CN105046987A - Pavement traffic signal lamp coordination control method based on reinforcement learning - Google Patents
Pavement traffic signal lamp coordination control method based on reinforcement learning Download PDFInfo
- Publication number
- CN105046987A CN105046987A CN201510338644.6A CN201510338644A CN105046987A CN 105046987 A CN105046987 A CN 105046987A CN 201510338644 A CN201510338644 A CN 201510338644A CN 105046987 A CN105046987 A CN 105046987A
- Authority
- CN
- China
- Prior art keywords
- phase state
- crossing
- traffic
- remote server
- under
- 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.)
- Granted
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 road traffic Control of coordinated signals method based on intensified learning that the invention discloses a kind of, monitoring device is equipped with including each crossing of correspondence, each monitoring device is connect through network module with remote server, its control method are as follows: (1) remote server calculates waiting time S by receiving vision signal; (2) for remote server at each phase state ai, analysis obtains congestion in road situation; (3) remote server obtains the Feasible degree under phase state ai
, when wagon flow can by when be expressed as unobstructed, Feasible degree
It is 1, is otherwise congestion, Feasible degree
It is 0; (4) waiting time S and Feasible degree
, calculate in the optimal driving phase state ai in the crossing; (5) adjustment signal lamp. The present invention is based on the video informations obtained in real time to improve traffic traffic efficiency, maximize the regional traffic flow, alleviate road traffic jam situation by coordinating and controlling multiple crossroad traffic signal lamps in a region.
Description
Technical field
The present invention relates to a kind of road traffic Signalized control method, particularly relate to a kind of road traffic Control of coordinated signals method based on intensified learning.
Background technology
Traffic is the basis of modern society, is the lifeblood of human social economy, and social action and the traffic of people are closely bound up.In a city, motor vehicle, bicycle recoverable amount are large, crossing and section situation numerous and complicated, process such one in large scale, dynamic, there is highly probabilistic distributed system, effectively controlling, is a very complicated job.When not increasing traffic route newly, by rational traffic control, improve the utilization ratio of road, and then improve a kind of effective way that traffic traffic efficiency is quick solution urban transport problems.
But, now congested in traffic, latch up phenomenon is day by day serious.Cause the reason of traffic problems, because vehicle gets more and more on the one hand, traffic programme is delayed with design, be a lot of traffic signal control system comparatively backwardness on the other hand, traffic lights fail to regulate the magnitude of traffic flow according to real-time traffic situation well, play the effect improving traffic traffic efficiency.Help transport solution problem to be more and more subject to people's attention by computing technique and machine intelligence, become trend.
In recent years, a large amount of road traffic watch-dog comes into operation, and real-time traffic video data is transferred to vehicle supervision department incessantly.How to make full use of these traffic video data, improved the control of road traffic signal lamp, to improve road traffic traffic efficiency, caused increasing concern.
There are some intelligent traffic control systems to obtain application at present, but do not solved well at the congestion problems that actual traffic controls in a traffic zone sheet facing between adjacent intersection.Region road traffic cooperation control can process this problem preferably.Region road traffic Signalized control, by considering that the traffic passage situation at multiple crossing in a traffic zone carries out Traffic signal control, can obtain than only considering that the traffic passage situation at single crossing just carries out the higher traffic traffic efficiency of traffic signalization.Road traffic Signalized control method as " green wave band " formula is exactly on the traffic route of specifying; when after the motor vehicle speed scope providing section; require that signal controlling machine is according to section distance; motor vehicle each crossing green light initial time of process; do corresponding adjustment, so, during to guarantee that motor vehicle arrives each crossing; just in time run into " green light ", thus make the motor vehicle of this circuit obtain the highest traffic traffic efficiency.
But the method cannot make actual adjustment according to real-time road traffic situation, make region road traffic signal control to play its advantage, perform practically no function.Such as, when morning peak and evening peak, need the many factors considered, as public transport near bus station is assembled, the quick-fried increasing when going to school and classes are over of school area traffic, etc.These factors can cause some crossing current not smooth, even paralyse.At present, a lot of vehicle supervision department can only rely on the mode floor manager of manpower, the change of direct manual control signal lamp.And manual type regulates the traffic, signal lamp easily causes careless omission; Simultaneously, the manual type signal lamp that regulates the traffic generally can only manage the signal lamp at single crossing, is difficult to the cooperation control accomplishing regional signal lamp, although be likely that traffic participant have passed certain crossing, but because upcoming traffic flow is large, cause still meeting with the difficult situation blocked up.If now consider, regional traffic is coordinated, and stops clearance being likely preferred plan.Therefore, the how current existing real-time traffic video data of maximum using and equipment, feasible region traffic coordinated control, real-time adaptation road condition change, alleviating the workload of vehicle supervision department, alleviate traffic congestion, is our current problem in urgent need to solve.
Summary of the invention
The present invention seeks to: a kind of road traffic Control of coordinated signals method based on intensified learning is provided, by gathering real time video data, based on vehicle-state migration, the traffic lights in a certain region of automatic regulating and control, improve the efficiency that traffic participant is current, alleviate traffic congestion situation, and then alleviate the workload of vehicle supervision department.
Technical scheme of the present invention is: a kind of road traffic Control of coordinated signals method based on intensified learning, and comprise corresponding each crossing and be provided with watch-dog, described in each, watch-dog is connected with remote server through mixed-media network modules mixed-media, and its control method is:
(1) remote server is by receiving the vision signal of watch-dog transmission, and calculate the stand-by period S of vehicle on each road, corresponding crossing car, waiting time is vehicle down time under red light and green light situation;
Using each for intersection red-combination of the corresponding lane mode of green light is as a phase state a
i, remote server is at each phase state a
iunder, according to step (1) in stand-by period of drawing analyze and obtain congestion in road situation;
(3) according to current phase state a
iunder, its green light can the wagon flow passage situation of through lane, and remote server obtains this phase state a
iunder Feasible degree
when wagon flow can by time be expressed as unobstructed, Feasible degree
be 1, otherwise for blocking up, Feasible degree
be 0;
Remote server by step (1) in obtain stand-by period S and step (3) in obtain Feasible degree
this crossing each phase state a is judged in analysis
iunder driving situation, by record and the renewal of certain hour driving situation data, calculated at this crossing optimum driving phase state a by program software analysis meter
i;
(5) according to optimum driving phase state a
i, the red light green light adjusting this crossing combines the time lighted, and obtains maximum driving flow.
In technique scheme, described phase state a
ifor garage's state in each track under the red light green light assembled state of road traffic signal, on the track of corresponding green light, vehicle allows to keep straight on and arrives opposing lane by crossing, and right-turn lane is also allowed to pass through simultaneously, only have when keeping straight on and turning right under all transitable state, described step (3) in Feasible degree
be 1, otherwise be considered as blocking up, Feasible degree
be 0; On the track of corresponding red light, vehicle is dead ship condition.
In technique scheme, the described stand-by period comprises the down time on this track under vehicle red light phase, and fails under green light phase down time of moving ahead.
In technique scheme, according to the vehicle flowrate needs of primary and secondary arterial highway or bus zone, the weighted value in corresponding track is set
In technique scheme, described step (4) in " program software analytical calculation " be kernel function, contrast existing driving situation by kernel function and be deposited in the similarity between the known driving situation in database before, consider the driving situation under the multiple phase state in crossing, the phase state that prioritizing selection does not perform for a long time and important phase state, perform that this phase state can make all vehicles being in waiting status in the difference of red light and green light " stand-by period " with maximum; Described important phase state is the phase state that major trunk roads or bus zone pass through, by arranging the weighted value in corresponding track
initial value realize.
In technique scheme, described mixed-media network modules mixed-media is Ethernet wired module or wireless data transmission network module.
Because technique scheme is used, the present invention compared with prior art has following advantages:
1. the present invention is by obtaining the video information of being included by watch-dog, extract the vehicle flowrate situation under out of phase signal lamp variable condition in video, server adjusts the change of signal lamp in real time according to road traffic situation, and oral sex through-current capacity of giving way maximizes, and minimizing is blocked up;
2. collects real time video data, based on vehicle-state migration, calculate the stand-by period of vehicle, use the accounting method selected phase state of intensified learning, find out a phase state that all vehicle stand-by period can be made the shortest, the change of real-time adjustment signal lamp, meets the fast changing of road traffic situation;
3. consider the primary and secondary in various track and the singularity of driving vehicle in the present invention, weighted value is set
initial value, namely each track arranges different weighted values, when server is selected, pays the utmost attention to the current of these tracks, as major trunk roads or bus zone, optimizes whole road traffic control system.
Accompanying drawing explanation
Fig. 1 is phase state 1 time track of embodiment one in the present invention and the arrangement schematic diagram of parking stall;
Fig. 2 is the phase state 1-4 schematic diagram of embodiment one in the present invention;
Fig. 3 is the phase state 5-8 schematic diagram of embodiment one in the present invention;
Fig. 4 is the network structure topological diagram of a certain traffic zone of embodiment one in the present invention;
Fig. 5 is the network structure topological diagram at a certain crossing of embodiment one in the present invention.
Embodiment
Below in conjunction with drawings and Examples, the invention will be further described:
Embodiment one: shown in Fig. 1 ~ 5, a kind of road traffic Control of coordinated signals method based on intensified learning, comprise corresponding each crossing and be provided with watch-dog, watch-dog described in each is connected with remote server through Ethernet wired network module (or wireless network module), and its control method is:
(1) remote server is by receiving the vision signal of watch-dog transmission, and calculate the stand-by period S of vehicle on each road, corresponding crossing car, waiting time is vehicle down time under red light and green light situation;
Using each for intersection red-combination of the corresponding lane mode of green light is as a phase state a
i, remote server is at each phase state a
iunder, according to step (1) in stand-by period of drawing analyze and obtain congestion in road situation;
(3) according to current phase state a
iunder, its green light can the wagon flow passage situation of through lane, and remote server obtains this phase state a
iunder Feasible degree
when wagon flow can by time be expressed as unobstructed, Feasible degree
be 1, otherwise for blocking up, Feasible degree
be 0; Phase state as shown in Figure 11 time, exit lane is exactly track 1,2,5,6,9,10,13,14, and when they are all unimpeded, then the Feasible degree of phase state 1 is 1.
Remote server by step (1) in obtain stand-by period S and step (3) in obtain Feasible degree
this crossing each phase state a is judged in analysis
iunder driving situation, by record and the renewal of certain hour driving situation data, calculated at this crossing optimum driving phase state a by program software analysis meter
i;
(5) according to optimum driving phase state a
i, the red light green light adjusting this crossing combines the time lighted, and obtains maximum driving flow.
As Figure 2-3, be 8 kinds of phase state figure of the intersection of Four-Lane Road, dotted arrow represents transitable direction, i.e. the track of green light phase, and solid arrow represents the direction of impassability, i.e. the track of red light phase.
Rate-determining steps is as follows:
(1) the Q value look-up table of all intersections server in initialization road traffic network, and deposit Q (s in Q table, action) value, wherein s=< pos, l >, pos refer to as the vehicle location in Fig. 1, and pos=3+5*n (n=1,2 ... Nl/5-1), l refers to as the track in Fig. 1.Q shows intermediate value initial value and is set as 0.Initialization discount factor γ, learning rate α.The phase place a of initialization Servers-all
iweight
each server origination action ai, the as=(a of random initializtion
1, a
2... a
n), and perform.The initial value of simulation time step t is 0.
(2) each crossing server passes through formula
Calculate during all car state s ' show with Q the s existed " k value, and be kept at K table in.Whether wherein whether s ' .l and s.l is similar refers between two tracks similar, and such as, in Fig. 1, track 3 is similar to track 11.S ' and s " track whether Rotational Symmetry, it is 1 that Ι Ψ () to represent in bracket that condition meets result, otherwise is 0; S
s 'represent the state set approximate relevant to state s '.
(3) t ← t+1, each crossing its entrance lane of server observes, the observed data according to connected crossing upgrades
value, if exit lane blocks up
otherwise,
according to formula
upgrade weight, when t is the integral multiple of 500, according to formula
the value of renewal learning rate α, wherein % is complementation symbol.
In system the vehicle that each server arrives alone according to the observation state transition, Q table and K table, pass through
Upgrade the Q value that state s that Q value table and reality exists jointly and phase place ai decomposes the ta of the action of concrete road traffic lamp.Wherein when s=s ' time R (s, s ')=0, otherwise, R (s, s ')=0.
(5) the value that in system, each server is shown according to Q table and K, according to formula
Select the action ai ' with maximum return value, wherein
by two parameters weighting that phase place is relevant
and parameter of blocking up
long-time that do not perform and the outlet selecting those weights large does not have the phase place of jam situation to perform, parameter in addition
consider the jam situation of other crossings when making server do decision-making, achieve the cooperation sharing road traffic situation between server.The car that Selecting phasing preferentially can choose car body long preferentially leaves, vl
srepresent the length over ends of body of car s, namely bus is preferential.Q (s, ta
red)-Q (s, ta
green) represent and wait for that vehicle s is red light at road traffic lamp, and road traffic lamp is the difference of green light income.Take all car income difference sums being in waiting status of certain phase bit motion maximum, and illustrate that this phase place can allow the average latency of vehicle the shortest, just consistent with our final purpose like this, oral sex through-current capacity of giving way maximization, minimizing is blocked up.
(6) each server of system is according to performing the phase place ai ' chosen, and as=(a1 ', a2 ' ... an '), adjustment road traffic signal lamp.Turn (3).
Claims (6)
1., based on a road traffic Control of coordinated signals method for intensified learning, comprise corresponding each crossing and be provided with watch-dog, described in each, watch-dog is connected with remote server through mixed-media network modules mixed-media, and its control method is:
(1) remote server is by receiving the vision signal of watch-dog transmission, and calculate the stand-by period S of vehicle on each road, corresponding crossing car, waiting time is vehicle down time under red light and green light situation;
Using each for intersection red-combination of the corresponding lane mode of green light is as a phase state a
i, remote server is at each phase state a
iunder, according to step (1) in stand-by period of drawing analyze and obtain congestion in road situation;
(3) according to current phase state a
iunder, its green light can the wagon flow passage situation of through lane, and remote server obtains this phase state a
iunder Feasible degree
when wagon flow can by time be expressed as unobstructed, Feasible degree
be 1, otherwise for blocking up, Feasible degree
be 0;
Remote server by step (1) in obtain stand-by period S and step (3) in obtain Feasible degree
this crossing each phase state a is judged in analysis
iunder driving situation, by record and the renewal of certain hour driving situation data, calculated at this crossing optimum driving phase state a by program software analysis meter
i;
(5) according to optimum driving phase state a
i, the red light green light adjusting this crossing combines the time lighted, and obtains maximum driving flow.
2. the road traffic Control of coordinated signals method based on intensified learning according to claim 1, is characterized in that: described phase state a
ifor garage's state in each track under the red light green light assembled state of road traffic signal, on the track of corresponding green light, vehicle allows to keep straight on and arrives opposing lane by crossing, and right-turn lane is also allowed to pass through simultaneously, only have when keeping straight on and turning right under all transitable state, described step (3) in Feasible degree
be 1, otherwise be considered as blocking up, Feasible degree
be 0; On the track of corresponding red light, vehicle is dead ship condition.
3. the road traffic Control of coordinated signals method based on intensified learning according to claim 1, is characterized in that: the described stand-by period comprises the down time on this track under vehicle red light phase, and fails under green light phase down time of moving ahead.
4. the road traffic Control of coordinated signals method based on intensified learning according to claim 1, is characterized in that: according to the vehicle flowrate needs of primary and secondary arterial highway or bus zone, arrange the weighted value in corresponding track
5. the road traffic Control of coordinated signals method based on intensified learning according to claim 1, it is characterized in that: described step (4) in " program software analytical calculation " be kernel function, contrast existing driving situation by kernel function and be deposited in the similarity between the known driving situation in database before, consider the driving situation under the multiple phase state in crossing, the phase state that prioritizing selection does not perform for a long time and important phase state, perform that this phase state can make all vehicles being in waiting status in the difference of red light and green light " stand-by period " with maximum, described important phase state is the phase state that major trunk roads or bus zone pass through, by arranging the weighted value in corresponding track
initial value realize.
6. the road traffic Control of coordinated signals method based on intensified learning according to claim 1, is characterized in that: described mixed-media network modules mixed-media is Ethernet wired module or wireless data transmission network module.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510338644.6A CN105046987B (en) | 2015-06-17 | 2015-06-17 | A kind of road traffic Control of coordinated signals method based on intensified learning |
| PCT/CN2016/075265 WO2016202009A1 (en) | 2015-06-17 | 2016-03-01 | Road traffic light coordination and control method based on reinforcement learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510338644.6A CN105046987B (en) | 2015-06-17 | 2015-06-17 | A kind of road traffic Control of coordinated signals method based on intensified learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105046987A true CN105046987A (en) | 2015-11-11 |
| CN105046987B CN105046987B (en) | 2017-07-07 |
Family
ID=54453489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510338644.6A Active CN105046987B (en) | 2015-06-17 | 2015-06-17 | A kind of road traffic Control of coordinated signals method based on intensified learning |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN105046987B (en) |
| WO (1) | WO2016202009A1 (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105513376A (en) * | 2015-11-20 | 2016-04-20 | 小米科技有限责任公司 | Traffic light adjustment method and device |
| CN105654744A (en) * | 2016-03-10 | 2016-06-08 | 同济大学 | Improved traffic signal control method based on Q learning |
| WO2016202009A1 (en) * | 2015-06-17 | 2016-12-22 | 苏州大学张家港工业技术研究院 | Road traffic light coordination and control method based on reinforcement learning |
| CN106910351A (en) * | 2017-04-19 | 2017-06-30 | 大连理工大学 | A kind of traffic signals self-adaptation control method based on deeply study |
| CN106991707A (en) * | 2017-05-27 | 2017-07-28 | 浙江宇视科技有限公司 | A kind of traffic lights image intensification method and device based on imaging features round the clock |
| CN108961788A (en) * | 2018-07-20 | 2018-12-07 | 张鹏 | Traffic lights wisdom transform method |
| CN109035812A (en) * | 2018-09-05 | 2018-12-18 | 平安科技（深圳）有限公司 | Control method, device, computer equipment and the storage medium of traffic lights |
| CN109215355A (en) * | 2018-08-09 | 2019-01-15 | 北京航空航天大学 | A kind of single-point intersection signal timing optimization method based on deeply study |
| CN109979191A (en) * | 2017-12-28 | 2019-07-05 | 杭州海康威视系统技术有限公司 | Traffic signal control method, device, electronic equipment and computer readable storage medium |
| CN110060475A (en) * | 2019-04-17 | 2019-07-26 | 清华大学 | A kind of multi-intersection signal lamp cooperative control method based on deeply study |
| CN110246345A (en) * | 2019-05-31 | 2019-09-17 | 闽南师范大学 | A kind of signal lamp intelligent control method and system based on HydraCNN |
| CN111047884A (en) * | 2019-12-30 | 2020-04-21 | 西安理工大学 | Traffic light control method based on fog calculation and reinforcement learning |
| CN111260937A (en) * | 2020-02-24 | 2020-06-09 | 武汉大学深圳研究院 | Cross traffic signal lamp control method based on reinforcement learning |
| CN111710177A (en) * | 2020-05-11 | 2020-09-25 | 华东师范大学 | Intelligent traffic signal lamp networking cooperative optimization control system and control method |
| CN112863206A (en) * | 2021-01-07 | 2021-05-28 | 北京大学 | Traffic signal lamp control method and system based on reinforcement learning |
| CN113393679A (en) * | 2021-06-10 | 2021-09-14 | 中南大学 | Regional traffic guidance method and system based on traffic intersection traffic flow identification and statistics |
| CN113487902A (en) * | 2021-05-17 | 2021-10-08 | 东南大学 | Reinforced learning area signal control method based on vehicle planned path |
| CN113763730A (en) * | 2020-06-05 | 2021-12-07 | 杭州海康威视数字技术股份有限公司 | Method and device for determining utilization rate of green wave bandwidth |
| CN113874923A (en) * | 2020-01-21 | 2021-12-31 | 深圳元戎启行科技有限公司 | Traffic signal lamp control method and device, computer equipment and storage medium |
| CN113870589A (en) * | 2021-09-03 | 2021-12-31 | 复旦大学 | Intersection signal lamp and variable lane combined control system and method |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738860B (en) * | 2019-09-18 | 2021-11-23 | 平安科技（深圳）有限公司 | Information control method and device based on reinforcement learning model and computer equipment |
| US11080602B1 (en) | 2020-06-27 | 2021-08-03 | Sas Institute Inc. | Universal attention-based reinforcement learning model for control systems |
| CN113487891B (en) * | 2021-06-04 | 2022-07-01 | 东南大学 | Intersection joint signal control method based on Nash Q learning algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006260313A (en) * | 2005-03-18 | 2006-09-28 | Koito Ind Ltd | Traffic signal control device, and traffic signal system using it |
| CN103280114A (en) * | 2013-06-24 | 2013-09-04 | 电子科技大学 | Signal lamp intelligent control method based on BP-PSO fuzzy neural network |
| CN104077918A (en) * | 2014-07-02 | 2014-10-01 | 上海理工大学 | Urban traffic junction signal lamp self-adaption control method based on vehicle-mounted data |
| CN104575035A (en) * | 2015-01-22 | 2015-04-29 | 大连理工大学 | Intersection self-adaptation control method based on car networking environment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000031707A1 (en) * | 1998-11-23 | 2000-06-02 | Nestor, Inc. | Non-violation event filtering for a traffic light violation detection system |
| CN1936999A (en) * | 2006-10-17 | 2007-03-28 | 大连理工大学 | City area-traffic cooperative control method based wireless sensor network |
| CN101901550B (en) * | 2010-06-24 | 2012-08-29 | 北京航空航天大学 | Vehicle flow detection system and traffic lamp control method |
| CN102142197B (en) * | 2011-03-31 | 2013-11-20 | 汤一平 | Intelligent traffic signal lamp control device based on comprehensive computer vision |
| CN104008659B (en) * | 2014-06-12 | 2016-01-27 | 北京易华录信息技术股份有限公司 | A kind of system and method precisely monitoring intersection signal control device control effects |
| CN105046987B (en) * | 2015-06-17 | 2017-07-07 | 苏州大学 | A kind of road traffic Control of coordinated signals method based on intensified learning |
-
2015
- 2015-06-17 CN CN201510338644.6A patent/CN105046987B/en active Active
-
2016
- 2016-03-01 WO PCT/CN2016/075265 patent/WO2016202009A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006260313A (en) * | 2005-03-18 | 2006-09-28 | Koito Ind Ltd | Traffic signal control device, and traffic signal system using it |
| CN103280114A (en) * | 2013-06-24 | 2013-09-04 | 电子科技大学 | Signal lamp intelligent control method based on BP-PSO fuzzy neural network |
| CN104077918A (en) * | 2014-07-02 | 2014-10-01 | 上海理工大学 | Urban traffic junction signal lamp self-adaption control method based on vehicle-mounted data |
| CN104575035A (en) * | 2015-01-22 | 2015-04-29 | 大连理工大学 | Intersection self-adaptation control method based on car networking environment |
Non-Patent Citations (2)
| Title |
|---|
| XIAOKE ZHOU ET AL.: "A Sarsa-Based Control Model for Real-Time Traffic", 《THE SCIENTIFIC WORLD JOURNAL》 * |
| 李良友等: "一种综合优化的路面交通监控系统", 《数采与监测》 * |
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016202009A1 (en) * | 2015-06-17 | 2016-12-22 | 苏州大学张家港工业技术研究院 | Road traffic light coordination and control method based on reinforcement learning |
| CN105513376A (en) * | 2015-11-20 | 2016-04-20 | 小米科技有限责任公司 | Traffic light adjustment method and device |
| CN105654744A (en) * | 2016-03-10 | 2016-06-08 | 同济大学 | Improved traffic signal control method based on Q learning |
| CN105654744B (en) * | 2016-03-10 | 2018-07-06 | 同济大学 | A kind of improvement traffic signal control method based on Q study |
| CN106910351B (en) * | 2017-04-19 | 2019-10-11 | 大连理工大学 | A kind of traffic signals self-adaptation control method based on deeply study |
| CN106910351A (en) * | 2017-04-19 | 2017-06-30 | 大连理工大学 | A kind of traffic signals self-adaptation control method based on deeply study |
| CN106991707A (en) * | 2017-05-27 | 2017-07-28 | 浙江宇视科技有限公司 | A kind of traffic lights image intensification method and device based on imaging features round the clock |
| CN106991707B (en) * | 2017-05-27 | 2020-02-18 | 浙江宇视科技有限公司 | Traffic signal lamp image strengthening method and device based on day and night imaging characteristics |
| CN109979191A (en) * | 2017-12-28 | 2019-07-05 | 杭州海康威视系统技术有限公司 | Traffic signal control method, device, electronic equipment and computer readable storage medium |
| CN109979191B (en) * | 2017-12-28 | 2022-02-11 | 杭州海康威视系统技术有限公司 | Traffic signal control method, traffic signal control device, electronic equipment and computer-readable storage medium |
| CN108961788A (en) * | 2018-07-20 | 2018-12-07 | 张鹏 | Traffic lights wisdom transform method |
| CN109215355A (en) * | 2018-08-09 | 2019-01-15 | 北京航空航天大学 | A kind of single-point intersection signal timing optimization method based on deeply study |
| CN109035812A (en) * | 2018-09-05 | 2018-12-18 | 平安科技（深圳）有限公司 | Control method, device, computer equipment and the storage medium of traffic lights |
| CN109035812B (en) * | 2018-09-05 | 2021-07-27 | 平安科技（深圳）有限公司 | Traffic signal lamp control method and device, computer equipment and storage medium |
| CN110060475A (en) * | 2019-04-17 | 2019-07-26 | 清华大学 | A kind of multi-intersection signal lamp cooperative control method based on deeply study |
| CN110246345A (en) * | 2019-05-31 | 2019-09-17 | 闽南师范大学 | A kind of signal lamp intelligent control method and system based on HydraCNN |
| CN110246345B (en) * | 2019-05-31 | 2020-09-29 | 闽南师范大学 | Signal lamp intelligent control method and system based on HydraCNN |
| CN111047884A (en) * | 2019-12-30 | 2020-04-21 | 西安理工大学 | Traffic light control method based on fog calculation and reinforcement learning |
| CN113874923A (en) * | 2020-01-21 | 2021-12-31 | 深圳元戎启行科技有限公司 | Traffic signal lamp control method and device, computer equipment and storage medium |
| CN111260937A (en) * | 2020-02-24 | 2020-06-09 | 武汉大学深圳研究院 | Cross traffic signal lamp control method based on reinforcement learning |
| CN111710177A (en) * | 2020-05-11 | 2020-09-25 | 华东师范大学 | Intelligent traffic signal lamp networking cooperative optimization control system and control method |
| CN111710177B (en) * | 2020-05-11 | 2021-07-27 | 华东师范大学 | Intelligent traffic signal lamp networking cooperative optimization control system and control method |
| CN113763730A (en) * | 2020-06-05 | 2021-12-07 | 杭州海康威视数字技术股份有限公司 | Method and device for determining utilization rate of green wave bandwidth |
| CN112863206A (en) * | 2021-01-07 | 2021-05-28 | 北京大学 | Traffic signal lamp control method and system based on reinforcement learning |
| CN113487902A (en) * | 2021-05-17 | 2021-10-08 | 东南大学 | Reinforced learning area signal control method based on vehicle planned path |
| CN113487902B (en) * | 2021-05-17 | 2022-08-12 | 东南大学 | Reinforced learning area signal control method based on vehicle planned path |
| CN113393679A (en) * | 2021-06-10 | 2021-09-14 | 中南大学 | Regional traffic guidance method and system based on traffic intersection traffic flow identification and statistics |
| CN113870589A (en) * | 2021-09-03 | 2021-12-31 | 复旦大学 | Intersection signal lamp and variable lane combined control system and method |
| CN113870589B (en) * | 2021-09-03 | 2023-05-02 | 复旦大学 | Intersection signal lamp and variable lane joint control system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105046987B (en) | 2017-07-07 |
| WO2016202009A1 (en) | 2016-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105046987A (en) | Pavement traffic signal lamp coordination control method based on reinforcement learning | |
| CN109544945B (en) | Regional control phase timing optimization method based on lane saturation | |
| CN104933876B (en) | A kind of control method of adaptive smart city intelligent traffic signal | |
| CN106297326A (en) | Based on holographic road network tide flow stream Lane use control method | |
| CN103578281B (en) | A kind of main line of communication signal lamp optimal control method and device | |
| CN104464314B (en) | A kind of Bus Priority method of bus special lane crossing | |
| CN110264717A (en) | A kind of municipal intelligent traffic regulator control system | |
| CN107025792A (en) | The method of adjustment and device in track and signal lamp cycle based on vehicle queue length | |
| CN106297329A (en) | A kind of signal timing dial adaptive optimization method of networking signals machine | |
| CN109887289A (en) | A kind of network vehicle flowrate maximization approach of urban traffic network model | |
| CN109584580B (en) | Intelligent control method and intelligent control system for urban traffic lights | |
| CN104036644A (en) | Intelligent traffic light control system and implementing method thereof | |
| CN105489034A (en) | Main line full traffic control system and method | |
| CN107730886A (en) | Dynamic optimization method for traffic signals at urban intersections in Internet of vehicles environment | |
| CN110136456A (en) | Traffic lights anti-clogging control method and system based on deeply study | |
| CN107886744A (en) | One kind is used for subway station adjacent to intersection public transport priority signal control method | |
| CN107293135B (en) | A kind of crossing self-adapting vehicle U-turn at intersection traffic guidance system and control method | |
| CN109461314B (en) | Urban intelligent lane change method and intelligent lane change system | |
| CN111932916B (en) | Control method and control system for dynamic emergency lane of urban road | |
| CN109920244A (en) | Changeable driveway real-time control system and method | |
| CN103942968B (en) | A kind of intelligent traffic light dispatching system based on traffic flow perception and dispatching method thereof | |
| CN105427630A (en) | Traffic control system based on traffic flow | |
| CN112017449B (en) | Traffic light intelligent control system and method based on Internet of things and letter fusion cloud platform | |
| WO2023035666A1 (en) | Urban road network traffic light control method based on expected reward estimation | |
| CN111009140A (en) | Intelligent traffic signal control method based on open-source road condition information |
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 |