JP3475243B2 - Traffic flow simulator - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a traffic flow simulator for evaluating the influence of a change in road structure on a traffic flow and for evaluating the influence of various measures of ITS on the traffic flow.

[0002]

2. Description of the Related Art In a conventional traffic flow simulator, a simulation is performed by a micro model or a macro model alone. For example, when controlling an individual vehicle such as an AHS, a simulation of a micro model is performed in a narrow area, the results are converted into parameters of a macro model, a macro simulation of a wide area is performed, and traffic flow is performed. I was evaluating the impact. However, the area where it is necessary to describe the operation in detail by connecting the macro simulator and the micro simulator to, for example, the user service of AHS is the micro behavior, and the behavior of the macro in a wide area around the area where the behavior spreads. Hybridization has been considered in which the simulation is carried out. Then, at that time, regarding the connection of the micro model described to define the micro behavior degree and the macro model describing the wide range behavior, there is a match between the macro models expressing the behavior of one unit and the behavior of multiple units. There was no sex.

[0003]

However, the conventional traffic flow simulator described above has the following problems. That is, the conventional system has a problem that the behavior of one vehicle and the behavior of a plurality of vehicles are not consistent between the macro model and the micro model, and connection is difficult.

Therefore, an object of the present invention is to solve the problems of the conventional traffic flow simulator as described above and to provide a traffic flow simulator having both the behaviors of a macro model and a micro model.

[0005]

In order to achieve the above object, the present invention describes a map database in which digitized road information is described, the number of vehicles generated per unit time, the location of occurrence, and the destination location. OD database, a condition database describing simulation conditions, a model function describing vehicle behavior, a simulation control function for controlling the vehicle behavior by generating vehicles according to a random number sequence based on the OD database, and inputting the simulation conditions. In a traffic flow simulator consisting of an input function and a display function that displays simulation results,
An application interface for connecting multiple models that describes multiple behaviors, and individual vehicle behaviors
Representation of a micro model that describes the vehicle and multiple vehicle behaviors
The macro model and the micro model from the macro model.
A machine that separates a group of vehicles moving to
And a data synchronization function between models that is capable of
As a function to separate the
It is characterized in that the vehicle is generated by a random number depending on .
That is, a buffer region is provided at a connecting portion between the macro model and the micro model, and a function to obtain consistency by disassembling and integrating a vehicle and a vehicle group in the buffer region is characterized.

[0006]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to FIGS.
This will be described in detail with reference to FIG. FIG. 1 shows the overall configuration of the present invention. That is, the traffic flow simulator system includes a traffic flow simulator body 1, an input function 2 for inputting simulation conditions, a display function 3 for displaying a simulation result, and an external data processing device for processing traffic volume measurement, external data, and the like. Composed of 15

The traffic simulation body 1 includes an input function 2, a display function 3 and an external data processing device 15.
And an interface function 4 for connecting the traffic simulator main body 1, a simulation condition database 6 in which various conditions and parameters for executing a simulation are described, and a map held as a digitized map of the alignment of the road on which the vehicle runs Database 7, describing the origin and disappearance points of vehicles, the type and number of vehicles
D (Origin: Origination point, Destination: Disappearance point) database 8, a simulation based on the various databases 6, 7 and 8 described above to control the timing of activating various models described later that describe the behavior of the vehicle A control function 5, a micro model group 11 that describes vehicle behavior based on individual vehicle behavior, a macro model group 12 that regulates the behavior of a plurality of vehicles as a group, the simulation control function 5 and the micro model group 11 Application program interface for connection (hereinafter,
Abbreviated as "API". ) Is a micro model API function 9, a macro model API function 10 connecting the macro model group 12 and the simulation control function 5, and a model for exchanging data between the macro model group 12 and the micro model group 11. It is composed of inter-data synchronization function 13.

The micro-model group 11 includes a follow-up model 17 that describes the follow-up movement of two or more vehicles, a split-and-merge model 18 that describes the split-and-join behavior of the vehicles at the split-and-join points, and vehicles approaching intersections and split-and-join points. A route selection model 19 for determining the direction of travel and a lane change model in which movements such as lane changes are described are installed in parallel and controlled by the simulation control function 5 via the micro model API function 9. It

On the other hand, the macro model group 12 does not represent the traffic flow by the behavior of each vehicle, but considers it as a fluid from the overall flow and controls the fluid model 20 and a plurality of vehicles. A vehicle group model 21 that represents the flow of a vehicle by considering it as a block and controlling the behavior of that block.
These models are macro models API
It is controlled by the simulation control function 5 via the function 10, and the result is the simulation database function 22.
Be stored in.

An outline of the simulation control function 5 will be described with reference to FIG. It should be noted that the outline described in FIG. 2 describes the function of controlling the activation of the various models, which is a part of the simulation control function 5, and omits the occurrence of a vehicle. Further, the simulation control function 5 has a start factor determination function 51 that determines the start timing and the start method based on the information from the micro model API function 9 and the macro model API function 10.
This function is activated from the result of this activation factor judgment function 51.
For example, an event activation function 53 that starts processing when a vehicle group reaches an intersection, a model described by a differential equation in the model of the micro model group 11, a vehicle group model or a fluid model, and Periodic activation function 54, which is activated when the position of a vehicle group or fluid is calculated at time intervals.
The start timing is determined by the event start function 53 or the fixed cycle start function 54, the start time and the start processing to be started at the start time, and the start that is a flag for determining whether the start is the event start or the fixed cycle start Type, and start queue database 52 consisting of the elapsed time of the current simulation, by starting various processing included in the micro model group 11 or the macro model group 12 based on the information of the start queue database 52. Start control function to process 55
And a simulation database function 22

An outline of processing of the inter-model data synchronization function 5 will be described with reference to FIG. The various databases 6, 7,
8 information is input to the interface 13. It is input through the micro API function 9. The individual vehicle information for the buffer area from the micro model is input to the vehicle integration processing 132 together with the information from the simulation control function 5 and the information from the interface 131. In this vehicle integration processing 132, blocks are generated by grouping vehicles by vehicle type and OD in the area. The generated block is subjected to macro model processing via the macro model API 10 activated by the simulation control function 5.

On the other hand, the individual vehicle information for the buffer area from the macro model, which is input via the macro API function 10, is combined with the information from the simulation control function 5 and the information from the interface 131 to decompose the vehicle group. Input to the process 134. And this car group disassembly processing 1
In 34, a vehicle management table for area i is generated based on the vehicle IDs included in the vehicle group. The generated vehicle has a micro model API 9 activated by the simulation control function 5.
The processing of the micro model is executed via.

Referring to FIG. 4, a database for vehicle management used by the micro model group 11 and the macro model group 12 will be described. The vehicle generated by the OD table is managed by the overall vehicle management table. For example, when a vehicle reaches the area i, it is registered in the area i vehicle management table 135, and when it leaves the area i, it is deleted. In this case, this data is used when performing a simulation with the micro model group 12 in the region i. When the macro model group is used in the area i, the vehicle group of the area i vehicle group management table 133 is generated at the time when the block is generated in the vehicle integration processing 132, and the block number and the vehicle ID corresponding to the block number are generated. Table is created.

FIG. 5 shows a flow chart of the vehicle integration process 132 according to the present invention. The vehicle integration process 132 is a process in which individual vehicles, which are the outputs of the micro model, are integrated and delivered to the macro model when they enter the connection area, and are composed of the following processing steps.

Extraction processing 1321 of vehicles entering the vehicle integration area
Is a process of determining whether or not a vehicle has entered the vehicle integrated area and registering the inflowing vehicle in the area i vehicle management table based on the vehicle ID. A process step 1322 of creating a destination of an inflow area from a destination of each vehicle is performed by using the vehicle ID as a key, from the entire vehicle management table, using a vehicle-specific generation point (O) and a destination point (D), and then traveling. This is a process of determining an in-area OD which is a combination of the generation point (O) and the destination point (D) of the area i to be registered and registering it in the area i vehicle management table. The processing step 1323 is a processing step of storing the vehicle in the FIFO for each destination / vehicle type, and is a processing step for storing the vehicle ID in the FIFO based on the vehicle ID obtained in the processing step 1321. In processing step 1324, the FIFOs for each destination and vehicle type are checked, and the specified number is extracted from the FIFOs that have reached the specified number (collection of the number handled by the macro model) for each vehicle type.
The arrival time to the area i is cut out from the first vehicle ID of the FO, the block number of the area i is generated, and the vehicle ID included therein, the area OD, and the arrival time are stored in the area i vehicle group management table. It is a process to do. Processing step 1325
Is a vehicle ID included in the vehicle group when the vehicle group is generated.
Based on the above, the block number of the vehicle management table related to the area i is updated.

The processing step 1326 is a processing step in which the processing from the processing step 1321 to the processing step 1326 is determined to be repeated until there is no vehicle flowing into the vehicle integration area.

FIG. 6 shows a procedure for disassembling a vehicle group used in the macro model and generating individual vehicles in a form usable in the micro model.

The vehicle group disassembly processing 134 is composed of the following processing steps 1341 to 1346.

A vehicle group arrival determination processing step 1341 determines whether or not a vehicle group from the macro model has arrived at the contact point between the macro model and the micro model, and when the vehicle group has arrived,
This is a determination processing step in which the ID and arrival time of the arrived vehicle group are stored and the processing is passed to the next step.

In processing step 1342, the time at which a vehicle group of the same vehicle type as the arrived vehicle group next arrives at the contact point from the macro model to the micro model is determined by the speed of the vehicle group and the position to the contact point. Is a processing step for predicting the time of arrival at the contact point.

The processing step 1343 is a processing step for calculating the vehicle group time interval from the difference between the vehicle group arrival time detected in the processing step 1341 and the estimated arrival time detected in the processing step 1342.

In processing step 1343, a random number having a statistical index such as distribution of arrival time intervals of vehicle generation is used to generate vehicles for the number of vehicle groups + 1 vehicle. The reason why the number of vehicles is +1 is that the leading vehicle occurrence time of the next vehicle group is necessary, and the occurrence time interval is adjusted so that it becomes the arrival time of the target vehicle group and the arrival time interval of the next vehicle group. That is, all the generation time intervals of the individual cars are added to obtain the total time. This calculation can be executed by dividing the individual occurrence time interval by the total time and multiplying by the vehicle group arrival time interval to obtain the corrected occurrence time interval.

The processing step 1346 includes the above step 343.
This is a processing step in which the generation time of the vehicle is obtained by sequentially adding to the arrival time of the vehicle group at the corrected generation time interval obtained in step 1 and stored in the area i vehicle management table 222.

FIG. 7 shows a processing outline of the activation correction control function 55 of FIG. The start correction control function 55 is a cyclic start function 55.
1, a vehicle generation function 552, a vehicle movement function 553, and a database update function 554 that updates the vehicle database for each simulation step.

The periodic activation function is activated for each OD resolution. For example, if the OD is 5 minutes, it is activated every 5 minutes after the simulation time elapses.

When the cooperation function is working, the vehicle generation function 552 extracts the arrival vehicle in the time zone in which the simulation is to be performed in unit time of OD from the area i vehicle management table 222 and stores it in the vehicle generation queue. Alternatively, when the cooperation function does not work, it is a function of extracting the number of vehicles generated from the OD database 8 in the unit time of OD in the time zone in which the simulation is to be performed, and generating a vehicle with a random number.

The vehicle moving function 553 is a process for moving the vehicle at each simulation time step, for example,
It is a function that solves differential equations such as Euler and Rugekutta.

The database updating function 554 is a function of updating the state quantity such as the position and speed of the vehicle updated by the vehicle moving function 553 and updating the database 22.

FIG. 8 is a diagram showing the details of the vehicle generating function 552. The vehicle generation function 552 is a function 5521 that determines whether there is a cooperative operation of a plurality of models such as a micro model and a macro model or is operating independently.
Process steps 5522 to 55 for generating a vehicle from the OD table
In the case of 24 steps, cooperative operation, the generation time is referenced from the area i vehicle management table, the vehicle ID and the generation time of the target time pair are acquired, and the vehicle ID is stored in the vehicle generation queue 5526.
And a processing step for storing the occurrence time.

A processing step 5522 for acquiring the number of vehicles generated from the OD table is a processing for acquiring the number of vehicles generated in the simulation target time zone (unit time of OD).

In processing step 5523, a vehicle is generated in accordance with a random number sequence that defines the number of vehicles + 1 obtained in processing step 5522 and the number of vehicles is determined, and the total time which is the cumulative time required for the generation is calculated. Is.

In the processing step 5524, the OD table defines the number of vehicles generated per unit time. However, when a vehicle is actually generated using a random number, the number of vehicles generated per unit time increases or decreases. In order to prevent increase / decrease, it is a process of dividing the generation time interval of each vehicle by the total time by the total time and multiplying the unit time by uniting the number of units generated per unit time.

With the above structure, the present invention allows the micro model and the macro model to be joined together and the vehicle to be replaced.

[0034]

According to the present invention, it is possible to operate an event start having a different start timing and a fixed cycle start with the same mechanism, and it is possible to easily exchange a vehicle between a micro model and a macro model. The effect is that you can do it. Further, there is an effect that a plurality of micro models or macro models can absorb the difference and enable the simulation by using the same data on the same platform.

[Brief description of drawings]

FIG. 1 is an example of an overall configuration diagram of a traffic flow simulator according to the present invention.

FIG. 2 is an example of an overall configuration diagram of a simulation control function according to the present invention.

FIG. 3 is an example of a configuration diagram of an inter-model data exchange function according to the present invention.

FIG. 4 is an example of a configuration diagram of a vehicle management table according to the present invention.

FIG. 5 is an example of a flowchart of vehicle integration processing according to the present invention.

FIG. 6 is an example of a procedure diagram for disassembling a vehicle group and generating individual vehicles according to the present invention.

FIG. 7 is an example of a processing outline diagram of a startup correction control function according to the present invention.

FIG. 8 is an example of a detailed view of a vehicle generating function according to the present invention.

[Explanation of symbols]

1 Traffic flow simulator body 2 input function 3 display function 5 Simulation control function 9 Micro model API function 10 Macro model API function 11 Micro model group 12 Macro model group 13 Data synchronization function between models

Continuation of the front page (56) Reference JP-A-10-222791 (JP, A) JP-A-11-144182 (JP, A) JP-A-10-334389 (JP, A) JP-A-9-147285 (JP , A) (58) Fields investigated (Int.Cl. ⁷ , DB name) G08G 1/00

Claims (3)

(57) [Claims] 1. A map database in which digitized road information is described, an OD database in which the number of vehicles generated per unit time and the location of occurrence, a destination location, a condition database in which simulation conditions are described, and vehicle behavior are described. Traffic flow simulator including a model function, a simulation control function for controlling a vehicle behavior by generating a vehicle according to a random number sequence based on the OD database, an input function for inputting the simulation condition, and a display function for displaying a simulation result. The application interface for connecting multiple models, which describes multiple behaviors, and the behavior of each vehicle
Representation of a micro model that describes the vehicle and multiple vehicle behaviors
The macro model and the micro model from the macro model.
A machine that separates a group of vehicles moving to
It has a data synchronization function between models, which is a function for separating the vehicles.
A traffic flow simulator characterized by generating vehicles with random numbers that depend on the statistics of . 2. The traffic flow simulator according to claim 1 , further comprising: an inter-model data synchronization function that is a function of combining individual vehicles moving from the micro model to the macro model into a group of vehicles. . 3. The traffic flow simulator according to claim 2 , wherein the function of combining the vehicles into groups is performed for each vehicle type.