JP2008052671A - Traffic jam prediction device - Google Patents

Abstract

[PROBLEMS] To provide a traffic condition prediction device for predicting a traffic condition in an area where the traffic volume is unknown.
A traffic jam condition prediction device V that predicts a traffic jam situation on a road that connects two areas and whose traffic volume is unknown is a traffic jam situation in the two areas and a correlation between the two areas regarding the traffic jam situation. The traffic condition predicting means 12 for predicting the traffic condition of a road whose traffic volume is unknown is provided. Furthermore, the traffic condition prediction means 12 predicts the traffic condition of a road whose traffic volume is unknown based on the correlation between the roads related to the traffic condition. Note that the correlation between the two areas is calculated for each direction and time zone based on the traffic jam situation in the two areas, and the correlation between the roads is set based on the road attributes.
[Selection] Figure 1

Description

  The present invention relates to a congestion situation prediction apparatus, and more particularly to a congestion situation prediction apparatus that predicts a congestion situation in an area including a road whose traffic volume is unknown.

  2. Description of the Related Art Conventionally, a traffic volume prediction device that predicts a change in traffic volume in a wide area road network, and affects a change in traffic volume between points using past traffic volume data in the wide area road network. A range division function unit that divides a range in the network using range division reference data, a traffic data storage memory that accumulates a daily pattern of traffic for the range divided by the range division function unit, and a traffic The observed traffic data memory that stores the traffic data on the forecasted volume date, the traffic pattern in the traffic data storage memory is searched using the data in the observed traffic data memory, and similar traffic patterns are extracted, and A traffic volume prediction function unit that predicts a traffic volume after a predetermined time from the prediction time using the extracted traffic volume pattern; Measuring device is known (e.g., see Patent Document 1.).

This traffic volume forecasting device specifies the traffic volume pattern to be used for prediction by pattern comparison between the traffic volume accumulation pattern and the current day observation pattern, so that it can be used in a short period of time as in the case of using the average value of past traffic volume data. The prediction accuracy can be improved in the prediction of the traffic volume after a relatively short period from the prediction time without burying the fluctuation of the traffic volume.
JP 2002-298281 A

  However, since the traffic volume predicting device described in Patent Document 1 predicts the traffic volume after a predetermined time based on the current traffic volume detected using the vehicle detection sensor, the vehicle detection sensor is not installed. It is not possible to accurately predict the traffic situation on a road where the current traffic volume is unknown.

  In view of the above-described problems, an object of the present invention is to provide a traffic jam situation prediction apparatus that predicts a jam situation in an area including a road whose traffic volume is unknown.

  In order to achieve the above-described object, a traffic jam situation prediction apparatus according to a first aspect of the present invention is a traffic jam situation prediction apparatus for forecasting a traffic jam situation of a road that connects two areas and whose traffic volume is unknown. And a traffic jam situation prediction means for forecasting a traffic jam situation of a road whose traffic volume is unknown based on a traffic jam situation in the two areas and a correlation between the two areas related to the traffic jam situation.

  Further, the second invention is a traffic jam situation prediction apparatus according to the first invention, wherein the traffic jam situation prediction means further includes a traffic jam of a road whose traffic volume is unknown based on a correlation between roads concerning the jam situation. It is characterized by predicting the situation.

  Further, the third invention is a traffic jam condition prediction apparatus according to the first or second invention, wherein the correlation between the two areas is based on the traffic jam situation in the two areas, according to direction, time zone. It is characterized by being set separately.

  According to a fourth aspect of the present invention, there is provided a traffic jam situation prediction apparatus according to the second aspect of the present invention, wherein the correlation between the roads is set based on road attributes.

  With the above-described means, the present invention can provide a traffic jam situation prediction apparatus that predicts a traffic jam situation in an area including a road whose traffic volume is unknown.

  Hereinafter, the best mode for carrying out the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram illustrating a configuration example of a traffic jam situation prediction system according to the present invention. The traffic jam situation prediction system S includes a traffic jam situation prediction device V and a communication center C. The traffic jam situation prediction device V includes a control device 1, a communication device 2, a storage device 3, and a display device 4. The control device 1 is based on information from the communication device 2 and the storage device 3 on a road whose traffic jam status is unknown. The traffic jam situation is predicted, and then the traffic jam situation of the entire area including the road where the traffic jam situation is unknown is predicted and the prediction result is output to the display device 4.

  The communication center C is a device that receives a signal from a vehicle detection sensor installed on a road, generates information related to a traffic jam status, and transmits the generated information to the traffic jam status prediction device V. An installed computer server or the like.

  The “congestion situation” is a congestion situation at a predetermined point, and is indicated by, for example, the congestion length of a predetermined section at a predetermined time, and the congestion situation of the entire predetermined area is indicated as a cumulative total of the congestion lengths.

  FIG. 2 is a diagram illustrating an example of a traffic jam situation in a predetermined area, and FIG. 2A shows a traffic jam situation by arranging circles having different sizes according to the traffic jam length in an area including areas A to D. In the figure, the largest circle indicates a traffic jam length of 5 km, the next largest circle indicates a traffic jam length of 2.5 km, and the smallest circle indicates a traffic jam length of 0.5 km. FIG. 2B is a graph in which the cumulative traffic jam length of area A is plotted on the vertical axis and the time is plotted on the horizontal axis, and the transition of the traffic jam situation for a predetermined time before the current time is shown. FIG. It is the graph extracted from the apparatus 3, Comprising: It is a graph which has a transition pattern most approximated to the transition pattern of FIG.2 (b). 2 is created based on information transmitted from the communication center C.

  Areas A to D are predetermined areas in which the traffic volume can be measured by the vehicle detection sensor. For example, the areas A to D indicate a city area in which the traffic volume can be measured by the vehicle detection sensor. Roads between urban areas that connect Routes 1 and 2 are roads where vehicle detection sensors are not installed and traffic volume cannot be measured, and routes 3 are roads where vehicle detection sensors are installed and traffic volume can be measured.

  In FIG. 1, a control device 1 is a computer including a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), etc., and includes an area information acquisition unit 10, a road information acquisition unit 11, and a traffic jam. Programs corresponding to the situation prediction means 12 are stored in the ROM, the programs are expanded on the RAM, and the corresponding processing is executed by the CPU.

  Further, the control device 1 may store information acquired via the communication device 2 in the storage device 3. This is because information transmitted from the communication center C is sequentially stored in the storage device 3 and accumulated as past data.

  The communication device 2 is a device for controlling the communication between the traffic jam situation prediction device V and the communication center C. For example, the communication center C uses the mobile phone frequency or the specific low-power wireless communication frequency. Receive information to send.

  The communication device 2 also receives VICS (Vehicle Information and Communication System) information transmitted from optical beacons, radio wave beacons, and FM stations, or transmits it from a communication device for road-to-vehicle communication embedded in a traveling road. Information received.

  The storage device 3 is a device for storing data necessary for predicting a traffic jam situation. For example, the storage device 3 is a non-volatile storage medium such as a hard disk or a semiconductor memory, and includes an inter-area correlation database 30 and a correlation between roads. A relational database 31 is stored.

  The inter-area correlation database 30 is a systematic configuration of the correlation related to the traffic situation between the two areas. For example, the traffic situation in the Tokyo area calculated based on past information (for example, cumulative traffic jam) The correlation coefficient between the traffic jam situation in the Yokohama area and the Yokohama area is maintained for each direction (for example, four directions of east, west, south, and north) and for each time zone (for example, every 30 minutes). The closer to 1, the stronger the positive correlation is, and one shows that if one is congested, the other is also congested. On the other hand, the closer to -1, the stronger the negative correlation is. It shows that. Further, the correlation coefficient indicates that the closer to 0 (zero), the weaker the correlation, and one traffic jam does not affect the other.

  The inter-area correlation coefficient is based on sample data (referred to as data used to calculate the correlation coefficient) based on the day of the week, holidays, season, time zone, weather, etc. (hereinafter referred to as “day of the week”). For example, if the current day is the first Monday in July, the traffic congestion status (cumulative traffic jam length) of the past Monday (excluding Tuesday through Sunday) in the two areas It may be calculated based on the past, or may be calculated based on the traffic condition (cumulative traffic jam length) of the past July weekday (excluding Saturday and Sunday). This is because there may be regularity in the traffic situation classified as a day of the week.

  Further, a correlation coefficient between the traffic jam situation at a predetermined time in the Tokyo area and the traffic jam situation in the Yokohama area a predetermined time before or after the predetermined time may be held. This is because there may be a time lag between the traffic conditions in the two areas with relevance.

  As a result, the inter-area correlation database 30 indicates, for example, that “the traffic congestion in the downward direction in the Tokyo area from 10:00 am to 11:00 am on Monday in July is the traffic congestion in the downward direction in the Yokohama area 30 minutes later. It is possible to provide information relating to the correlation of the traffic jam situation between areas, such as “having a positive correlation between”.

  Note that the correlation between areas is based on the population, number of cars, traffic capacity, etc. for each area, regardless of past data. For example, “the cumulative traffic jam length in area B is 70 times the cumulative traffic jam length in area A”. % ”May be determined in advance. The “traffic capacity” refers to the maximum number of cars that pass through a cross section of a target road within a unit time.

  The correlation database 31 between roads is a systematic configuration of the correlation regarding the traffic jam situation between two roads. For example, the route 1 (for example, city) connecting the area A and the area B in FIG. Correlation between traffic conditions on sub-routes such as roads and traffic conditions on route 3 (for example, main lines such as expressways) by direction (up or down direction), by time zone (for example, 30 minutes) (Unit) in the form of the coefficient between roads (for example, the ratio to the traffic congestion length of the reference road).

  The coefficient between roads is determined based on road attributes (referring to road type, road width, number of lanes, number of intersections, presence / absence of sidewalks, degree of parallelism between roads, etc.) regardless of past data. However, if both of the target routes have data related to past traffic conditions, similar to the inter-area correlation coefficient, the sample data may be calculated as a correlation coefficient after classifying it into days of the week, etc. Good.

  Thereby, the correlation database 31 between roads can provide the information regarding the correlation of the congestion condition between roads.

  The display device 4 is an in-vehicle device for displaying information. For example, the display device 4 is a liquid crystal display used in a car navigation system, and displays a prediction result by the traffic jam situation prediction device V. The traffic jam situation prediction device V may output the prediction result by voice using a voice output device.

  Next, various units included in the control device 1 will be described.

  The area information acquisition unit 10 is a unit for acquiring information on an area in which traffic volume can be measured. The area information is a predetermined area in which traffic volume can be measured (for example, a rectangular area having a length of 5 km and a width of 5 km). Information including current and past traffic conditions in a predetermined area (for example, the traffic location included in the VICS information, the vertical direction, the traffic length, etc.), the inter-area correlation coefficient, and the like. . Areas A to D in FIG. 2A correspond to predetermined areas.

  The area information acquisition means 10 receives information transmitted from the communication center C via the communication device 2 or the inter-area phase relationship of the target area group from the inter-area correlation database 30 stored in the storage device 3. The area information is acquired by extracting the number.

  The road information acquisition means 11 is means for acquiring road information. The road information is information related to a road connecting two adjacent or neighboring areas, and is obtained through the communication device 2. It includes information such as past traffic conditions, current traffic conditions calculated based on traffic demand and traffic capacity, and coefficient between roads. “Traffic demand” refers to the number of cars that attempt to pass a predetermined section or a predetermined section on the road per unit time, and the traffic demand on the road connecting the two areas is based on the traffic congestion in the two areas. It may be derived, or may be preset for each time zone based on the population for each area, the number of cars owned, and the like.

  The road information acquisition unit 11 receives information transmitted from the communication center C via the communication device 2, or roads connecting the target area groups from the inter-road correlation database 31 stored in the storage device 3. The road information is acquired by extracting an interval coefficient or the like.

  The traffic condition prediction means 12 is a means for predicting the traffic condition of an area including a plurality of areas. For example, a road (Fig. 1) that can measure traffic volume among a group of roads connecting two adjacent or neighboring areas. Based on road information of route 3) of 2 (a) and area information of two areas, the traffic situation of roads with unknown traffic volume (route 1 and route 2 of FIG. 2 (a)) is predicted. Then, the traffic congestion situation of the entire area including the two areas and the road group connecting the areas is predicted.

  Here, with reference to FIG. 3, the flow of processing in which the traffic jam condition predicting means 12 predicts the traffic jam status (total traffic jam length) in a predetermined area will be described.

  First, the traffic jam condition predicting means 12 is most similar to the cumulative traffic jam length transition pattern (see FIG. 2B) of each area received from the communication center C by the area information acquisition means 10 via the communication device 2. A past accumulated traffic length transition pattern (see FIG. 2C) of each area to be extracted is extracted from the storage device 3 by a pattern matching process or the like (step S1), and a predetermined time of each area based on the extracted transition pattern The subsequent cumulative traffic jam length (for example, 6 stages from 30 minutes to 3 hours later) is predicted (step S2).

  For example, the traffic jam condition predicting means 12 acquires a transition pattern of the cumulative traffic jam length in the area A retroactively for a predetermined time from the current time (see FIG. 2B), and a past transition pattern that most closely approximates the acquired transition pattern. Is extracted from the storage device 3 by pattern matching processing, and the cumulative congestion length transition pattern after the time corresponding to the current time of the extracted transition pattern (see FIG. 2C) (thick line portion in FIG. 2C) Based on the above, the accumulated traffic jam length after a predetermined time has elapsed is predicted. The traffic jam condition predicting means 12 similarly predicts the cumulative traffic jam length for the other areas B to D included in the predetermined area.

  After that, the traffic jam condition prediction means 12 determines whether the area A and the area B have a correlation coefficient that satisfies a predetermined condition (for example, 0.8 or more or −0.8 or less) from the inter-area correlation database 30 of the storage device 3. A correlation between them is extracted (step S3).

  For example, the traffic jam condition predicting means 12 extracts a correlation such as “the cumulative traffic jam length at 5 pm in area A has a positive correlation with the cumulative traffic jam length at 5:30 pm in area B”. Similarly, the traffic jam situation prediction means 12 also extracts correlations between area A and area C, area B and area C, and area B and area D.

  Thereafter, the traffic jam condition predicting means 12 corrects the cumulative traffic jam length of each area predicted in step S2 based on the correlation coefficient extracted from the inter-area correlation database 30 (step S4).

  For example, the traffic situation prediction unit 12 determines that the relationship between the cumulative traffic length of area A after 1 hour and the cumulative traffic length of area B after 1 hour is predicted from the inter-area correlation database 30 as predicted in step S2. When the extracted correlation (for example, the relationship in which the cumulative congestion length in area A is 80% of the cumulative congestion length in area B) is not met (for example, the cumulative congestion length in area A is 95% of the cumulative congestion length in area B) If the difference of 10% or more occurs), the estimated cumulative traffic jam length is corrected (for example, the area so that the cumulative traffic jam length of area A is 90% of the cumulative traffic jam length of area B) Reduce 5% of A's cumulative traffic jam length.)

  Similarly, when it is predicted in step S2 that the cumulative traffic jam length in area A after one hour becomes 65% of the cumulative traffic jam length in area B after one hour, The cumulative traffic jam length is increased by 5% so that the cumulative traffic jam length of area B becomes 70% (the difference from the relationship indicated by the inter-area correlation database 30 is 10%).

  Thereafter, the traffic jam condition prediction means 12 predicts the cumulative traffic jam length of the route connecting the areas after a predetermined time based on the cumulative traffic jam length prediction of each area after the predetermined time corrected in step S4 (step S5).

  For example, the traffic jam condition predicting means 12 is based on the total traffic jam length prediction in the up direction in the area A (from the area B to the area A) and the total traffic jam length prediction in the up direction in the area B. The cumulative traffic jam length in the upward direction on the route to B is predicted. In this case, the traffic jam condition predicting unit 12 treats a plurality of routes as one group, and predicts the cumulative traffic jam length of the grouped routes from the relationship between the traffic capacity and the traffic demand.

  Further, the traffic jam condition prediction means 12 uses the transition pattern of the cumulative traffic jam length of the route 3 in which the traffic volume can be measured in the route group between the area A and the area B, and is the most approximate as in step S1. The transition pattern to be extracted may be extracted from past information stored in the storage device 3 to predict the cumulative traffic jam length of the route 3 after a predetermined time has elapsed from the present time. In this case, a value obtained by subtracting the cumulative traffic jam length of route 3 from the cumulative traffic jam length of the route group collected together is the sum of the cumulative traffic jam lengths of route 1 and route 2.

  Thereafter, the traffic jam condition predicting means 12 extracts an inter-road coefficient from the inter-road correlation database 31 (Step S6), and corrects the cumulative traffic jam length of each route predicted in Step S5 based on the extracted inter-road coefficient ( Step S7).

  For example, for the routes 1 to 3 that connect the area A and the area B, the traffic jam condition prediction means 12 has a route 1 cumulative traffic length of 80% of the route 3 cumulative traffic jam length, and the route 2 cumulative traffic jam A road-to-road coefficient indicating that the length is 70% of the total traffic jam length of route 3 is extracted, and the cumulative traffic jam length (for example, 5 km) predicted in step S5 is calculated according to the road-to-road coefficient. .6 km), route 2 (cumulative traffic jam length of 1.4 km) and route 3 (cumulative traffic jam length of 2.0 km).

  Further, the traffic condition prediction means 12 assumes that the influence degree of the traffic condition of the area A decreases as the distance from the area A decreases, and divides the route connecting the area A and the area B into a plurality of sections. The cumulative traffic jam length may be predicted. For example, in the case of the above-described example, the cumulative traffic jam length may be set to 1.6 km in the section close to the area A of the route 1, and the prediction that no traffic jam may occur in other sections may be performed.

  When the cumulative traffic jam length of route 3 is predicted based on past data, it is predicted when the deviation from the correlation extracted from the inter-road correlation database 31 is large, similar to the correction in step S4. The accumulated congestion length of route 3 is corrected so as to be within an allowable range (for example, the deviation width is within ± 10%).

  With the above configuration, the traffic jam condition prediction device V adds the current and past traffic jam conditions in a predetermined area group in which the traffic volume can be measured, as well as the correlation and each area between the statistically calculated traffic jam situations. Can be predicted with high accuracy after a predetermined time has elapsed from the current time on a road where the traffic volume cannot be measured among the roads connecting the areas.

  In addition, the traffic situation prediction device V predicts with high accuracy the traffic situation of roads where traffic volume cannot be measured, so that it can accurately predict the traffic situation after a predetermined time in an area including roads where traffic volume cannot be measured. it can.

  In addition, since the traffic condition prediction device V predicts the traffic condition of a road where traffic volume cannot be measured with high accuracy, a new route is opened and the traffic volume of the existing route changes and past data cannot be used. Even in this case, it is possible to predict the traffic situation with high accuracy.

  In addition, the traffic condition prediction device V predicts the traffic condition of a road where the traffic volume cannot be measured with high accuracy, so that the total traffic jam length of the road between areas including the road where the traffic volume cannot be measured or each The accumulated traffic jam length on the road can be output in various modes such as comparison with the previous day and comparison with several hours ago.

  The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above-described embodiments, and various modifications and substitutions can be made to the above-described embodiments without departing from the scope of the present invention. Can be added.

  For example, in the above-described embodiment, the area information acquisition unit 10, the road information acquisition unit 11, the traffic jam situation prediction unit 12, the inter-area correlation database 30 and the inter-road correlation database 31 are all provided on the in-vehicle device side. A part or all of them may be provided on the communication center C side.

It is a figure which shows the structural example of a traffic condition prediction system. It is a figure which shows the example of the traffic congestion condition in a predetermined area | region. It is a flowchart which shows the flow of the process which a traffic condition prediction means estimates a traffic condition.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Control apparatus 2 Communication apparatus 3 Storage apparatus 4 Display apparatus 10 Area information acquisition means 11 Road information acquisition means 12 Congestion condition prediction means 30 Inter-area correlation database 31 Inter-road correlation database C Communication center S Congestion condition prediction system V Congestion condition Prediction device

Claims (4)

A traffic situation prediction device that predicts the traffic situation of a road that connects two areas and whose traffic volume is unknown.
A traffic condition prediction means for predicting the traffic condition of a road with unknown traffic volume based on the traffic condition in the two areas and the correlation between the two areas regarding the traffic condition;
A traffic situation prediction apparatus characterized by comprising: The traffic condition prediction means further predicts the traffic condition of a road whose traffic volume is unknown based on the correlation between the roads related to the traffic condition.
The traffic jam situation prediction apparatus according to claim 1, wherein: The correlation between the two areas is set for each direction and each time zone based on the traffic jam situation in the two areas.
The traffic jam situation prediction apparatus according to claim 1 or 2. The correlation between the roads is set based on road attributes.
The traffic jam situation prediction apparatus according to claim 2, wherein:

Images