WO2019061933A1 - 交通信号泛弦控制方法及其系统 - Google Patents
交通信号泛弦控制方法及其系统 Download PDFInfo
- Publication number
- WO2019061933A1 WO2019061933A1 PCT/CN2018/000332 CN2018000332W WO2019061933A1 WO 2019061933 A1 WO2019061933 A1 WO 2019061933A1 CN 2018000332 W CN2018000332 W CN 2018000332W WO 2019061933 A1 WO2019061933 A1 WO 2019061933A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- intersection
- time
- phase
- traffic
- vehicle
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0133—Traffic data processing for classifying traffic situation
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
- G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/081—Plural intersections under common control
- G08G1/082—Controlling the time between beginning of the same phase of a cycle at adjacent intersections
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/081—Plural intersections under common control
- G08G1/083—Controlling the allocation of time between phases of a cycle
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/095—Traffic lights
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0418—Architecture, e.g. interconnection topology using chaos or fractal principles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
Definitions
- the invention relates to the field of traffic signal mode control. Specifically, it is a control method and system that can adjust the signal time according to traffic conditions.
- trunk road technology that was created to provide smooth traffic will cause the road surface to widen year by year due to the gathering of surrounding traffic flow. It is unbearable to be burdened by the burden and the waste of surrounding roads.
- the origin of the trunk road technology that solves the small urban economic traffic demand has the advantages of the original “trunk priority” with the common surface road traffic demand of modern urban economies of scale.
- the fading becomes a defect inherent in the innate adaptation of the technique that causes the vicious cycle of "widening the main road to lead to heavier congestion.”
- the recently invented time differential ratio technology solves the waste problem of small traffic broad-spectrum load green light and space in the whole network.
- the discovery of the principle minimizes waste of passage traffic.
- the string mode needs to solve the traffic flow response problem distributed in the area to further improve the actual control effect of the string technology. Traffic flow prediction is the key to signal dynamic control.
- Current researches are mostly neural networks, chaotic timing, wavelets and other methods.
- the object of the present invention is to solve the problem of low-cost optimization of signal response to traffic flow distribution.
- the present invention proposes a solution to achieve the above object, including a newly created road network traffic prediction mathematical model with a signal spacing-team length redundancy formula trq as the core, and an analytical-intelligent predictive control based on the model including an intelligent learning module package.
- a road network traffic signal pan-chord control method referred to as "A-A” method, whose features include step note 1:
- S1 acquires signal parameters and road network parameters
- S2 detects the number of detained vehicles in each direction of each intersection, or the number of vehicles entering or leaving the vehicle, that is, the number of vehicles entering or exiting, that is, the number of exits, or the number of outbound vehicles, ie, the outflow amount x, or / and the head position q0 Information and phase variable sub-time ⁇ t Th0 is the differential time;
- Each intersection element of the S3 prediction layer predicts the direction of the next period or/and the phase fleet Q and its change ⁇ Q, the number of vehicles exiting x and the remaining green signal time, ie, the remaining time
- the S4 analysis layer initial judgment meta-analysis determines the fluctuation of the time difference of the signal between the intersections in the next period, or / and the initial difference of the origin of the time difference caused by the change of the two cross-flows of the regional traffic flow, that is, the initial judgment of the shift, etc., or/and the combination Relevant intersections Together with the queuing Q and its change ⁇ Q, the initial determination of the solitary wave of the signal is temporarily determined, or / and the section of the trunk green wave channel in which the super-threshold is found is subjected to fluctuation, solitary wave initial judgment, or / and discovery Dimensional flow changes exceed the threshold, or / and find no fleet can differentiate the intersection, or / and the discovery area green letter ratio changes exceed the threshold, and so on;
- the S5 decision-making layer co-ordinates the initial judgments, the order, the time, and the issuance of instructions, and no car team crossing directly enters S7;
- S6 signal time adjustment (1) Super-threshold intersection: According to the instruction decoding configuration, the chord time difference: 1) Fluctuation time difference tgw: According to the macro fluctuation/metamorphism command, the fluctuation time difference tgw, configure the transition period, and send it To the intersection of the rise and fall and all the downstream intersections of the green wave, 2) Solitary wave optimization: the solitary wave source and its flow path intersection timing scheme are made to modify the phase time and time limit of each intersection to the corresponding intersection system, and seamlessly switch It enters the solitary wave state; 3) configures the two-dimensional origin movement transition period, and (2) the non-threshold intersection: the intersection of the fleet but the threshold directly performs the next step S5;
- S7 execution (1) transition period control: before the new cycle, first run the transition period or/and temporary schedule of each sub-area mode; or / and (2) differential control [patent application 201710224791.X" pan-green wave "-Right 1-S5]: Install the micro-inductance intersection or decide whether to enable the differential control according to the command: analyze the intersection phase to obtain the head position q0 position, decide to transfer the differential (ie quantum phase change) green wave control: when q0 is at a safe distance When the vehicle-free direction phase signal is green, a differential time (ie, the phase variable is used) ⁇ t Th0 is transferred to the vehicle q0 phase occupancy and the differential is prohibited; the differential operation judges: “Nes” is returned to S3;
- the vehicle queuing Q measurement unit is the meter or standard number of cars, the standard car number captain includes the car spacing, the conversion rate with the meter length, the non-standard car and the standard car conversion rate, by which the fleet can be converted into the standard number of cars. Or m;
- the number of vehicles refers to the equivalent standard number of vehicles after conversion
- the lower period includes multiple signal periods C, such as 1C, 2C, 4C, 8C, which are common to any signal network fleet prediction.
- a panning method according to the present invention characterized in that said S2 further comprises:
- the tail information includes the last vehicle position of the traffic flow and the flow direction intersection distance represents the traffic flow team length q
- the head information includes the traffic flow front position and the flow direction intersection position.
- Distance q0 the tail information can be obtained by real-time meter-level precision big data, such as running vehicle positioning device, on-board mobile phone positioning plug-in, common traffic sensing device, such as video, microwave radar, etc. can be measured in real time.
- the device of the last car in the car flow, the team head information can be obtained by any device that can measure the first car in real time, such as traffic high-traffic traffic video analysis device, microwave, mobile positioning big data, etc.;
- a panning method according to the present invention characterized in that said S2 further comprises:
- phase variable sub-time ⁇ t represents the minimum safe green light response time used by the time differential ratio method, which is recommended for roads below the city's 60 km speed limit. Available less than or equal to 6 seconds, the corresponding traffic flow head q0 response distance ranges from 40 meters to 60 meters, or / and directly calculated by the controlled road segment flow rate limit speed;
- the S3 further includes:
- 'd2,j2' indicates the direction of the direction of the left-turning d-flow direction
- 'd3,j3' indicates the direction of the direction of the right-turning d-flow direction.
- 'd1, j1' or 'd, j' represents the direction of the straight direction of the flow direction d
- j 2 left turn
- j 3 right turn
- (c) represents the next time period
- ( C-1) represents the previous period, and the following is also applied to each variable;
- the traffic flow S d (c) of the vehicle source is a predicted value, and the vehicle source flow intelligent function is used. According to the measured value, S d (c-1) is obtained, and the intelligent function is obtained. It is obtained by statistical learning or other intelligent methods and past data training or online learning;
- the direction phase traffic volume shared by the lane is still determined by the phase distribution ⁇ d (c);
- the road vehicle source includes multiple vehicle sources, and the time difference between them flows to the vehicle source to the intersection to determine the time difference, and the common average or 0 is estimated;
- the intelligent method includes comprehensive use of neural network ann, chaotic time series, wavelet theory, statistical regression and support vector machine svm, genetic optimization ga, particle swarm optimization pso, fuzzy analysis and information granulation, and the like, intelligent learning and timing analysis methods, It is said that the intelligent method is the same;
- the outflow amount x ⁇ 0,d,j ,x ⁇ 1,d1,j1 ,x ⁇ 1,d2,j2 ,x ⁇ 1,d3,j3, etc. can be predicted or/and retrofitted by the following method 4
- the direction phase exit detector is actually measured;
- the S3-1 further includes:
- the phase flow distribution ⁇ d (c) of the S3-1-1 intersection is a predicted value, and is an intelligent function of the phase flow distribution using the intersection direction. Predicted according to the measured value ⁇ d (c-1);
- the step of measuring the value ⁇ d (c-1) comprises: (1) subtracting the queuing change ⁇ Q d,j (c-1) of the previous period by using the measured phase queuing amount measured in the first two periods, (2) By multiplying the phase green time ⁇ d,j by the phase outflow rate v d,j, the outflow time x d,j (c-1) is obtained, and the outflow predicted by the pre-period is replaced by the “measured” outflow of the cycle.
- phase outflow rate v d,j refers to the number of vehicles per second from the traffic light control line
- the S3-1 further includes:
- the method of obtaining the predicted value is to measure the outflow amount x ⁇ k,d,j (c) of each intersection from the predicted intersection to the upstream to the k-segment.
- the outflow Etc. can be predicted by the following method 5;
- the S3-1 further includes:
- the fleet amount is calculated by dividing the fleet amount by the phase outflow rate v d,j by the time interval tq0 d,j (c);
- the time interval trq ⁇ (k-1 (c) of the upstream k intersection fleet and the preceding traffic flow is divided by the distance D ⁇ (k-1) of the vehicle from the intersection k to the predicted intersection section divided by the vehicle prescribed speed v d,l minus Go to the front intersection of the vehicle quantity q ⁇ (k-1) (c) and the product of the team disturbance factor ⁇ .
- the team dynamic coefficient ⁇ is the time from the start of the first team to the start of the tail vehicle in unit unit length, in units of seconds/meter, and its valuation range is 0.14 to 0.22, which is 0.18, which can be adjusted according to experience;
- the signal time difference ⁇ c ⁇ i, dc initial value is divided by the segment length i ⁇ D by the specified vehicle speed v d, l , that is, tv0 ⁇ i ;
- the S3-1 further includes:
- the queuing Q m,n,d,j (c) of the S3-1-4 intersection direction phase prediction and its variation ⁇ Q m,n,d,j (c) will generate the intersection element output signal when the following control threshold is exceeded and The relevant traffic information is sent for further analysis.
- the control thresholds include a minimum team change threshold ⁇ Q Th0 , an abnormal threshold Q ThC , a minimum solitary relative team threshold ⁇ Q ThS , and a minimum solitary absolute team threshold Q ThS ;
- the minimum team change threshold ⁇ Q Th0 refers to a minimum change in the length of the fleet length over a period of time
- the metamorphic threshold Q ThC refers to the queue length of the traffic flow reaching the signal green wave flow direction commutation value or balance value
- the minimum solitary wave relative team change threshold ⁇ Q ThS refers to a team length relative to other phase fleet minimum team length difference value
- the minimum solitary wave absolute team threshold value Q ThS refers to the minimum solitary wave identification value of the traffic queue length
- the S3-1 further includes:
- each port element to obtain data time the signal is not a green wave system, each intersection synchronizes to obtain data before the start of the cycle, and the green wave is asynchronous before the start of the respective cycle of the intersection;
- the S3-1 further includes:
- the range K d of the intersection element obtained from S3-1-6 from other intersection elements depends on the intersection time of the intersection signal green light time.
- the analysis layer fluctuation initial judgment element of S4-1 is based on receiving the intersection direction queuing Q from the corresponding row-column intersection element of the prediction layer and the change ⁇ Q exceeds the control thresholds ⁇ Q Th0 and Q ThC to determine whether the intersection is in the same flow direction, If the number of links in the downstream intersection of the road segment of the same road segment or column exceeds the threshold ⁇ Q Th0 exceeds the control line threshold M Th0 or the column threshold N Th0 , if it is exceeded, the initial fluctuation is made, and the green light time is too short or the road spacing is Excessively large areas are not judged by the row or column thresholds M Th0 or N Th0 , and are independently judged by fluctuations;
- the S4 further includes:
- the analysis layer flow direction initial judgment element according to the direction of each intersection received from the prediction layer intersection element Q and its change ⁇ Q exceeds the control threshold ⁇ Q Th0 , Q ThC , and the total traffic volume of each intersection flow in the calculation area or / And queuing amount And its total changes Greater than the control value According to the two maximum flow directions, determine the two-dimensional origin time difference table of the reset area, and make a preliminary judgment;
- the S4 further includes:
- the analysis layer solitary wave initial judgment element according to the intersection direction direction Q received from the prediction layer intersection element and its change ⁇ Q exceeds the control thresholds ⁇ Q ThS and Q ThS , and the prediction calculation determines the flow direction to the intersection and the intersection thereof When there is no time left at each intersection Can be used, can generate solitary waves, yes, then the solitary wave initial judgment;
- the S4-3 further includes:
- the S5 further includes:
- the decision-making layer coordination decision rules mentioned in S5-1 include: (1) There is no conflict rule between the initial waves of solitary waves: the path between the solitary waves is parallel or the path between the solitary waves does not meet the intersection point, (2) the solitary wave-incremental initial judgment Collision-free rules: the solitary wave flows to the upstream of the path fluctuation green wave, (3) the large solitary wave priority rule: the solitary wave priority is given when the solitary wave conflicts, (4) the solitary wave priority occurs when the fluctuation conflicts, (5) the solitary wave Stage management, solitary waves pass through at most Intersections, the development of intelligent instructions and solitary path prohibition of re-isphaned path time limit instructions, intelligent instructions
- the S3-1 further includes:
- phase arrival amount Q ⁇ 0,d,j (c) in each direction of the intersection of S3-1-7 using the obtained x ⁇ 1,d,1 (c),x ⁇ 1,d2,2 (c), x ⁇ 1, d3, 3 (c), S d (c), ⁇ d (c), the steps are as follows;
- the S3-1 feature further includes:
- Intelligent function described in S3-1-1-1 Use q m,n,d,j (c) or its changes ⁇ qm,n,d,j (c), s m,n,d,j (c) and train with intelligent learning method to find a phase timing of the intersection Distribution function
- the steps are as follows:
- the S4 further includes:
- Coordination review and coordination (decision-making level): review, coordinate conflicts with other intelligent instructions, and send intelligent instructions
- Decoding execution fluctuations (execution layer): According to the macro fluctuation/metamorphism command and the fluctuation time difference tgw, configure the transition period and send it to the intersection of the fluctuation and all the downstream intersections of the green wave;
- the S3-1 further includes:
- the pan-chord mode control method according to the present invention is characterized in that the S6 further comprises:
- the signal time adjustment mentioned in S6-1 includes the fluctuation time difference tgw of the fluctuation team, the time-tbl of the solitary wave timing scheme, the time difference t-dmd of the two-dimensional origin, and the like, 1) the fluctuation tgw counts the tgw into the intersection and the downstream thereof.
- the time difference ⁇ c of each intersection the fluctuation time difference tgw is configured to be sent to the fluctuation intersection and all the downstream intersections of the green wave during the transition period, and 2) the solitary wave timing scheme time-tbl makes the solitary wave source and its flow path intersection timing scheme
- the command to modify the phase time limit of each intersection direction is sent to the corresponding intersection system to seamlessly switch to the solitary wave state, and 3) the two-dimensional origin movement time difference o-tmd will be made into the intersection of the intersections tmp-p to the corresponding intersection;
- the pan-chord mode control method according to the present invention is characterized in that the S7 further comprises:
- a road network traffic signal panning control system characterized by a predictive control package that runs the "AA” method, that is, an "AA” package, a traffic positioning data center or/and a fleet, a stuck number of cars detector, and a traffic signal control intersection machine. , or / and the roadside source into and out of the detector, or another intersection to exit the vehicle detector, or / and the road segment into the vehicle detector,
- the "AA” package predicts the amount of traffic in the next period and determines the signal time scheme based on the amount of traffic received by the vehicle, or/and the roadside vehicle access detector from the traffic location data center or/and the fleet detector.
- “AA” packages can be central or distributed or run in parallel, with software deployment or / and hardware deployment;
- the traffic location data center detects the direction of the location intersection or/and the vehicle position at the end of the phase fleet to determine the length of the fleet, and the positioning data is from any positionable mobile terminal including vehicle positioning navigation, vehicle bound mobile phone positioning navigation, and the like;
- the fleet length detector refers to any device that can detect the length of the phase fleet in the direction of the intersection, such as a video analysis device, ultrasound, microwave, infrared, coil set, and the like;
- the roadside vehicle source access detector detects the number of inbound and outbound vehicles flowing in the intersection, including roadside time charging devices, residential areas, parking lots, alleys, non-traffic light control intersections, highway access detectors, and multiple roadsides.
- the source of the vehicle can be integrated into one vehicle source estimate according to the average distance from the relevant flow direction to the intersection;
- the exiting vehicle detector detects the number of vehicles leaving at the exits of intersections, road sections, communities, highways, etc.;
- the driving vehicle detector detects the number of vehicles entering at the entrance of a road section, a cell, a highway, and the like;
- the vehicle number detector includes, for example, available coils, piezoelectrics, magnetic induction, infrared, video, or/and any other measurable acquisition by the vehicle through the counting device;
- the "AA" packet feature includes at least a prediction layer including a prediction module called a junction element for predicting the amount of traffic at the intersection of the time interval based on the measured traffic volume, and includes threshold information transmitted from the intersection element.
- the analysis layer of the analysis module called the initial judgment element, which contains the decision-making layer that the overall analysis layer firstly transmits the various threshold information functions;
- pan-string system "A-A" package according to the present invention: its intersection element module features include:
- intersection element prediction module and the actual intersection are corresponding to each other, and the intersection elements are dynamically adjusted according to the needs of the respective measured/predicted traffic information;
- intersection element module features include:
- intersection element inputs the direction or/and phase pre-period fleet, or/and its flow direction, the number of exit vehicles, the output is predicted the next time period direction and/or the phase remaining green time, or/and the exit vehicle Information such as the number, or/and the change in the fleet, or/and the length of the fleet, etc., is transmitted to the analysis layer related initial element module;
- intersection element module features include:
- the intersection element includes a neural network and an intelligent method module such as statistical learning and timing analysis;
- the initial element module features include:
- the initial element module input is a threshold value of the intersection element and its related information
- the output is a signal time difference or/and a preliminary judgment of the signal temporary time table, and is transmitted to the decision layer for coordinating;
- the decision-making layer features include:
- the decision-making layer coordinating module inputs the initial judgment transmitted from the analysis layer, and the output is a decision on the timing, timing, time, etc. of the initial judgment conflict condition, and is sent to the execution layer signal time instruction;
- the advantages of the invention are as follows: 1) The mathematical model of traffic prediction of Xinlu.com is closer to the actual situation.
- the "AA” method and its system provide reliable theoretical support for predicting and solving the problem of queue congestion at urban intersections. 2) The method can only come from Large data such as queues in the cloud are easy to implement, and 3) medium-large load-guided green wave, near-saturated-saturated load-resolved green wave, and small-load broad-spectrum differential green wave (adaptive “0”) Red light technology) provides a series of continuous solution tools for signal control in order to resolve the congestion core, initial congestion, and delay the arrival of the captain's large-scale congestion.
- the quantization step is always ⁇ Q d, j Th0
- the signal network time is taken to reduce the loss caused by the queuing growth to the lowest state, from the medium-small load-guided state to the saturated load's blocked state, which avoids the redundant stop-start and the per-segment per-segment per cycle.
- the 7-step structure of the pan-chord control method naturally includes the following transformations: 1) When the signal is 0 time difference, the pan-chord method naturally becomes the “differential green wave” method, and 2) when the instruction “does not enable differential green” When the wave S5" or road network system does not have the corresponding sensing and data acquisition device installed and the "differential control" in step "S7" cannot be used, the pan-chord method naturally has no "differential green wave” function and is often in a non-differential state. .
- Figure 1 is a flow chart of the pan-chord control method
- Figure 2 chord mode quadrilateral road network structure, flow direction and traffic time distribution
- Figure 3 is a three-layer structure diagram of the pan-string-predictive control
- Figure 4 shows the relationship between the number of detained vehicles and the flow rate of the road network and the principle diagram of the "A-A" method
- Figure 5 Signal, queuing, prediction and fluctuation-deformation, solitary-phase-optimal two-state diagram at 630 seconds of each road network;
- Figure 2 The left-handed worm-type string mode runs four sub-zones of four sides, with four two-dimensional guided green wave origins Q1(0,5), Q2(4,9), Q3(9,5) and Q4( 5,0); 1 - network intersection node code identification starting point (0,0) is the lower left corner of the road network, 2-way network symbol ⁇ (0,0), (9,9) ⁇ represents the origin is (0 , 0), vertical and horizontal maximum coordinate increments (9, 9) are 9, 3-way, 4-signal, 5-drive fleet, 6-road signal control, 7-Internet cloud, 8-center control system, The 9-sub-region mark 4 ⁇ (5,0),(4,4) ⁇ represents the 4th sub-region, the sub-region coordinate starting point is (5,0), and the vertical and horizontal maximum coordinate increments (4,4) are each The 4,10-solid hollow arrows represent the main direction and its channel group green wave pointing east-right, the dotted arrow represents the secondary flow direction and its channel group green wave, 11-two-dimensional origin mark Q and small oc
- Figure 3 1-predictive layer, 2-channel element basic data receiving and prediction module, 3-predictive layer and analysis layer data connection, 4-analysis layer-sub-region initial judgment, one sub-area module, 5-sub-area 1 intersection column rise and fall metamorphosis initial judgment element, 6-sub-zone 0th intersection line fluctuation metamorphosis initial judgment element, 7-sub-area 0th intersection solitary wave phase optimization initial judgment element, 8-analysis layer and decision layer data connection, 9-analysis layer-sub-area macro analysis module, 10-sub-area micro-ratio control, 11-sub-area strong flow direction analysis, 12-sub-area strong flow direction combination, 13-analysis layer and decision-making layer data connection, 14-decision layer- Sub-area 4,15-single wave collision determination, 16-fluctuation conflict determination, 17-coordination decision, 18-single wave management, 19-decision and intelligent command connection, 20-origin movement decision,
- the arrow represents the existing fleet Q d (c-1), which is obtained by actual measurement.
- the dotted line represents the predicted fleet ⁇ Q d (c). It is predicted that the part entering the intersection represents the outbound fleet X d (c).
- Figure 5 1-4 ⁇ (5,0),(4,4) ⁇ is the road network mark: it indicates that the sub-area 4 coded marker starting point (5,0) is the lower left corner of the road network, and the time difference of the green wave is calculated.
- Two-dimensional origin Q4 the main green wave flows to the east, the secondary flow to the north, the 2-way spacing - the traffic time is recorded as #-#/#: meters-seconds/second, if the value indicates the road-line (0, 1) )
- the distance D 125 meters
- the speed is 45 kilometers
- “*” represents the road sections or intersections of the channel.
- the “spacing-traffic time is recorded as #-#/#” for each road segment, 3-the group is marked with six “East”, “West”, “South” and “North” directions around the intersection.
- the number indicates that the straight, left and right phases of the azimuth are measured by the length of the intersection and the predicted fleet length separated by double angle brackets "", such as "East 1/0/0"0/1/0" indicates the intersection On the east side, etc., the actual test team of the west line goes straight to phase 1, turn left 0, turn right 0, predict long "0/1/0, the direction is opposite to the flow direction, the unit is the number of cars, the number of cars is q, the number of seconds is the standard car.
- the length of 6.25 meters is converted into the captain of the meter and reused.
- the length of the secondary flow to the green wave covers the intersection of the green wave period, such as the secondary green wave running from the intersection (7, 0) to the intersection (7, 1) for 8 seconds, the main green wave from the intersection (7, 2) ) Run to intersection (9, 2) for 18 seconds and remaining for 2 seconds, 5-circle represents the source of the vehicle, the representative of the road head connects to the vehicle source of other districts or highways, and the representative parking lot and the community in the road section are in the warehouse. Vehicle source, there is another roadside parking space ⁇ mark on each road section (the net input and exit of this average period is 0, slightly), and the figure is the number of vehicles that predict the outflow related flow;
- a traffic signal control system software is used to control the road network shown in Fig. 2 [already applied for 201710137495.6 embodiment], the pan-string system includes six parts marked with quotation marks, The "roadway machine system” is marked with the “fleet length video traffic detector” or / and the “roadway exit vehicle number detector” as shown in Fig. 2, and the “vehicle source s is shown in Fig. 2, which is marked with 13 and is driven out of the vehicle.
- the number detector "connects the communication network through 232/485/wifi as shown in Figure 2, marking the data to the "traffic data center", through the communication network as shown in Figure 2, by the central control system, as shown in Figure 2, labeled 8 or from the cloud traffic data.
- the center obtains the mobile positioning big data custom meter-level, sub-meter-level cross-section phase fleet length q data, the number of vehicles entering and exiting the vehicle source s, and the prediction of the center's "predictive control software" is shown in Figure 3. 1.
- the tail information q is taken from the vehicle-mounted big data center or/and the intersection traffic video every 10 seconds, and the vehicle source and outlet data s is used to drive out the detector to obtain big data like the vehicle. It can be used in various vehicle sources through the special setting of the center.
- the roadside parking uses its charging device as the access data, or / and the installed intersection machine coil measures the outflow x per cycle, the head information q0 from the intersection real-time traffic
- Fig. 3 The corresponding intersection system of each intersection of S3 is as shown in Fig. 3, which is marked with the operation of the "AA" algorithm module.
- the basic principle of calculation is shown in Fig. 4.
- the phase fleet fleet q predicts the periodic queuing Q and its variation ⁇ Q, Flow X and the rest And according to the specified captain small change threshold ⁇ Q Th0 , critical threshold Q ThC , large change threshold Q ThS determination and other information through the information channel as shown in Figure 3 mark 3 to push the relevant initial elements as shown in Figure 3 mark 4 to mark 7, mark 9 to mark 12 Wait;
- the S4 initial judgment element is further analyzed according to the threshold information pushed from the predictions according to the predictions, such as the label 5 to the mark 7, the mark 10 to the mark 12, etc., and the same green line flow direction, the same road line, that is, the adjacent between the adjacent lines.
- the same road line that is, the adjacent between the adjacent lines
- the queue change ⁇ Q of the intersection at the downstream of the green wave exceeds the threshold
- the intersection of the 0th intersection of the line 6 can generate north and south.
- the two flow directions fluctuate.
- the road section refers to the north section of the intersection.
- the road section refers to the south section of the intersection.
- the fluctuations are abnormal, all the sections of the same section or column are raised or shared together.
- S5 is co-ordinated as shown in Figure 3, labeled by the solitary wave collision rule as shown in Figure 3, the solitary wave fluctuation conflict rule, as shown in Figure 3, labeled 16, solitary time management rules, as shown in Figure 3, mark 18, the original rule, as shown in Figure 3
- the solitary wave fluctuation conflict rule as shown in Figure 3
- solitary time management rules as shown in Figure 3
- the original rule as shown in Figure 3
- S6 signal time adjustment (1) Super-threshold intersection: Execution of the intelligent command configuration transition period, 1) "Falling", 2) "Solitary wave", (2) Non-threshold intersection: The intersection of the fleet but the threshold directly executes the next step S7 According to S3 "no wisdom instruction", enter S7;
- channel-line 4 The five intersections of channel-line 4 are all ⁇ North 0/0/0" 0/0/0 East 0/0/0 "0/0/0 South 0/0/0" 0/0/0 West 0 /0/0 "0/0/0 ⁇ , the remaining intersections due to the increase in traffic load, waiting for the fleet to automatically restore their ratio control: the left-hand phase of each intersection is 0;
- step S7 After the non-differential state intersection operation step S7 is determined, the process returns to step S3;
- the following shows the predictive control software package to predict the next cycle captain Q(c) and signal residual time according to the previous cycle captain Q(c-1) with the "AA” method.
- Non-green wave Get the range of intersections ⁇ *v0 at the vehicle speed v0 within the green time ⁇ in each direction.
- Green wave the upstream of the wave to the origin, the downstream of the wave to the intersection of the signal time ⁇ 1 *w0,
- Non-green wave system green light time ⁇ d,j minus the time of the intersection of the team through time Q d,j (c-1)/v d,j , when there is time Then reduce the time interval of the upstream intersection and the front team or the team time difference trq ⁇ k,d,j (c) and the sum of the times Q ⁇ k,d,j (c-1)/v d,j until
- K d covers the westbound traffic of the westbound traffic to the green wave starting point including all K e of the intersection (3, 2), (2, 2), (1, 2) and the starting point (0, 2)
- K d covers the intersection to the north.
- the reverse wave car K d covers all k road segments and the queue is divided into discrete traffic.
- Trq ⁇ k,d,j ( ⁇ c dc ) Q ⁇ k,d,j (c-1)* ⁇ >0, and compact traffic trq ⁇ k,d,j ( ⁇ c dc ) ⁇ 0 (congestion),
- the outflow amount x d,j (c) is equal to the number of teams at the intersection. At the same time, add the upstream intersection to the intersection Divide the flow until Run out, rest time When the shortage of intersections includes the intersection and its outflow,
- the inverse wave comes ⁇ c ⁇ k,d : the middle d green wave flow direction is opposite to the traffic flow at the intersection, which is a negative value, using the pre-period prediction value Q(c) or / and the previous pairs of prediction/measurement Q(c)/Q ( C-1) an estimate of the trend (""setting" in the figure,
- the waiting time of 10 is 10 seconds of its turning phase, idle, can be used for 3 straight-line phase, the rest of the time;
- Solitary source signal timing East and West straight phase 20, turn 10, north-south straight phase 30, turn 0, time period is the same as the original period 14)
- the solitary wave amount 13 takes 26 seconds to pass.
- Driving (4,1) takes 8 seconds, and the waiting time is 44: At this time, the green wave of the proprietor has used the straight line for 16 seconds and 4 seconds, the turning phase for 10 seconds, the north-south straight for 18 seconds, and the turn for 8 seconds. 40 seconds remaining;
- Solitary wave source (4, 2) South flow direction intersection (4, 1) timing scheme: east and west straight phase 2, turn 6, north-south straight phase 26, turn 10, the time period is the same as the original cycle;
- Solitary wave source (4, 2) south flow direction intersection (4, 0) timing scheme: north-south straight line phase 24, turn 4, the period is the same as the original period;
- single wave (4, 2) south to -2 means solitary wave: source intersection (4, 2) south flow to two intersections;
- the solitary wave source (4, 2) and its south flow path intersection (4, 1) (4, 0) timing scheme are made to modify the phase time time limit of each intersection direction to the corresponding intersection system, and seamlessly switch Entering the solitary wave state
- intersections are marked with "North*/*/*"*/*/* East*/*/*”*/*/*South*/*/*”*/*/*West*/*/ * "*/*/*”, marked "direction” followed by three numbers representing the actual phase of the straight-phase, left-phase, and right-phase one-time test leader Q(c-1), followed by the symbol "" Three forecasting captains Q(c);
- the initial time difference matrix is shown in Figure 5: Set the previous prohibition of the fluctuation operation to maintain the initial value
- the initial time difference is reduced by 13 time difference matrix:
- Coordination review and coordination (decision-strategy-coordination): review, coordinate conflicts with other intelligent instructions, no conflicts, and intelligent instructions include fluctuations ⁇ *,0 ⁇ northward
- Decoding execution fluctuations (execution layer): According to the macro fluctuation/metamorphism command and the fluctuation time difference tgw, configure the transition period and send it to the intersection of the fluctuation and all the downstream intersections of the green wave;
Abstract
Description
Claims (20)
- 一种道路网络交通信号泛弦控制方法即“A-A”方法,其特征至少包括步骤:S1获取信号参数、路网参数;S2检测各路口各流向d或/和相位j车队尾q即滞留车数、或/和流向路段车源出入车数即出入量s、或还含路口驶出车数即流出量x、或/和队头位置q0信息及相变量子用时Δt Th0;S4各初判元分析确定下时段的路口间信号时间差变化的涨落、或/和区域车流两个交叉大流量变化引起的时间差原点移动初判,即初判等,或/和结合得到的相关各路口方向或/和和相位余时 一起进一步为上述车队Q及其变化ΔQ测算出信号临时配时的孤波初判;S5统筹各初判取舍、先后、时间,制发指令;S6根据指令和预测出信号时间差变化或/和临时配时决定下时段的信号配时;所述车队Q计量单位是米或标准车数,标准车数队长包括车间距,与米长有换算率,非标准车与标准车换算率,通过这些换算率可将车队折算成标准车数或米;所述车数指折算后等效标准车数;所述下时段包括多倍信号周期C,如1C、2C、4C、8C,通用任何信号路网车队预测。
- 根据权利要求1所述方法,其特征是所述步骤S3进一步包括:S3-1所述下时段各路口方向或/和相位车队Q及其变化ΔQ预测步骤:(1)将路口流向d上游实测的路段驶入车数 或/和该上游路口x ±1汇入该流向d的各方向相位流出量x ±1,d1,j1,x ±1,d2,j2,x ±1,d3,j3之和,加上该上游路段车源的出入车数S d(c),没有流向d车源的加0,得到路口预测流入车数a ±0,d,(2)再与该路口方向相位流量分布μ d(c)相乘得到该路口方向相位预测到达车数a d,j(c),(3)再用到达量a d,j(c)减去流出量x ±0,d,j得到各相位排队变化ΔQ,(4)将预测得的车队变化ΔQ加上时段的车队Q d,j(c-1)得到预测Q d(c);所述x ±0,d,j下角标:±k,d,j,按其位置顺序“±”号表示上游、表示k段上游路口、d车流方向、j信号相位,k=0段代表本路口,k=1相邻路口,k=2级是相邻后下一个上游等等,本路口简记为q d,j(c)或q(c)或q ±0(c)或q m,n,d,j(c),‘m,n’代表路口坐标位置,上游路口驶出车数x ±1,d1,j1,x ±1,d2,j2,x ±1,d3,j3中‘d#,j#’表示上游路口驶出汇入下游d流向的方向相位,‘d2,j2’表示左转汇入d流向的方向相位,‘d3,j3’表示右转汇入d流向的方向相位,‘d1,j1’或‘d,j’表示直行汇入d流向的方向相位,相位j=1表示直行、j=2左转、j=3右转,其(c)代表下时段、(c-1)代表前时段,以下依此类推及至各变量;所述方向相位交通量共用车道的,仍用相位分布μ d(c)决定;所述路段车源包括多车源的按它们流向到车源到路口平均距离决定其时间差,常用平均值或0估计;所述智能方法包括综合使用神经网络ann、混沌时序、小波理论、统计回归与支撑向量机svm、遗传优化ga、粒子群优化pso、模糊分析与信息粒化等等智能学习及时序分析方法,以下提到智能方法均同此意;所述流出量x ±0,d,j、x ±1,d1,j1,x ±1,d2,j2,x ±1,d3,j3等可以用下面权利4方法预测得到;或加装路口方向相位驶出检测器实测得到。
- 根据权利要求2所述方法,其特征是所述步骤S3-1进一步包括:所述测算值μ d(c-1)的步骤包括,(1)用前两个时段实测得的相位车队相减得到前时段的排队变化ΔQ d,j(c-1),(2)用相位绿灯时间τ d,j乘以相位流出速率v d,j得到前时段流出量x d,j(c-1),轻载时用前周期预测的流出量代替作为本周期“实测”流出量,或直接用实测流出量,(3)将预测车队变化ΔQ d,j(c-1)和流出量x d,j(c-1)相加得到相位到达量a d,j(c-1),(4)分别将相位到达量除以前面三个相位的到达量之和得到相位流量分布比μ d,j(c-1);所述相位流出速率v d,j指车流驶离交通灯控制线每秒车辆数;
- 根据权利要求2所述方法,其特征是所述步骤S3-1进一步包括:S3-1-2路口方向相位流出量x d,j(c)获得预测值的方法是从预测路口开始向流向上游逐k路段测算每个路口流出量x ±k,d,j(c)及其路段车源流出量s ±k,d,j(c)达到通过预测路口用时及预测路口余时 k=0,1,2…,有余时的路口,该路口流出量计入预测路口的流出量,直至预测路口余时不足,余时不足的各路口包括本路口及其流出量时,其流出量x ±k,d,j(c)是该方向相位相均排队量-车数乘以该余时与该排队量的通过用时比,这里的路口流出量x ±k,d,j(c)以该路口流向的等待车队q ±k,d,j(c-1)测算,k=0,1,2…;或用加装路口驶出车数检测器实测替代计算预测x d,j(c)。
- 根据权利要求2所述方法,其特征是所述步骤S3-1进一步包括:S3-1-3路口方向相位余时 获得预测值的方法是预测路口绿灯时间从预测路口开始向流向上游逐k路段减去每个路口车队量q ±k,d,j(c)及其路段车源流出量s ±k,d,j(c)的相位分流量对预测路口通过的前面路段车流的时距trq ±k(c)及其通过用时tq0 ±k,d,j(c),即余时 k=0,1,2…,直减至余时 为0;所述车队量通过路口用时tq0 d,j(c)以该车队量除以相位流出速率v d,j计;所述上游k路口车队与前面车流的时距trq ±(k-1)(c)以车辆从路口k驶达预测路口路段距离D ±(k-1)除以车辆规定车速v d,l再减去前面路口车队量q ±(k-1)(c)与队扰因子β的乘积计,当信号是绿波系统时,信号时间差|δc ±i,dc|>0,顺波来车流,trq ±(k-1)(c)=-β×q ±(k-1)(c),逆波来车流,其trq ±(k-1)(δc)=2×tv0 ±(k-1)(0)-β×q ±(k-1)(c);所述队扰因子β=1/v d,l+α,是规定车速的倒数与队动系数α之和;所述队动系数α是单位车队长度从车队首车启动到尾车启动的时间,单位秒/米,其估值范围0.14至0.22,取中0.18,可根据经验调整;所述信号时间差δc ±i,dc初始值以路段i长度D ±i除以规定车速v d,l计,即tv0 ±i。
- 根据权利要求2所述方法,其特征是所述步骤S3-1进一步包括:S3-1-4路口方向或和相位预测车队Q m,n,d,j(c)及其变化ΔQ m,n,d,j(c)超过下列控制阈值时,将产生路口元输出并将相关交通信息送出,这些控制阈值包括最小队变阈值ΔQ Th0、变态阈值Q ThC、最小孤波相对队变阈值ΔQ ThS、最小孤波绝对队变阈值Q ThS;所述最小队变阈值ΔQ Th0指一段时间内车队长度最小变化认定值;所述变态阈值Q ThC指车队长度达到信号绿波流向换向值或说平衡值;所述最小孤波相对队变阈值ΔQ ThS指一段时间内车队长度相对其它相位车队最小队长差认定值;所述最小孤波绝对队变阈值Q ThS指车流排队长度最小孤波认定值。
- 根据权利要求2所述方法,其特征是所述步骤S3-1进一步包括S3-1-5预测层各路口元获取数据时间:信号不是绿波的系统,各路口同步在周期开始前获取数据,是绿波的则异步在路口各自周期开始前。
- 根据权利要求1所述方法,其特征是所述步骤S4进一步包括:S4-1所述分析层涨落初判元根据从预测层相应行-列路口元收到路口方向或/和相位车队Q及其变化ΔQ超过控制阈值ΔQ Th0、Q ThC,判断是否超过该路口所在同流向、同路段行或列的路段下游路口的排队变化Δq超过阈值ΔQ Th0的路段数超过控制行阈值M Th0或列阈值N Th0,是超过,则作涨落初判,对绿灯时间过短或道路间距过大的区域,不用行或列阈值M Th0或N Th0判定,分别独立作涨落初判。
- 根据权利要求1所述方法,其特征是所述步骤S5进一步包括:S5-1所述决策层统筹判定规则包括:(1)孤波初判间无冲突规则:孤波间路径平行或孤波间路径流向没有交汇点,(2)孤波-涨落初判间无冲突规则:孤波流向路径涨落绿波的上游,(3)大孤波优先规则:孤波冲突时大孤波量优先,(4)涨落冲突时孤波优先,(5)孤波阶段管理,孤波每波最多经过n LimS个路口,制定智指令及孤波路径禁止再孤波路径时限指令,发智指令。
- 一种道路网络交通信号泛弦控制系统,其特征至少包括运行“A-A”方法的预测控制包即“A-A”包、交通定位数据中心或/和车队即滞留车数检测器、交通路网及其信号灯控制路 口机、或/和路边车源出入检测器、或另含路口驶出车辆检测器、或/和路段驶入车辆检测器;所述“A-A”包根据从交通定位数据中心或/和车队检测器获取的各路口方向或/和相位滞留车数、或/和路边车源出入检测器获取的出入车数预测下时段交通量、确定信号时间方案,方法包可以是中心式或分布式或并行运行,用软件展开或/和硬件展开实现;所述交通定位数据中心检测定位路口各方向或/和相位车队末尾车辆位置确定车队,定位数据来自包括车载定位导航、车辆绑定手机定位导航等任何可定位的移动终端;所述车队检测器指任何可以检测路口方向或/和相位车队长度的装置,如视频分析装置、超声、微波、红外、线圈组等等;所述路边车源出入检测器检测该路口流向的出入车辆数,包括路边计时收费器、小区、停车场、小巷、非交通灯控制路口、高速公路出入检测器,路边有多个车源的可按相关流向到路口的平均距离整合成1个车源估算;所述驶出车辆检测器在路口、路段、小区、高速公路等出口处检测驶离的车辆数;所述驶入车辆检测器在路段、小区、高速公路等入口处检测驶入路段的车辆数;所述车辆检测器包括如可用线圈、压电、磁感、红外、视频或/和其它任何可进行车辆通过计数装置实测获取。
- 根据权利要求14所述“A-A”包,其特征至少包括含有根据实测交通量预测下时段路口交通量的称为路口元的预测模块的预测层,含有分析路口元传来的阈值等信息的称为初判元的分析模块的分析层,含有统筹分析层初判元传来各种阈值信息功能的决策层。
- 根据权利要求15所述路口元模块,其特征至少包括所述路口元预测模块与实际路口是对应关系,路口元之间按需动态相调各自实测/预测交通信息。
- 根据权利要求15所述路口元模块,其特征至少包括路口元输入的是方向或/和相位前时段滞留车数、或/和其路段车源驶出车数,输出是预测出的下时段方向或/和相位剩余绿灯时间、或/和驶出车数、或/和车队变化、或/和车队长度等阈值判定等信息,传送给分析层相关初判元模块。
- 根据权利要求15所述路口元模块,其特征至少包括路口元包括神经元网络及统计学习及时序分析等智能方法模块。
- 根据权利要求15所述初判元模块,其特征至少包括初判元模块输入是路口元传来阈值及其相关信息、输出是作出的信号时间差或/和信号临时配时表的初判,传送给决策层统筹。
- 根据权利要求15所述决策层,其特征至少包括决策层统筹模块输入从分析层传来的初判,输出是对这些初判冲突情况决定取舍、先后、时间等,发出给执行层信号时间指令。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/649,652 US20210375129A1 (en) | 2017-09-28 | 2018-09-18 | Traffic Signal Pan-String Control Method and Its System |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710897777.6 | 2017-09-28 | ||
CN201710897777.6A CN108665714A (zh) | 2017-09-28 | 2017-09-28 | 交通信号泛弦控制方法及其系统 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019061933A1 true WO2019061933A1 (zh) | 2019-04-04 |
Family
ID=63784879
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2018/000332 WO2019061933A1 (zh) | 2017-09-28 | 2018-09-18 | 交通信号泛弦控制方法及其系统 |
Country Status (3)
Country | Link |
---|---|
US (1) | US20210375129A1 (zh) |
CN (1) | CN108665714A (zh) |
WO (1) | WO2019061933A1 (zh) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110135436A (zh) * | 2019-04-30 | 2019-08-16 | 中国地质大学(武汉) | 一种用智能小车识别闪烁信标灯的方法、设备及存储设备 |
CN110782658A (zh) * | 2019-08-16 | 2020-02-11 | 华南理工大学 | 一种基于LightGBM算法的交通量预测方法 |
CN111785042A (zh) * | 2020-07-01 | 2020-10-16 | 浙江工商大学 | 基于ZigBee的车流量检测并自动调节红绿灯时长的方法 |
CN112489456A (zh) * | 2020-12-01 | 2021-03-12 | 山东交通学院 | 基于城市主干线车辆排队长度的信号灯调控方法及系统 |
CN114120655A (zh) * | 2022-01-26 | 2022-03-01 | 苏州映赛智能科技有限公司 | 道路状态预测方法、系统、设备及存储介质 |
CN114446062A (zh) * | 2022-01-28 | 2022-05-06 | 刘银环 | 基于大数据服务的动态调配系统及方法 |
CN114582166A (zh) * | 2022-03-07 | 2022-06-03 | 清华大学 | 交叉路口环境下的多车运动规划方法及装置 |
CN114677846A (zh) * | 2022-04-13 | 2022-06-28 | 合肥工业大学 | 纯网联自动驾驶环境下信号交叉口进口道的动态分配方法 |
EP4040420A3 (en) * | 2021-06-15 | 2022-12-21 | Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd. | Method and apparatus for acquiring traffic volume and electronic device |
CN116819986A (zh) * | 2023-08-28 | 2023-09-29 | 温州宝翔科技有限公司 | 一种稳压排插的远程控制方法及系统 |
US11875371B1 (en) | 2017-04-24 | 2024-01-16 | Skyline Products, Inc. | Price optimization system |
Families Citing this family (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020039538A1 (ja) * | 2018-08-23 | 2020-02-27 | 三菱電機株式会社 | 通信装置、通信方法及び通信プログラム |
CN109544945B (zh) * | 2018-11-30 | 2021-06-01 | 江苏智通交通科技有限公司 | 基于车道饱和度的区域控制相位配时优化方法 |
US11403938B2 (en) | 2019-04-04 | 2022-08-02 | Geotab Inc. | Method for determining traffic metrics of a road network |
CN110263987A (zh) * | 2019-06-06 | 2019-09-20 | 太原理工大学 | 一种适用于半干旱半湿润地区的洪水预报方法 |
CN111127894B (zh) * | 2019-12-30 | 2021-07-09 | 江苏中路工程技术研究院有限公司 | 一种基于监测手段的交通运输能耗控制方法及其检测方法 |
US20230108068A1 (en) * | 2020-03-02 | 2023-04-06 | Weiping Meng | Traffic Signal Polarized Green-Wave Control Method |
CN111402615B (zh) * | 2020-04-08 | 2021-10-26 | 王爱伶 | 一种基于导航信息的可变车道控制方法 |
CN111696348B (zh) * | 2020-06-05 | 2022-08-09 | 南京云创大数据科技股份有限公司 | 一种多功能的智能信号控制系统和方法 |
US11731651B2 (en) * | 2020-09-30 | 2023-08-22 | Baidu Usa Llc | Automatic parameter tuning framework for controllers used in autonomous driving vehicles |
US11823565B2 (en) * | 2020-12-18 | 2023-11-21 | Yunex Llc | System and method for vehicle queue length detection in a connected vehicle infrastructure environment |
CN112785858B (zh) * | 2021-02-19 | 2023-06-16 | 阿里巴巴集团控股有限公司 | 一种交通控制方法、装置及电子设备 |
US20220375340A1 (en) * | 2021-05-20 | 2022-11-24 | Blyncsy, Inc. | Machine-learning based control of traffic operation |
CN113450569A (zh) * | 2021-06-30 | 2021-09-28 | 阿波罗智联(北京)科技有限公司 | 确定路口状态的方法、装置、电子设备和存储介质 |
CN113823099B (zh) * | 2021-09-10 | 2023-03-21 | 上海商汤智能科技有限公司 | 信控方案调整方法、装置、设备及计算机可读存储介质 |
CN114863677B (zh) * | 2022-04-26 | 2023-04-21 | 公安部交通管理科学研究所 | 一种停车场上游路口信号控制动态优化方法 |
CN114898576B (zh) * | 2022-05-10 | 2023-12-19 | 阿波罗智联(北京)科技有限公司 | 交通控制信号的生成方法、目标网络模型的训练方法 |
CN115662121A (zh) * | 2022-10-19 | 2023-01-31 | 吉林大学 | 混行信控交叉口以排队方式估算通行能力上下限的方法 |
CN116229740B (zh) * | 2023-05-10 | 2023-10-10 | 湖北大学 | 一种红绿灯配时控制方法及系统 |
CN117198071B (zh) * | 2023-11-03 | 2024-01-09 | 合肥工业大学 | 一种交通信号相位自适应调整方法、系统、设备 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003016582A (ja) * | 2001-07-03 | 2003-01-17 | Matsushita Electric Ind Co Ltd | 信号制御装置および方法 |
CN101281685A (zh) * | 2008-01-30 | 2008-10-08 | 吉林大学 | 区域混合交通自适应信号协调控制方法 |
CN101615344A (zh) * | 2009-07-21 | 2009-12-30 | 青岛海信网络科技股份有限公司 | 一种基于线协调的区域交通控制方法 |
CN101877169A (zh) * | 2009-11-13 | 2010-11-03 | 北京交通大学 | 干线多交叉口交通流均衡性控制的数据融合系统及方法 |
CN104464310A (zh) * | 2014-12-02 | 2015-03-25 | 上海交通大学 | 城市区域多交叉口信号协同优化控制方法及系统 |
CN105390000A (zh) * | 2015-12-18 | 2016-03-09 | 天津通翔智能交通系统有限公司 | 一种基于路况交通大数据的交通信号控制系统及方法 |
CN106355885A (zh) * | 2016-11-24 | 2017-01-25 | 深圳市永达电子信息股份有限公司 | 基于大数据分析平台的交通信号动态控制方法及系统 |
-
2017
- 2017-09-28 CN CN201710897777.6A patent/CN108665714A/zh active Pending
-
2018
- 2018-09-18 US US16/649,652 patent/US20210375129A1/en not_active Abandoned
- 2018-09-18 WO PCT/CN2018/000332 patent/WO2019061933A1/zh active Application Filing
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003016582A (ja) * | 2001-07-03 | 2003-01-17 | Matsushita Electric Ind Co Ltd | 信号制御装置および方法 |
CN101281685A (zh) * | 2008-01-30 | 2008-10-08 | 吉林大学 | 区域混合交通自适应信号协调控制方法 |
CN101615344A (zh) * | 2009-07-21 | 2009-12-30 | 青岛海信网络科技股份有限公司 | 一种基于线协调的区域交通控制方法 |
CN101877169A (zh) * | 2009-11-13 | 2010-11-03 | 北京交通大学 | 干线多交叉口交通流均衡性控制的数据融合系统及方法 |
CN104464310A (zh) * | 2014-12-02 | 2015-03-25 | 上海交通大学 | 城市区域多交叉口信号协同优化控制方法及系统 |
CN105390000A (zh) * | 2015-12-18 | 2016-03-09 | 天津通翔智能交通系统有限公司 | 一种基于路况交通大数据的交通信号控制系统及方法 |
CN106355885A (zh) * | 2016-11-24 | 2017-01-25 | 深圳市永达电子信息股份有限公司 | 基于大数据分析平台的交通信号动态控制方法及系统 |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11875371B1 (en) | 2017-04-24 | 2024-01-16 | Skyline Products, Inc. | Price optimization system |
CN110135436A (zh) * | 2019-04-30 | 2019-08-16 | 中国地质大学(武汉) | 一种用智能小车识别闪烁信标灯的方法、设备及存储设备 |
CN110135436B (zh) * | 2019-04-30 | 2020-11-27 | 中国地质大学(武汉) | 一种用智能小车识别闪烁信标灯的方法、设备及存储设备 |
CN110782658A (zh) * | 2019-08-16 | 2020-02-11 | 华南理工大学 | 一种基于LightGBM算法的交通量预测方法 |
CN111785042A (zh) * | 2020-07-01 | 2020-10-16 | 浙江工商大学 | 基于ZigBee的车流量检测并自动调节红绿灯时长的方法 |
CN112489456A (zh) * | 2020-12-01 | 2021-03-12 | 山东交通学院 | 基于城市主干线车辆排队长度的信号灯调控方法及系统 |
CN112489456B (zh) * | 2020-12-01 | 2022-01-28 | 山东交通学院 | 基于城市主干线车辆排队长度的信号灯调控方法及系统 |
EP4040420A3 (en) * | 2021-06-15 | 2022-12-21 | Apollo Intelligent Connectivity (Beijing) Technology Co., Ltd. | Method and apparatus for acquiring traffic volume and electronic device |
CN114120655A (zh) * | 2022-01-26 | 2022-03-01 | 苏州映赛智能科技有限公司 | 道路状态预测方法、系统、设备及存储介质 |
CN114446062B (zh) * | 2022-01-28 | 2022-09-27 | 深圳市八度云计算信息技术有限公司 | 基于大数据服务的动态调配系统及方法 |
CN114446062A (zh) * | 2022-01-28 | 2022-05-06 | 刘银环 | 基于大数据服务的动态调配系统及方法 |
CN114582166A (zh) * | 2022-03-07 | 2022-06-03 | 清华大学 | 交叉路口环境下的多车运动规划方法及装置 |
CN114677846A (zh) * | 2022-04-13 | 2022-06-28 | 合肥工业大学 | 纯网联自动驾驶环境下信号交叉口进口道的动态分配方法 |
CN114677846B (zh) * | 2022-04-13 | 2022-12-06 | 合肥工业大学 | 纯网联自动驾驶环境下信号交叉口进口道的动态分配方法 |
CN116819986A (zh) * | 2023-08-28 | 2023-09-29 | 温州宝翔科技有限公司 | 一种稳压排插的远程控制方法及系统 |
CN116819986B (zh) * | 2023-08-28 | 2023-11-24 | 温州宝翔科技有限公司 | 一种稳压排插的远程控制方法及系统 |
Also Published As
Publication number | Publication date |
---|---|
CN108665714A (zh) | 2018-10-16 |
US20210375129A1 (en) | 2021-12-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019061933A1 (zh) | 交通信号泛弦控制方法及其系统 | |
CN103927887B (zh) | 结合离散速度模型的阵列式fpga交通状态预测及控制系统 | |
CN101894477B (zh) | 一种城市信号灯控制道路网络交通自锁控制方法 | |
CN102867422B (zh) | 基于车辆自组网的单点十字交叉口信号灯实时控制方法 | |
CN101308604B (zh) | 大范围战略交通协调控制方法 | |
CN103927890B (zh) | 一种基于动态o-d矩阵估计的干线协调信号控制方法 | |
CN108470444A (zh) | 一种基于遗传算法优化的城市区域交通大数据分析系统与方法 | |
Chang et al. | A study on traffic signal control at signalized intersections in vehicular ad hoc networks | |
KR101385057B1 (ko) | Its 데이터를 활용한 도심 구간의 정체도 예측 방법 | |
CN104778834A (zh) | 一种基于车辆gps数据的城市道路交通拥堵判别方法 | |
CN103236164B (zh) | 一种保障公交车辆优先通行的车辆调控方法 | |
CN102708679A (zh) | 一种城市信号交叉口短时交通流预测方法 | |
CN106408984B (zh) | 基于物联网的城市交通流诱导系统 | |
CN101894461A (zh) | 城市地面道路行程时间预测方法 | |
CN108427273A (zh) | 一种减少交通拥堵现象的反馈控制设计方法 | |
CN103106789A (zh) | 一种交通诱导系统与信号控制系统的协同方法 | |
Liu et al. | A generalized integrated corridor diversion control model for freeway incident management | |
CN201262784Y (zh) | 基于数据特征的城市信号控制路口交通状态检测和评价系统 | |
CN111009140B (zh) | 一种基于开源路况信息的智能交通信号控制方法 | |
CN112926768A (zh) | 基于时空注意力机制的地面道路车道级交通流预测方法 | |
CN110211392A (zh) | 一种针对可变公交进口道的信号优先控制方法 | |
Ma et al. | Gating control for a single bottleneck link based on traffic load equilibrium | |
CN112258856A (zh) | 一种区域交通信号数据驱动控制模型建立方法 | |
CN110021178B (zh) | 一种基于路口相似度的信号配时方案推荐方法 | |
Lu et al. | Research on the intelligent assignment model of urban traffic planning based on optimal path optimization algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18863275 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18863275 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 120121) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 18863275 Country of ref document: EP Kind code of ref document: A1 |