JP4003828B2 - Road control method, road control system, and recording medium - Google Patents

Description

[0001]
[Technical field to which the invention belongs]
The present invention relates to a congestion elimination system using traffic flow control means, and more particularly to a system that processes data from a traffic sensor by case-based reasoning and controls appropriate traffic flow.
[0002]
[Prior art]
The elimination of congestion in road traffic is a well-known problem. For example, on a highway, traffic jams mainly occur at junctions on highways such as toll gates and parking area exits, junctions with ordinary roads, slopes, sudden accidents, road construction, and the like.
[0003]
In recent years, various methods for resolving such traffic jams have been attempted, such as providing traffic jam information by television and radio broadcasting, and notifying traffic conditions using electric bulletin board road signs.
[0004]
[Problems to be solved by the invention]
On highways, so-called ETC, which collects tolls by wireless communication, has been introduced as a measure to eliminate such traffic congestion. As shown in Fig. 1, as a result, traffic jam 1 at the upstream of the toll booth can be resolved 3, but traffic jam 2 at the junction with the general road after passing the toll gate, hills, accidents, etc. has been resolved. Absent. For example, a vehicle that has passed through the ETC joins a vehicle that has entered from a junction with the previous general road and a vehicle that is staying in the congestion section 4 to form a large congestion section. Once a traffic jam occurs (each vehicle stops and starts repeatedly), the amount of traffic that can pass per unit time decreases, and the traffic time is (number of cars) x (response time per car) It is said that it takes a long time to clear a traffic jam. Therefore, the issue is how to prevent congestion. As one of the countermeasures, the adoption of ETC for toll gates is being promoted. However, ETC cannot solve the traffic congestion due to road junctions, slopes, sudden accidents, road construction, etc. as described above. Or even a greater traffic jam.
[0005]
In other words, with the introduction of the ETC, the traffic flow bottleneck has only moved from the toll booth to the further section, and the actual situation is that the congestion has not been essentially eliminated.
[0006]
[Means for Solving the Problems]
In view of the above problems, the present invention is a road control method using an inference process, in which a simulation is performed with various control amounts in advance and a calculation result is stored in a case database. acquires flow data, based on the simulation data subjected to traffic flow data and simulation after a predetermined time together with the control amount and the second step of saving the case database, the traffic flow data obtained by said second step A third example of searching for similar cases from the case database, and an example of handling the control amount used in the traffic flow data of the second step corresponding to one or more similar examples searched in the third step Simulation result and control obtained by the second process when a similar case is not found in the fourth process to be presented and the third process A road control method characterized by having a fifth process of presenting a quantity is proposed.
[0007]
In the above method, the subsequent predicted traffic flow corresponding to the current traffic flow control is the traffic flow after about 5 to 10 minutes.
[0008]
DETAILED DESCRIPTION OF THE INVENTION
As described above, the present invention actively controls the traffic flow itself using a traffic flow control means such as an ETC gate. Data used for traffic flow control is data from a traffic sensor and data obtained by a surveillance camera or the like installed in the vicinity of the sensor. Based on this data, the traffic flow control means is operated by a simulation and inference system. . In the initial stage of operation of the system, no data is stored in the case database. Therefore, the results calculated in advance under various conditions using a traffic flow simulator are stored in the case database. Next, traffic flow control is performed using the data, and the resulting change in traffic flow is measured and stored in the database as an example. Furthermore, after sufficient data is stored in the case database, traffic flow control is performed mainly using case-based reasoning, and the control results are similarly stored in the database as cases. Hereinafter, the present invention will be described in more detail.
[0009]
As described above, since the case is not stored in the case database at the initial stage when the system operates, the results calculated under various conditions in advance are stored in the case database using the traffic flow simulator. In addition, as shown in FIG. 3, the traffic flow in the near future of about 10 minutes is predicted by the traffic flow simulator from the data 24 acquired by the traffic detector and the camera, for example, the traffic volume, the average passing speed, the number of staying, etc. Then, using the data 29 calculated in advance, the operation of the traffic flow control means is determined by the case-based reasoning so that the traffic flow is in a predetermined state, and the traffic flow control is performed.
[0010]
Next, a change in traffic flow caused by the operation is acquired by the traffic sensor and camera. That is, data before and after traffic flow control is acquired at the same location. At this time, the traffic flow is measured at least on the upstream side and the downstream side of the traffic jam location. Furthermore, the acquired data is stored in the case database as a case. By repeating this operation, cases are accumulated in the database.
[0011]
When a predetermined amount of data is accumulated in the case database by repeating the above process, traffic flow control is performed by case-based reasoning. In the following, traffic flow control using case-based reasoning will be described in three phases: a data acquisition phase, a simulation phase, and a case-based reasoning phase.
[0012]
(Phase 1)
The first phase is to acquire traffic flow data. As shown in FIG. 3, the traffic flow around the ETC toll gate 23 is measured at regular intervals by the traffic detectors 20 and 22, for example. Although not shown, a camera for measuring the length of congestion is also installed near the toll booth. With these devices, data such as the passing speed of vehicles near the ETC toll gate, the number of passing vehicles for each time zone, the staying length and the staying number can be acquired.
[0013]
(Phase 2)
A schematic diagram of Phase 2 is shown in FIG. Using the data acquired in phase 1, a traffic flow in the near future of about 10 minutes when the traffic flow control device is operated by a predetermined amount is predicted. As described above, the values of the passing traffic volume, the number of stays, the staying length, the passing average speed, etc. 102 acquired by the traffic detector are input to the traffic flow simulator 26, and the traffic flow around the ETC toll booth and further ahead section. 101 is predicted.
[0014]
(Phase 3)
A schematic diagram of Phase 3 is shown in FIG. The data 202 from the traffic detector acquired in phase 1 is input to the case-based reasoning system to perform inference, and the traffic flow control device is operated at a predetermined value 205. In case-based reasoning, the current situation is compared with the situation stored in the database, and similar cases are searched. Next, a difference between the similar case and the current situation is calculated, and based on this difference, the control amount stored in the database is corrected to match the current situation, and a countermeasure case 204 is shown. Next, traffic flow control is performed 205 using this result. If a similar example is not found by inference, a phase 2 simulation is used. Furthermore, the traffic flow after control is measured, and the results are stored in the database as examples.
[0015]
Further, for example, sudden phenomena such as accidents and construction are simulated and controlled by the traffic flow simulator, and the control results are similarly stored in the case database.
[0016]
The above is the flow of traffic flow control according to the present invention. The above-described series of flows is shown in a block diagram in FIG. However, the present invention is not limited to the above examples. For example, when sufficient data is stored in the case database, the traffic flow control can be performed using only the measurement data and the case-based reasoning without performing the traffic flow simulation of the phase 2.
[0017]
For traffic flow control, the number of vehicles passing through the gate per unit time can be controlled by controlling the opening and closing time of the ETC gate. This controls the traffic flow. Furthermore, dynamic signal control or control by speed limit may be used. At present, an automatic driving system, so-called AHS, has been developed, and traffic flow can be controlled by the AHS.
[0018]
In the present invention, case-based reasoning is used in the reasoning system. As is obvious to those skilled in the art, in case-based reasoning, cases are represented by their attribute vectors. Examples of the attribute include traffic volume, passing average speed, number of staying, staying length, and others. The similarity between the input case and the past case is calculated using this attribute vector. Case-based reasoning memorizes the problem-solving process as a single event, so it can save the process of constructing the knowledge base like knowledge-based reasoning, and it can be used for complex problem-solving processes that cannot be described by knowledge-based reasoning. Can respond.
[0019]
As described above, the present invention is a method and system for eliminating congestion related to road traffic. However, the present invention can be used not only for road traffic but also for marine traffic, air traffic, and the like. Moreover, since case-based reasoning is used, it can be applied to more complex control such as road traffic around the station and the flow of pedestrians.
[0020]
In addition, although some time (time value) was specified as an example of said description, the value of time does not have a special meaning and should just be made into an appropriate value in implementation.
[0021]
【The invention's effect】
According to the present invention, it is possible to reduce the occurrence of traffic congestion over the entire road, and it is possible to avoid a decrease in passing traffic volume.
[Brief description of the drawings]
FIG. 1 is a schematic diagram showing conventional traffic conditions on a highway and traffic conditions after introduction of ETC.
FIG. 2 is a block diagram of the present invention.
FIG. 3 is a schematic diagram of a traffic flow control system at an early stage and the installation location of a traffic detector at a toll gate on a highway.
FIG. 4 is a diagram showing a traffic situation prediction in phase 2;
FIG. 5 is a schematic diagram showing presentation of countermeasure cases using case-based reasoning in phase 3 and traffic flow control.
[Explanation of symbols]
10 Data acquired from traffic detector 15 Case database 26 Traffic flow simulator 204 Presentation of countermeasures by case-based reasoning 205 Control of traffic flow control means

Claims (10)

A road control method using an inference process,
A first process of performing simulations with various control amounts in advance and storing the calculation results in a case database;
A second process of acquiring traffic flow data from a traffic detector, storing traffic flow data after a predetermined time together with a control amount and simulation data obtained by simulation in the case database ;
A third process of searching for similar cases from the case database based on the traffic flow data acquired by the second process;
A fourth process of presenting a control amount handling example used in the traffic flow data of the second process corresponding to the one or more similar examples retrieved by the third process;
A fifth step of presenting a simulation result and a control amount obtained by the second step when a similar case is not found in the third step;
The road control method characterized by having. Furthermore, the traffic detector was the value of the traffic flow data and traffic flow control obtained from the road control of claim 1 including the step of adding to the value of the control and the past measurement data stored in the case database Method. 3. The method according to claim 1, further comprising a step of simulating the measurement data and traffic flow control value of the traffic detector and the predicted traffic flow thereafter, and supplementing the past measurement data and control value by the simulation. The road control method described in 1. 4. The road control method according to claim 1, wherein the traffic flow control is performed using an automatic toll collection system, that is, ETC, a road speed limit display device, a traffic light, and an automatic driving system, that is, AHS. . The road control method according to claim 1, wherein the inference process is case-based inference. The road control method according to any one of claims 1 to 4, wherein the traffic flow data by the traffic detector is at least upstream and downstream traffic flow of a traffic jam location. The road control method according to claim 1, wherein the data is at least traffic flow before and after traffic flow control. The difference between the similar case retrieved by the third process and the current situation is calculated, and based on this difference, the control amount stored in the case database is corrected to match the current situation, and the countermeasure case A sixth step of presenting 1 The road control method described in 1. A computer-readable recording medium storing a software program for executing road control using an inference process,
A first process of performing simulations with various control amounts in advance and storing the calculation results in a case database;
Get the traffic flow data from the traffic sensor, and a second process of storing and simulation data subjected to the traffic flow data and simulation after a predetermined time together with the control amount to the case database,
A third process for searching for a similar case from the case database based on the traffic flow data acquired by the second process;
A fourth process for presenting a control amount handling example used in the traffic flow data of the second process corresponding to the one or more similar examples retrieved by the third process;
A fifth process for presenting a simulation result and a control amount obtained by the second process when a similar example is not found in the third process;
A computer-readable storage medium storing a program for causing a computer to execute the program. A road control system using an inference process,
A first means for performing simulations with various control amounts in advance and storing the calculation results in a case database;
A second means for acquiring traffic flow data from a traffic detector, storing traffic flow data after a predetermined time together with a control amount and simulation data obtained by simulation in the case database ;
Third means for retrieving a similar case from the case database based on the traffic flow data acquired by the second means;
A fourth means for presenting a control amount handling example used in the traffic flow data of the second means corresponding to the one or more similar examples searched by the third means;
Fifth means for presenting a simulation result and a control amount obtained by the second means when a similar case is not found in the third means;
A road control system comprising:

Images