CN109727470A - A kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene - Google Patents
A kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene Download PDFInfo
- Publication number
- CN109727470A CN109727470A CN201910015069.4A CN201910015069A CN109727470A CN 109727470 A CN109727470 A CN 109727470A CN 201910015069 A CN201910015069 A CN 201910015069A CN 109727470 A CN109727470 A CN 109727470A
- Authority
- CN
- China
- Prior art keywords
- intersection
- vehicle
- speed
- decision
- traffic
- 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
Landscapes
- Traffic Control Systems (AREA)
Abstract
A kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene proposed by the present invention, belongs to field of intelligent transportation technology.It is different from tradition rule-based or intensified learning no signal lamp intersection decision-making technique, it violates the traffic regulations testing mechanism invention introduces vehicle, realizes regular merging with intensified learning method.Judge whether intersection has vehicle to violate the traffic regulations using the vehicle violation detection algorithm of Implicit Curves race first;When vehicle is in compliance with traffic rules, vehicle pass-through decision in intersection is carried out using rule and method;It is on the contrary then method based on intensified learning is used to carry out intersection vehicle pass-through decision.The present invention is realized to abide by and independently be passed through decision with the no signal lamp intersection intelligent network connection automobile under the vehicles in complex traffic scene that mixes of violating the traffic regulations, and provides technical support to improve traffic safety under complex environment and intersection passing efficiency.
Description
Technical field
The invention belongs to field of intelligent transportation technology, in particular to a kind of Distributed Intelligent Network connection automobile intersection is complicated
Scene passage decision-making technique.
Background technique
Correlation report show China in 2016 because caused by road traffic accident casualties position it is at the forefront in the world, directly
Caused by economic loss it is up to billions of.Document (urban road cross junction Research on Safety Assessment of the Ren Shuhang based on TCT
[D] Chang An University, 2016.) the traffic accident data survey carried out to representative cities, China is had recorded, investigation result is aobvious
Show that city intersection is the high-incidence section of traffic accident first.The document by adjusting 15, certain city intersection on the spot
It grinds, it is traffic accident is high-incidence at intersection basic reason that pedestrian and driver, which violate the traffic regulations, as the result is shown.
Intelligent network connection automobile (Intelligent Connected Vehicle, ICV) refers to equipped with advanced vehicle-mounted biography
The devices such as sensor, controller, actuator, and merge V2X communication (Vehicle-to-Everything, V2X refer to vehicle and outer
Boundary carries out information exchange), it realizes that the information exchanges such as vehicle and people, vehicle, road are shared, realizes safety, comfortable, energy conservation, efficiently traveling, and
Final alternative people is come the young mobile that operates.ICV is the inevitable development trend of automobile industry.Intersection passing decision is
Refer to that ICV passes through information search, processing, finally makes the decision that vehicle passes through intersection with what speed.Carry out current decision
It is top priority of the ICV by intersection.Therefore, research is abided by and the complex cross road mixed under scene that violates the traffic regulations
The current decision of mouth ICV, has very important theory significance and engineering value.
One Main Topics of intersection passing decision are that the distributed cross crossing passage of non-mandrel roller is determined
Plan, rule-based distributed cross crossing decision have been achieved with a series of research achievements, and main includes being based on acceptability risk
The method of model, the method based on dynamic game, resource locking method based on collision table etc..But rule-based distributed friendship
Cross road mouth decision only can be only achieved desired effect when all vehicles are in compliance with traffic rules, disobey when intersection exists
When method traffic rules vehicle, rule-based distributed cross crossing decision-making results is not very ideal.
Intensified learning is also known as reinforcement function, is a kind of important machine learning method.Intensified learning is also applied in recent years
Passage decision in intersection.But intensified learning needs relatively large calculation amount it is difficult to ensure that ICV whole process online processing.
It is also achieved for the violate the traffic regulations detection method of (People's Republic of China's traffic route safety law) of vehicle
Certain research achievement (such as bibliography 1,2,3).The detection method to violate the traffic regulations at present is mainly used in as public security traffic
Management board, which identifies traffic offence, provides auxiliary reference, but does not directly apply to the automatic Pilot of intersection
Among vehicle is made decisions on one's own.
Summary of the invention
The purpose of the present invention is being directed to the demand, a kind of Distributed Intelligent Network connection automobile intersection complex scene is proposed
Current decision-making technique, the method for the present invention, for no signal lamp intersection, vehicle are violated and is handed under conditions of full-automatic driving
Among the automatic driving vehicle that the testing mechanism of drift then is introduced into intersection is made decisions on one's own, rule and learning decision are realized
The fusion of method, for improve complex environment under traffic safety and intersection passing efficiency technical support is provided.
To achieve the goals above, the present invention adopts the following technical scheme:
A kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene proposed by the present invention, feature
It is, this method includes that 1) system platform is built with parameter setting, 2) vehicle violates the traffic regulations to detect and determines, 3) is rule-based
The intersection passing decision of method determines, 4) the intersection passing decision based on intensified learning determines four parts, specific real
Existing process is as follows:
1) system platform is built and parameter setting, is specifically included:
1-1) system platform is built:
Vehicle-state detection and V2X broadcast system are built in intersection, including trackside server, video camera and V2X lead to
Believe equipment;Wherein, video camera is mounted on the top of each lane direction to the car in intersection, guarantees that all video camera coverings include
The all areas of intersection;The trackside server is installed on the trackside of intersection, connect with each video camera, for connecing
Receive the image of all video camera acquisitions, processing includes all data including image;V2X communication equipment is loaded in intersection model
On each intelligent network connection automobile in enclosing, for receiving the intersection map and the intersection that pass through trackside server transport
Global vehicle-state in range, the overall situation vehicle-state include the license plate number of all vehicles within the scope of intersection, vehicle
Coordinate, car speed, headstock are towards angle and acquisition time;
1-2) parameter setting:
Set cycle period T=1/f, wherein f is that intelligent network joins automobile by V2X communication equipment reception intersection
The frequency of figure and global vehicle-state;
Intersection earth coordinates are set to be from West to East X-axis, from south orientation using intersection geometric center as zero point
North is Y-axis, the right-handed Cartesian rectangular coordinate system established;
Set following four kinds of speeds: a) speed 1 be according to actual traffic situation, a speed of selection be greater than 10km/h and
Speed less than 20km/h;B) speed 2 is according to actual traffic situation, and a speed of selection is greater than 20km/h and is less than road
The speed of Maximum speed limit;C) speed 3 is instructing from vehicle speed for the intersection passing decision output based on learning method, this refers to
It enables and is more than or equal to 0km/h and is less than or equal to 20km/h;D) it stops, i.e., car speed is 0km/h;
2) data acquisition violates the traffic regulations with vehicle detects judgement, specifically includes the following steps:
2-1) clock starts a new cycle period with frequency f generation trigger signal;
When 2-2) new cycle period starts, intersection map and complete is received with frequency f by V2X communication equipment from vehicle
Office's vehicle-state;
2-3) it will be used as and input by the intersection map and overall situation vehicle-state that step 2-2) is obtained, by intersection
Whether each vehicle in range observes traffic rules and regulations as output, and according to violating the traffic regulations, detection algorithm judges intersection
Whether there is vehicle to violate the traffic regulations in range;When all vehicles within the scope of intersection are in compliance with traffic rules, execute
Step 3-1);When at least violating the traffic regulations in the presence of a vehicle within the scope of intersection, step 4-1 is executed);Wherein, institute
Stating the detection algorithm that violates the traffic regulations includes the vehicle violation detection algorithm of Implicit Curves race, the vehicle based on computer vision technique
Peccancy detection algorithmic method and the traffic violation detection method based on track of vehicle;
3) the intersection passing decision of rule-based approach determines, specifically includes the following steps:
3-1) by from coordinate of the vehicle in the earth coordinates of intersection and intersection map judge from vehicle whether
Within the scope of intersection: within the scope of intersection, being exported from vehicle speed command speed 1, this cycle period meter when from vehicle coordinate
Terminate, intelligent network joins automobile and travels according to command speed 1, return step 2-1), it waits clock to generate trigger signal and starts newly
Cycle period;When from vehicle coordinate not within the scope of intersection, execute step 3-2);
3-2) under the earth coordinates of intersection, definition threshold time is hanging down from vehicle coordinate and intersection stop line
Straight distance is divided by from vehicle speed, and for from vehicle coordinate close to intersection the case where definition threshold time is less than 2/f, otherwise is known as
From vehicle coordinate not close to intersection;Whether judge close to intersection from vehicle coordinate: when from vehicle coordinate close to handing over
When cross road mouth, output terminates from vehicle speed command speed 2, the calculating of this cycle period, and intelligent network joins automobile according to command speed 2
Traveling, return step 2-1), it waits clock to generate trigger signal and starts new cycle period;When from vehicle coordinate close to intersection
When, execute step 3-3);
3-3) according to step 2-2) the intersection map received and global vehicle-state, using rule-based approach
Intersection decision carries out intersection passing decision to from vehicle;Then according to the intersection decision knot of rule-based approach
Fruit, exporting the speed instruction from vehicle is speed 1 or parking, and the calculating of this cycle period terminates, and intelligent network joins automobile according to instruction speed
1 traveling of degree or parking, return step 2-1), it waits clock to generate trigger signal and starts new cycle period;Wherein, the base
In the intersection decision of rule and method include the method based on acceptability risk model, the control method based on dynamic game,
Resource locking method based on collision table and it is based on vehicle pass-through rule base method;
4) the intersection passing decision based on intensified learning, specifically includes the following steps:
4-1) according to step 2-2) the intersection map received and global vehicle-state, whether handing over from vehicle coordinate
Judged within the scope of cross road mouth: when from vehicle coordinate not within the scope of intersection, execute step 4-2);Exist when from vehicle coordinate
Within the scope of intersection, step 4-3 is executed);
4-2) use step 3-2) defined in from vehicle coordinate whether close to intersection, to from vehicle coordinate whether close to handing over
Cross road mouth is judged: when being not close to intersection from vehicle coordinate, being exported from vehicle speed command speed 2, this cycle period
Calculating terminates, and intelligent network joins automobile and travels according to command speed 2, return step 2-1), it waits clock to generate trigger signal and starts
New cycle period;When from vehicle coordinate close to intersection, output is instructed from vehicle speed and is stopped, i.e., speed is 0m/s, is originally followed
The calculating of ring period terminates, and intelligent network joins parking of automobile, return step 2-1), it waits clock to generate trigger signal and starts new circulation
Period;
4-3) according to step 2-2) the intersection map received and global vehicle-state, using based on Timing Difference side
The intensified learning of method carries out intersection passing decision to from vehicle;Then it is obtained according to the intensified learning based on Timing Difference method
The intersection result of decision, exporting from the instruction of the speed of vehicle is speed 3, and the calculating of this cycle period terminates, and intelligent network joins automobile
Travelled according to instruction speed 3, return step 2-1), it waits clock to generate trigger signal and starts new cycle period;Wherein, described
Intensified learning based on Timing Difference method includes the Q-learning algorithm of the Sarsa algorithm and different strategy with strategy.
Feature of the present invention and the utility model has the advantages that with traditional rule-based distributed cross crossing decision or intensified learning is based on
Intersection decision-making technique it is different, the present invention is under conditions of full-automatic driving, for no signal lamp intersection, by vehicle
Among the automatic driving vehicle that the testing mechanism to violate the traffic regulations is introduced into intersection is made decisions on one's own, a kind of distribution is proposed
Formula intelligent network joins the current decision-making technique of automobile intersection complex scene, realizes merging for rule and learning decision method.It adopts
The whether illegal traffic rules of intersection vehicle are judged with the vehicle testing mechanism that violates the traffic regulations;When vehicle is observed traffic rules and regulations
When, vehicle pass-through decision in intersection is carried out using rule and method;When intersection has the vehicle to violate the traffic regulations, adopt
Intersection vehicle pass-through decision is carried out with learning method.The present invention, which realizes, to be abided by and the complicated friendship mixed that violates the traffic regulations
No signal lamp intersection intelligent network under logical scene joins automobile independently current decision, to improve the traffic safety under complex environment
Technical support is provided with intersection passing efficiency.
Detailed description of the invention
Fig. 1 is a kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene of the embodiment of the present invention
Flow diagram.
Specific embodiment
To technical solution of the present invention, detailed description are as follows with reference to the accompanying drawings and embodiments:
A kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene proposed by the present invention, process
Figure is as shown in Figure 1, this method includes that 1) system platform is built with parameter setting, 2) vehicle violates the traffic regulations to detect and determine, 3)
The intersection passing decision of rule-based approach determines, 4) the intersection passing decision based on intensified learning determines four
Point, the specific implementation process is as follows:
1) system platform is built and parameter setting, is specifically included:
1-1) system platform is built:
Vehicle-state detection and V2X broadcast system are built in intersection, including trackside server, video camera and V2X lead to
Believe equipment;Wherein, video camera is mounted on the top of each lane direction to the car in intersection, guarantees that all video camera coverings include
The all areas of intersection;The trackside server is installed on the trackside of intersection, connect with each video camera, for connecing
Receive the image of all video camera acquisitions, processing includes all data including image;V2X communication equipment is loaded in intersection model
On each intelligent network connection automobile in enclosing, for receiving the intersection map and the intersection that pass through trackside server transport
Global vehicle-state in range, the overall situation vehicle-state include the license plate number of all vehicles within the scope of intersection, vehicle
Coordinate, car speed, headstock are towards angle and acquisition time;
By the system platform built, for global vehicle-state (i.e. all vehicles within the scope of real-time detection intersection
License plate number, vehicle coordinate, car speed, headstock towards the information such as angle and acquisition time, and pass through V2X communications facility
It is communicated with ICV, intersection map (i.e. the coordinate of the travelable road of intersection vehicle) is wide with global vehicle-state
It broadcasts to each ICV within the scope of intersection.Each ICV within the scope of intersection is fitted with vehicle-mounted V2X equipment,
For receiving the license plate number of all vehicles, vehicle coordinate, car speed, headstock court within the scope of intersection map and intersection
To the information such as angle and acquisition time.
1-2) parameter setting:
Set cycle period T=1/f, wherein f, which is ICV, receives intersection map and complete by vehicle-mounted V2X communication equipment
The frequency of office's vehicle-state, the value of cycle period T selects according to the actual situation, general value range 10ms and 200ms it
Between.
Intersection earth coordinates are set to be from West to East X-axis, from south orientation using intersection geometric center as zero point
North is Y-axis, the right-handed Cartesian rectangular coordinate system established.
Set following four kinds of speeds: a) speed 1 be according to actual traffic situation, a speed of selection be greater than 10km/h and
Speed less than 20km/h;B) speed 2 is according to actual traffic situation, and a speed of selection is greater than 20km/h and is less than road
The speed of Maximum speed limit;C) speed 3 is that the intersection passing decision based on learning method determines instructing from vehicle speed for output,
The instruction is more than or equal to 0km/h and is less than or equal to 20km/h;D) it stops, i.e., car speed is 0km/h.
2) data acquisition violates the traffic regulations with vehicle detects judgement, specifically includes the following steps:
2-1) clock starts a new cycle period with frequency f generation trigger signal.
When 2-2) new cycle period starts, intersection map and complete is received with frequency f by V2X communication equipment from vehicle
Office's vehicle-state;
2-3) it will be used as and input by the intersection map and overall situation vehicle-state that step 2-2) is obtained, by intersection
Whether each vehicle in range observes traffic rules and regulations as output, and according to violating the traffic regulations, detection algorithm judges intersection
Whether there is vehicle to violate the traffic regulations (i.e. violation People's Republic of China's traffic route safety law) in range;When intersection model
When all vehicles in enclosing are in compliance with traffic rules, step 3-1 is executed);When within the scope of intersection at least exist a vehicle
When violating the traffic regulations, step 4-1 is executed);Wherein, the detection algorithm that violates the traffic regulations mainly includes Implicit Curves race
Vehicle violation detection algorithm (for details, reference can be made to bibliography 1), the vehicle violation detection algorithm side based on computer vision technique
Method (for details, reference can be made to bibliography 2), (for details, reference can be made to bibliography for the traffic violation detection method based on track of vehicle
3) etc.;
The present embodiment uses the vehicle violation detection algorithm of Implicit Curves race, and specific implementation is as follows:
2-3-1) in intersection, geodetic coordinates is fastened, and according to intersection actual conditions, is described using Implicit Curves race F
Different zones on the map of intersection, wherein F={ fi(x, y) }, i=1,2 ..., n, i indicate the on the map of intersection
I region, (x, y) indicate vehicle location, Implicit Curves fi(x,y)>0、fi(x, y)=0, fi(x, y) < 0 respectively indicates vehicle
In i-th of Implicit Curves fiIn (x, y), vehicle is in i-th of Implicit Curves fiOn (x, y), vehicle is in i-th of Implicit Curves fi
(x, y) outside;
All rule Rule (j) violating the regulations 2-3-2) are set according to People's Republic of China's traffic route safety law, wherein j
Indicate the jth kind vehicle violation situation of People's Republic of China's traffic route safety law defined;Implicit Curves when setting 1 frame
Initial value be 0, i.e. F={ fi(x, y)=0 }, i=1,2 ..., n;Then the intersection map received according to step 2-2)
With global vehicle-state, the value of the Implicit Curves race F of this cycle period is calculated, judges last time cycle period and this circulating cycle
Whether the value of the Implicit Curves race F of phase meets Rule (j);Corresponding traffic rules are violated if meeting Rule (j), it is on the contrary
Corresponding traffic rules are not violated then;
Now by straight trip turn left it is violating the regulations for be illustrated: setting Implicit Curves f1(x, y) indicates that region 1 only allows to keep straight on,
Set Implicit Curves f2(x, y) indicates the region 2 that enters if vehicle is turned left from region 1, setting straight trip turn left break rules and regulations be
Rule (1): f1(x, y) >=0 → f2(x, y) >=0, i.e., Implicit Curves when jumping to this cycle period from last time cycle period
Value occur change;Then according to step 2-2) the intersection map received and global vehicle-state, judge that last time is followed
Whether ring period and this cycle period vehicle, which meet vehicle straight trip, is turned left Rule violating the regulations (1);Vehicle straight trip is left if meeting
Turn it is violating the regulations, it is on the contrary then without vehicle straight trip turn left it is violating the regulations;
3) the intersection passing decision of rule-based approach determines, specifically includes the following steps:
3-1) by from coordinate of the vehicle in the earth coordinates of intersection and intersection map judge from vehicle whether
Within the scope of intersection: within the scope of intersection, being exported from vehicle speed command speed 1, this cycle period meter when from vehicle coordinate
Terminate, then intelligent network connection automobile is travelled according to command speed 1, return step 2-1), it waits clock to generate trigger signal and opens
Begin new cycle period;When from vehicle coordinate not within the scope of intersection, execute step 3-2);
3-2) under the earth coordinates of intersection, definition threshold time is hanging down from vehicle coordinate and intersection stop line
Straight distance is divided by from vehicle speed, and for from vehicle coordinate close to intersection the case where definition threshold time is less than 2/f, otherwise is known as
From vehicle coordinate not close to intersection.Whether judge close to intersection from vehicle coordinate.It is handed over when from vehicle coordinate is not close
When cross road mouth, output terminates from vehicle speed command speed 2, the calculating of this cycle period, and then intelligent network connection automobile is according to instruction speed
2 traveling of degree, return step 2-1), it waits clock to generate trigger signal and starts new cycle period;Intersect when from vehicle coordinate is close
When crossing, step 3-3 is executed);
3-3) according to step 2-2) the intersection map received and global vehicle-state, using rule-based approach
Intersection decision carries out intersection passing decision to from vehicle;Then according to the intersection decision knot of rule-based approach
Fruit, exporting the speed instruction from vehicle is speed 1, and the calculating of this cycle period terminates, and then intelligent network joins automobile according to command speed 1
Traveling or parking, return step 2-1), it waits clock to generate trigger signal and starts the new Zhou Xunhuan phase;Wherein, described based on rule
Then the intersection decision of method mainly include the method based on acceptability risk model, the control method based on dynamic game,
Resource locking method based on collision table, based on vehicle pass-through rule base method etc.;
The intersection decision of the present embodiment rule-based approach uses the vehicle pass-through rule base method that is based on, specific implementation
Mode is as follows: the priority highest of setting ambulance and fire fighting truck, and the priority of remaining type vehicle is all identical and lower than rescue
The priority of vehicle and fire fighting truck,;Speed instruction after setting vehicle deceleration is speed 1;Then it is advised using based on vehicle pass-through
Then library method is calculated (based on vehicle pass-through rule base method for details, reference can be made to bibliography 4, page 129, Fig. 6 .9);
4) the intersection passing decision based on intensified learning, specifically includes the following steps:
4-1) according to step 2-2) the intersection map received and global vehicle-state, whether handing over from vehicle coordinate
Judged within the scope of cross road mouth.When from vehicle coordinate not within the scope of intersection, execute step 4-2);Exist when from vehicle coordinate
Within the scope of intersection, step 4-3 is executed);
4-2) use step 3-3) defined in from vehicle coordinate whether close to intersection, to from vehicle coordinate whether close to handing over
Cross road mouth is judged.When being not close to intersection from vehicle coordinate, export from vehicle speed command speed 2, this cycle period
Calculating terminates, and then intelligent network connection automobile is travelled according to command speed 2, return step 2-1), wait clock to generate trigger signal
Start new cycle period;When from vehicle coordinate close to intersection, output is instructed from vehicle speed and is stopped, i.e., speed is 0m/s,
The calculating of this cycle period terminates, and intelligent network joins parking of automobile, return step 2-1), wait clock generation trigger signal to start new
Cycle period;
4-3) according to step 2-2) the intersection map received and global vehicle-state, using based on Timing Difference side
The intensified learning of method carries out intersection passing decision to from vehicle;Then it is obtained according to the intensified learning based on Timing Difference method
The intersection result of decision, exporting from the instruction of the speed of vehicle is speed 3, and the calculating of this cycle period terminates, and intelligent network joins automobile
Travelled according to instruction speed 3, return step 2-1), it waits clock to generate trigger signal and starts new cycle period;Wherein, described
Intensified learning based on Timing Difference method includes the Q-learning algorithm of the Sarsa algorithm and different strategy with strategy;
The present embodiment uses the Sarsa algorithm with strategy, and specific implementation is as follows:
With a meters for distance interval, 0.5m≤a≤2m, edge is parallel to intersection earth coordinates X-axis respectively and Y-axis is drawn
The intersection map partitioning received is multiple grids, by the collection cooperation of all grids within the scope of intersection by straight line
For state set S;Using bkm/h as speed interval, 0km/h≤b≤10km/h chooses multiple speed between 0km/h to 20km/h
The set of composition is as behavior aggregate A;Definition sets track to meet People's Republic of China's traffic route safety law and making intelligence
One group of tracing point that connection automobile passes through intersection can be netted.From vehicle along setting the return of track as r1,0 < r1 < 100;From
Vehicle, which deviates, sets the return of track as r2, -50 < r2 < 0;The return to collide from vehicle and other vehicles is set as r3, -100
< r3 < -50;Set decay factor γ, 0 < γ < 1;Assessment strategy and action strategy are set as greedy strategy;Then according to same
The Sarsa algorithm of strategy carries out operation (with strategy Sarsa algorithm for details, reference can be made to bibliography 5, page 79, Fig. 5 .6), general
Calculated result is set as exporting the speed instruction from vehicle, that is, speed 3.
In one cycle, this patent method at most exports a speed instruction, i.e. parking, speed 1, speed 2, speed 3
In one instruction;Then ICV responds speed instruction, according to instruction institute to speed driving ICV by intersection, until ICV
Leave intersection region.
With traditional rule-based distributed cross crossing decision or the intersection decision-making technique based on intensified learning not
Together, the present invention is under conditions of full-automatic driving, for no signal lamp intersection, the detection machine that vehicle is violated the traffic regulations
Among the automatic driving vehicle that system is introduced into intersection is made decisions on one's own, a kind of Distributed Intelligent Network connection automobile crossroad is proposed
Mouthful current decision-making technique realizes merging for rule and learning decision method.The present invention, which realizes, to be abided by and violates the traffic regulations
No signal lamp intersection intelligent network under the vehicles in complex traffic scene mixed joins automobile independently current decision, to improve complex environment
Under traffic safety and intersection passing efficiency provide technical support.
Scheme example described above only preferred embodiments of the invention are described, not to the scope of the present invention into
Row limits, and without departing from the spirit of the design of the present invention, those of ordinary skill in the art do technical solution of the present invention
Modification and improvement out, should fall within the scope of protection determined by the claims of the present invention.
Bibliography explanation:
Bibliography 1: Ma Liwen, a kind of vehicle violation detection algorithm using Implicit Curves race of Guo Yukun, Li Jinping
[J] Xian Electronics Science and Technology University journal (natural science edition), 2016,43 (2): 139-144.
Bibliography 2: recognizer research [D] the violating the regulations Harbin Institute of Technology in intelligent transportation, 2014.
Bibliography 3: Jia Yonghua, a kind of traffic violation detection method [J] based on track of vehicle of Zhang Xiao, Zhu Jiang
Chinese public safety, 2012 (11): 196-200.
Bibliography 4: Lu Guangquan, Wang Yunpeng, Tian great Xin collaborative truck safety control technology [M] Science Press,
2014. (page 129, Fig. 6 .9 no signal intersection driving behavior flow chart)
Bibliography 5: Guo Xian, Fang Yongchun are explained the profound in simple terms intensified learning [M], Electronic Industry Press, and 2018 the (the 79th
Page, Fig. 5 .6 is with strategy Sarsa nitrification enhancement).
Claims (5)
1. a kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene, which is characterized in that this method packet
Include 1) system platform build with parameter setting, 2) vehicle violate the traffic regulations detection determine, 3) crossroad of rule-based approach
The current decision of mouth determines, 4) the intersection passing decision based on intensified learning determines four parts, the specific implementation process is as follows:
1) system platform is built and parameter setting, is specifically included:
1-1) system platform is built:
Vehicle-state detection and V2X broadcast system are built in intersection, including trackside server, video camera and V2X communication are set
It is standby;Wherein, video camera is mounted on the top of each lane direction to the car in intersection, guarantees all video camera coverings comprising intersecting
The all areas at crossing;The trackside server is installed on the trackside of intersection, connect with each video camera, for receiving
There are the image of video camera acquisition, processing to include all data including image;V2X communication equipment is loaded within the scope of intersection
Each intelligent network connection automobile on, for receiving through the intersection map of trackside server transport and intersection range
Interior global vehicle-state, the overall situation vehicle-state include the license plate numbers of all vehicles within the scope of intersection, vehicle coordinate,
Car speed, headstock are towards angle and acquisition time;
1-2) parameter setting:
Set cycle period T=1/f, wherein f be intelligent network join automobile by V2X communication equipment receive intersection map with
The frequency of global vehicle-state;
Intersection earth coordinates are set to be from West to East X-axis using intersection geometric center as zero point, are from south orientation north
Y-axis, the right-handed Cartesian rectangular coordinate system established;
Set following four kinds of speeds: a) speed 1 is according to actual traffic situation, and a speed of selection is greater than 10km/h and is less than
The speed of 20km/h;B) speed 2 is according to actual traffic situation, and a speed of selection is greater than 20km/h and is less than road highest
The speed of speed limit;C) speed 3 is instructing from vehicle speed for the intersection passing decision output based on learning method, and the instruction is big
In equal to 0km/h and less than or equal to 20km/h;D) it stops, i.e., car speed is 0km/h;
2) data acquisition violates the traffic regulations with vehicle detects judgement, specifically includes the following steps:
2-1) clock starts a new cycle period with frequency f generation trigger signal;
When 2-2) new cycle period starts, intersection map and global vehicle are received with frequency f by V2X communication equipment from vehicle
State;
2-3) it will be used as and input by the intersection map and overall situation vehicle-state that step 2-2) is obtained, by intersection range
Whether interior each vehicle observes traffic rules and regulations as output, judges intersection range according to the detection algorithm that violates the traffic regulations
Inside whether there is vehicle to violate the traffic regulations;When all vehicles within the scope of intersection are in compliance with traffic rules, step is executed
3-1);When at least violating the traffic regulations in the presence of a vehicle within the scope of intersection, step 4-1 is executed);Wherein, described to disobey
Anti- traffic rules detection algorithm includes that the vehicle violation detection algorithm of Implicit Curves race, vehicle based on computer vision technique are disobeyed
Chapter detection algorithm method and traffic violation detection method based on track of vehicle;
3) the intersection passing decision of rule-based approach determines, specifically includes the following steps:
3-1) by judging whether intersecting from vehicle with intersection map from coordinate of the vehicle in the earth coordinates of intersection
Within the scope of crossing: when from vehicle coordinate, within the scope of intersection, output calculates knot from vehicle speed command speed 1, this cycle period
Beam, intelligent network join automobile and travel according to command speed 1, return step 2-1), it waits clock to generate trigger signal and starts new follow
The ring period;When from vehicle coordinate not within the scope of intersection, execute step 3-2);
3-2) under the earth coordinates of intersection, definition threshold time be from vehicle coordinate it is vertical with intersection stop line away from
From divided by from vehicle speed, and for from vehicle coordinate close to intersection the case where definition threshold time is less than 2/f, otherwise it is known as from vehicle
Coordinate is not close to intersection;Whether judge close to intersection from vehicle coordinate: when from vehicle coordinate not close to crossroad
When mouth, output terminates from vehicle speed command speed 2, the calculating of this cycle period, and intelligent network joins automobile and travels according to command speed 2,
Return step 2-1), it waits clock to generate trigger signal and starts new cycle period;When from vehicle coordinate close to intersection, hold
Row step 3-3);
3-3) according to step 2-2) the intersection map received and global vehicle-state, using the intersection of rule-based approach
Crossing decision carries out intersection passing decision to from vehicle;Then defeated according to the intersection result of decision of rule-based approach
The speed instruction for coming from vehicle is speed 1 or parking, and the calculating of this cycle period terminates, and intelligent network joins automobile according to 1 row of command speed
It sails or stops, return step 2-1), wait clock to generate trigger signal and start new cycle period;Wherein, described rule-based
The intersection decision of method includes the method based on acceptability risk model, the control method based on dynamic game, based on punching
The resource locking method of prominent table and it is based on vehicle pass-through rule base method;
4) the intersection passing decision based on intensified learning, specifically includes the following steps:
The intersection map and global vehicle-state 4-1) received according to step 2-2), to from vehicle coordinate whether in crossroad
Judged in mouthful range: when from vehicle coordinate within the scope of intersection, execution step 4-2);Intersecting when from vehicle coordinate
Within the scope of crossing, step 4-3 is executed);
4-2) use step 3-2) defined in from vehicle coordinate whether close to intersection, to from vehicle coordinate whether close to crossroad
Mouth is judged: when being not close to intersection from vehicle coordinate, output is calculated from vehicle speed command speed 2, this cycle period
Terminate, intelligent network joins automobile and travels according to command speed 2, return step 2-1), wait clock generation trigger signal to start new
Cycle period;When from vehicle coordinate close to intersection, output is instructed from vehicle speed and is stopped, i.e., speed is 0m/s, this circulating cycle
Phase calculating terminates, and intelligent network joins parking of automobile, return step 2-1), it waits clock to generate trigger signal and starts the new circulating cycle
Phase;
4-3) according to step 2-2) the intersection map received and global vehicle-state, using based on Timing Difference method
Intensified learning carries out intersection passing decision to from vehicle;Then the friendship obtained according to the intensified learning based on Timing Difference method
The cross road mouth result of decision, exporting from the instruction of the speed of vehicle is speed 3, and the calculating of this cycle period terminates, intelligent network join automobile according to
Speed 3 is instructed to travel, return step 2-1), it waits clock to generate trigger signal and starts new cycle period;Wherein, described to be based on
The intensified learning of Timing Difference method includes the Q-learning algorithm of the Sarsa algorithm and different strategy with strategy.
2. the current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene as described in claim 1, feature
It is, step 2-3) in, the detection algorithm that violates the traffic regulations uses the vehicle violation detection algorithm of Implicit Curves race, specifically
Realization process is as follows:
2-3-1) in intersection, geodetic coordinates is fastened, and according to intersection actual conditions, describes to intersect using Implicit Curves race F
Different zones on the map of crossing, wherein F={ fi(x, y) }, i=1,2 ..., n, i indicate i-th on the map of intersection
Region, (x, y) indicate vehicle location, Implicit Curves fi(x,y)>0、fi(x, y)=0, fi(x, y) < 0 respectively indicates vehicle i-th
A Implicit Curves fiIn (x, y), vehicle is in i-th of Implicit Curves fiOn (x, y), vehicle is in i-th of Implicit Curves fi(x, y) outside;
2-3-2) all rule Rule (j) violating the regulations set according to People's Republic of China's traffic route safety law, wherein j table
Show the jth kind vehicle violation situation of People's Republic of China's traffic route safety law defined;Implicit Curves when setting 1 frame
Initial value is 0, i.e. F={ fi(x, y)=0 }, i=1,2 ..., n;Then the intersection map received according to step 2-2) with
Global vehicle-state calculates the value of the Implicit Curves race F of this cycle period, judges last time cycle period and this cycle period
The value of Implicit Curves race F whether meet Rule (j);Violate corresponding traffic rules if meeting Rule (j), it is on the contrary then
Corresponding traffic rules are not violated.
3. the current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene as described in claim 1, feature
It is, step 3-3) in, the intersection decision of the rule-based approach, which uses, is based on vehicle pass-through rule base method,
In, the priority highest of ambulance and fire fighting truck is set, the priority of remaining type vehicle is all identical and is lower than ambulance and disappears
The priority of anti-vehicle, and setting the instruction of the speed after vehicle deceleration is the speed 1.
4. the current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene as described in claim 1, feature
It is, step 4-3) in, the intensified learning based on Timing Difference method uses the Sarsa algorithm with strategy, specific implementation
Mode is as follows:
With a meters for distance interval, 0.5m≤a≤2m, edge is parallel to intersection earth coordinates X-axis respectively and Y-axis draws straight line,
It is multiple grids by the intersection map partitioning received, using the set of all grids within the scope of intersection as state
Collect S;Using bkm/h as speed interval, 0km/h≤b≤10km/h chooses multiple speed compositions between 0km/h to 20km/h
Set is used as behavior aggregate A;Definition sets track to meet People's Republic of China's traffic route safety law and joining intelligent network
Automobile passes through one group of tracing point of intersection;From vehicle along setting the return of track as r1,10 < r1 < 100;Deviate from vehicle
The return of track is set as r2, -50 < r2 < 0;The return to collide from vehicle and other vehicles is set as r3, -100 < r3
< -50;Set decay factor γ, 0 < γ < 1;Assessment strategy and action strategy are set as greedy strategy;Then according to strategy
Sarsa algorithm carry out operation, calculated result is set as the speed 3.
5. the Distributed Intelligent Network connection automobile intersection complex scene passage as described in any one of Claims 1 to 4 is determined
Plan method, which is characterized in that step 1-2) in, the cycle period T value range of setting is between 10ms and 200ms.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910015069.4A CN109727470B (en) | 2019-01-08 | 2019-01-08 | Complex scene passing decision method for distributed intelligent network-connected automobile intersection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910015069.4A CN109727470B (en) | 2019-01-08 | 2019-01-08 | Complex scene passing decision method for distributed intelligent network-connected automobile intersection |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109727470A true CN109727470A (en) | 2019-05-07 |
CN109727470B CN109727470B (en) | 2020-09-11 |
Family
ID=66298875
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910015069.4A Active CN109727470B (en) | 2019-01-08 | 2019-01-08 | Complex scene passing decision method for distributed intelligent network-connected automobile intersection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109727470B (en) |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110111605A (en) * | 2019-06-12 | 2019-08-09 | 吉林大学 | Automatic driving vehicle entrance ring road based on dynamic game travels decision-making technique |
CN110263709A (en) * | 2019-06-19 | 2019-09-20 | 百度在线网络技术(北京)有限公司 | Driving Decision-making method for digging and device |
CN110288847A (en) * | 2019-06-28 | 2019-09-27 | 浙江吉利控股集团有限公司 | A kind of automatic Pilot decision-making technique, device, system, storage medium and terminal |
CN110444015A (en) * | 2019-07-08 | 2019-11-12 | 清华大学 | Intelligent network based on no signal crossroad subregion joins car speed decision-making technique |
CN110473419A (en) * | 2019-09-09 | 2019-11-19 | 重庆长安汽车股份有限公司 | A kind of passing method of automatic driving vehicle in no signal lamp intersection |
CN111833614A (en) * | 2020-09-18 | 2020-10-27 | 得威科技(浙江)有限公司 | Traffic violation automatic identification method, system, electronic equipment and storage medium |
CN112349094A (en) * | 2020-09-27 | 2021-02-09 | 北京博研智通科技有限公司 | Method and system for evaluating motor vehicle passing efficiency of signal lamp-free control intersection |
CN112373472A (en) * | 2021-01-14 | 2021-02-19 | 长沙理工大学 | Method for controlling vehicle entering time and running track at automatic driving intersection |
CN113205693A (en) * | 2020-01-31 | 2021-08-03 | 奥迪股份公司 | Method for operating a traffic signal system |
CN113276884A (en) * | 2021-04-28 | 2021-08-20 | 吉林大学 | Intelligent vehicle interactive decision passing method and system with variable game mode |
CN113313957A (en) * | 2021-05-30 | 2021-08-27 | 南京林业大学 | Signal lamp-free intersection vehicle scheduling method based on enhanced Dijkstra algorithm |
CN114268369A (en) * | 2021-12-20 | 2022-04-01 | 东南大学 | Inter-vehicle network communication system based on visible light |
CN114919581A (en) * | 2022-05-11 | 2022-08-19 | 中南大学 | Behavior decision method and computer device for intelligent vehicle unordered intersection |
CN116168550A (en) * | 2022-12-30 | 2023-05-26 | 福州大学 | Traffic coordination method for intelligent network-connected vehicles at signalless intersections |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE3405269C2 (en) * | 1984-02-15 | 1986-09-25 | Battelle-Institut E.V., 6000 Frankfurt | Method and device for the automatic guidance of vehicles along guidelines |
US20130018572A1 (en) * | 2011-07-11 | 2013-01-17 | Electronics And Telecommunications Research Institute | Apparatus and method for controlling vehicle at autonomous intersection |
CN103177596A (en) * | 2013-02-25 | 2013-06-26 | 中国科学院自动化研究所 | Automatic intersection management and control system |
CN103996312A (en) * | 2014-05-23 | 2014-08-20 | 北京理工大学 | Pilotless automobile control system with social behavior interaction function |
CN104590274A (en) * | 2014-11-26 | 2015-05-06 | 浙江吉利汽车研究院有限公司 | Driving behavior self-adaptation system and method |
CN104754011A (en) * | 2013-12-31 | 2015-07-01 | 中国移动通信集团公司 | Internet of Vehicles multi-party coordination control method and Internet of Vehicles coordination platform |
CN104882008A (en) * | 2015-06-03 | 2015-09-02 | 东南大学 | Method for vehicle cooperative control at non-signaled intersection in vehicle networking environment |
CN105321362A (en) * | 2015-10-30 | 2016-02-10 | 湖南大学 | Intersection vehicle intelligent cooperative passage method |
JP5973447B2 (en) * | 2010-10-05 | 2016-08-23 | グーグル インコーポレイテッド | Zone driving |
US9459623B1 (en) * | 2015-04-29 | 2016-10-04 | Volkswagen Ag | Stop sign intersection decision system |
CN108009587A (en) * | 2017-12-01 | 2018-05-08 | 驭势科技(北京)有限公司 | A kind of method and apparatus based on intensified learning and the definite driving strategy of rule |
RU2016143345A (en) * | 2015-11-05 | 2018-05-10 | ФОРД ГЛОУБАЛ ТЕКНОЛОДЖИЗ, ЭлЭлСи | OFFLINE CROSSROADS DRIVING BASED ON PERCEPTION DATA |
CN108225364A (en) * | 2018-01-04 | 2018-06-29 | 吉林大学 | A kind of pilotless automobile driving task decision system and method |
CN108877269A (en) * | 2018-08-20 | 2018-11-23 | 清华大学 | A kind of detection of intersection vehicle-state and V2X broadcasting method |
-
2019
- 2019-01-08 CN CN201910015069.4A patent/CN109727470B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE3405269C2 (en) * | 1984-02-15 | 1986-09-25 | Battelle-Institut E.V., 6000 Frankfurt | Method and device for the automatic guidance of vehicles along guidelines |
JP5973447B2 (en) * | 2010-10-05 | 2016-08-23 | グーグル インコーポレイテッド | Zone driving |
US20130018572A1 (en) * | 2011-07-11 | 2013-01-17 | Electronics And Telecommunications Research Institute | Apparatus and method for controlling vehicle at autonomous intersection |
CN103177596A (en) * | 2013-02-25 | 2013-06-26 | 中国科学院自动化研究所 | Automatic intersection management and control system |
CN104754011A (en) * | 2013-12-31 | 2015-07-01 | 中国移动通信集团公司 | Internet of Vehicles multi-party coordination control method and Internet of Vehicles coordination platform |
CN103996312A (en) * | 2014-05-23 | 2014-08-20 | 北京理工大学 | Pilotless automobile control system with social behavior interaction function |
CN104590274A (en) * | 2014-11-26 | 2015-05-06 | 浙江吉利汽车研究院有限公司 | Driving behavior self-adaptation system and method |
US9459623B1 (en) * | 2015-04-29 | 2016-10-04 | Volkswagen Ag | Stop sign intersection decision system |
CN104882008A (en) * | 2015-06-03 | 2015-09-02 | 东南大学 | Method for vehicle cooperative control at non-signaled intersection in vehicle networking environment |
CN105321362A (en) * | 2015-10-30 | 2016-02-10 | 湖南大学 | Intersection vehicle intelligent cooperative passage method |
RU2016143345A (en) * | 2015-11-05 | 2018-05-10 | ФОРД ГЛОУБАЛ ТЕКНОЛОДЖИЗ, ЭлЭлСи | OFFLINE CROSSROADS DRIVING BASED ON PERCEPTION DATA |
CN108009587A (en) * | 2017-12-01 | 2018-05-08 | 驭势科技(北京)有限公司 | A kind of method and apparatus based on intensified learning and the definite driving strategy of rule |
CN108225364A (en) * | 2018-01-04 | 2018-06-29 | 吉林大学 | A kind of pilotless automobile driving task decision system and method |
CN108877269A (en) * | 2018-08-20 | 2018-11-23 | 清华大学 | A kind of detection of intersection vehicle-state and V2X broadcasting method |
Non-Patent Citations (5)
Title |
---|
CHAIRIT WUTHISHUWONG ET AL.: "Vehicle to Infrastructure based Safe Trajectory Planning for Autonomous Intersection Management", 《IEEE》 * |
E. ONIEVA ET AL.: "Amulti-objective evolutionary algorithm for the tuning of fuzzy rule bases for uncoordinated intersections in autonomous driving", 《EL SEVIER》 * |
GAOCHAO WANG ET AL.: "TLB-VTL: 3-Level Buffer based Virtual Traffic Light Scheme for Intelligent Collaborative Intersections", 《IEEE》 * |
ZHANG TAO ET AL: "Multi-branch Model of Intersection for Vehicle Navigation", 《IEEE》 * |
卓福庆: "面向自驾联网车的交叉口协同控制方法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110111605A (en) * | 2019-06-12 | 2019-08-09 | 吉林大学 | Automatic driving vehicle entrance ring road based on dynamic game travels decision-making technique |
CN110111605B (en) * | 2019-06-12 | 2021-08-31 | 吉林大学 | Automatic driving vehicle entrance and exit ramp driving decision method based on dynamic game |
CN110263709A (en) * | 2019-06-19 | 2019-09-20 | 百度在线网络技术(北京)有限公司 | Driving Decision-making method for digging and device |
CN110263709B (en) * | 2019-06-19 | 2021-07-16 | 百度在线网络技术(北京)有限公司 | Driving decision mining method and device |
CN110288847A (en) * | 2019-06-28 | 2019-09-27 | 浙江吉利控股集团有限公司 | A kind of automatic Pilot decision-making technique, device, system, storage medium and terminal |
CN110288847B (en) * | 2019-06-28 | 2021-01-19 | 浙江吉利控股集团有限公司 | Automatic driving decision method, device and system, storage medium and terminal |
CN110444015A (en) * | 2019-07-08 | 2019-11-12 | 清华大学 | Intelligent network based on no signal crossroad subregion joins car speed decision-making technique |
CN110444015B (en) * | 2019-07-08 | 2020-10-09 | 清华大学 | Intelligent network-connected automobile speed decision method based on no-signal intersection partition |
CN110473419A (en) * | 2019-09-09 | 2019-11-19 | 重庆长安汽车股份有限公司 | A kind of passing method of automatic driving vehicle in no signal lamp intersection |
CN113205693A (en) * | 2020-01-31 | 2021-08-03 | 奥迪股份公司 | Method for operating a traffic signal system |
CN111833614A (en) * | 2020-09-18 | 2020-10-27 | 得威科技(浙江)有限公司 | Traffic violation automatic identification method, system, electronic equipment and storage medium |
CN112349094A (en) * | 2020-09-27 | 2021-02-09 | 北京博研智通科技有限公司 | Method and system for evaluating motor vehicle passing efficiency of signal lamp-free control intersection |
CN112373472A (en) * | 2021-01-14 | 2021-02-19 | 长沙理工大学 | Method for controlling vehicle entering time and running track at automatic driving intersection |
CN113276884A (en) * | 2021-04-28 | 2021-08-20 | 吉林大学 | Intelligent vehicle interactive decision passing method and system with variable game mode |
CN113313957A (en) * | 2021-05-30 | 2021-08-27 | 南京林业大学 | Signal lamp-free intersection vehicle scheduling method based on enhanced Dijkstra algorithm |
CN114268369A (en) * | 2021-12-20 | 2022-04-01 | 东南大学 | Inter-vehicle network communication system based on visible light |
CN114919581A (en) * | 2022-05-11 | 2022-08-19 | 中南大学 | Behavior decision method and computer device for intelligent vehicle unordered intersection |
CN114919581B (en) * | 2022-05-11 | 2024-04-26 | 中南大学 | Behavior decision method and computer device for unordered intersection of intelligent vehicle |
CN116168550A (en) * | 2022-12-30 | 2023-05-26 | 福州大学 | Traffic coordination method for intelligent network-connected vehicles at signalless intersections |
Also Published As
Publication number | Publication date |
---|---|
CN109727470B (en) | 2020-09-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109727470A (en) | A kind of current decision-making technique of Distributed Intelligent Network connection automobile intersection complex scene | |
CN109767630B (en) | A kind of traffic signal control system based on bus or train route collaboration | |
CN108986488A (en) | Ring road imports collaboration track and determines method and apparatus under a kind of truck traffic environment | |
CN101308604B (en) | Traffic coordinating and controlling method with strategy of big range | |
CN104637317B (en) | A kind of crossing based on real-time vehicle track actuated signal control method | |
CN107798861A (en) | A kind of vehicle cooperative formula formation running method and system | |
CN104504918A (en) | Urban highway bus signal priority method | |
CN106997690A (en) | The optional lane-change control method of vehicle on highway under a kind of car networking environment | |
CN104077918A (en) | Urban traffic junction signal lamp self-adaption control method based on vehicle-mounted data | |
CN107274684A (en) | A kind of single-point integrative design intersection policy selection method under bus or train route cooperative surroundings | |
CN107248297A (en) | A kind of rasterizing signal phase duration calculation method in intersection under bus or train route cooperative surroundings | |
CN106875710A (en) | A kind of intersection self-organization control method towards net connection automatic driving vehicle | |
CN104794919A (en) | Real-time optimal configuration method for lane resources of multi-lane expressway | |
CN103886764A (en) | Bus transit lane time division sharing method | |
CN109410561A (en) | A kind of even heterogeneous formation travel control method of highway vehicle | |
CN112365714B (en) | Traffic signal control method for intersection of intelligent rail passing main branch road | |
CN110047306A (en) | A kind of intelligent traffic network | |
CN104778832B (en) | A kind of multimode Urban Road Cross-section resource coordination collocation method | |
CN110444015A (en) | Intelligent network based on no signal crossroad subregion joins car speed decision-making technique | |
CN104933874A (en) | Traffic light self-adaptive control method and device | |
CN105869422A (en) | Traffic control system | |
Pan et al. | Research on the behavior decision of connected and autonomous vehicle at the unsignalized intersection | |
CN114495547B (en) | Signal intersection cooperative passing method for automatically-driven automobile | |
CN109765801A (en) | The implementation method of car networking desin speed adjustment based on VISSIM emulation | |
Pi et al. | Automotive platoon energy-saving: A review |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20191231 Address after: 100083 401a, floor 4, building 6, yard 1, Zhongguancun East Road, Haidian District, Beijing Applicant after: Beijing Chaoxing Future Technology Co., Ltd Address before: 100084 Haidian District Tsinghua Yuan Beijing No. 1 Applicant before: Tsinghua University |
|
GR01 | Patent grant | ||
GR01 | Patent grant |