedmaxIndicate cluster head node away from base station
Maximum distance, dminIndicate cluster head node away from base station minimum range, d (gi, BS) indicate i-th cluster head node to base station BS away from
From c is control parameter, c ∈ (0,1)；
Crossroad is divided into eight regions according to road middle line and vehicle heading, it is common in each region
Node can only be communicated with the cluster head node in same area；The condition for validity of ordinary node are as follows: ordinary node and cluster head section
The distance of point is less than or equal to cluster head and collects information radius Rc, i.e.,
dist(sj(xj,yj,zj),gi(xi,yi,zi))≤Rc
Wherein, sj(xj,yj,zj) indicate j-th of ordinary node coordinate, j ∈ { 1,2 ... N-M }, gi(xi,yi,zi) indicate
The coordinate of i-th of cluster head node, i ∈ { 1,2 ... M }, N-M indicate the number of ordinary node, and M indicates the number of cluster head node；
Use CtIt indicates the number of the ordinary node of the cluster head node covering of t moment whole, common section is covered with cluster head node
Fitness function is arranged in the more a height of target of the coverage rate of point, and fitness function is embodied as:
The value of the fitness function fit of node is bigger, more meets the requirement of cluster head node.
Further, the step 3:
Step 3.1: initialization population
The coordinate information and velocity information of M cluster head node are generated at random, and M cluster head node is considered as an individual, each
Individual includes D=3*M data, this step needs to be implemented NP times, ultimately generates NP group cluster head node data information；
Step 3.2: being encoded using decimal coded mode
The matrix of a NP*3M is established, every a line represents one group of feasible solution, and every a line is M subgroup, every subgroup from left to right
The data for indicating a cluster head node are made of 3 column, respectively indicate abscissa value, the ordinate of the cluster head node from left to right
Value and velocity amplitude；In genetic algorithm, each behavior item chromosome of the matrix, corresponding one group of feasible solution；Appointing in a line
A subgroup of anticipating is a gene, including three column, respectively corresponds abscissa value, ordinate value and the velocity amplitude of a cluster head node；
Step 3.3: crossover operation
Maximum two parent chromosomes of fitness function value are chosen, crosspoint is determined using Logistic chaos sequence
Position, and determining cross term is intersected using 3 interleaved modes, generate child chromosome；
Step 3.4: mutation operation
Two integers between 1~NP are randomly selected, corresponding two chromosome of the two integers can generate variation
Gene；The position that change point on this two chromosomes is determined using Logistic chaos sequence will be made a variation on this two chromosomes
Corresponding abscissa value, ordinate value and velocity amplitude change new abscissa value, ordinate value and velocity amplitude at point position；New
Abscissa value, ordinate value and velocity amplitude are generated by the method generated at random in step 3.1；
Step 3.5: choosing optimal cluster head node
The fitness function value of two chromosome after calculating variation, it is suitable with the chromosome of the chromosome before variation, that is, originally
Response functional value is compared, if the fitness function value of the chromosome after variation is big, with the chromosome replacement after variation
Chromosome originally keeps original chromosome constant if the chromosome fitness function value after variation is small；
The NP chromosome after intersecting, making a variation is sent to step 3.3, in this way completion an iteration；The dyeing of NP item
Body need to undergo step 3.3,3.4 and 3.5 again, until the number of iterations reaches Gm；If by intersecting, being selected after mutation operation
Node coordinate at there is no the node of actual distribution, by such node be known as " dummy node ", by dummy node playback to away from it most
Closely and at the actual node of the same area, then actual node coordinate at this time is the coordinate of optimal cluster head node.
Further, the step 4: after the completion of optimal cluster head node is chosen, ordinary node is added in its communication range
Distance is minimum and in the cluster head node of the same area, forms M cluster, ordinary node passes the coordinate information of itself and velocity information
It is defeated by corresponding cluster head node, cluster head node collects its cluster interior nodes data information, and calculates by the velocity information of cluster interior nodes
Average vehicle speed in cluster calculates the average distance of common vehicle in cluster to cluster head vehicle by the coordinate informations of cluster interior nodes,
Obtain information of vehicles；The average distance of common vehicle in average vehicle speed in information of vehicles, that is, cluster, cluster to cluster head vehicle is believed
Breath is transferred to traffic lights.
Further, the step 5:
Input: the average distance of average vehicle speed, common vehicle to cluster head vehicle in cluster；
Output: traffic light system time；
When east-west direction is green light:
IfThen east-west direction green light extends 10s, north and south
Direction red light extends 10s；
IfThen east-west direction green light extends 5s, North and South direction
Red light extends 5s；
IfThen east-west direction green light becomes amber light, becomes red light after amber light 2s,
North and South direction red light becomes amber light, becomes green light after amber light 2s；
IfThen east-west direction green light becomes amber light, becomes red light, north and south after amber light 5s
Direction red light becomes amber light, becomes green light after amber light 5s；
Long green light time does not change in the case of other；
When east-west direction is red light:
IfThen North and South direction green light extends 10s, east
West extends 10s to red light；
IfThen east-west direction red light extends 5s, North and South direction
Green light extends 5s；
IfThen east-west direction red light becomes amber light, becomes green light after amber light 2s,
North and South direction green light becomes amber light, becomes red light after amber light 2s；
IfThen east-west direction red light becomes amber light, becomes green light, north and south after amber light 5s
Direction green light becomes amber light, becomes red light after amber light 5s；
Red light duration does not change in the case of other；
Wherein:Speed indicates automobile from west to east in the speed at crossing；Indicate that automobile from west to east is driven out to road
The speed of mouth；Indicate automobile from south to north in the speed at crossing；The automobile of expression from south to north is driven out to the vehicle at crossing
Speed；Empty (NS_IN) indicates that North and South direction drives without vehicle；Empty (EW_IN) indicates that east-west direction drives without vehicle；VmaxIt indicates
The maximum passage speed of road, VminIndicate the minimum passage speed of road；RNS_ time indicates North and South direction red light remaining time,
REW_ time indicates North and South direction red light remaining time.
The invention has the following beneficial effects:
The present invention uses decimal coded mode in coding, improves the processing speed of intersection and variation；When intersecting,
The position intersected is determined using Logistic chaos sequence, uses 3 interleaved modes, it is smaller to original solution change, it can
Problem is buffeted to avoid the optimizing that genetic algorithm generates in Combinatorial Optimization application problem；In variation, the present invention is without as passing
System genetic algorithm like that swaps the gene order inside chromosome, but directly generates new gene, such chromosome
Variation it is more obvious, ensure that the optimal solution of searching has jumped out local optimum, to global optimum develop；The present invention and improve before
Genetic algorithm compare, improved genetic algorithm have faster convergence rate and higher convergence precision；The present invention proposes
The information of cluster head is transmitted to traffic signal lamp system, the description of real-time monitoring traffic lights is conducive to the high efficiency of traffic, promotes
The realization of intelligent transportation.
Detailed description of the invention
Fig. 1 is overall flow schematic diagram of the embodiment of the present invention；
Fig. 2 is the node distribution map in the embodiment of the present invention by the MATLAB vehicle generated at random；
Fig. 3 is that crossroad is divided into eight areas according to road middle line and vehicle heading in the embodiment of the present invention
The schematic diagram in domain；
Fig. 4 is the cluster head node schematic diagram that the embodiment of the present invention generates after improved adaptive GA-IAGA optimizing；
Fig. 5 is that the optimal solution value of improved adaptive GA-IAGA and traditional genetic algorithm changes with the number of iterations in the embodiment of the present invention
Variation tendency compare figure.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawing and specific implementation
Invention is further described in detail for example.
Fig. 1 to Fig. 5 is please referred to, the present invention is a kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA,
Its step includes:
Step 1: establishing crossroad environment model and parameter and population is set
Crossroad environment is simulated with MATLAB, vehicle is considered as node, generates the coordinate of N number of node at random, and random
The corresponding speed of these nodes is generated, the coordinate of node meets the coordinates restriction condition of crossroad, gives egress simulation distribution
Figure, the node simulation distribution figure in the present embodiment is referring to Fig.2, velocity restraint condition are as follows: velocity interval be -60km/h≤v≤
60km/h, node direction only consider whether node moves towards signal lamp, are positive if it is speed, are negative if not speed；
The pre- of information radius is collected to population scale NP, all node number N, cluster head node number M, maximum number of iterations Gm, cluster head
If constant R0Assign initial value；Wherein, the ordinary node number is N-M；In the emulation of the embodiment of the present invention, NP=50, N=150,
M=20, Gm=400, R0=150 meters；
Step 2: defining fitness function
Crossroad center, that is, traffic lights position is set as base station, since communication uses multi-hop transmission between cluster
Mechanism can consume bigger energy so the cluster head node close apart from base station needs to forward the data from other cluster head nodes；For
Energy consumption between balance cluster head node, with closer to the cluster head node of base station, communication range is smaller to be arranged cluster head for principle
Collect information radius Rc, specific radius formula is as follows
Wherein, R0For predetermined constant, the maximum communication radius of cluster head node, d are indicatedmaxIndicate cluster head node away from base station
Maximum distance, dminIndicate cluster head node away from base station minimum range, d (gi, BS) indicate i-th cluster head node to base station BS away from
From c is control parameter, c ∈ (0,1)；
Since crossroad environment has certain particularity, in different lanes, the driving direction of automobile is different, by cross
Crossing is divided into eight regions according to road middle line and vehicle heading, participates in Fig. 3, the ordinary node in each region is only
It can be communicated with the cluster head node in same area；The condition for validity of ordinary node is that there are the clusters that ordinary node can be added
Head needs to meet: the distance of cluster head node and ordinary node is less than or equal to cluster head and collects information radius Rc, ordinary node ability
Cluster head is added the ordinary node is considered as effective node, is added when ordinary node joined any one in 20 cluster head nodes
Enter communication network；The distance of ordinary node and cluster head node is less than or equal to cluster head and collects information radius Rc, i.e.,
dist(sj(xj,yj,zj),gi(xi,yi,zi))≤Rc
Wherein, sj(xj,yj,zj) indicate j-th of ordinary node coordinate, j ∈ { 1,2 ... 130 }, gi(xi,yi,zi) indicate
The coordinate of i-th of cluster head node, i ∈ { 1,2 ... 20 }；
The same ordinary node may be in simultaneously in the communication range of multiple cluster head nodes, use CtIndicate that t moment is complete
The number of the ordinary node of the cluster head node covering in portion, with the more a height of target setting of the coverage rate of cluster head node covering ordinary node
Fitness function, fitness function are embodied as:
Cluster head node select permeability, which can be converted into from N number of node, chooses M node, in this implementation, cluster head node selection
Problem can be converted into from 150 nodes 20 nodes of selection, keep the value of last fitness function fit as big as possible, then this
When one group of cluster head coordinate be to be best suitable for the cluster head node position of requirement；
Step 3: choosing optimal cluster head node based on Revised genetic algorithum and according to fitness function；
Step 3.1: initialization population
The coordinate information and velocity information of M cluster head node are generated at random, and M cluster head node is considered as an individual, each
Individual includes D=3*M data, this step needs to be implemented NP times, ultimately generates NP group cluster head node data information；The present embodiment
In generate the coordinate information and velocity information of 20 nodes at random, 20 nodes are considered as an individual, and each individual includes D=3*
M=60 data, this step need to be implemented 50 times, ultimately generate 50 groups of cluster head node data informations；
Step 3.2: being encoded using decimal coded mode
The coding of node is made of the cross, ordinate and speed of cluster head node, establishes a NP*3M (50*60)
Matrix, every a line represent one group of feasible solution, and every a line is M subgroup from left to right, and every subgroup indicates the data of a cluster head node,
It is made of 3 column, respectively indicates abscissa value, ordinate value and the velocity amplitude of the cluster head node from left to right；
For example, the matrix of generation is when cluster head node quantity M is 2, when population quantity NP=2
The 2*6 matrix indicates the individual information of two cluster head node set of initialization, 2 cluster head sections in the first group information
The coordinate of point is respectively (485,515) and (503,492), and velocity magnitude is respectively 60km/h and 55km/h；2 in second group information
The coordinate of a cluster head node is respectively (517,488) and (491,504), and velocity magnitude is respectively 30km/h and 40km/h；With this
Analogize, the matrix of the corresponding 50*60 of 50 groups of feasible solutions of 20 cluster head node coordinates and speed；
In genetic algorithm, each behavior item chromosome of the 50*60 matrix corresponds to appointing in one group of feasible solution a line
A subgroup of anticipating is a gene, including three column, respectively corresponds abscissa value, ordinate value and the velocity amplitude of a cluster head node；
Step 3.3: crossover operation
It chooses maximum two parent chromosomes of fitness function value to be intersected, generates child chromosome；Utilize chaos
Sequence determines the position in crosspoint, and is intersected using 3 interleaved modes to determining cross term, and concrete operations are as follows: false
If parent chromosome A1, A2 are intersected,The random number on one (0,1) is taken to make
For initial value, once generated on (0, a 1) section using Logistic chaos sequence x (n+1)=4x (n) [1-x (n)] iteration
Chaos value, save the above chaos value as the chaos iteration initial value for generating next-generation cross term, then this value multiplied by 60 so
After be rounded；If this number is 34, then carrying out three point friendships to gene corresponding in A1, A2 using 34,35 and 36 as crosspoint
Fork, obtains new chromosome (A '1,A′2), i.e.,
Due to cluster head node coordinate be successively arrange by subgroup (first be classified as abscissa value, first be classified as ordinate value,
First is classified as velocity amplitude), thus intersect when should also intersect by subgroup, and it is noted that sequence, such as if generate number be 33, that
It shouldWithExchange；This 3 cross-pair original chromosomes change is smaller, can be to avoid something lost
Propagation algorithm buffets problem in the optimizing that combinatorial optimization problem application generates, and improves algorithmic statement precision.
Step 3.4: mutation operation
Variation is to realize a kind of means of population diversity, is to jump out local optimum, the important guarantee of global optimizing；This hair
The variation design of bright embodiment is as follows: randomly selecting two integers between 1~50 first, the two integers are two corresponding
Chromosome can generate the gene of variation；Step 3.3 is copied to generate the chaos on (0, a 1) section using Logistic sequence again
Value, then then this value is rounded multiplied by 60, it determines the position of change point on this two chromosomes, will become on this two chromosomes
Corresponding abscissa value, ordinate value and velocity amplitude change new abscissa value, ordinate value and velocity amplitude at dissimilarity position, false
If this number is 34, then the corresponding abscissa value in 34,35,36 positions, ordinate value and velocity amplitude on two chromosomes are changed
Abscissa value, ordinate value and the velocity amplitude of Cheng Xin；New abscissa value, ordinate value and velocity amplitude by step 1 with
The method that machine generates generates, and also needs to meet crossroad boundary condition and velocity restraint condition；
Step 3.5: choosing optimal cluster head node
The fitness function value of two chromosome after calculating variation, it is suitable with the chromosome of the chromosome before variation, that is, originally
Response functional value is compared, if the fitness function value of the chromosome after variation is big, with the chromosome replacement after variation
Chromosome originally keeps original chromosome constant if the chromosome fitness function value after variation is small；
The NP chromosome after intersecting, making a variation is sent to step 3.3, in this way completion an iteration；NP=50 item
Chromosome need to undergo step 3.3,3.4 and 3.5 again, until the number of iterations reaches GmUntil=400；If by intersecting, making a variation
At the node coordinate selected after operation may not no actual distribution node, such node is known as " dummy node ", by dummy section
Point playback is to away from it, recently and at the actual node of the same area, then actual node coordinate at this time is that cluster head node is sat
Mark；Fig. 4 shows the cluster head node schematic diagram that the embodiment of the present invention generates after improved adaptive GA-IAGA optimizing；
Step 4: sub-clustering being carried out to ordinary node and its coordinate information and velocity information are passed into corresponding cluster head node, cluster
First node, which is integrated and obtains information of vehicles, is transferred to traffic lights, specifically: after the completion of optimal cluster head node is chosen, commonly
It is minimum and in the cluster head node of the same area that node is added in its communication range distance, forms 20 clusters, ordinary node by itself
Coordinate information and velocity information be transferred to corresponding cluster head node, cluster head node collects its cluster interior nodes data information, and passes through
The velocity information of cluster interior nodes calculates average vehicle speed in cluster, calculates common vehicle in cluster by the coordinate information of cluster interior nodes
To the average distance of cluster head vehicle, information of vehicles is obtained；The form of information of vehicles data frame is stored, and passes through multi-hop communication
Mode be transferred to traffic lights；The content of data is included, byte number and byte length are as shown in the table:
1 data format of table
Design data frame storage above data as follows:
MATLAB simulates crossroad environment, and when generating node at random, all ordinary nodes and cluster head node are assigned
Unique ID, cluster head ID acquisition when node generates in data frame；
Fig. 5 shows genetic algorithm and improves front and back, the variation tendency that optimal solution value changes with the number of iterations；Although holding every time
State is random before line program, but from figure general trend it can be seen that improved genetic algorithm compared with original genetic algorithm,
Has many advantages, such as convergence rate faster, convergence precision is higher；
Step 5: traffic lights are regulated and controled according to information of vehicles
Input: the average distance of average vehicle speed, common vehicle to cluster head vehicle in cluster；
Output: traffic light system time；
When east-west direction is green light:
IfThen east-west direction green light extends 10s, north and south
Direction red light extends 10s；
IfThen east-west direction green light extends 5s, North and South direction
Red light extends 5s；
IfThen east-west direction green light becomes amber light, becomes red light after amber light 2s,
North and South direction red light becomes amber light, becomes green light after amber light 2s；
IfThen east-west direction green light becomes amber light, becomes red light, north and south after amber light 5s
Direction red light becomes amber light, becomes green light after amber light 5s；
Long green light time does not change in the case of other；
When east-west direction is red light:
IfThen North and South direction green light extends 10s, east
West extends 10s to red light；
IfThen east-west direction red light extends 5s, North and South direction
Green light extends 5s；
IfThen east-west direction red light becomes amber light, becomes green light after amber light 2s,
North and South direction green light becomes amber light, becomes red light after amber light 2s；
IfThen east-west direction red light becomes amber light, becomes green light, north and south after amber light 5s
Direction green light becomes amber light, becomes red light after amber light 5s；
Red light duration does not change in the case of other；
Wherein:Speed indicates automobile from west to east in the speed at crossing；Indicate that automobile from west to east is driven out to road
The speed of mouth；Indicate automobile from south to north in the speed at crossing；The automobile of expression from south to north is driven out to the vehicle at crossing
Speed；
Empty (NS_IN) indicates that North and South direction drives without vehicle；Empty (EW_IN) indicates that east-west direction drives without vehicle；Vmax
Indicate the maximum passage speed of road, VminIndicate the minimum passage speed of road；RNS_ time indicates that North and South direction red light is remaining
Time, REW_ time indicates North and South direction red light remaining time.
In the emulation experiment of the present embodiment, environment is crossroad region, is the road vehicle situation of closer to reality, setting
Node total number N=150, the energy of each node are 1J, when the death nodes in whole network reach the 30% of total node number
When, system can not work on.
The embodiment of the present invention is using a crossroad as background, and road width is 50 meters, the vehicle passed through at the parting of the ways
150 nodes of middle random distribution, one vehicle of each node on behalf choose several and are used as cluster head node, select according to ordinary node
It selects cluster head node nearest in its communication range and the principle being added carries out sub-clustering, its information is passed to respective cluster by ordinary node
Head, these cluster heads receive surrounding RcInformation of vehicles (the distance between speed, vehicle and the vehicle of ordinary node in rice distance
Driving direction), and information of vehicles is sent to signal lamp by way of multi-hop, signal lamp analyzes these data
And processing, the time of red light, green light and amber light is dynamically changed, realizes the real-time monitoring of traffic signal system；Choosing cluster head
When node, fitness function is arranged in the more a height of target of coverage rate covered with ordinary node by cluster head node, then utilizes improvement
Genetic algorithm, optimal cluster head is selected under environment at the parting of the ways.The present invention is based on the carry out sub-clusterings of improved adaptive GA-IAGA, with biography
Sub-clustering result under system genetic algorithm is compared, and is substantially increased the convergence precision of genetic algorithm, is accelerated convergence rate, makes algorithm
Optimal solution can be converged to greater probability；In addition to this, by the energy consumption and biography of car networking system under improved adaptive GA-IAGA
The energy consumption of car networking system is compared under system leach agreement, and the car networking sub-clustering under improved adaptive GA-IAGA is in energy-saving square
Face has greater advantages.
The above content is specific embodiment is combined, further detailed description of the invention, and it cannot be said that this hair
Bright specific implementation is only limited to these instructions.For those of ordinary skill in the art to which the present invention belongs, it is not taking off
Under the premise of from present inventive concept, a number of simple deductions or replacements can also be made, all shall be regarded as belonging to protection of the invention
Range.
Claims (6)
1. a kind of intelligent traffic lamp based on improved adaptive GA-IAGA regulates and controls method, which comprises the steps of:
Step 1: establishing crossroad environment model and parameter and population is set；
Step 2: defining fitness function；
Step 3: choosing optimal cluster head node based on Revised genetic algorithum and according to fitness function；
Step 4: sub-clustering being carried out to ordinary node and its coordinate information and velocity information are passed into corresponding cluster head node, cluster head section
O'clock sharp merges acquisition information of vehicles and is transferred to traffic lights；
Step 5: traffic lights are regulated and controled according to information of vehicles.
2. the intelligent traffic lamp according to claim 1 based on improved adaptive GA-IAGA regulates and controls method, which is characterized in that
The step 1: simulating crossroad environment with MATLAB, vehicle be considered as node, generates the coordinate of N number of node at random, and with
Machine generates the corresponding speed of these nodes；To population scale NP, all node number N, cluster head node number M, greatest iteration time
Number Gm, cluster head collect the predetermined constant R of information radius0Assign initial value；Wherein, ordinary node number is N-M.
3. the intelligent traffic lamp according to claim 1 based on improved adaptive GA-IAGA regulates and controls method, which is characterized in that
The step 2: traffic lights position is set as base station, setting cluster head collects information radius Rc, specific radius formula is such as
Under
Wherein, R0For predetermined constant, the maximum communication radius of cluster head node, d are indicatedmaxIndicate cluster head node away from base station it is maximum away from
From dminIndicate cluster head node away from base station minimum range, d (gi, BS) and i-th of cluster head node is indicated to the distance of base station BS, c is
Control parameter, c ∈ (0,1)；
Crossroad is divided into eight regions according to road middle line and vehicle heading, the ordinary node in each region
It can only be communicated with the cluster head node in same area；The condition for validity of ordinary node are as follows: ordinary node and cluster head node
Distance is less than or equal to cluster head and collects information radius Rc, i.e.,
dist(sj(xj,yj,zj),gi(xi,yi,zi))≤Rc
Wherein, sj(xj,yj,zj) indicate j-th of ordinary node coordinate, j ∈ { 1,2 ... N-M }, gi(xi,yi,zi) indicate i-th
The coordinate of a cluster head node, i ∈ { 1,2 ... M }, N-M indicate the number of ordinary node, and M indicates the number of cluster head node；
Use CtThe number of the ordinary node of the cluster head node covering of t moment whole is indicated, with covering for cluster head node covering ordinary node
Fitness function is arranged in the more a height of target of lid rate, and fitness function is embodied as:
The value of the fitness function fit of node is bigger, more meets the requirement of cluster head node.
4. the intelligent traffic lamp according to claim 1 based on improved adaptive GA-IAGA regulates and controls method, which is characterized in that
The step 3:
Step 3.1: initialization population
The coordinate information and velocity information of M cluster head node are generated at random, and M cluster head node is considered as an individual, each individual
Comprising D=3*M data, this step is needed to be implemented NP times, ultimately generates NP group cluster head node data information；
Step 3.2: being encoded using decimal coded mode
The matrix of a NP*3M is established, every a line represents one group of feasible solution, and every a line is M subgroup from left to right, and every subgroup indicates
The data of one cluster head node are made of 3 column, respectively indicate from left to right the abscissa value of the cluster head node, ordinate value and
Velocity amplitude；In genetic algorithm, each behavior item chromosome of the matrix, corresponding one group of feasible solution；It is any one in a line
Subgroup is a gene, including three column, respectively corresponds abscissa value, ordinate value and the velocity amplitude of a cluster head node；
Step 3.3: crossover operation
Maximum two parent chromosomes of fitness function value are chosen, the position in crosspoint is determined using Logistic chaos sequence
It sets, and determining cross term is intersected using 3 interleaved modes, generate child chromosome；
Step 3.4: mutation operation
Two integers between 1~NP are randomly selected, corresponding two chromosome of the two integers can generate the gene of variation；
The position that change point on this two chromosomes is determined using Logistic chaos sequence, by change point position on this two chromosomes
Locate corresponding abscissa value, ordinate value and velocity amplitude and changes new abscissa value, ordinate value and velocity amplitude into；New abscissa
Value, ordinate value and velocity amplitude are generated by the method generated at random in step 3.1；
Step 3.5: choosing optimal cluster head node
The fitness function value of two chromosome after calculating variation, the fitness with the chromosome of the chromosome before variation, that is, originally
Functional value is compared, original with the chromosome replacement after variation if the fitness function value of the chromosome after variation is big
Chromosome, if variation after chromosome fitness function value it is small, keep original chromosome it is constant；
The NP chromosome after intersecting, making a variation is sent to step 3.3, in this way completion an iteration；NP chromosome needs
Step 3.3,3.4 and 3.5 are undergone again, until the number of iterations reaches Gm；If by the section for intersecting, being selected after mutation operation
Point coordinate at there is no the node of actual distribution, by such node be known as " dummy node ", by dummy node playback to away from its recently and
At the actual node of the same area, then actual node coordinate at this time is the coordinate of optimal cluster head node.
5. the intelligent traffic lamp according to claim 1 based on improved adaptive GA-IAGA regulates and controls method, which is characterized in that
The step 4: after the completion of optimal cluster head node is chosen, it is minimum and in same area that distance in its communication range is added in ordinary node
The cluster head node in domain forms M cluster, and the coordinate information of itself and velocity information are transferred to corresponding cluster head node by ordinary node,
Cluster head node collects its cluster interior nodes data information, and calculates average vehicle speed in cluster by the velocity information of cluster interior nodes,
Common vehicle in cluster, which is calculated, by the coordinate information of cluster interior nodes obtains information of vehicles to the average distance of cluster head vehicle；By vehicle
Average vehicle speed in information, that is, cluster, the average distance information of common vehicle to cluster head vehicle is transferred to traffic signals in cluster
Lamp.
6. the intelligent traffic lamp according to claim 1 based on improved adaptive GA-IAGA regulates and controls method, which is characterized in that
The step 5:
Input: the average distance of average vehicle speed, common vehicle to cluster head vehicle in cluster；
Output: traffic light system time；
When east-west direction is green light:
IfThen east-west direction green light extends 10s, North and South direction
Red light extends 10s；
IfThen east-west direction green light extends 5s, North and South direction red light
Extend 5s；
IfThen east-west direction green light becomes amber light, becomes red light, north and south after amber light 2s
Direction red light becomes amber light, becomes green light after amber light 2s；
IfThen east-west direction green light becomes amber light, becomes red light, North and South direction after amber light 5s
Red light becomes amber light, becomes green light after amber light 5s；
Long green light time does not change in the case of other；
When east-west direction is red light:
IfThen North and South direction green light extends 10s, east-west direction
Red light extends 10s；
IfThen east-west direction red light extends 5s, North and South direction green light
Extend 5s；
IfThen east-west direction red light becomes amber light, becomes green light, north and south after amber light 2s
Direction green light becomes amber light, becomes red light after amber light 2s；
IfThen east-west direction red light becomes amber light, becomes green light, North and South direction after amber light 5s
Green light becomes amber light, becomes red light after amber light 5s；
Red light duration does not change in the case of other；
Wherein:Speed indicates automobile from west to east in the speed at crossing；Indicate that automobile from west to east is driven out to crossing
Speed；Indicate automobile from south to north in the speed at crossing；The automobile of expression from south to north is driven out to the speed at crossing；
Empty (NS_IN) indicates that North and South direction drives without vehicle；Empty (EW_IN) indicates that east-west direction drives without vehicle；VmaxIndicate road
Maximum passage speed, VminIndicate the minimum passage speed of road；RNS_ time indicates North and South direction red light remaining time, REW_
Time indicates North and South direction red light remaining time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910441394.7A CN110164147A (en) | 2019-05-24 | 2019-05-24 | A kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910441394.7A CN110164147A (en) | 2019-05-24 | 2019-05-24 | A kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110164147A true CN110164147A (en) | 2019-08-23 |
Family
ID=67632808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910441394.7A Pending CN110164147A (en) | 2019-05-24 | 2019-05-24 | A kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110164147A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112562363A (en) * | 2020-11-09 | 2021-03-26 | 江苏大学 | Intersection traffic signal optimization method based on V2I |
| CN112885117A (en) * | 2021-01-13 | 2021-06-01 | 长安大学 | Network communication control intersection control system and method |
| CN113870592A (en) * | 2021-10-20 | 2021-12-31 | 温州大学 | Traffic light improved timing method based on DEEC clustering |
| CN114120648A (en) * | 2021-12-03 | 2022-03-01 | 东软集团股份有限公司 | Traffic signal lamp timing method and device, storage medium and electronic equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5898389A (en) * | 1996-10-11 | 1999-04-27 | Electro-Tech's | Blackout backup for traffic light |
| WO2001018767A1 (en) * | 1999-09-02 | 2001-03-15 | Siemens Aktiengesellschaft | Control device for traffic light crossroads |
| CN101557636A (en) * | 2009-05-15 | 2009-10-14 | 广东工业大学 | Wireless sensor network routing method |
| CN102685688A (en) * | 2012-06-12 | 2012-09-19 | 西华大学 | Wireless sensor network clustering method based on first-price-sealed bid auction |
| CN103150911A (en) * | 2013-02-07 | 2013-06-12 | 江苏大学 | Method for optimizing signal timing of single intersection based on genetic algorithm |
| CN103337180A (en) * | 2013-06-28 | 2013-10-02 | 上海宽岱电讯科技发展有限公司 | Intelligent traffic light control system based on WSN (wireless sensor network) |
| US20150019128A1 (en) * | 2003-06-19 | 2015-01-15 | Here Global B.V. | Method and System for Representing Traffic Signals in a Road Network Database |
| CN107705589A (en) * | 2017-11-06 | 2018-02-16 | 西南交通大学 | Bicyclic signal timing optimization method based on self-adapted genetic algorithm |
| CN109360431A (en) * | 2018-11-15 | 2019-02-19 | 华东师范大学 | A kind of self-adapting traffic signal light control algolithm and system based on speed monitoring |
-
2019
- 2019-05-24 CN CN201910441394.7A patent/CN110164147A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5898389A (en) * | 1996-10-11 | 1999-04-27 | Electro-Tech's | Blackout backup for traffic light |
| WO2001018767A1 (en) * | 1999-09-02 | 2001-03-15 | Siemens Aktiengesellschaft | Control device for traffic light crossroads |
| US20150019128A1 (en) * | 2003-06-19 | 2015-01-15 | Here Global B.V. | Method and System for Representing Traffic Signals in a Road Network Database |
| CN101557636A (en) * | 2009-05-15 | 2009-10-14 | 广东工业大学 | Wireless sensor network routing method |
| CN102685688A (en) * | 2012-06-12 | 2012-09-19 | 西华大学 | Wireless sensor network clustering method based on first-price-sealed bid auction |
| CN103150911A (en) * | 2013-02-07 | 2013-06-12 | 江苏大学 | Method for optimizing signal timing of single intersection based on genetic algorithm |
| CN103337180A (en) * | 2013-06-28 | 2013-10-02 | 上海宽岱电讯科技发展有限公司 | Intelligent traffic light control system based on WSN (wireless sensor network) |
| CN107705589A (en) * | 2017-11-06 | 2018-02-16 | 西南交通大学 | Bicyclic signal timing optimization method based on self-adapted genetic algorithm |
| CN109360431A (en) * | 2018-11-15 | 2019-02-19 | 华东师范大学 | A kind of self-adapting traffic signal light control algolithm and system based on speed monitoring |
Non-Patent Citations (4)
| Title |
|---|
| 刘静: "分簇数据收集的协同分布式Q学习交通信号配时优化", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 李聪: "WSN中基于能量均衡的路由算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李金洋等: "基于车速的自适应交通信号灯控制系统", 《计算机技术与发展》 * |
| 王校锋等: "基于混沌遗传算法的TSP问题求解", 《第九届全国数学建模教学与应用会议》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112562363A (en) * | 2020-11-09 | 2021-03-26 | 江苏大学 | Intersection traffic signal optimization method based on V2I |
| CN112885117A (en) * | 2021-01-13 | 2021-06-01 | 长安大学 | Network communication control intersection control system and method |
| CN113870592A (en) * | 2021-10-20 | 2021-12-31 | 温州大学 | Traffic light improved timing method based on DEEC clustering |
| CN114120648A (en) * | 2021-12-03 | 2022-03-01 | 东软集团股份有限公司 | Traffic signal lamp timing method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110164147A (en) | A kind of intelligent traffic lamp regulation method based on improved adaptive GA-IAGA | |
| CN103593535B (en) | Urban traffic complex self-adaptive network parallel simulation system and method based on multi-scale integration | |
| Schroth et al. | Simulating the traffic effects of vehicle-to-vehicle messaging systems | |
| Boucetta et al. | Review of mobility scenarios generators for vehicular ad-hoc networks simulators | |
| Rapelli et al. | A distributed V2V-based virtual traffic light system | |
| Li et al. | Cellular automata model for unsignalized T-shaped intersection | |
| Gopi et al. | Intelligent DoS Attack Detection with Congestion Control Technique for VANETs. | |
| Dong et al. | Analysis and control of intelligent traffic signal system based on adaptive fuzzy neural network | |
| Zhang et al. | Dynamical topology analysis of VANET based on complex networks theory | |
| Shang et al. | A new cellular automaton model for traffic flow considering realistic turn signal effect | |
| Jiang et al. | Evolution towards optimal driving strategies for large‐scale autonomous vehicles | |
| Kaur et al. | An Overview of Ad Hoc Networks Routing Protocols and Its Design Effectiveness | |
| Hormann et al. | Simulator for inter-vehicle communication based on traffic modeling | |
| Qasim et al. | Performance evaluation of ad-hoc on-demand distance vector protocol in highway environment in VANET with MATLAB | |
| Sahoo et al. | Connectivity modeling of vehicular ad hoc networks in signalized city roads | |
| Jin et al. | Trajectory-prediction based relay scheme for time-sensitive data communication in VANETs | |
| Wang et al. | A Reinforcement Learning Approach to CAV and Intersection Control for Energy Efficiency | |
| Neudecker et al. | Verification and evaluation of fail-safe virtual traffic light applications | |
| Zhang et al. | A scalable unsignalized intersection system for automated vehicles and semi-physical implementation | |
| Song et al. | Path planning in urban environment based on traffic condition perception and traffic light status | |
| Gao et al. | Multi-Vehicles Decision-Making in Interactive Highway Exit: A Graph Reinforcement Learning Approach | |
| Brahmia et al. | Investigating fading impact on AODV, OLSR and DSDV protocols in NS3 | |
| Zhu et al. | State-of-the-Art and Technical Trends of Road-Based Autonomous Driving System | |
| Milo | Intersection Simulation and Path Estimation | |
| Zan et al. | A model based on cellular automata to simulate traffic flows moving in a circle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190823 |
|
| RJ01 | Rejection of invention patent application after publication |