CN102819954B - Traffic region dynamic map monitoring and predicating system - Google Patents
Traffic region dynamic map monitoring and predicating system Download PDFInfo
- Publication number
- CN102819954B CN102819954B CN201210309274.XA CN201210309274A CN102819954B CN 102819954 B CN102819954 B CN 102819954B CN 201210309274 A CN201210309274 A CN 201210309274A CN 102819954 B CN102819954 B CN 102819954B
- Authority
- CN
- China
- Prior art keywords
- module
- vehicle
- car
- traffic flow
- 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.)
- Expired - Fee Related
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 38
- 230000001413 cellular effect Effects 0.000 claims abstract description 22
- 238000013528 artificial neural network Methods 0.000 claims abstract description 20
- 238000013523 data management Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims description 30
- 230000000903 blocking effect Effects 0.000 claims description 14
- 230000006698 induction Effects 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 14
- 238000001514 detection method Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 10
- 230000001133 acceleration Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 8
- 241000209094 Oryza Species 0.000 claims description 6
- 235000007164 Oryza sativa Nutrition 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 230000008447 perception Effects 0.000 claims description 6
- 235000009566 rice Nutrition 0.000 claims description 6
- 238000012512 characterization method Methods 0.000 claims description 4
- 230000004075 alteration Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 238000007726 management method Methods 0.000 abstract description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 abstract 1
- 229910052799 carbon Inorganic materials 0.000 abstract 1
- 230000006870 function Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000004907 flux Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 125000002837 carbocyclic group Chemical group 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000013341 scale-up Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a traffic region dynamic map monitoring and predicating system and belongs to the field of traffic flow monitoring. The traffic region dynamic map monitoring and predicating system comprises a sensing module, a data management module, a map generation module, a map display module, a vehicle analog module, an authentication configuration module and a traffic flow parameter predicating module. By means of combination of a cellular automation model and an RBF (radial basis function) neural network short-term traffic flow predicating model, current jamming conditions and future jamming conditions of road sections can be computed to enable drivers to rapidly know road network situations so as to select proper roads, and accordingly running time is saved while carbon conservation and environment protection are realized. Further, traffic management personnel can know road network conditions in real time and be aware of specific traffic parameters and road condition videos through windows to take necessary measures.
Description
Technical field
The invention belongs to magnitude of traffic flow monitoring field, more particularly, relate to a kind of traffic zone dynamic map monitoring and controlling forecast system.
Background technology
Along with China's car owning amount grows with each passing day, road network flux density continues to increase, and vehicle congestion and traffic hazard have become a difficult problem for puzzlement society.Obviously, effectively the measure of guide car flow have real-time control road network flux density, in good time select suitable route, balance road network flow, rapidly know traffic events, and in time emergent troubleshooting process, relieve traffic congestion etc., but this depends on the modernization science perception means of road network both macro and micro.Although the awareness apparatus such as current monitoring camera, radar velocity measurement are generally laid, many vehicles have been installed GIS tour guide device, and traffic perception has subject matter to be solved to have:
1) the single monitoring camera of trackside can only provide the road conditions video of effective sighting distance, is difficult to the traffic of reflection section, road network integral body, or macro-traffic situation, the accurate information that can not provide in real time road induction or point duty to dredge;
2) a large amount of trackside shootings, manually watch, and except labor intensive, often can not understand in time traffic events;
3) road GIS device is housed, is loaded with static road data, do not reflect actual motion road conditions, can not select route from consuming time or unimpeded angle.
China Patent No. 201010290408.9, open day on 01 26th, 2011, disclose a name and be called the magnitude of traffic flow based on GPS and the patent document of road congestion detection system, it comprises being arranged on car-mounted terminal in monitor vehicle and in order to receive the information of car-mounted terminal the server that carries out crowded calculating; Described car-mounted terminal is connected with server communication by cordless communication network, and described car-mounted terminal comprises GPS module, the first communication module, display module and message processing module; Described server comprises, congestion status computing module, in order to the coordinate information returning according to different car-mounted terminals, by blocking up, function obtains Real-time Traffic Information: adopt Congestion function to calculate the crowding of each different sections of highway, if crowding is greater than default crowding threshold value, be judged to be congestion state; Otherwise, be judged to be unobstructed state; The second communication module, in order to send to car-mounted terminal by the traffic behavior of each different sections of highway.This invention is low-cost, practical, be easy to promote.
Chinese Patent Application No. 201110439431.4, open day on 06 13rd, 2012, the patent document that a name is called the collection of a kind of road traffic flow and Forecasting Methodology is disclosed, it comprises toroidal inductor, vehicle detection module, magnitude of traffic flow acquisition module, traffic flow data pre-service and prediction, the data pre-service of road traffic flow and forecasting software carry out on host computer (PC), and read the traffic flow data in acquisition module (SD card) by network interface.For improving forecasting reliability, in the data pre-service of road traffic flow and Forecasting Methodology, first adopt wavelet analysis, in conjunction with least square method, traffic flow data is carried out to noise eliminating; Then adopt improved BP neural network model of traffic flux forecast, realize the prediction to the magnitude of traffic flow, for optimizing control timing scheme and the traffic planninng of road traffic, provide foundation.This invention can obtain the road traffic parameter such as vehicle flowrate, average speed, occupation rate and traffic density in specified period, thereby realizes the prediction to road traffic flow, improves the accuracy of data acquisition and the prediction of road traffic flow.
The patent of the above-mentioned magnitude of traffic flow based on GPS and road congestion detection system a large amount of monitoring vehicle that requires to travel on road, or require a large amount of common vehicle that GPS module is installed and to server open message interface.The former can increase traffic pressure, and cost is huge, and it is large that the latter promotes difficulty, is difficult to carry out.The patent of the collection of above-mentioned a kind of road traffic flow and Forecasting Methodology has just described how traffic is carried out to data acquisition and prediction, do not mention the information how system statistical analysis being obtained and express better, and the effective Feedback form of information is only the direct mode that embodies that traffic analysis system is worth.
Summary of the invention
the problem solving
For existing traffic analysis system, to infrastructure, require high, Informational Expression mode is dull, content is single, the problem of shortage multi-level information feedback from macroscopic view to microcosmic, the invention provides a kind of traffic zone dynamic map monitoring and controlling forecast system, can allow driver understand rapidly road network situation, select suitable route, save working time, joint carbocyclic ring is protected; Traffic administration personnel can hold road network situation in real time, by window, know concrete traffic parameter and road conditions video, take necessary counter-measure, relieve traffic congestion in time.
technical scheme
In order to address the above problem, the technical solution adopted in the present invention is as follows:
A traffic zone dynamic map monitoring and controlling forecast system, comprises sensing module, data management module, map generation module, map exhibition module, vehicle simulation module, authenticated configuration module and traffic flow parameter prediction module;
Described sensing module is responsible for collection signal, information for detection of each porch, unit, section vehicle, via data management module analysis, input traffic flow parameter prediction module is processed, by map generation module, map, put on display module and vehicle simulation module composition dynamic map, reflection road network macroscopic view is to the multi-level situation of microcosmic;
The measuring-signal of described sensing module is mainly derived from the monitoring camera on highway, the section that native system be take in the middle of every two monitoring cameras is elementary cell, detect the essential information of each elementary cell porch vehicle, utilize traffic flow parameter prediction module prediction vehicle in the mode of travelling of inside, unit and carry out three-dimensional simulation demonstration, forming virtual region dynamic map;
Described map generation module carries out the maintenance of dynamic map platform, mainly carries out three-dimensional simulation, projection conversion, map registration, the processing such as sign of blocking up, and can from macroscopic view to microcosmic, show at many levels road network operation conditions;
Described map is put on display module has a color ribbon to characterize the degree of blocking up of corresponding road section at the other mark of road automatically;
Described vehicle simulation module is carried out the virtual demonstration of Vehicle Driving Cycle, carry out the processing such as three-dimensional modeling, speed of a motor vehicle adjusting, car look registration, vehicle classification of vehicle, the rule that the described speed of a motor vehicle regulates is according to specific traffic flow model, with less detection Data Representation road network operation conditions, and guarantee smoothly travelling of section, unit intersection model vehicle, described vehicle is detected and is obtained by monitor video, can be divided into middle minibus, motor bus, jubilee wagon, medium truck, high capacity waggon, super-huge lorry, seven kinds of vehicles of container car, corresponding seven kinds of three-dimensional models;
Described authenticated configuration module, mainly to obtain authority by authenticated configuration module, can retrieve for examination the local trackside video pictures that fixedly monitoring camera is taken, also can command the mobile video sources such as road inspection car specifying section follow shot, watch fixing monitoring to be difficult to the on-the-spot details or the road blind area that catch.
Further, described traffic flow parameter prediction module adopts the mode that cellular Automation Model combines with RBF neural network short-term forecasting traffic flow model;
Described cellular Automation Model is all done careful calculating to the rule of travelling of each car, is that system each car in segment unit that satisfies the need is done the theoretical foundation that three-dimensional simulation shows; In the traffic flow model of cellular automaton, on track the time of Vehicle Driving Cycle, space and speed by discretize,
in process, model is pressed following four regular parallel evolutionaries:
1) accelerate rule: in the larger distance of the place ahead, there is no car,
, corresponding to the characteristic that in reality, driver's expectation is travelled with maximal rate;
2) rule of slowing down: in the nearer distance in the place ahead, there is car,
, driver takes the measure of slowing down for fear of bumping with front truck;
3) random slowing down: with Probability p,
, represent various uncertain factors,, driver different phychology fluctuations bad such as pavement behavior etc., the vehicle deceleration that may cause, but the speed of a motor vehicle is greater than a rear car all the time;
4) motion:
, vehicle according to the speed after adjusting to overtake.
Here, Xn represents the position of n car; Vn represents the speed of n car, in this speed characterization system the every frame of dynamic map refresh after the distance in this car early stage; Dn represents the distance between n car and front truck n+1.
System is processed each car on each frame of dynamic map, reads information of vehicles, according to aforementioned four rules, calculate car in the position of next frame, the state such as speed, acceleration, and in next frame, carry out corresponding demonstration;
The numerical simulation that completes model also must be determined boundary condition, and the present invention adopts improved open boundary condition:
In road exit, a truck position exceeds road scope and assert that it rolls section away from, no longer shows, the second car of closelying follow thereafter becomes a new car, in road porch, vehicle was located before front truck produces position with certain probability to produce, folded with two car weights before and after preventing;
Described RBF neural network short-term forecasting traffic flow model is by the mode of sample training, the segment unit magnitude of traffic flow of satisfying the need has more in advance and prediction accurately, system arranges an entrance in each unit, utilize awareness apparatus to detect the situation that enters vehicle, system is M rice before each unit, section, vehicle travels according to original cellular Automation Model rule, at rear Y rice, system detects the magnitude of traffic flow A in dummy model, utilize RBF neural network short-term forecasting traffic flow model simultaneously, read the magnitude of traffic flow in front several sections, dope the now due magnitude of traffic flow B in this end, unit, section, A is compared with B, then adjust the parameter in cellular Automation Model, comprise acceleration distance, deceleration distance, the random chance of random slowing down, make vehicle flowrate in dummy model in the end tend to gradually the requirement of magnitude of traffic flow B in Y rice, the vehicle that reaches section intersection shows effect level and smooth and that traffic is constantly revised.
Further, described color ribbon is comprised of red, yellow, green, blue gradual change aberration, gets 7 kinds, conversion step, and the degree of blocking up uses the average velocity of corresponding road section vehicle as criterion.For example, with reference to international standard and Chinese practice traffic, should set corresponding road section average speed is being redness below 60 kilometers, and 100 kilometers is green above, and centre is a step by 8 kilometers, and color is gradual change successively.Also can get other colors, and the scope of average speed.
Further, described M is that 900, Y is 100.
Further, the measuring-signal of described sensing module also derive from ground induction coil, radar velocity measurement equimagnetic frequently, ripple frequently, piezoelectric type sensor, can the perception speed of a motor vehicle, vehicle number, vision-based detection sensing can obtain the information such as vehicle, car look in addition.
Further, also comprise and estimate induction module, utilize and estimate induction module, collecting on the basis of current Real-time Road information, the analysis such as carry out region selection, the judgement of blocking up, time are estimated, finally draws and arrives the shortest route induction information of destination required time.
When the present invention uses, the section of take in the middle of every two monitoring cameras is elementary cell, detect the essential information of porch, every unit vehicle, utilize the mode of travelling of one to two kilometer in inside, unit, traffic flow model prediction vehicle and carry out three-dimensional simulation demonstration, form virtual region dynamic map.
First, utilize ground induction coil, radar equimagnetic frequency, ripple frequency, piezoelectric type sensor, each section entrance of perception enters the speed of a motor vehicle, the vehicle number of vehicle; Utilize the unusual condition on vehicle, color and the current road surface of video detection vehicle, as ponding, sleet etc.; Utilize trackside monitoring camera and road inspection car follow shot condition of road surface, and these data are delivered to data management module.
Data management module carries out corresponding format conversion, classification by the parameter obtaining, and to vehicle attribute data, carries out the unification of unit, and then rejecting abnormalities data submit to forecasting traffic flow module and vehicle simulation module.To environmental data, carry out order of severity differentiation, submit to map generation module.To video data, carry out H264 coding, to language data, carry out AAC coding, submit to map and put on display module.
Traffic flow parameter prediction module receives after information of vehicles, the mode predicted link situation of utilizing cellular automaton to combine with RBF neural network short-term forecasting traffic flow model, and the information of forecasting of each car is submitted to map generation module.
Vehicle simulation module receives after the data such as the size, color of vehicle, contrasts with the auto model in self model storehouse, selects the most similar auto model, submits to map generation module.
Map generation module receives the data of traffic flow parameter prediction module and vehicle simulation module, utilize OPENGL technology, the map interface of generating three-dimensional, outside the road apart from three-dimensional simulation on interface, also have the driving vehicle corresponding with reality, then cartographic information is submitted to map and put on display module.
Map is put on display the information that module receives map generation module, and the evaluation speed of adding up driving vehicle on each road be take 60 kilometers below as red, 100 kilometers is green above, centre is a step by 8 kilometers, and color is the rule of gradual change successively, gives every section color matching.Can zoom out visual angle by selection, by watching colour bar to understand the congestion of whole road network macroscopic view, the reality that also can further, the section of amplification appointment, the driving details of watching microcosmic.
If traffic administration person or road management person use native system, when when macroscopic aspect is observed road network operation conditions, concrete microcosmic traffic situation to a certain section is interested, can obtain authority by authenticated configuration module, retrieve for examination the local trackside video pictures that fixedly monitoring camera is taken, also can command the mobile video sources such as road inspection car specifying section follow shot, watch fixing monitoring to be difficult to the on-the-spot details or the road blind area that catch, be convenient to dispatch control or on-site law-enforcing etc.
If driving driver uses native system, can utilize and estimate induction module, to collect on the basis of current Real-time Road information, the analysis such as carry out region selection, the judgement of blocking up, time are estimated, finally draws the shortest route induction information of arrival destination required time.
beneficial effect
Than prior art, beneficial effect of the present invention is:
(1) the present invention includes sensing module, data management module, map generation module, map exhibition module, vehicle simulation module, authenticated configuration module and traffic flow parameter prediction module; Utilize a small amount of road awareness apparatus, obtain the traffic parameter of each section mouth in road network, can hold in real time the traffic information of road network situation and timely predict future and select best route map to driver;
(2) traffic flow parameter prediction module of the present invention adopts the mode that cellular Automation Model combines with RBF neural network short-term forecasting traffic flow model, can calculate jam situation and following jam situation in section, driver understands rapidly road network situation, select suitable route, save working time, joint carbocyclic ring is protected; Traffic administration personnel can hold road network situation in real time, by window, know concrete traffic parameter and road conditions video, take necessary counter-measure;
(3) section that the present invention be take in the middle of every two monitoring cameras is elementary cell, takes full advantage of freeway surveillance and control device distribution characteristic, during use the system reform simple, cost is low;
(4) the present invention adopts and accelerates rule, deceleration rule, random slowing down and four rules of motion the prediction of vehicle flowrate, and Forecasting Methodology more approaches actual conditions, and it is more accurate to predict;
(5) the inventive method is simple, reasonable in design, is easy to realize.
Accompanying drawing explanation
Fig. 1 is system architecture schematic diagram of the present invention.
Embodiment
Below in conjunction with concrete accompanying drawing, describe the present invention.
As shown in Figure 1, a kind of traffic zone dynamic map monitoring and controlling forecast system, comprises sensing module, data management module, map generation module, map exhibition module, vehicle simulation module, authenticated configuration module and traffic flow parameter prediction module.
Sensing module is responsible for collection signal, information for detection of each porch, unit, section vehicle, the measuring-signal of sensing module also derives from ground induction coil, radar velocity measurement equimagnetic frequency, ripple frequency, piezoelectric type sensor, can the perception speed of a motor vehicle, vehicle number, vision-based detection sensing can obtain the unusual condition on vehicle, Che Se and current road surface in addition, as information such as ponding, sleet.Via data management module analysis, data management module carries out corresponding format conversion, classification by the parameter obtaining, to vehicle attribute data, carry out the unification of unit, rejecting abnormalities data, then submitting to traffic flow parameter prediction module processes, to environmental data, carry out order of severity differentiation, submit to map generation module, map exhibition module and vehicle simulation module composition dynamic map, for video data, carry out H264 coding, to language data, carry out AAC coding, submit to map and put on display module.
The measuring-signal of sensing module is mainly derived from the monitoring camera on highway, the section that native system be take in the middle of every two monitoring cameras is elementary cell, detect the essential information of each elementary cell porch vehicle, utilize traffic flow parameter prediction module prediction vehicle in the mode of travelling of inside, unit and carry out three-dimensional simulation demonstration, forming virtual region dynamic map.
Map generation module carries out the maintenance of dynamic map platform, mainly carries out three-dimensional simulation, projection conversion, map registration, the processing such as sign of blocking up, and can from macroscopic view to microcosmic, show at many levels road network operation conditions.
Map is put on display module has a color ribbon to characterize the degree of blocking up of corresponding road section at the other mark of road automatically; Color ribbon is comprised of red, yellow, green, blue gradual change aberration, and desirable 7 kinds, conversion step, the degree of blocking up uses the average velocity of corresponding road section vehicle as criterion.With reference to international standard and Chinese practice traffic, should set corresponding road section average speed is being redness below 60 kilometers, and 100 kilometers is green above, and centre is a step by 8 kilometers, and color is gradual change successively.
Vehicle simulation module is carried out the virtual demonstration of Vehicle Driving Cycle, carry out the processing such as three-dimensional modeling, speed of a motor vehicle adjusting, car look registration, vehicle classification of vehicle, the rule that the described speed of a motor vehicle regulates is according to specific traffic flow model, with less detection Data Representation road network operation conditions, and guarantee smoothly travelling of section, unit intersection model vehicle, vehicle is detected and is obtained by monitor video, can be divided into middle minibus, motor bus, jubilee wagon, medium truck, high capacity waggon, super-huge lorry, seven kinds of vehicles of container car, corresponding seven kinds of three-dimensional models; Vehicle simulation module receives after the data such as the size, color of vehicle, contrasts with the auto model in self model storehouse, selects the most similar auto model, submits to map generation module.
If traffic administration person or road management person use native system, when when macroscopic aspect is observed road network operation conditions, concrete microcosmic traffic situation to a certain section is interested, can obtain authority by authenticated configuration module, retrieve for examination the local trackside video pictures that fixedly monitoring camera is taken, also can command the mobile video sources such as road inspection car specifying section follow shot, watch fixing monitoring to be difficult to the on-the-spot details or the road blind area that catch, be convenient to dispatch control or on-site law-enforcing etc.
The mode that traffic flow parameter prediction module adopts cellular Automation Model to combine with RBF neural network short-term forecasting traffic flow model;
Cellular Automation Model is all done careful calculating to the rule of travelling of each car, is that system each car in segment unit that satisfies the need is done the theoretical foundation that three-dimensional simulation shows; In the traffic flow model of cellular automaton, on track the time of Vehicle Driving Cycle, space and speed by discretize,
in process, model is pressed following four regular parallel evolutionaries:
1) accelerate rule: in the larger distance of the place ahead, there is no car,
, corresponding to the characteristic that in reality, driver's expectation is travelled with maximal rate;
2) rule of slowing down: in the nearer distance in the place ahead, there is car,
, driver takes the measure of slowing down for fear of bumping with front truck;
3) random slowing down: with Probability p,
, represent various uncertain factors,, driver different phychology fluctuations bad such as pavement behavior etc., the vehicle deceleration that may cause, but the speed of a motor vehicle is greater than a rear car all the time;
4) motion:
, vehicle according to the speed after adjusting to overtake.
Here, Xn represents the position of n car; Vn represents the speed of n car, in this speed characterization system the every frame of dynamic map refresh after the distance in this car early stage; Dn represents the distance between n car and front truck n+1.
System is processed each car on each frame of dynamic map, reads information of vehicles, according to aforementioned four rules, calculate car in the position of next frame, the state such as speed, acceleration, and in next frame, carry out corresponding demonstration.
The numerical simulation that completes model also must be determined boundary condition, and the present invention adopts improved open boundary condition:
In road exit, a truck position exceeds road scope and assert that it rolls section away from, no longer shows, the second car of closelying follow thereafter becomes a new car, in road porch, vehicle was located before front truck produces position with certain probability to produce, folded with two car weights before and after preventing.
RBF neural network short-term forecasting traffic flow model is by the mode of sample training, the segment unit magnitude of traffic flow of satisfying the need has more in advance and prediction accurately, system arranges an entrance in each unit, utilize awareness apparatus to detect the situation that enters vehicle, first 900 meters in each unit, section of systems, vehicle travels according to original cellular Automation Model rule, at rear 100 meters of, system detects the magnitude of traffic flow A in dummy model, utilize RBF neural network short-term forecasting traffic flow model simultaneously, read the magnitude of traffic flow in front several sections, dope the now due magnitude of traffic flow B in this end, unit, section, A is compared with B, then adjust the parameter in cellular Automation Model, comprise acceleration distance, deceleration distance, the random chance of random slowing down, make vehicle flowrate in dummy model in the end tend to gradually the requirement of magnitude of traffic flow B in 100 meters, the vehicle that reaches section intersection shows effect level and smooth and that traffic is constantly revised.
Traffic flow parameter prediction module receives after information of vehicles, the mode predicted link situation of utilizing cellular automaton to combine with RBF neural network short-term forecasting traffic flow model, and the information of forecasting of each car is submitted to map generation module.Map generation module receives the data of traffic flow parameter prediction module and vehicle simulation module, utilize OPENGL technology, the map interface of generating three-dimensional, outside the road apart from three-dimensional simulation on interface, also have the driving vehicle corresponding with reality, then cartographic information is submitted to map and put on display module.
If driving driver uses native system, can utilize and estimate induction module, to collect on the basis of current Real-time Road information, the analysis such as carry out region selection, the judgement of blocking up, time are estimated, finally draws the shortest route induction information of arrival destination required time.
The monitoring camera spacing of domestic highway is generally 1 to 2 kilometer, and the section that native system be take in the middle of every two monitoring cameras is elementary cell.
The flow of traffic flow refers at moment t extremely
in,
for shorter time span, be generally and get 5--15min, by the vehicle number of a certain section observation station.Traffic flow forecasting ultimate principle is as described below:
If:
for certain observation station α on i bar section in road network is in moment section
to the integrated flow in t,
for predetermined period, in the prediction of short-term traffic flow, general Δ t≤15 min, same,
,
represent respectively the flow of its forward and backward n in the period.The problem concrete to certain, due to
fix with a, therefore will in following presents
,
brief note is
,
.
historical statistical data for the identical period of same place; With i each m section of upstream and downstream section that section is adjacent, its label is i ± j, j=1, and 2 ..., m.
Short-time traffic flow forecast is exactly to know i section and i+ j section P flow constantly in the past according to oneself
And relevant statistics, refer to this prediction section traffic flow data of same time period in the past,
Obtain interior flow of following k time period of i section
estimated value, and these
predicted value is called predictor.
Predictor mainly comprises the data of time and two aspects, space: temporal data refer to flow and the history average in several time intervals in past in i section; Data on space refer to the flow in the current of the up and down section adjacent with i section and each moment in past.
Native system is introduced the algorithm of RBF neural network.RBF neural network full name radial basis function (Radial Basis Function RBF) neural network is 3 layers of feedforward network with single hidden layer.RBF network analog locally in human brain adjust, mutually cover the acceptance domain neural network structure of (or claiming receptive field, Receptive Field), proved that RBF network can be competent at arbitrary accuracy and approach arbitrary continuation function.
To i bar section in road network, the flow value of first 15 minutes each k bar section entrances of front and back of system statistics, goes out the current due flow value in i bar section in conjunction with RBF Neural Network Prediction
and be designated as A; Meanwhile, calculate the magnitude of traffic flow B of the traffic flow last generation in section generating by cellular Automation Model; Compare A and B.
Get X=A-B.X has characterized the gap of the magnitude of traffic flow algorithm that the magnitude of traffic flow that cellular Automation Model simulates and neural network algorithm study draws.
In system, simulated automotive travels and goes out to 100 meters, end, section, calculates X, and adjusts acceleration distance and deceleration distance parameter in cellular Automation Model.To arbitrary car, establishing current acceleration distance is that deceleration distance is, after adjustment
For scale-up factor, and be greater than 0, according to X with
,
unit adjusts.After adjusting, simulating vehicle travels by new parameter.
It is right that value depends in system
,
, unit selection.The arbitrary unit of above-mentioned parameter is changed, can in system initialization process, select in advance K bar section, and order equals 1, then the above-mentioned adjustment formula of substitution is verified.The magnitude of traffic flow after adjusting if find is greater than RBF neural network prediction value,
otherwise,,
.After repeatedly adjusting, make both differences minimum, so can be worth.
Claims (6)
1. a traffic zone dynamic map monitoring and controlling forecast system, is characterized in that: comprise sensing module, data management module, map generation module, map exhibition module, vehicle simulation module, authenticated configuration module and traffic flow parameter prediction module;
Described sensing module is responsible for collection signal, information for detection of each porch, unit, section vehicle, via data management module analysis, input traffic flow parameter prediction module is processed, by map generation module, map, put on display module and vehicle simulation module composition dynamic map, reflection road network macroscopic view is to the multi-level situation of microcosmic;
The measuring-signal of described sensing module derives from the monitoring camera on highway, the section that native system be take in the middle of every two monitoring cameras is elementary cell, detect the essential information of each elementary cell porch vehicle, utilize traffic flow parameter prediction module prediction vehicle in the mode of travelling of inside, unit and carry out three-dimensional simulation demonstration, forming virtual region dynamic map;
Described map generation module carries out the maintenance of dynamic map platform, carries out three-dimensional simulation, projection conversion, map registration, the characterization process of blocking up, and can from macroscopic view to microcosmic, show at many levels road network operation conditions;
Described map is put on display module has a color ribbon to characterize the degree of blocking up of corresponding road section at the other mark of road automatically;
Described vehicle simulation module is carried out the virtual demonstration of Vehicle Driving Cycle, carrying out three-dimensional modeling, speed of a motor vehicle adjusting, car look registration, the vehicle classification of vehicle processes, the rule that the described speed of a motor vehicle regulates is according to specific traffic flow model, with less detection Data Representation road network operation conditions, and guarantee smoothly travelling of section, unit intersection model vehicle, described vehicle is detected and is obtained by monitor video, can be divided into middle minibus, motor bus, jubilee wagon, medium truck, high capacity waggon, super-huge lorry, seven kinds of vehicles of container car, corresponding seven kinds of three-dimensional models;
Described authenticated configuration module, by authenticated configuration module, obtain authority, retrieve for examination the local trackside video pictures that fixedly monitoring camera is taken, section follow shot is being specified in commander mobile video source, watches fixing monitoring to be difficult to the on-the-spot details or the road blind area that catch.
2. traffic zone dynamic map monitoring and controlling forecast system according to claim 1, is characterized in that: the mode that described traffic flow parameter prediction module adopts cellular Automation Model to combine with RBF neural network short-term forecasting traffic flow model;
Described cellular Automation Model is all done careful calculating to the rule of travelling of each car, is that system each car in segment unit that satisfies the need is done the theoretical foundation that three-dimensional simulation shows; In the traffic flow model of cellular automaton, on track, the time of Vehicle Driving Cycle, space and speed are by discretize, and in t → t+1 process, model is pressed following four regular parallel evolutionaries:
1) accelerate rule: in the larger distance of the place ahead, there is no car, V
n→ min (V
n+1, V
max), corresponding to the characteristic that in reality, driver's expectation is travelled with maximal rate;
2) rule of slowing down: have car, V in the nearer distance in the place ahead
n→ min (V
n, D
n), driver takes the measure of slowing down for fear of bumping with front truck;
3) random slowing down: with Probability p, V
n→ V
n-1, represent various uncertain factors;
4) motion: X
n=X
n+ V
n, vehicle according to the speed after adjusting to overtake;
Here, Xn represents the position of n car; Vn represents the speed of n car, in this speed characterization system the every frame of dynamic map refresh after the distance in this car early stage; Dn represents the distance between n car and front truck n+1;
System is processed each car on each frame of dynamic map, reads information of vehicles, according to aforementioned four rules, calculates car in position, speed, the acceleration condition of next frame, and in next frame, carries out corresponding demonstration;
The numerical simulation that completes model also must be determined boundary condition, and the present invention adopts improved open boundary condition:
In road exit, a truck position exceeds road scope and assert that it rolls section away from, no longer shows, the second car of closelying follow thereafter becomes a new car, in road porch, vehicle was located before front truck produces position with certain probability to produce, folded with two car weights before and after preventing;
Described RBF neural network short-term forecasting traffic flow model is by the mode of sample training, the segment unit magnitude of traffic flow of satisfying the need has more in advance and prediction accurately, system arranges an entrance in each unit, utilize awareness apparatus to detect the situation that enters vehicle, system is M rice before each unit, section, vehicle travels according to original cellular Automation Model rule, at rear Y rice, system detects the magnitude of traffic flow A in dummy model, utilize RBF neural network short-term forecasting traffic flow model simultaneously, read the magnitude of traffic flow in front several sections, dope the now due magnitude of traffic flow B in this end, unit, section, A is compared with B, then adjust the parameter in cellular Automation Model, comprise acceleration distance, deceleration distance, the random chance of random slowing down, make vehicle flowrate in dummy model in the end tend to gradually the requirement of magnitude of traffic flow B in Y rice, the vehicle that reaches section intersection shows effect level and smooth and that traffic is constantly revised.
3. traffic zone dynamic map monitoring and controlling forecast system according to claim 1, it is characterized in that: described color ribbon is comprised of red, yellow, green, blue gradual change aberration, get 7 kinds, conversion step, the degree of blocking up uses the average velocity of corresponding road section vehicle as criterion.
4. traffic zone dynamic map monitoring and controlling forecast system according to claim 2, is characterized in that: described M is that 900, Y is 100.
5. traffic zone dynamic map monitoring and controlling forecast system according to claim 1, it is characterized in that: the measuring-signal of described sensing module also derives from magnetic frequency, ripple frequency, piezoelectric type sensor, can the perception speed of a motor vehicle, vehicle number, vision-based detection sensing can obtain vehicle, car look information in addition.
6. traffic zone dynamic map monitoring and controlling forecast system according to claim 5, it is characterized in that: also comprise and estimate induction module, induction module is estimated in utilization, collecting on the basis of current Real-time Road information, carry out region selection, the judgement of blocking up, time Predict analysis, finally draw and arrive the shortest route induction information of destination required time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210309274.XA CN102819954B (en) | 2012-08-28 | 2012-08-28 | Traffic region dynamic map monitoring and predicating system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201210309274.XA CN102819954B (en) | 2012-08-28 | 2012-08-28 | Traffic region dynamic map monitoring and predicating system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102819954A CN102819954A (en) | 2012-12-12 |
CN102819954B true CN102819954B (en) | 2014-11-12 |
Family
ID=47304049
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201210309274.XA Expired - Fee Related CN102819954B (en) | 2012-08-28 | 2012-08-28 | Traffic region dynamic map monitoring and predicating system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102819954B (en) |
Families Citing this family (38)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103095815B (en) * | 2012-12-31 | 2015-12-09 | 普天新能源有限责任公司 | Positioning of mobile equipment method and apparatus |
CN103699761B (en) * | 2014-01-13 | 2015-08-05 | 中国人民解放军国防科学技术大学 | Based on general artificial social product battle state display method and the device of data stream |
CN103839410B (en) * | 2014-03-10 | 2016-05-11 | 北京掌城科技有限公司 | Transport information conversion method between a kind of different grain size map |
CN104301673B (en) * | 2014-09-28 | 2017-09-05 | 北京正安维视科技股份有限公司 | A kind of real-time traffic analysis and panorama visual method based on video analysis |
CN104317583A (en) * | 2014-10-21 | 2015-01-28 | 合肥星服信息科技有限责任公司 | Road congestion optimization algorithm based on grid theory |
US9824580B2 (en) * | 2015-12-17 | 2017-11-21 | International Business Machines Corporation | Method, computer readable storage medium and system for producing an uncertainty-based traffic congestion index |
CN106017496A (en) * | 2016-05-25 | 2016-10-12 | 南京信息职业技术学院 | Real-time navigation method based on road condition |
CN105825677B (en) * | 2016-05-31 | 2018-10-09 | 武汉大学 | A kind of urban traffic blocking prediction technique based on improvement BML models |
CN106205125A (en) * | 2016-07-27 | 2016-12-07 | 安徽聚润互联信息技术有限公司 | A kind of ambulance arrival time real-time estimate system and method |
CN106781464A (en) * | 2016-12-06 | 2017-05-31 | 广州市科恩电脑有限公司 | A kind of congestion in road situation method of testing |
CN108279018A (en) * | 2017-01-06 | 2018-07-13 | 北京四维图新科技股份有限公司 | A kind of traffic information display method, device and hybrid navigation equipment |
CN106846812B (en) * | 2017-03-14 | 2019-01-25 | 北京航空航天大学 | A kind of magnitude of traffic flow appraisal procedure of crossroad |
US10481044B2 (en) * | 2017-05-18 | 2019-11-19 | TuSimple | Perception simulation for improved autonomous vehicle control |
US10823574B2 (en) * | 2017-06-01 | 2020-11-03 | Panasonic Intellectual Property Corporation Of America | Communication method, roadside unit, and communication system |
CN108648480B (en) * | 2018-05-20 | 2022-09-02 | 中佳盛建设股份有限公司 | Road isolation model display method based on mixed reality and storage medium |
CN108831168B (en) * | 2018-06-01 | 2020-10-02 | 武汉泰坦智慧科技有限公司 | Traffic signal lamp control method and system based on visual identification of associated intersection |
CN108961767B (en) * | 2018-07-24 | 2021-01-26 | 河北德冠隆电子科技有限公司 | Highway inspection chases fee alarm system based on four-dimensional outdoor traffic simulation |
CN108919256B (en) * | 2018-07-24 | 2021-03-26 | 河北德冠隆电子科技有限公司 | Four-dimensional real-scene traffic simulation vehicle overspeed whole-course tracking detection alarm system and method |
CN109255953B (en) * | 2018-09-19 | 2020-11-03 | 江苏本能科技有限公司 | Region-based movable vehicle distribution display method and system |
CN109767028A (en) * | 2018-12-07 | 2019-05-17 | 东南大学 | Recognition methods, Traffic volume forecasting method and the system of road virtual blind area monitoring point |
CN109637135A (en) * | 2018-12-29 | 2019-04-16 | 江苏工程职业技术学院 | A kind of circumstance video monitoring early warning system based on computer network |
CN109886126B (en) * | 2019-01-23 | 2020-10-23 | 长安大学 | Regional vehicle density estimation method based on dynamic sampling mechanism and RBF neural network |
CN109830102A (en) * | 2019-02-14 | 2019-05-31 | 重庆邮电大学 | A kind of short-term traffic flow forecast method towards complicated urban traffic network |
CN109872013A (en) * | 2019-03-26 | 2019-06-11 | 重庆同济同枥信息技术有限公司 | Freeway traffic time map construction method, apparatus and system |
WO2020210960A1 (en) * | 2019-04-15 | 2020-10-22 | 华为技术有限公司 | Method and system for reconstructing digital panorama of traffic route |
CN110044451A (en) * | 2019-04-30 | 2019-07-23 | 湖南行必达网联科技有限公司 | Weighing method, device and the electronic equipment of vehicle load |
JP7298285B2 (en) * | 2019-05-10 | 2023-06-27 | トヨタ自動車株式会社 | Information processing device and information processing program |
EP4273836A3 (en) * | 2019-09-17 | 2024-01-17 | Mobileye Vision Technologies Ltd. | Systems and methods for predicting blind spot incursions |
CN110930698B (en) * | 2019-11-18 | 2021-05-04 | 北京交通大学 | Front road transparent calculation method under 5G Internet of vehicles environment |
CN111179598B (en) * | 2020-01-18 | 2020-12-29 | 杭州后博科技有限公司 | Vehicle running speed calculation method and system based on intelligent lamp pole |
CN113382384B (en) * | 2021-06-13 | 2022-08-02 | 西北工业大学 | Vehicle local networking oriented road region dividing method based on multi-source sensing technology |
CN113724504B (en) * | 2021-08-06 | 2023-04-07 | 之江实验室 | Urban area traffic prediction system and method oriented to vehicle track big data |
CN113870553B (en) * | 2021-08-20 | 2023-08-29 | 西安电子科技大学 | Road network running state detection system and method for mixed traffic flow |
CN113905215B (en) * | 2021-12-07 | 2022-03-01 | 江西佳铭特实业有限公司 | Bus safe driving monitoring system |
CN114387555B (en) * | 2022-01-19 | 2024-09-06 | 广东技术师范大学 | Cellular automaton model correction method and device based on unmanned aerial vehicle video image |
CN116824870B (en) * | 2023-08-31 | 2023-11-24 | 国交空间信息技术(北京)有限公司 | Road segment flow prediction method, device, equipment and storage medium |
CN116913097B (en) * | 2023-09-14 | 2024-01-19 | 江西方兴科技股份有限公司 | Traffic state prediction method and system |
CN118313179B (en) * | 2024-06-12 | 2024-08-16 | 宁波朗达科技有限公司 | Vehicle load effect simulation method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000242884A (en) * | 1999-02-23 | 2000-09-08 | Toyota Central Res & Dev Lab Inc | Traffic flow simulation system |
CN1611917A (en) * | 2003-10-28 | 2005-05-04 | 日本先锋公司 | Device, system and method for reporting a traffic condition and program and recording medium |
EP1657693A2 (en) * | 2004-11-16 | 2006-05-17 | Microsoft Corporation | Traffic forecasting employing modeling and analysis of probabilistic interdependencies and contextual data |
CN101127158A (en) * | 2006-08-18 | 2008-02-20 | 株式会社查纳位资讯情报 | Predictive traffic information creating method, predictive traffic information creating apparatus, and traffic information display terminal |
CN101465058A (en) * | 2009-01-05 | 2009-06-24 | 天津大学 | Method for making decision of inductive unit of intelligent dynamic route inductive system |
KR20110124905A (en) * | 2010-05-12 | 2011-11-18 | 한국항공대학교산학협력단 | Data processing method for transportation and communication integrated simulation |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4783414B2 (en) * | 2008-09-12 | 2011-09-28 | 株式会社東芝 | Traffic situation prediction system |
-
2012
- 2012-08-28 CN CN201210309274.XA patent/CN102819954B/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2000242884A (en) * | 1999-02-23 | 2000-09-08 | Toyota Central Res & Dev Lab Inc | Traffic flow simulation system |
CN1611917A (en) * | 2003-10-28 | 2005-05-04 | 日本先锋公司 | Device, system and method for reporting a traffic condition and program and recording medium |
EP1657693A2 (en) * | 2004-11-16 | 2006-05-17 | Microsoft Corporation | Traffic forecasting employing modeling and analysis of probabilistic interdependencies and contextual data |
CN101127158A (en) * | 2006-08-18 | 2008-02-20 | 株式会社查纳位资讯情报 | Predictive traffic information creating method, predictive traffic information creating apparatus, and traffic information display terminal |
CN101465058A (en) * | 2009-01-05 | 2009-06-24 | 天津大学 | Method for making decision of inductive unit of intelligent dynamic route inductive system |
KR20110124905A (en) * | 2010-05-12 | 2011-11-18 | 한국항공대학교산학협력단 | Data processing method for transportation and communication integrated simulation |
Non-Patent Citations (4)
Title |
---|
基于WLAN路况视频监控系统设计;邢业平等;《计算机工程与设计》;20081216(第23期);第6130-6132页 * |
基于监控视频的运动车辆行为分析算法;李勃等;《仪器仪表学报》;20060615;第2118-2120页、第2142页 * |
李勃等.基于监控视频的运动车辆行为分析算法.《仪器仪表学报》.2006, * |
邢业平等.基于WLAN路况视频监控系统设计.《计算机工程与设计》.2008,(第23期), * |
Also Published As
Publication number | Publication date |
---|---|
CN102819954A (en) | 2012-12-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102819954B (en) | Traffic region dynamic map monitoring and predicating system | |
Treiber et al. | Traffic flow dynamics | |
Yang et al. | Eco-driving system for connected automated vehicles: Multi-objective trajectory optimization | |
US20200406925A1 (en) | Comfort-based self-driving planning method | |
CN104778834B (en) | Urban road traffic jam judging method based on vehicle GPS data | |
CN109859500A (en) | A kind of high speed merging area safe early warning method based on bus or train route collaboration | |
Ma et al. | Modeling bus travel time reliability with supply and demand data from automatic vehicle location and smart card systems | |
CN102136190A (en) | Dispatching management system and method for event emergency response of urban bus passenger transport | |
Anya et al. | Application of AIMSUN microsimulation model to estimate emissions on signalized arterial corridors | |
CN109389845A (en) | A kind of multifactor integration high speed highway dynamic vehicle speed managing and control system | |
Yang et al. | A novel approach for fine-grained traffic risk characterization and evaluation of urban road intersections | |
Shankaran et al. | Intelligent Transport Systems and Traffic Management | |
Ahn et al. | Traffic flow theory and characteristics | |
Rakha et al. | Traffic signal control enhancements under vehicle infrastructure integration systems. | |
Oliskevych et al. | Optimization of vehicle speed forecasting horizont on the intercity highway | |
Scora et al. | Real-time roadway emissions estimation using visual traffic measurements | |
Hadi et al. | Framework for multi-resolution analyses of advanced traffic management strategies. | |
Coelho et al. | A numerical tool for estimating pollutant emissions and vehicles performance in traffic interruptions on urban corridors | |
CN115226064A (en) | 5G and CV 2X-based vehicle road cloud collaborative intelligent bus networking platform | |
Avrenli et al. | Four-regime speed–flow relationships for work zones with police patrol and automated speed enforcement | |
Islam | Intergrading Connected Vehicle Data into the Transportation Performance Measurement Process | |
Khan | Connected and automated vehicles in urban transportation cyber-physical systems | |
Tyagi et al. | Intelligent transportation system: Need, working, and tools | |
Davis et al. | Moving array traffic probes | |
Ren et al. | Traffic flow characteristics and traffic conflict analysis in the downstream area of expressway toll station based on vehicle trajectory data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141112 Termination date: 20210828 |
|
CF01 | Termination of patent right due to non-payment of annual fee |