CN112133086B - Regional traffic signal data driving control method based on multi-agent network - Google Patents
Regional traffic signal data driving control method based on multi-agent network Download PDFInfo
- Publication number
- CN112133086B CN112133086B CN202010798063.1A CN202010798063A CN112133086B CN 112133086 B CN112133086 B CN 112133086B CN 202010798063 A CN202010798063 A CN 202010798063A CN 112133086 B CN112133086 B CN 112133086B
- Authority
- CN
- China
- Prior art keywords
- intersection
- space occupancy
- state
- error
- matrix
- 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
- 238000000034 method Methods 0.000 title claims abstract description 16
- 239000011159 matrix material Substances 0.000 claims description 22
- 239000003795 chemical substances by application Substances 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 9
- 238000013461 design Methods 0.000 claims description 4
- 238000011217 control strategy Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000007704 transition Effects 0.000 claims description 3
- 230000000737 periodic effect Effects 0.000 abstract description 4
- 239000002699 waste material Substances 0.000 abstract description 3
- 238000001514 detection method Methods 0.000 description 4
- 238000007667 floating Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005312 nonlinear dynamic Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/081—Plural intersections under common control
Abstract
The invention relates toA regional traffic signal data driving control method based on a multi-agent network is characterized in that a storage forwarding model is established, constraint conditions for green light time of an intersection are adjusted, a time-varying control signal period C (k) is introduced, and a brand-new green light time constraint condition is obtained, so that green light time waste is avoided, the constraint conditions for breaking a fixed period are considered, and the space occupancy rate is guaranteed to meet x all the timeiAnd (k +1) is not less than 0, and the established regional traffic signal control model can describe three traffic state forms of undersaturation, critical saturation and supersaturation at the intersection at the same time, so that the applicability is stronger. The control algorithm is a dynamic feedback control algorithm, can respond to the change of periodic traffic flow, estimates and dynamically adjusts a periodic timing scheme based on the number of vehicles in the current period, and realizes the balance of space occupancy in each direction. The method has the advantages that the saturation degree in all directions is balanced, so that the possibility of supersaturation in certain directions is reduced, the traffic fairness is ensured, and the overall traffic operation efficiency is improved.
Description
Technical Field
The invention relates to the technical field of intelligent traffic signal control, in particular to a regional traffic signal data driving control method based on a multi-agent network.
Background
The method comprises the steps of obtaining intersection flow parameters through an estimation method on the basis of floating car data (typical mobile detection data sources), constructing a periodic vehicle number estimation model on the basis of a storage-forwarding modeling method, and constructing a regional traffic signal data driving control model which aims at space occupancy balance on the basis of a multi-agent network.
In the traditional analysis of a store-and-forward model, generally, only the traffic signal timing problem in the oversaturated traffic state is considered, more green light time is allocated to a certain direction and a certain period of an undersaturated intersection, and the problem of green light time waste, namely the idle discharge phenomenon, exists; in fact, at this time, the traffic capacity can be guaranteed only by allocating less green light time. However, most traffic signal controls are periodic control, and the constraint of the maximum and minimum green time of the phase is also required to be met, so the adjustable range of the green time in the traditional signal control is limited, and the control problems of three traffic states of undersaturation, critical saturation and oversaturation cannot be perfectly compatible.
In order to solve the problem, in the establishment of a store-and-forward model, constraint conditions for green light time of an intersection are adjusted, a time-varying control signal period C (k) is introduced, and a brand-new green light time constraint condition is obtained, so that green light time waste is avoided, the constraint conditions for breaking a fixed period are considered, and the space occupancy rate is guaranteed to meet x all the timeiAnd (k +1) is not less than 0, and the established regional traffic signal control model can describe three traffic state forms of undersaturation, critical saturation and supersaturation at the intersection at the same time, so that the applicability is stronger.
The invention content is as follows:
the method is based on the practical problems of urban regional traffic detection and signal control, high-precision low-occupancy-ratio floating car data are used as data sources, intersection flow parameters are obtained through an estimation method, a multi-intersection multi-directional vehicle conservation equation is established by combining the networking property, the expansibility and the reproducibility of distributed single-intersection multi-directional space occupancy balance control with a positive system model, and a regional traffic signal data driving control method based on a multi-agent network and aiming at space occupancy balance is provided.
With the rapid development of intelligent vehicles and internet traffic and communication technologies, the scale, quality, accuracy, instantaneity and the like of mobile detection data are greatly improved. The signal control method of the invention is based on floating car data (typical mobile detection data source), does not depend on any model information and fixed detector input, and can provide a better theoretical and application basis for the optimization design of a novel traffic control system under the future road environment. The invention adopts the following technical scheme:
a regional traffic signal data driving control method based on a multi-agent network comprises the following steps:
(1) creating a component-positive system model
Wherein x ism(k) N represents the state of the component system element m at the time k, wherein m is equal to or more than 0, and m is equal to 1; a ismnM ≠ n denotes the state quantity transfer proportionality coefficient of the component system element m to the component system element n in the sampling period T > 0; i ism(k) Not less than 0 and Om(k) ≧ 0 represents input and output of Compartment System element m within the sampling period, respectively, and ammThe proportional coefficient is output for the state quantity of more than or equal to 0;
describing the component system model into a vector form:
x(k+1)=Ax(k)+I(k)
in the formula: x (k) ═ x1(k),...,xN(k)]T∈RNIs a system state vector;
I(k)=[I1(k),...,IN(k)]T∈RNinputting for the outside of the system; a is an element of RN×NIs a system state matrix and has
Assuming that the current state quantity of any component is greater than or equal to the total quantity of state transitions in the sampling period, diagonal elements of the state matrix A are all non-negative, and the column sum satisfies:
(2) establishing a traffic area multi-intersection variable-period multi-direction space occupancy model with green light time constraint
The global balanced dynamic model of its space occupancy can be described as:
wherein the global space occupancy state vector is x ═ x1,...,xM]T∈RM,xi∈RNThe global nonlinear dynamic vector is f ═ f1,...,fM]T∈RM,fi∈RNControlling the input green time to ui∈RNM represents the number of urban regional intersections, and N represents the number of each direction of a single intersection;
the distributed consistency coordination error for defining the space occupancy rate of the ith direction of the mth intersection is as follows:
wherein the content of the first and second substances,representing a connection matrix of a multi-agent system formed by a single cross port and multiple directions,connection matrix representing multi-intersection constituting multi-agent system, biThe connection coefficient between the space occupancy of the ith direction of the mth intersection and the mean value of the expected space occupancy is represented;
the distributed global consistency coordination error of the space occupancy of each direction of the single intersection is described as follows:
according to the distributed consistency coordination error, a consistency coordination control item v in the following form is selectedm,i(k),
Or
Wherein c > 0 represents an error learning gain,
distributed green light time coordination control strategy for each direction of mth intersectionThe design is as follows,
or
Where μ > 0 denotes the coordinated control gain, pm,i> 0, and the parameter learning rate is designed as follows:
or
Wherein, Fi=ΠiIs greater than 0; κ > 0 represents the parameter learning gain;
the control and learning gains satisfy the following conditions:
wherein P ═ PT∈RN×N>0,Q=QT∈RN×N> 0 is a positive definite matrix;
wherein the global consistency coordination error vector e (k) e RNThe consistency converges to a bounded neighborhood near the zero value, and the neighborhood upper bound of the consistency coordination error is reduced by increasing the error learning gain c.
Drawings
FIG. 1 is a schematic diagram of the component model.
Detailed Description
A positive system with n ≧ 2 components is shown in FIG. 1, in which xm(k) N represents the state of the component system element m at the time k, wherein m is equal to or more than 0, and m is equal to 1; a ismnM ≠ n denotes the state quantity transfer proportionality coefficient of the component system element m to the component system element n in the sampling period T > 0; i ism(k) Not less than 0 and Om(k) ≧ 0 represents input and output of Compartment System element m within the sampling period, respectively, and ammAnd more than or equal to 0 is a state quantity output proportional coefficient. Thus, the state of the component system element m satisfies the following conservation equation:
further, equation (4-93) is written in vector form:
x(k+1)=Ax(k)+I(k)
in the formula: x (k) ═ x1(k),...,xN(k)]T∈RNIs a system state vector;
I(k)=[I1(k),...,IN(k)]T∈RNinputting for the outside of the system; a is an element of RN×NIs a system state matrix and has
Assuming that the current state quantity of any component is greater than or equal to the total quantity of state transitions in the sampling period, diagonal elements of the state matrix A are all non-negative, and the column sum satisfies:thus, the state matrix A is a non-negative matrix, so that the component system is a type of positive system, and the matrix A is referred to as the component matrix.
The traffic area multi-intersection variable-period multi-direction space occupancy model considering the green light time constraint is as follows:
the global balanced dynamic model of its space occupancy can be described as:
wherein the global space occupancy state vector is x ═ x1,...,xM]T∈RM,xi∈RNThe global nonlinear dynamic vector is f ═ f1,...,fM]T∈RM,fi∈RNControlling the input green time to ui∈RNM represents the number of urban regional intersections, and N represents the number of each direction of a single intersection.
The control target of the space occupancy equilibrium control is to make the distribution consistency coordination error of the space occupancy of each intersection in each direction zero. The distributed consistency coordination error for defining the space occupancy rate of the ith direction of the mth intersection is as follows:
wherein the content of the first and second substances,representing a connection matrix of a multi-agent system formed by a single cross port and multiple directions,connection matrix representing multi-intersection constituting multi-agent system, biAnd the connection coefficient between the space occupancy of the ith direction of the mth intersection and the mean value of the expected space occupancy is shown.
The distributed global consistency coordination error of the space occupancy of each direction of the single intersection can be described as follows:
Considering a multi-agent multi-intersection multidirectional space occupancy network system, all direction communication directed graphs at each intersection are strongly connected, and at least one b existsiNot equal to 0, and selecting a consistency coordination control item v in the following form according to the distributed consistency coordination errorm,i(k),
Or
Wherein c > 0 represents an error learning gain,
distributed green light time coordination control strategy for each direction of mth intersectionThe design is as follows,
or
Where μ > 0 denotes the coordinated control gain, pm,i> 0, and the parameter learning rate is designed as follows:
or
Wherein, Fi=ΠiIs greater than 0; κ > 0 represents the parameter learning gain.
The control and learning gains satisfy the following conditions:
wherein P ═ PT∈RN×N>0,Q=QT∈RN×N> 0 is a positive definite matrix;
then there is a global consistency coordination error vector e (k) e RNThe method is characterized in that the method is uniformly converged in a bounded neighborhood near a zero value, the space occupancy of each direction of a single intersection is uniformly converged to a desired space occupancy, and the upper boundary of the neighborhood of the consistency coordination error can be reduced by increasing the error learning gain c.
Claims (1)
1. A regional traffic signal data driving control method based on a multi-agent network is characterized by comprising the following steps:
(1) creating a component-positive system model
Wherein x ism(k) N represents the state of the component system element m at the time k, wherein m is equal to or more than 0, and m is equal to 1; a ismnAnd m ≠ n denotes Compertent System elements m through CompaThe state quantity of the moment system element n in the sampling period T is greater than 0 and is transferred to a proportionality coefficient; i ism(k) Not less than 0 and Om(k) More than or equal to 0 respectively represents the input and the output of the component system element m in the sampling period;
describing the component system model into a vector form:
x(k+1)=Ax(k)+I(k)
in the formula: x (k) ═ x1(k),...,xN(k)]T∈RNIs a system state vector;
I(k)=[I1(k),...,IN(k)]T∈RNinputting for the outside of the system; a is an element of RN×NIs a system state matrix and has
Assuming that the current state quantity of any component is greater than or equal to the total quantity of state transitions in the sampling period, diagonal elements of the state matrix A are all non-negative, and the column sum satisfies:
(2) establishing a traffic area multi-intersection variable-period multi-direction space occupancy model with green light time constraint
The global balanced dynamic model of its space occupancy can be described as:
wherein the global space occupancy state vector is x ═ x1,...,xM]T∈RM,xi∈RNGlobal non-linear dynamicsThe vector is f ═ f1,...,fM]T∈RM,fi∈RNControlling the input green time to ui∈RNM represents the number of urban regional intersections, and N represents the number of each direction of a single intersection;
the distributed consistency coordination error for defining the space occupancy rate of the ith direction of the mth intersection is as follows:
wherein the content of the first and second substances,representing a connection matrix of a multi-agent system formed by a single cross port and multiple directions,connection matrix representing multi-intersection constituting multi-agent system, bm,iThe connection coefficient between the space occupancy of the ith direction of the mth intersection and the mean value of the expected space occupancy is represented;
the distributed global consistency coordination error of the space occupancy of each direction of the single intersection is described as follows:
according to the distributed consistency coordination error, a consistency coordination control item v in the following form is selectedm,i(k),
Or
Wherein c > 0 represents an error learning gain,
distributed green light time coordination control strategy for each direction of mth intersectionThe design is as follows,
or
Where μ > 0 denotes the coordinated control gain, pm,i> 0, and the parameter learning rate is designed as follows:
or
Wherein, Fi=ΠiIs greater than 0; κ > 0 represents the parameter learning gain;
the control and learning gains satisfy the following conditions:
wherein P ═ PT∈RN×N>0,Q=QT∈RN×N> 0 is a positive definite matrix;
wherein the global consistency coordination error vector e (k) e RNThe consistency converges to a bounded neighborhood near the zero value, and the neighborhood upper bound of the consistency coordination error is reduced by increasing the error learning gain c.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010798063.1A CN112133086B (en) | 2020-08-10 | 2020-08-10 | Regional traffic signal data driving control method based on multi-agent network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010798063.1A CN112133086B (en) | 2020-08-10 | 2020-08-10 | Regional traffic signal data driving control method based on multi-agent network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112133086A CN112133086A (en) | 2020-12-25 |
CN112133086B true CN112133086B (en) | 2022-01-18 |
Family
ID=73851759
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010798063.1A Active CN112133086B (en) | 2020-08-10 | 2020-08-10 | Regional traffic signal data driving control method based on multi-agent network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112133086B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112698634B (en) * | 2020-12-28 | 2021-09-21 | 南京邮电大学 | Event trigger-based traffic intelligent system fixed time dichotomy consistency method |
CN113192318B (en) * | 2021-01-29 | 2022-09-02 | 安徽科力信息产业有限责任公司 | Data drive control regional traffic signal dynamic optimization method and system |
CN113284354B (en) * | 2021-06-18 | 2022-01-14 | 北京航空航天大学 | Traffic elasticity regulation and control method and system based on reinforcement learning |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101281685A (en) * | 2008-01-30 | 2008-10-08 | 吉林大学 | Coordination control method for area mixed traffic self-adaption signal |
CN102110371A (en) * | 2011-03-04 | 2011-06-29 | 哈尔滨工业大学 | Hierarchical multi-agent framework based traffic signal control system |
CN105931474A (en) * | 2016-02-29 | 2016-09-07 | 南京航空航天大学 | City road intersection group local overflow control method with quantum decision |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7742845B2 (en) * | 2002-07-22 | 2010-06-22 | California Institute Of Technology | Multi-agent autonomous system and method |
TWI326859B (en) * | 2007-03-30 | 2010-07-01 | Ind Tech Res Inst | System and method for intelligent traffic control using wireless sensor and actuator networks |
CN101650877A (en) * | 2009-08-31 | 2010-02-17 | 吉林大学 | Method for setting crossing self-adapting changeable driveway |
KR101601958B1 (en) * | 2014-10-14 | 2016-03-09 | 공주대학교 산학협력단 | Roundabout signal metering operation apparatus and method by considering approach lane's degree of saturation |
CN108648446B (en) * | 2018-04-24 | 2020-08-21 | 浙江工业大学 | Road network traffic signal iterative learning control method based on MFD |
CN109785619B (en) * | 2019-01-21 | 2021-06-22 | 南京邮电大学 | Regional traffic signal coordination optimization control system and control method thereof |
-
2020
- 2020-08-10 CN CN202010798063.1A patent/CN112133086B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101281685A (en) * | 2008-01-30 | 2008-10-08 | 吉林大学 | Coordination control method for area mixed traffic self-adaption signal |
CN102110371A (en) * | 2011-03-04 | 2011-06-29 | 哈尔滨工业大学 | Hierarchical multi-agent framework based traffic signal control system |
CN105931474A (en) * | 2016-02-29 | 2016-09-07 | 南京航空航天大学 | City road intersection group local overflow control method with quantum decision |
Also Published As
Publication number | Publication date |
---|---|
CN112133086A (en) | 2020-12-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112133086B (en) | Regional traffic signal data driving control method based on multi-agent network | |
CN110347155B (en) | Intelligent vehicle automatic driving control method and system | |
Cong et al. | Ant colony routing algorithm for freeway networks | |
CN104010863A (en) | Method and module for controlling vehicle speed based on rules and/or costs | |
CN110111562A (en) | A kind of urban transportation macro-regions boundary control method | |
Aslani et al. | Continuous residual reinforcement learning for traffic signal control optimization | |
Nilsson et al. | A micro-simulation study of the generalized proportional allocation traffic signal control | |
Wang et al. | Optimal control design for connected cruise control with edge computing, caching, and control | |
CN112258856B (en) | Method for establishing regional traffic signal data drive control model | |
Han et al. | An extended linear quadratic model predictive control approach for multi-destination urban traffic networks | |
CN114937366A (en) | Traffic flow calculation method based on multi-scale traffic demand and supply conversion | |
Chen et al. | Duality between density function and value function with applications in constrained optimal control and markov decision process | |
CN110456790B (en) | Intelligent networking electric automobile queue optimization control method based on adaptive weight | |
CN109886126B (en) | Regional vehicle density estimation method based on dynamic sampling mechanism and RBF neural network | |
CN113870588B (en) | Traffic light control method based on deep Q network, terminal and storage medium | |
Lu et al. | Stability and fuel economy of nonlinear vehicle platoons: A distributed economic MPC approach | |
CN110011929B (en) | Distributed predictive control method for improving network congestion phenomenon | |
Como et al. | On the well-posedness of deterministic queuing networks with feedback control | |
Pettersson et al. | Stability of hybrid systems using LMIs—a gear-box application | |
CN112258855A (en) | Single-intersection multi-direction space occupancy balance control method | |
Como et al. | On the well-posedness of dynamical flow networks with feedback-controlled outflows | |
Ying et al. | Infrastructure-Assisted cooperative driving and intersection management in mixed traffic conditions | |
CN114137831B (en) | Longitudinal control method and device in intelligent network automobile queue system | |
Ji et al. | Data-driven adaptive cooperative control for urban traffic signal timing in multi-intersections | |
Chen et al. | Heterogeneous Vehicle Platoon Control Based on Predictive Constant Time Headway Strategy |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |