JP3473478B2 - Inflow ramp control device and inflow ramp control method - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inflow lamp control device and an inflow lamp control method, and more particularly, in an expressway system in an urban area where traffic congestion increases year by year, in order to prevent traffic on the main line of the target expressway from congesting The present invention relates to an inflow lamp control device and an inflow lamp control method that minimize the inflow amount from each inflow lamp in consideration of future traffic demand.

[0002]

2. Description of the Related Art On-ramp (in-ramp) inflow control is performed in order to prevent the occurrence of traffic congestion on highways. LP (linear programming; linear programming) control is a typical method for determining the inflow control pattern.

As a document relating to LP control, for example, "Development of a dynamic inflow control model for an urban expressway network" (civil engineering planning research / collection No. 12; 19)
August 1995) and "Dynamic LP-type inflow control model of urban expressways using observation data" (Civil engineering research /
Proceedings No. 13; August 1996) and "quasi-dynamic LP control problem considering time fluctuation of inflow demand" (15th Traffic Engineering Research Presentation Report, November 1995).

[0004]

However, the conventional techniques have the following problems.

(1) The conventional technology does not sufficiently grasp the dynamic characteristics of traffic and the momentary change in traffic flow of the main road. In other words, dynamic changes in parameters and traffic data are not included in the model, and predictive control that considers future changes in traffic demand and main line traffic is not performed. Therefore, a sufficient dynamic model is not considered, and accurate control cannot be performed.

(2) Since the conventional technology is intended for measures to prevent congestion, it is premised that the main road is not congested, and there is no theoretical expandability when congestion occurs. In reality, congestion may occur, even if only partially, due to misprediction or miscontrol. In such a case, it is not realistic that the subsequent control cannot be performed at all.

(3) Real-time information such as traffic volume and speed of main roads, a part of travel time, the number of vehicles derived from these, or data of OD traffic volume that will be obtained in future will be obtained and predicted. There is no idea in the related art to learn and estimate parameters in the control model and control model, and there is a problem that the realistic operability of the model is not sufficiently taken into consideration.

(4) Congestion at the boundary between the target expressway and another road network, that is, the most downstream portion of the expressway main line, or the bolt-neck congestion on the general road of the outflow ramp extends to the expressway main line. There was no one that constructed a dynamic model of the lamp control in consideration of the influence of the case. In practice, this is considered to be the cause of congestion in many cases, and thus the conventional technology has not been sufficiently considered for practical use.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an inflow lamp control and an inflow lamp control method capable of appropriately controlling the inflow of a vehicle.

[0010]

According to one aspect of the present invention for achieving the above object, a road inflow ramp control device determines that the number of vehicles existing in each section of a target road flows into the section. Of the number of existing vehicles, the traffic volume of vehicles flowing out of the section, and the existing vehicle dynamics model represented by the number of existing vehicles before a certain time, the number of existing vehicles E or density K, and the traffic volume Q. E, showing the relationship
(K) -Q model, and an existing vehicle number predicting unit that predicts the number of existing vehicles up to a certain time point in each section from the forecast data of the inflow traffic demand from the outside for the target road. Based on the total inflow amount of inflow lamps and the critical existing vehicle number of each section, the existing vehicle number of each vehicle in the section up to a certain time point is determined to be the critical existing vehicle number by using the prediction result of the existing vehicle number prediction means. And a determining unit that determines the inflow amount of each of the lamps up to a certain time after the condition of not exceeding. The inflow lamp control device is
We formulate Dell by linear combination of inflow lamp inflow,
Determine the inflow of each lamp up to a certain time ahead by the planning method
It is characterized by

According to the present invention, the existing vehicle number is predicted based on the existing vehicle dynamic characteristic model, the E (K) -Q model, and the forecast data of the inflow traffic demand amount, and the inflow amount is determined based on the existing vehicle number. It As a result, it is possible to provide an inflow lamp control device capable of appropriately controlling the inflow of the vehicle. In addition, at a certain time by linear programming
It is possible to determine the inflow amount of each lamp so far.
It

[0012]

[0013]

Preferably, the E (K) -Q model is characterized by utilizing a linear characteristic when there is no traffic jam.

According to this, it becomes possible to appropriately control the inflow lamp when the traffic is not congested. According to another aspect of the present invention, the inflow lamp control device determines the number of vehicles existing for each section of the target road, the traffic volume of vehicles flowing into the section, the traffic volume of vehicles flowing out of the section, and For a target road, an E (K) -Q model showing the relationship between the number of existing vehicles dynamics model represented by the number of existing vehicles before a certain time, the number or density of vehicles, and the traffic volume Vehicle presence forecasting means for predicting the number of vehicles existing in each section up to a certain time, and forecasting the number of existing vehicles in each section from the forecast data of the inflow traffic demand from outside And a determination unit that determines the inflow amount of each lamp up to a certain time ahead so that the number of vehicles existing in each section up to a certain time ahead approaches the critical existing number using the prediction result of the means.

According to the present invention, the existing vehicle number is predicted based on the existing vehicle dynamic characteristic model, the E (K) -Q model, and the forecast data of the inflow traffic demand amount, and the inflow amount is determined. It is possible to provide the inflow lamp control capable of accurately controlling the inflow of the vehicle.

Preferably, the inflow lamp control device formulates a dynamic model of the number of existing lamps by a linear combination of inflow amounts of inflow lamps, and determines an inflow amount of each lamp up to a certain time point by a linear programming method or a quadratic programming method. It is characterized by doing.

According to this, it is possible to determine the inflow amount of each lamp up to a certain time point by the linear programming method or the quadratic programming method.

Preferably, the E (K) -Q model is characterized by utilizing a linear characteristic when there is no traffic jam and a linear characteristic when there is traffic jam.

According to this, it becomes possible to perform control using the linear characteristics when there is no traffic and when there is traffic.

Preferably, the number of existing vehicles is determined every time from traffic data on the main line and speed data or travel time data, which are separately obtained by a system.

According to this, it becomes easy to obtain the number of existing vehicles. Preferably, the inflow traffic demand prediction data is determined by adjusting the statistical value of the OD traffic volume by the traffic volume of each inflow ramp for each time obtained by a system or the OD traffic volume of the actual value of the day.

According to this, it becomes possible to obtain more accurate forecast data of traffic demand. Preferably, the forecast data of the inflow traffic demand is the statistical value of the OD demand traffic,
It is determined separately by adjusting the demanded traffic volume of each inflow lamp for each time obtained by the system or the OD demanded traffic volume of the actual value on the day.

According to this, it becomes possible to obtain more accurate forecast data of traffic demand. Preferably, the actual inflow control is performed based on only the inflow amount up to one time ahead of the inflow amount of each lamp obtained up to a certain time ahead, and feedback control is performed in which the control amount determination process is repeated every time. .

According to this, it becomes possible to execute a more accurate control amount determination process. Preferably, the actual inflow control uses a traffic light and adjusts its blue time so that the actual inflow amount matches the inflow amount determined by the determining means.

According to this, it becomes possible to perform the inflow control by the traffic light. Preferably, in the actual inflow control, a breaker is used, and the interruption time is adjusted so that the actual inflow amount is matched with the inflow amount determined by the determining means.

According to this, the inflow amount can be controlled by the breaker. Preferably, the actual inflow control is performed by opening and closing the inflow booth, or by reducing the number of booths when there are a plurality of booths, whereby the inflow amount is actually matched with the inflow amount determined by the determining means.

According to this, it becomes possible to control the inflow amount by the booth. Preferably, in the inflow lamp control device, when the inflow booth is an automatic ticket issuing machine, the actual inflow control is performed by the maximum number of tickets issued per unit time or the average ticket issuing interval.

According to this, the inflow control can be performed by the maximum number of tickets issued per unit time or the average ticket interval.

Preferably, in the inflow lamp control device, when the inflow booth is a manned toll collection point, the actual inflow control is performed by the maximum number of vehicles collected per unit time or the average fee collection interval.

According to this, the inflow control can be performed according to the maximum number of tolls collected per unit time or the average toll collection interval.

Preferably, in the actual inflow control, an on-ramp or a guide information plate near the on-ramp is used, and the information providing time is adjusted so that the actual inflow amount matches the inflow amount determined by the determining means. .

According to this, it becomes possible to control the inflow amount by the guide information board. Preferably, the actual inflow control is performed by information guidance by communication with a vehicle information terminal such as a navigation device, and the inflow amount determined by the determining means by adjusting the vehicle to be provided, the point to be provided, and the providing time. To match the actual inflow.

According to this, it becomes possible to control the inflow amount by a navigation device or the like.

According to another aspect of the present invention, the inflow lamp control device determines the number of vehicles existing for each section of the target road, the traffic volume of vehicles flowing into the section, and the traffic of vehicles flowing out of the section. E (K), which shows the relationship between the traffic volume model and the number-of-vehicles existing model represented by the number of vehicles and the number of vehicles existing before a certain time
An existing vehicle number predicting unit for predicting the number of vehicles existing in each section up to a certain time point based on the Q model and the forecast data of the inflow traffic demand from the outside for the target road; Based on the total inflow amount of the vehicle and the critical existing vehicle number of each section, using the prediction result of the existing vehicle number prediction means, it is said that the number of existing vehicles of each vehicle in the section up to a certain time ahead does not exceed the critical existing vehicle number. Using the condition, the first determining means for determining the inflow amount of each lamp until a certain time ahead and the road utilization efficiency are selected based on the critical existing vehicle number of each section, and the predicting means of the existing vehicle number is predicted. Using the result, a second determining means for determining the inflow amount of each lamp up to a certain time ahead such that the number of existing vehicles in each section up to a certain time ahead approaches the critical existing number, and the target road is provided. There is a section that is congested If or when there is a section in which road becoming crowded has immediately before the congestion using the second determination means, in other cases utilizing first determination means.

According to this, it is possible to provide an inflow lamp control device capable of performing control in consideration of traffic congestion.

According to still another aspect of the present invention, an inflow ramp control method is configured so that the number of vehicles existing in each section of a target road is the traffic volume of vehicles flowing into the section and the traffic of vehicles flowing out of the section. E, which shows the relationship between the number of existing vehicles and the number of existing vehicles before and after a certain time
(K) -Q model, and an existing vehicle number predicting step of predicting the number of existing vehicles up to a certain time point in each section from the forecast data of the inflowing traffic demand from the outside for the target road. , Based on the total inflow amount of inflow lamps and the critical existing vehicle number of each section, using the prediction result of the existing vehicle number prediction step, the existing vehicle number of each vehicle of the section up to a certain time ahead is determined as the critical existing vehicle number. And a determination step of determining the inflow amount of each of the lamps up to a predetermined time point, using the condition of not exceeding. Inflow lamp control method
The in-vehicle dynamic characteristic model is linearly combined with the inflow amount of the inflow lamp.
Formulated, and the flow of each lamp up to a certain time ahead by linear programming
It is characterized by determining the amount to be received.

According to this, the existence number dynamic characteristic model,
The number of existing vehicles is predicted from the E (K) -Q model and the prediction data of the inflow traffic demand, and the inflow amount is determined based on that. As a result, it is possible to provide an inflow lamp control method capable of accurately controlling the inflow of the vehicle. In addition, each ramp up to a certain time ahead by linear programming
It is possible to determine the inflow amount of.

According to still another aspect of the present invention, an inflow ramp control method determines the number of existing vehicles for each section of a target road, the traffic volume of vehicles entering the section, and the number of vehicles exiting the section. A dynamic model of the number of existing vehicles, which is represented by the traffic volume and the number of vehicles existing before a certain time, and the relationship between the number or density of vehicles and the traffic volume, E
(K) -Q model, and an existing vehicle number predicting step of predicting the number of existing vehicles up to a certain time point in each section from the forecast data of the inflowing traffic demand from the outside for the target road. , Based on the critical existing vehicle number of each section, using the prediction result of the existing vehicle number prediction step, so that the existing vehicle number of each section up to a certain time ahead approaches the critical existing vehicle number of each lamp until a certain time ahead And a determining step of determining the inflow amount.

According to the present invention, the existing vehicle number is predicted from the existing vehicle dynamic characteristic model, the E (K) -Q model, and the forecast data of the inflow traffic demand amount, and the inflow amount is determined based on that. As a result, it is possible to provide an inflow lamp control method capable of appropriately performing inflow control.

[0041]

BEST MODE FOR CARRYING OUT THE INVENTION First, an outline of a traffic flow control system (inflow lamp control device) in the present embodiment will be described.

(1) In the present embodiment, as a dynamic characteristic of a traffic flow, a Q (traffic volume) -E (number of existing vehicles) curve (where the number of vehicles existing per unit distance is Q (traffic volume) -K)
A dynamic traffic flow model using (equivalent to (density) curve) is used. This makes it possible to respond to the ever-changing traffic flow.

In addition, the accuracy of the control model is ensured, and the expandability for dealing with the case of congestion of the main line or congestion at the boundary with other road networks, feedback control effect by real-time data, and learning correction of parameters. It can have an adjusting effect.

The traffic flow model is devised so that linearity is maintained for ease of control and stability, and the solution of the control method considers the entire target road up to some time ahead. A linear programming method or a quadratic programming method is used as the predicted optimal control.

(2) Basically, since the purpose is to control the inflow amount of the inflow lamp so that the main line is not congested, there is a condition that congestion does not occur (that is, the density is below the limit value). With, the control evaluation criteria are set so that the total inflow is maximized.

However, in actuality, there is a possibility that traffic congestion will occur due to deviations in prediction and control. In this case, the control evaluation criterion is set so that the density is as close as possible to the limit value in both the congested section and the non-congested section, that is, the heavy congestion that causes an extreme decrease in speed is avoided. This also includes the effect of increasing the total inflow amount as much as possible.

Therefore, the evaluation criterion for controlling the inflow is changed depending on whether or not a traffic jam is occurring or immediately before the traffic jam is occurring.

(3) As real-time data, the current value of the number of existing sections is obtained from the traffic flow and the speed of the vehicle, or from the traffic volume and the travel time, and the traffic is constantly changed in the dynamic traffic flow model. Reflect the situation correctly.
Further, the data used for the Q-E curve and the Q-K curve can be constantly collected in real time to improve the accuracy.

In addition, the current and future traffic flow into the ramp,
In order to obtain the traffic demand at the uppermost point of the main line, real-time data on the traffic congestion length (congestion number) of inflow ramps is used, and OD (on / off ramp) traffic volume (from each inflow ramp to the outflow ramp). Traffic volume for each time zone), OD
The statistical value or actual value of the demand traffic volume (the demand traffic volume for each time period from each inflow ramp to each outflow ramp) is used.

(4) When a bottleneck on a portion other than the target road affects the target road, the outflow probability at the outflow ramp is changed.

(5) Actual physical control is performed by a traffic signal,
This is done by using breakers, opening and closing booths, reducing the number of booths, information on guide boards, and information guidance through communication with navigation devices.

FIG. 1 is a block diagram showing the configuration of a traffic flow control system in one of the embodiments of the present invention.

Referring to the figure, the traffic flow control system is roughly comprised of a traffic control center 100, a traffic information collection device 200 connected to the traffic control center, a road condition collection device 300, an inflow control device 400, Information providing device 50
It consists of 0 and.

The traffic control center 100 is composed of a traffic monitoring system, an information providing system, a road monitoring system, an unexpected event detection system, and a lamp inflow control system.

The traffic information collecting device comprises a vehicle detector (loop type, ultrasonic type, etc.), an image processing device, a data communication device (road beacon, telephone, etc.), and an ETC device.

The vehicle detector measures the traffic volume at the place where the vehicle is installed, the speed of the vehicle, the occupancy rate of the vehicle, and the traffic congestion length when traffic congestion occurs.

The image processing apparatus processes the image obtained by the camera or the like to measure the traffic volume at that point, the space velocity of the vehicle, the length of the traffic jam and the number of vehicles present. Further, the image processing device detects the plate number of the vehicle, measures the travel time by matching the characteristic amount of the vehicle, obtains OD traffic information, monitors the traffic, and detects an unexpected event. It does detection.

The data communication device measures the travel time of the vehicle within the predetermined section by obtaining the up ring data from the vehicle.

The ETC device obtains OD traffic information and travel time. The road condition collection device 300 includes a snow cover, a rain gauge, a fog meter, and an anemometer, and transmits the road condition to the traffic control center 100.

The inflow control device 400 is composed of a traffic light control device for controlling a traffic light, a breaker control device for controlling a breaker, a booth opening / closing device for controlling booth opening / closing, and the like.

The information providing device 500 is composed of an information board control device for controlling the display on the information board and a data communication device composed of a road beacon and a telephone.

The traffic control center 100, the traffic information collecting device 200, the road condition collecting device 300, the inflow control device 400, and the information providing device 500 are connected to each other by a dedicated wired line or wireless.

FIG. 2 is a plan view showing an installation state of the traffic information collecting device 200 of FIG. Referring to the figure, the road is divided into a plurality of sections (i-1) to (i + 1). Here, it is assumed that 1 ≦ i ≦ n.

At the starting point of each section (the part on the left side in the drawing), vehicle detectors 203 (i-1) to 203 (i + 1) for measuring the traffic volume Q [vehicles / minute] in that section.
Is provided.

Further, in each section, a camera (image processing device) for measuring the number of vehicles existing in that section
201 (i-1) to 201 (i) are provided.

In the section where the on-ramp is present, the vehicle detectors 207 (i-1) and 207 (i) for measuring the number U of vehicles (inflow traffic demand) U that have reached the on-ramp entrance, and the section actually from the on-ramp Vehicle detectors 209 (i-1) and 209 (i) for measuring the number u of vehicles entering are provided.

Further, the on-ramp is composed of a traffic signal, a breaker, or a booth opening / closing device, and an inflow control device 400 (i-1) for controlling the inflow amount of the vehicle from the on-ramp,
400 (i) is provided.

When the number of vehicles waiting on-ramp (the number of vehicles staying on-ramp) is D, the relationship of u ≦ U + D is established.

The variables listed above are summarized as follows.

[0070]

[Equation 1]

In the present embodiment, the measured E,
Based on the values of Q, u, and U, the change in traffic flow over time is taken into consideration, and the most preferable inflow amount (inflow control amount) of the vehicle in each on-ramp is obtained, and the inflow control device 400 is obtained.
Controls (i-1) and 400 (i) are performed.

The relationship between the traffic volume Q and the density K (or the number of existing vehicles E) (Q-E (or K) curve) is a bridge in considering the temporal change of the traffic flow. Is. Although the traffic volume is determined by the density, the density is not determined by the traffic volume and is binary. These relationships are generally expressed by a quadratic equation.

FIG. 3 is a diagram showing a specific example of the QK curve. Referring to the figure, when the vehicle density is k ₀ , the traffic volume q becomes maximum. When the density of vehicles is less than or equal to k ₀ , the traffic volume increases as the density increases. This state is called non-congestion time.

On the other hand, when the vehicle speed exceeds k ₀ , the traffic volume decreases as the density increases. This state is called traffic jam.

However, in actual field data, there is a clear correlation when there is no traffic, but it is normal to invent a considerable variation when there is traffic. At present, this temporal fluctuation phenomenon has not been clarified, and it cannot be said that there is a sufficient relationship.

However, if there is no temporal correlation in the fluctuation phenomenon, that is, if there is an appropriate variation around the approximate curve, then Q and E should be taken as the average relationship of the traffic flow even during a traffic jam. You can As a result, Q-
The E-curve can be used for modeling.

However, since it is basically a characteristic of one point, it is desirable to use the section as short as possible when it is used as a section.

In the present embodiment, as shown in FIG. 4, it is considered to use a model in which a non-congestion time and a congestion time are respectively linearly approximated. Although the model of FIG. 4 is symmetrical, it may not be symmetrical actually.

The specific method of obtaining the inflow control amount will be described below. (1) When there is no traffic congestion, the number of existing sections in the section i at time t can be calculated by the following equation (a). Since Q and E have the relationship shown in FIG. 4, Q in equation (a) is
It can be obtained by the equation (b).

Here, α ⁱ is the slope of the Q-E curve of section i when there is no traffic (see FIG. 4). Further, the outflow amount v ⁱ _t from the off-ramp in the section i can be obtained by the equation (c). Β ⁱ _t in the equation (c) is the probability that the vehicle will leave the off-ramp in the section i from the time (t−1) to the time t.

The number of stayed lamps D ⁱ _t is
It can be obtained by the equation (d).

[0082]

[Equation 2]

Equation (e) can be derived from equations (a) to (d), and equation (f) can be derived from equation (e ').

[0084]

[Equation 3]

However, there are the following constraints (g) to (i). In the equation (g), E _max indicates the number of vehicles (maximum traffic capacity or the number of critically existing vehicles) present when the traffic flow flows most efficiently in one section.

The equation (h) is a constraint condition for preventing the inflow amount u from the on-ramp from exceeding the number of vehicles arriving at the on-ramp.

The expression (i) is a constraint condition for preventing the number of stayed on-lamps from exceeding the predetermined number D _max .

When the time t = 0 is considered to be the current time, the above equation is an equation showing the relationship between the inflow control amounts u _{1 to} u _t from time t = 1 to _t and the traffic flow.

[0089]

[Equation 4]

Since it is desired to insert as many cars as possible from each on-ramp, the maximum u is obtained by using the linear programming according to the equation (j). Further, when the toll is not uniform and is proportional to the distance, weighting may be performed on the on-ramp into which a vehicle having a long OD average distance flows.

This is to allow as many cars as possible from a certain on-ramp when a car entering from a certain on-ramp often goes far. In this case, equation (k) may be used instead of equation (j).

[0092]

[Equation 5]

That is, the model formula shown by the formula (f) is
While satisfying the conditions of the expressions (g) to (i), maxΣΣu
ⁱ _t , the future lamp inflow control amount (u _{1 to} u
_t ) are determined at the same time. However, how far in the future the time t should be considered is determined depending on the size of the road network, the predictability of traffic conditions, and the like. In the actual control, it is realistic to find u up to some extent and use only u ₁ one hour ahead. In the next cycle, u will be recalculated to some extent.
This makes it possible to always apply the latest u while anticipating the traffic situation to some extent.

(2) When the traffic jam is occurring in the section (i-1) and when the section i is currently in the traffic jam, the equation (b) is changed to the equation (l).

[0095]

[Equation 6]

The expression (l) is an expression obtained by linearly approximating the portion during the traffic jam shown in FIG. If there is a blockage at the boundary of the road network, β ⁱ _{t in} equation (c) and Λ ⁿ ₁ and Λ ⁿ ₂ in equation (l) may be changed.

As an evaluation criterion, with respect to the maximum number of inflowing traffic and the number of existing traffic congestion sections, the constraint of equation (g) is used so that the number of existing traffic vehicles approaches a constant value (usually the maximum value E ⁱ _max ). In addition, it is necessary to prevent congestion even in the non-congested section, but it is necessary to increase the traffic volume within the range. From this point of view, also in the non-congested section, the number of existing vehicles is made to approach a fixed value (usually the maximum value E ⁱ _max ). Note that E _max is E corresponding to q _max in FIG. 3 or FIG.

Only when all of the target sections 1 to n are non-congested, the E (K) -Q curve is set to (b).
A linear programming method in which the evaluation formula is approximated by the formula (j) (or (k)) is used. If there is traffic congestion or just before traffic congestion even in one section, the E (K) -Q curve is approximated by a combination of the equations (b) and (l) in all the zones, and the evaluation equation is defined as the equation (m). The next programming method is used.

[0099]

[Equation 7]

In the equation (m), since the evaluation criterion includes the quadratic equation, it is necessary to use the quadratic programming method.

Next, a specific control method will be described using a flowchart. FIG. 5 is a flowchart for explaining the operation of the traffic flow control system in this embodiment.

In step S1, it is judged whether or not there is a request for ending the control. If YES, this routine is ended, and if NO, the routine proceeds to step S2.

[0103] In step S2, collecting traffic Q ¹ ₀ in the most upstream section of the road. This is what the vehicle detector actually collects.

At step S3, Q ¹ of the most upstream section
Predict _t sequentially. This prediction is predicted by another system. For example, Q ¹ _t changes according to the time zone of the day, and thus can be predicted from the empirical value of the traffic flow during the time zone.

In step S4, the flow in each section is
Traffic demand Uⁱ ₀To collect. Then on another system
Therefore, in step S5, the future inflow traffic demand U of each section
ⁱ _tPredict. However, Uⁱ _tTo directly predict
Is difficult in reality. Therefore an alternative is needed
However, this will be described in detail later.

In step S6, the number of existing vehicles E ⁱ ₀ in each section is collected. In step S7, it is determined whether or not there is traffic congestion in the n section. If NO, the process proceeds to step S12 to perform non-traffic congestion control, Y
If it is ES, the process proceeds to step S8 to perform control during traffic jam.

In this step S7, if even one of the n sections is judged to be congested, the process proceeds to step S8. Further, not only when there is congestion at the present time, but also when the prediction result of the controlled object up to time t is congestion (or there is a high possibility of congestion), the control measurement during congestion is used.

For example, the following criteria can be used as the criterion for determining whether the traffic is jammed or not.

(A) When Q ⁱ calculated and predicted in the previous cycle is equal to or greater than a certain value. (B) When E ⁱ calculated and predicted in the previous cycle is equal to or larger than a certain value.

(C) The speed of the vehicle in the section is below a certain value (when a speed sensor is provided in the section).

In step S9, it is determined whether or not a solution for control exists, and if YES, step S10
In step S11, the control result is transmitted to each on-ramp entrance, and the inflow amount is actually controlled in step S11. afterwards,
Return to step S1.

If NO in step S9, an error message is output in step S14, and this routine ends.

FIG. 6 is a flow chart showing the contents of the non-traffic jam control (S12) of FIG. Referring to the figure, in step S21, a relational expression (f) between E ⁱ _t and u ⁱ _{t at} each time is derived.

Next, in step S22, E
and ⁱ _t, relational expression of the u ⁱ _t (f), the constraint equation (g),
From (h) and (i) and the evaluation function (j), u ⁱ _t is calculated using linear programming. When the toll is proportional to the distance, the equation (k) is used instead of the equation (j).

FIG. 7 is a control during traffic jam (S8) in FIG.
It is a flowchart which shows the content of.

Referring to the figure, in step S31, a relational expression (f) 'between E ⁱ _t at each time and u ⁱ _t is derived.
The expression (f) 'is a modification of the expression (f) by substituting the expressions (l) and (b) into the expression (a).

[0117] In step S32, a relational expression between E ⁱ _t and u ⁱ _t (f) ', constraints (g), from (h), (i) and the evaluation function (m), using a quadratic programming Calculate u ⁱ _t .

Note that β in the equation (c) is determined based on the average outflow rate at that time calculated from the time of each fixed number of days, the traffic volume and the number of outflows in that section. Note that β may be determined depending on the day of the week, the presence / absence of an event, or the like.

The number E of existing vehicles is measured as follows.
You may make it correct from a value. (A) Method using traffic volume Q and speed V   When not in traffic By finding the Q / V of a specific point at every time t, that point
The density of the vehicle at can be determined. According to
If the target section is a single road, 1 representing that section
From the Q / V at one or more points and its range distance L
Number of existing sections E _tCan be estimated.

At this time, if there are inflow / outflow lamps in the target section, the number of existing sections can be estimated from the above data, inflow traffic volume, and outflow traffic volume.

For example, referring to FIG. 8, when the upstream and downstream power ranges are considered to be the same, the number of existing vehicles E _{t in the} section is expressed by the equation (n).

Referring to FIG. 9, when there is an inflow / outflow ramp, when changing the upstream and downstream power ranges by the ramp, it is preferable to use the equation (o).

In the case of a traffic jam, the traffic condition inside the section becomes unbalanced, and if there are few points where data can be obtained, it is not possible to determine the section of the influence range.

Therefore, the difference in the inflow and outflow traffic is accumulated from the last time point (tk) when there is no traffic congestion, instead of using the traffic congestion length separately obtained by the system or the correction using only the data at each time t. Then take the method of seeking.

More specifically, referring to FIG. 10, assuming that a traffic jam of length a occurs on a single road,
Since it is necessary to change the upstream and downstream power ranges depending on the presence or absence of traffic jam, E _t can be calculated by the equation (p).

Further, referring to FIG. 11, when there is an inflow / outflow ramp, when the upstream and downstream power ranges are changed depending on the presence or absence of traffic jam, it is preferable to use the equation (q).

[0127]

[Equation 8]

(B) Method of using traffic volume and travel time When the travel time T _t is obtained relatively frequently, the number of existing vehicles is calculated from the travel time T _t and the past traffic volume at the time t when this is obtained. Can be estimated.

For example, assuming a road with only one lane, the traffic volume Q _u (a, b), Q _d (a,
b), the travel time T (t) at time t (based on the time when the vehicle reaches the downstream point), and the number of vehicles E (t) between the upstream point and the downstream point at time t , The following relations hold.

E (t) = E (t−dt) + Q _u (t−dt, t) −Q _d (t−dt, t) (r) E (t) = Q _u (t−T (t ), T) or ... (s) E (t-T (t)) = _Qd (t-T (t), t) (t) Relatively frequently, the travel time T and its reliability (accuracy or (Variation) is obtained by filtering from dynamic characteristics including the traffic volume measurement error related to the number of existing vehicles according to the above equation (r) and the observation equation including the reliability according to the above equation (s) or (t). By the processing (for example, the processing using the Kalman filter), the number of existing vehicles can be easily estimated.

In the case of multiple lanes, the total number of vehicles in each lane may be considered. If the inflow traffic demand (demand traffic) U cannot be predicted directly, the following means is used instead.

(A) When the statistical value of the OD traffic volume is known: As the statistical value of the OD traffic volume, if the statistical data already investigated by the competent department or obtained from ETC in the future,
The following method is used.

When the number of staying vehicles D is known, the inflow traffic demand U from the initial time 0 to the current time t is given by the equation (w). If the traffic demand traffic volume at that time obtained from the statistical value is bar U _t , the current time t
From k to k, the predicted value is obtained by the equation (y). If only congestion length can be obtained, calculate the number of staying vehicles by converting from the mixture ratio of large vehicles or the average vehicle length.

When the number of staying vehicles D is not known, the inflow traffic demand U is ∞, so no prediction is necessary.
At this time, the expression (h) becomes like the expression (x), and the expression (i) becomes invalid.

(B) When the statistical value of OD traffic volume is unknown: The inflow demand traffic volume is made proportional to all outflow traffic volumes, and O
Estimate D.

[0136]

[Equation 9]

The actual inflow control can be performed as follows. (1) A method of adjusting the green time by using a traffic signal so as to match the actual inflow amount with the inflow amount determined thereby.

(2) A method in which a circuit breaker is used and its circuit break time is adjusted to match the actual inflow rate with the determined inflow rate.

(3) A method of adjusting the actual inflow amount to the determined inflow amount by opening / closing the inflow booth or reducing the number of booths when there are a plurality of booths.

(4) A method of adjusting the actual inflow amount to the determined inflow amount by using the on-ramp or the plan information plate near the on-ramp and adjusting the information providing time.

(5) Information guidance by communication with a vehicle information terminal (including an in-vehicle personal computer) such as a navigation device is used to adjust the vehicle to be provided with information, the providing location, the providing point, and the providing time, and to make the decision. A method to match the actual inflow with the inflow.

(6) When the inflow booth is an automatic ticket issuing machine,
The actual inflow control is performed by the maximum number of tickets issued per unit time or the average ticket interval.

(7) When the inflow booth is a manned toll collection point, the actual inflow control is performed by the maximum number of tolls collected per unit time or the average fee collection interval.

The embodiments disclosed this time are to be considered as illustrative in all points and not restrictive. The scope of the present invention is shown not by the above description but by the claims, and is intended to include meanings equivalent to the claims and all modifications within the scope.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a traffic flow control system according to an embodiment of the present invention.

FIG. 2 is a plan view showing an installation state of a traffic information collecting device.

FIG. 3 is a diagram showing a specific example of a QK curve.

FIG. 4 is a diagram showing a model in which a QK curve is linearly approximated.

FIG. 5 is a flowchart showing the operation of the traffic flow control system.

FIG. 6 is a flowchart showing the contents of non-traffic jam control (S12) of FIG.

FIG. 7 is a flowchart showing the contents of traffic jam control (S8) in FIG.

FIG. 8 is a first diagram showing an example of correcting the existing number E.

FIG. 9 is a second diagram showing an example of correcting the number of existing vehicles E.

FIG. 10 is a third diagram illustrating a correction example of the existing number E.

FIG. 11 is a fourth diagram illustrating a correction example of the existing number E.

[Explanation of symbols]

100 Traffic Control Center 200 Traffic Information Collection Device 201 camera (image processing device) 203 vehicle detector 207,209 Vehicle detector 400 Inflow control device 500 Information providing device

Continuation of the front page (56) Reference JP-A-9-147285 (JP, A) JP-A-55-105799 (JP, A) JP-A-10-160498 (JP, A) JP-A-9-62679 (JP , A) JP 63-208196 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G08G 1/00-1/16

Claims (19)

(57) [Claims] 1. An inflow lamp control device for a road, comprising: the number of vehicles existing in each section of a target road, the traffic volume of vehicles flowing into the section, the traffic volume of vehicles flowing out of the section, and For a target road, an E (K) -Q model showing the relationship between the number of existing vehicles dynamics model represented by the number of existing vehicles before a certain time, the number or density of vehicles, and the traffic volume The number of existing vehicles predicting the number of vehicles existing in each of the sections up to a certain time point from the forecast data of the inflow traffic demand from the outside, the total inflow amount of the inflow lamps, and the Based on the critical existing vehicle number, using the prediction result of the existing vehicle number predicting means, using the condition that the number of existing vehicles in each of the sections up to a certain time ahead does not exceed the critical existing vehicle number, a certain time La to the end And a determining means for determining a respective inflow of flops, the presence volume dynamic characteristic model, the entry ramp inflow of linear
Formulated by connection, each run up to a certain time ahead by linear programming
An inflow ramp control device , characterized in that it determines an inflow amount of a pump . 2. The inflow lamp control device according to claim 1 , wherein the E (K) -Q model uses a linear characteristic when there is no traffic congestion. 3. An inflow lamp control device for a road, wherein the number of vehicles existing in each section of the target road is defined as the traffic volume of vehicles flowing into the section, the traffic volume of vehicles flowing out of the section, and For a target road, an E (K) -Q model showing the relationship between the number of existing vehicles dynamics model represented by the number of existing vehicles before a certain time, the number or density of vehicles, and the traffic volume From the forecast data of the inflow traffic demand from the outside, the existing vehicle number predicting unit that predicts the number of vehicles existing until a certain time in each of the sections, and the critical vehicle number of each of the sections as a reference, Using the prediction result of the existing number prediction means, so that the existing number of each section until a certain time ahead approaches the critical existing number, equipped with a determining means for determining the inflow amount of each lamp until a certain time ahead, Inflow lamp control Location. 4. The inflow quantity of each lamp up to a certain time point is determined by a quadratic programming method by formulating the existing number dynamic model of the existing vehicles by linear combination of the inflow quantity of the inflowing lamps. Inflow lamp control device according to item 3 . 5. The inflow ramp control device according to claim 4 , wherein the E (K) -Q model uses a linear characteristic when there is no traffic jam and a linear characteristic when there is traffic jam. 6. presence number of the vehicle, the inflow according to separate and traffic data on the obtained main line in the system, for each time determined from the rate data or travel time data, any one of claims 1 to 5 Lamp control device. 7. The forecast data of the inflow traffic demand is O
The inflow according to any one of claims 1 to 5 , wherein the statistical value of the D traffic volume is determined by adjusting the traffic volume of each inflow ramp for each time obtained by a system or the OD traffic volume of the actual value on the day. Lamp control device. 8. The forecast data of inflow traffic demand is O
Statistics D demand traffic is determined by adjusting a separate OD demand traffic demand traffic or on the day of actual values of each entry ramp each time obtained by the system, in any one of claims 1 to 5 Inflow lamp control device as described. 9. The actual inflow control is performed based on only the inflow amount up to one time ahead of the inflow amount of each lamp obtained up to a certain time ahead, and the feedback control in which the control amount determination process is repeated every time. performing, entry ramp control apparatus according to any one of claims 1 to 5. 10. The actual inflow control is performed by using a traffic light and adjusting the blue time so that the actual inflow amount is matched with the inflow amount determined by the determining means. > Inflow lamp control device according to any one of 5 above. 11. The actual inflow control is performed by using a breaker, and adjusting the shutoff time so that the actual inflow amount is matched with the inflow amount determined by the determining means. > Inflow lamp control device according to any one of 5 above. 12. The actual inflow control is performed by opening and closing an inflow booth,
Alternatively, when there are a plurality of booths, the number of booths is reduced so that the actual inflow amount is adjusted to match the actual inflow amount determined by the determining means, whereby the inflow lamp control according to any one of claims 1 to 5. apparatus. 13. When the inflow booth of an automatic ticketing machine, the actual inflow control is performed by the maximum ticket number or average ticketing distance per unit time, entry ramp control apparatus according to any one of claims 1 to 5. 14. When the inflow booth is a manned toll collection point,
The actual inflow control is performed by the maximum fare collection number or mean tolling intervals per unit time, claims 1-5
Inflow lamp control device according to any one of 1. 15. In the actual inflow control, an on-ramp or a guide information plate near the on-ramp is used, and the information providing time is adjusted so that the actual inflow amount matches the inflow amount determined by the determining means. entry ramp control apparatus according to any one of claims 1 to 5. 16. The actual inflow control is performed by information guidance by communication with a vehicle information terminal such as a navigation device, and the vehicle to be provided, the point to be provided, and the providing time are adjusted, and the determining means determines the result. so as match the actual inflow to inflow, inflow lamp control apparatus according to any one of claims 1 to 5. 17. A road inflow lamp control device, comprising: the number of vehicles existing in each section of a target road, the traffic volume of vehicles flowing into the section, the traffic volume of vehicles flowing out of the section, and For a target road, an E (K) -Q model showing the relationship between the number of existing vehicles dynamics model represented by the number of existing vehicles before a certain time, the number or density of vehicles, and the traffic volume The number of existing vehicles predicting the number of vehicles existing in each of the sections up to a certain time point from the forecast data of the inflow traffic demand from the outside, the total inflow amount of the inflow lamps, and the Based on the critical existing vehicle number, using the prediction result of the existing vehicle number predicting means, using the condition that the number of existing vehicles in each of the sections up to a certain time ahead does not exceed the critical existing vehicle number, a certain time Up to A first determining means for determining the respective inflow pump, the utilization efficiency of the road selected based on the critical presence number of each of said sections, by using the prediction result of the presence volume prediction means,
It is equipped with a second determining means for determining the inflow amount of each lamp up to a certain time ahead so that the number of vehicles existing in each section up to a certain time ahead approaches the critical existing vehicle number, and there is congestion on the target road. If there is a section that exists, or if there is a section that is just before traffic jam due to congestion of the road, the second determining means is used, otherwise, the first determining means is used. , Inflow lamp controller. 18. A road inflow ramp control method, comprising the number of vehicles existing in each section of a target road, the traffic volume of vehicles flowing into the section, the traffic volume of vehicles flowing out of the section, and For a target road, an E (K) -Q model showing the relationship between the number of existing vehicles dynamics model represented by the number of existing vehicles before a certain time, the number or density of vehicles, and the traffic volume From the forecast data of the inflow traffic demand from the outside of the vehicle, the existing vehicle number predicting step of predicting the number of vehicles existing until a certain time in each of the sections, the total inflow amount of the inflow lamps, and the Based on the critical existing vehicle number, using the prediction result of the existing vehicle number predicting step, using the condition that the existing vehicle number of each vehicle in the section up to a fixed time ahead does not exceed the critical existing vehicle number, a certain time And a determining step of determining a respective inflow of ramp up, the presence volume dynamic characteristic model, the entry ramp inflow of linear
Formulated by connection, each run up to a certain time ahead by linear programming
A method for controlling an inflow ramp , which comprises determining an inflow amount of a pump . 19. A road inflow ramp control method, comprising: the number of vehicles existing in each section of a target road, the traffic volume of vehicles flowing into the section, the traffic volume of vehicles flowing out of the section, and For a target road, an E (K) -Q model showing the relationship between the number of existing vehicles dynamics model represented by the number of existing vehicles before a certain time, the number or density of vehicles, and the traffic volume From the forecast data of the inflow traffic demand from the outside, the existing vehicle number predicting step of predicting the number of vehicles existing until a certain time in each of the sections, and the critical existing vehicle number of each of the sections as a reference, Using the prediction result of the existing number prediction step, so that the existing number of each section until a certain time ahead approaches the critical existing number, comprising a determining step of determining the inflow amount of each lamp until a certain time ahead, ON lamp control method.