Simulation apparatus and simulation program
Patent Information
- Application Number
- JP2023132085
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-08-14
- Publication Date
- 2025-02-27
AI Technical Summary
【0014】 本発明によれば、エージェントの行動の自由度を維持しつつ、効率的に高精度な人流及び車流の予測が可能となる。
Smart Images
Figure 2025027348000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a technique for implementing a multi-agent simulation representing the flow of people and vehicles. [Background technology]
[0002] Conventionally, simulators have been developed that predict the flow of pedestrians or vehicles for the purpose of planning development plans and construction projects after predicting and evaluating the impact on surrounding traffic flows. Furthermore, for example at the entrances and exits of facilities, the flow of people (people flow) and the flow of vehicles (vehicle flow) intersect, so there is a need for a method of evaluating the surrounding environment using multi-agent simulation that takes into account the interactions between people and vehicles in addition to the interactions between people and vehicles.
[0003] In multi-agent simulations, in order to avoid collisions between agents, processes such as checking each agent's position and collision detection are carried out. In this case, in order to represent the interaction between people and cars, the amount of calculation required is equal to the number of people x the number of cars, and the large processing load was an issue.
[0004] In such a situation, for example, Non-Patent Documents 1 and 2 propose methods to reduce the amount of calculation by grouping people, and Non-Patent Document 3 proposes methods to represent the range of people's activities in an extended one-dimensional space. [Prior art documents] [Non-patent literature]
[0005] [Non-Patent Document 1] Hideki Fujii et al., Development of an intelligent multi-agent traffic flow simulator MATES, Part 3: Development of a multi-level pedestrian model and simulation of mixed pedestrian and vehicular traffic, Journal of the Japan Society for Simulation Studies, Vol. 3, No. 3, pp. 70-78, 2011. [Non-Patent Document 2] Masashi Uda et al., Traffic Simulation Model Taking into Account Crossing Pedestrians, 55th Annual Conference of the Japan Society of Civil Engineers, IV-23, 2000. [Non-Patent Document 3] Hideki Fujii et al., Construction of an Extended 1D Pedestrian Model and Simulation of Mixed Pedestrian and Vehicle Traffic at Intersections, Transactions of Information Processing Society of Japan, Vol. 59, No. 3, pp. 874-881, 2018. [Non-Patent Document 4] Kazuhiro Amano et al., Incorporating a Queueing Model into a People Flow Simulator, Proceedings of the Architectural Institute of Japan, pp.51-52, 2021. Summary of the Invention [Problem to be solved by the invention]
[0006] However, conventional methods require processes such as grouping people into groups and calculating collision avoidance areas between the groups and vehicles each time, which still places a heavy processing load on the system, and further improvements are desired. In addition, although it is expected that the amount of calculations can be reduced by restricting people flow to an extended one-dimensional virtual link, the microscopic behavioral expression of each agent is restricted.
[0007] An object of the present invention is to provide a simulation device and a simulation program for pedestrian and vehicle flows that are capable of efficiently making highly accurate predictions while maintaining the freedom of action of agents. [Means for solving the problem]
[0008] The simulation device of the present invention includes a setting unit that sets an area where there is no real traffic light and where pedestrian flow and vehicular flow intersect as a fixed collision avoidance area, a people flow simulation unit that is provided with a people virtual signal for controlling whether people are allowed to enter the collision avoidance area and a people virtual stop line where people wait to enter the collision avoidance area, a vehicle flow simulation unit that is provided with a vehicle virtual signal for controlling whether vehicles are allowed to enter the collision avoidance area and a vehicle virtual stop line where vehicles wait to enter the collision avoidance area, and a virtual signal control unit that determines whether or not there are people or vehicles inside the collision avoidance area for each cycle based on the prediction results for each cycle by the people flow simulation unit and the vehicle flow simulation unit, and controls the people virtual signal and the vehicle virtual signal in accordance with the determination result.
[0009] The people flow simulation unit may include a virtual entry sensor for people that detects people entering the collision avoidance area, and a virtual exit sensor for people that detects people exiting from the collision avoidance area, the vehicle flow simulation unit may include a virtual entry sensor for vehicles that detects vehicles entering the collision avoidance area, and a virtual exit sensor for vehicles that detects vehicles exiting from the collision avoidance area, and the virtual signal control unit may determine whether or not there is a person or a vehicle inside the collision avoidance area based on the detection results by the virtual entry sensor for people, the virtual exit sensor for people, the virtual entry sensor for vehicles, and the virtual exit sensor for vehicles.
[0010] The virtual signal control unit may maintain a list of identifiers of people and vehicles within the collision avoidance area.
[0011] The setting unit may set the collision avoidance area to be wider than the intersection area by a predetermined margin for each of people and vehicles.
[0012] The people flow simulation unit may provide a queuing model on a route leading to the virtual stop line for people, the queuing model consisting of a plurality of cells with a specified capacity, in which people flow is restricted to movement to the previous cell, and the decision as to whether or not to enter the collision avoidance area may be limited to the leading cell.
[0013] A simulation program according to the present invention causes a computer to function as the simulation device. Effect of the Invention
[0014] According to the present invention, it is possible to efficiently predict the flow of people and vehicles with high accuracy while maintaining the freedom of action of agents. [Brief description of the drawings]
[0015] [Figure 1] FIG. 1 is a diagram comparing a human-vehicle interaction model according to an embodiment with a conventional model. [Diagram 2] 1 is a block diagram showing a functional configuration of a simulation device according to an embodiment; [Diagram 3] 11 is a diagram illustrating an example of an operation of a virtual signal control unit in the embodiment. [Figure 4A] 1 is a first flowchart showing a processing procedure of a simulation involving entry into a collision avoidance area in an embodiment. [Figure 4B] 10 is a second flowchart showing the processing procedure of a simulation involving entry into a collision avoidance area in the embodiment. [Figure 4C] 10 is a third flowchart showing the processing procedure of a simulation involving entry into a collision avoidance area in the embodiment. [Diagram 5] FIG. 2 illustrates a queuing model applied in an embodiment. [Figure 6] 10 is a flowchart showing a processing procedure of a people flow simulation when a queueing model in an embodiment is applied. [Figure 7A] FIG. 11 is a first diagram illustrating a simulation result in the embodiment. [Figure 7B] FIG. 11 is a second diagram illustrating a simulation result in the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0016] An example of an embodiment of the present invention will now be described. In the multi-agent simulation of this embodiment, in a signal-free area where collisions between people and cars need to be avoided, a model of interaction between people and cars is constructed by providing a virtual signal for cars that turns red when a person enters, and a virtual signal for people that turns red when a car enters.
[0017] FIG. 1 is a diagram comparing the human-vehicle interaction model according to this embodiment with the conventional one. For example, when simulating the movements of multiple people and vehicles at the unsignalized intersection shown in the figure, in conventional simulation methods, in addition to the interactions between the people and the vehicles, processes to avoid collisions are also carried out between the people and the vehicles.
[0018] On the other hand, in this embodiment, virtual signals for people and vehicles are provided in the collision avoidance area, and instead of the mutual processing required in the past to avoid collisions, people and vehicles each check only the signal and determine whether or not to enter the collision avoidance area according to its color.
[0019] FIG. 2 is a block diagram showing the functional configuration of the simulation device 1 in this embodiment. The simulation device 1 is an information processing device (computer) that includes a control unit 10, a storage unit 20, and also input / output and communication interfaces.
[0020] The control unit 10 is a part that controls the entire simulation device 1, and realizes each function in this embodiment by appropriately reading and executing various software including a simulation program stored in the storage unit 20. The control unit 10 may be a CPU.
[0021] The storage unit 20 is a storage area for storing various programs for causing the hardware group to function as the simulation device 1, various data, and the like, and may be a ROM, a RAM, a flash memory, a hard disk drive (HDD), or the like. Specifically, the storage unit 20 stores various software such as a simulation program, as well as information on agents (people and cars) that are the targets of the simulation, and the environment in which the agents act, as parameters.
[0022] The control unit 10 includes a setting unit 11, a people flow simulation unit 12, a vehicle flow simulation unit 13, and a virtual signal control unit 14.
[0023] The setting unit 11 sets an area in the environment to be simulated where there is no actual traffic light and where the flow of people and the flow of vehicles intersect as a fixed collision avoidance area. Here, an area where pedestrian flow and vehicular flow intersect may be one where the pedestrian flow and vehicular flow are controlled by traffic lights, such as at a pedestrian crossing. In this case, the pedestrian flow simulation unit 12 and the vehicular flow simulation unit 13 share the traffic light cycle with each other, thereby synchronizing with each other and making it possible to make predictions that avoid collisions between people and vehicles.
[0024] In this embodiment, the target is an environment where no traffic lights exist, such as an intersection without traffic lights, or a road in front of a facility where there is no pedestrian crossing but people are expected to cross. At a crossing without traffic lights, interactions occur in which cars stop to wait for people to cross. In order to deal with such a situation, first, the setting unit 11 sets, based on operation input or setting file input, etc., an area where the pedestrian flow and the vehicular flow intersect, or a fixed collision avoidance area that is further expanded by a predetermined margin for each of the pedestrians and the vehicular flows than this intersecting area.
[0025] The people flow simulation unit 12 is a simulator that predicts (reproduces) people flow in a virtual environment equipped with a virtual human signal for controlling whether people can enter a set collision avoidance area, and a virtual human stop line where people wait to enter the collision avoidance area.
[0026] This people flow simulation unit 12 applies a step-by-step behavior decision algorithm based on a multi-agent model to each agent (a person), and periodically predicts the position of each agent and outputs it in a time series. The behavior decision algorithm includes, for example, global route selection that selects an efficient route from links connecting a departure point, a destination, and intermediate points, and local course determination that selects a speed vector that will move the longest in a desired direction while avoiding obstacles.
[0027] Here, the virtual signal for people is installed in the collision avoidance area instead of a traffic light that does not actually exist, and at least two states, red and green, are switched by the virtual signal control unit 14. When the signal is red, the people flow simulation unit 12 makes people heading toward the collision avoidance area wait in front of the virtual stop line for people until the signal changes to green.
[0028] The vehicle flow simulation unit 13 is a simulator that predicts (reproduces) vehicle flow in a virtual environment that includes virtual signals for vehicles to control whether or not vehicles can enter a set collision avoidance area, and virtual stop lines for vehicles where vehicles wait before entering the collision avoidance area.
[0029] Like the pedestrian flow simulation unit 12, the vehicle flow simulation unit 13 also applies a behavior decision algorithm to each agent (car) based on a multi-agent model, and periodically predicts the position of each agent and outputs it in time series. In addition, a virtual traffic light for vehicles is installed in the collision avoidance area, and is switched between at least two states, red and green, by the virtual traffic light control unit 14. When the traffic light is red, the vehicle flow simulation unit 13 makes a vehicle heading toward the collision avoidance area wait in front of the virtual stop line for vehicles until the light changes to green.
[0030] Furthermore, the pedestrian flow simulation unit 12 may be provided with a virtual human entry sensor that detects a person entering the collision avoidance area and a virtual human exit sensor that detects a person exiting the collision avoidance area, and the vehicle flow simulation unit 13 may be provided with a virtual vehicle entry sensor that detects a vehicle entering the collision avoidance area and a virtual vehicle exit sensor that detects a vehicle exiting the collision avoidance area.
[0031] The virtual signal control unit 14 judges whether or not there are people or vehicles inside the collision avoidance area for each period based on the prediction results for each period by the pedestrian flow simulation unit 12 and the vehicle flow simulation unit 13, and controls the pedestrian virtual signal and the vehicle virtual signal according to the judgment result. The pedestrian flow simulation unit 12 and the vehicle flow simulation unit 13 perform periodic predictions alternately, and the virtual signal control unit 14 controls each virtual signal for each prediction result. This makes it possible to avoid problems such as people and vehicles entering the collision avoidance area at the same time.
[0032] That is, when there is a person in the collision avoidance area, the virtual signal control unit 14 switches the vehicle virtual signal to red, and as a result, the vehicle flow simulation unit 13 waits for the vehicle to enter the collision avoidance area. Similarly, when there is a car in the collision avoidance area, the virtual signal control unit 14 switches the pedestrian virtual signal to red, and as a result, the pedestrian flow simulation unit 12 waits for the person to enter the collision avoidance area.
[0033] Furthermore, when all people have exited the collision avoidance area, the virtual signal control unit 14 switches the vehicle virtual signal to green, which causes the vehicle flow simulation unit 13 to permit vehicles to enter the collision avoidance area. Similarly, when all cars have exited the collision avoidance area, the virtual signal control unit 14 switches the pedestrian virtual signal to green, which causes the pedestrian flow simulation unit 12 to permit people to enter the collision avoidance area.
[0034] Specifically, the virtual signal control unit 14 determines whether or not a person or a vehicle is inside the collision avoidance area based on the detection results from the aforementioned virtual entry sensor for people, virtual exit sensor for people, virtual entry sensor for vehicles, and virtual exit sensor for vehicles. For example, the virtual signal control unit 14 may count the number of people and cars inside the collision avoidance area based on the increase or decrease of people or cars determined by each sensor (+1 for entering, -1 for exiting).
[0035] In addition, for example, when multiple collision avoidance areas are close to or overlapping, it is possible that entry and exit will not be paired and will be erroneously detected, such as exit being detected by a sensor installed in a collision avoidance area different from the area at which entry was detected. Therefore, the virtual signal control unit 14 may determine the number of people and vehicles within the collision avoidance area by managing a list of identifiers of people and vehicles.
[0036] FIG. 3 is a diagram illustrating the operation of the virtual signal control unit 14 in this embodiment. This example shows a case where people and vehicles move from left to right and bottom to top, respectively, intersecting with each other, and this intersecting area is set as a collision avoidance area.
[0037] Outside the collision avoidance area, a virtual entry sensor for people, a virtual exit sensor for people, a virtual entry sensor for vehicles, and a virtual exit sensor for vehicles are provided. Furthermore, a virtual stop line for people is provided in front of the virtual entry sensor for people, and a virtual stop line for vehicles is provided in front of the virtual entry sensor for vehicles.
[0038] When a person (A) moving from left to right in the figure comes before the virtual stop line for people, the people flow simulation unit 12 checks the color of the virtual signal for people and then has the person enter the collision avoidance area according to the green signal. Then, when the person (A) who has entered the collision avoidance area passes the virtual entrance sensor for people, the virtual signal control unit 14 determines that there is a person with whom a collision should be avoided inside the collision avoidance area and switches the virtual signal for vehicles to red. As a result, when a car (B) heading from the bottom to the top of the figure comes in front of the virtual stop line for cars, the traffic flow simulation unit 13 checks the color of the virtual signal for cars and makes the car wait before entering the collision avoidance area in accordance with the red signal.
[0039] After that, when a person (A) exits the collision avoidance area and passes the virtual exit sensor for people, the virtual signal control unit 14 determines that there is no person in the collision avoidance area with whom a collision should be avoided, and switches the virtual signal for vehicles to green. As a result, the vehicle flow simulation unit 13 causes a vehicle (B) that has been waiting in front of the virtual waiting line for vehicles to enter the collision avoidance area. Then, when a car (B) enters the collision avoidance area and passes the vehicle virtual entry sensor, the virtual signal control unit 14 determines that there is a car with which a collision should be avoided inside the collision avoidance area, and switches the pedestrian virtual signal to red. As a result, the pedestrian flow simulation unit 12 checks the color of the virtual traffic light for people when the following person (C) comes before the virtual stop line for cars, and makes the person wait before entering the collision avoidance area according to the red light. Furthermore, when the car (B) reaches the virtual exit sensor for cars, the virtual traffic light for people is switched to green.
[0040] Here, the sensors that detect entry into and exit from the collision avoidance area are provided separately for entry and exit, but this is not limited to this, and a single sensor may be used to determine entry or exit depending on the direction of movement (passage direction) of a person or vehicle.
[0041] 4A, 4B, and 4C are flowcharts showing the processing procedure of a simulation involving entry into a collision avoidance area in this embodiment. This process is executed repeatedly at a predetermined cycle, and the positions of multiple people and vehicles are determined in time series.
[0042] In steps S1 to S3, the control unit 10 repeatedly executes the person movement process in step S2 (FIG. 4B) the number of times corresponding to the number (N) of people to be simulated. In steps S4 to S6, the control unit 10 repeatedly executes the car movement process in step S5 (FIG. 4C) the number of times corresponding to the number (M) of cars to be simulated. Here, the only difference between the human movement process (FIG. 4B) and the vehicle movement process (FIG. 4C) is the simulation target (agent). It should be noted that there is no limitation as to which of the simulations, pedestrian flow (steps S1 to S3) and vehicle flow (steps S4 to S6), is performed first. Here, as an example, a processing procedure that prioritizes pedestrian flow is shown.
[0043] In step S21 of Fig. 4B, the people flow simulation unit 12 determines the color of the virtual signal for people associated with the virtual stop line for people in front of the target person. If the signal is red, the process proceeds to step S22, and if the signal is green, the process proceeds to step S23.
[0044] In step S22, the people flow simulation unit 12 prohibits the target person from entering the collision avoidance area according to the color (red) of the traffic light, and makes the target person wait in front of the virtual stop line for people.
[0045] In step S23, the people flow simulation unit 12 moves the target people according to the color (green) of the traffic light. In step S24, the people flow simulation unit 12 judges whether a person has passed the virtual human entrance sensor (whether the sensor has reacted). If the judgment is YES, the process proceeds to step S25, and if the judgment is NO, the process proceeds to step S26.
[0046] In step S25, the virtual signal control unit 14 changes or maintains the vehicle virtual signal to red. The virtual signal control unit 14 also updates a list of identifiers (IDs) of people who have passed through the virtual human entry sensor and are in the collision avoidance area. Note that, although a method of managing a list of IDs is adopted here, as described above, the virtual signal control unit 14 may manage only the number of people in the collision avoidance area.
[0047] In step S26, the people flow simulation unit 12 judges whether a person has passed the virtual human exit sensor, i.e., whether the sensor has reacted. If the judgment is YES, the process proceeds to step S27, and if the judgment is NO, the process ends.
[0048] In step S27, the virtual signal control unit 14 updates the list of identifiers (IDs) of people inside the collision avoidance area and determines whether or not there are people. If there are people (determination is YES), the process ends, and if there are no people (determination is NO), the process proceeds to step S28. In step S28, the virtual signal control unit 14 changes the vehicle virtual signal to green.
[0049] Steps S51 to S58 in FIG. 4C (vehicle movement processing) are similar to those in FIG. 4B, with "people" replaced with "vehicles."
[0050] In this way, for example, if the number of people is N and the number of cars is M, the movement process is performed a total of N+M times for each time step. At this time, the collision avoidance area is set and fixed in advance, so the processing is very lightweight. On the other hand, in the conventional simulation method, it was necessary to set a collision avoidance area between the person and the vehicle individually a total of N×M times, and to judge whether or not the vehicle could enter the collision avoidance area. In this case, the processing load was very high because the positional relationship between the person and the vehicle fluctuated for each object and each time step.
[0051] As described above, the simulation device 1 can efficiently execute a simulation of pedestrian and vehicular flows intersecting in a collision avoidance area in an environment in which a fixed collision avoidance area can be set. Furthermore, the simulation device 1 can improve the efficiency of the simulation of the flow of people up to the collision avoidance area by applying, for example, the queueing model proposed in Non-Patent Document 4.
[0052] That is, the people flow simulation unit 12 may provide a queueing model on the route leading to the virtual stop line for people, which consists of a plurality of cells with a prescribed capacity, and limits people flow to movement to the previous cell.
[0053] FIG. 5 is a diagram illustrating a queue model applied in this embodiment. The shape of each cell constituting the queue model, the distance between the cells, and the capacity of each cell can be set arbitrarily. Therefore, a stay area of any shape can be configured to match the shape of the sidewalk restricted by walls, buildings, etc. in front of the collision avoidance area.
[0054] The people flow simulation unit 12 directs newly arrived people to the vacant cell at the end of the queue, and then, when a vacancy occurs in the cell ahead, moves them to the cell immediately before. By simplifying the people flow movement algorithm in this way, the people flow simulation unit 12 limits the determination of whether or not a cell can enter the collision avoidance area to the leading cell.
[0055] FIG. 6 is a flowchart showing the processing procedure of a people flow simulation when the queuing model in this embodiment is applied.
[0056] In step S11, the people flow simulation unit 12 judges whether or not the target person has arrived at the queue model. If the judgment is YES, the process proceeds to step S13, and if the judgment is NO, the process proceeds to step S12.
[0057] In step S12, the people flow simulation unit 12 performs normal movement processing without performing processing related to the collision avoidance area that is specific to this embodiment.
[0058] In step S13, the people flow simulation unit 12 judges whether or not the target person has arrived at the head cell. If the judgment is YES, the process proceeds to step S17, and if the judgment is NO, the process proceeds to step S14.
[0059] In step S14, the people flow simulation unit 12 judges whether or not there is a vacant cell ahead of the target person. If the judgment is YES, the process proceeds to step S15, and if the judgment is NO, the process proceeds to step S16.
[0060] In step S15, the people flow simulation unit 12 moves the target person who has become vacant in the forward cell to this vacant cell. In step S16, since there is no vacant cell ahead, the people flow simulation unit 12 does not move the target person, but makes the person wait in the current cell.
[0061] In step S17, the target person has reached the head cell of the queue model, i.e., in front of the collision avoidance area, so the people flow simulation unit 12 judges the color of the virtual traffic light for people. If the traffic light is green, the process proceeds to step S18, and if the traffic light is red, the process proceeds to step S19.
[0062] In step S18, the people flow simulation unit 12 moves the target person to the collision avoidance area according to the color (green) of the traffic light. In step S19, the people flow simulation unit 12 prohibits the target person from entering the collision avoidance area according to the color (red) of the traffic light, and makes the target person wait in front of the virtual stop line for people.
[0063] 7A and 7B are diagrams illustrating simulation results in this embodiment. These figures are quoted from "Naoki Mita et al., Development of an Integrated People and Traffic Flow Simulator Using a Pedestrian and Vehicle Interaction Model, 78th Annual Academic Conference of the Japan Society of Civil Engineers National Convention, VI-247, 2023." Here, we show the results of a simulation conducted around a certain facility for two cases: with pedestrians present and without pedestrians present.
[0064] Figure 7A shows the pedestrian flow and number of people, and the vehicle flow and number of vehicles that are the conditions for the simulation. Specifically, 600 vehicles are heading to the facility's parking lot, 200 vehicles are leaving the parking lot, and 360 pedestrians are traveling on the sidewalk outside the facility.
[0065] FIG. 7B shows the state 10 minutes after the start of the simulation. By applying the method of this embodiment, it was possible to express pedestrians waiting on the sidewalk for vehicles to pass at the facility exit when pedestrians were present. In addition, the number of vehicles waiting on the roadway was 18 when pedestrians were present and 8 when there were no pedestrians, resulting in a difference of 10 vehicles. In this way, the interaction between pedestrians and vehicles was adequately represented.
[0066] According to this embodiment, in a multi-agent simulation representing the flow of people and vehicles, the simulation device 1 sets a fixed collision avoidance area where the flow of people and the flow of vehicles intersect, and controls a virtual signal by determining whether or not there are people or vehicles inside this collision avoidance area. As a result, the simulation device 1 is able to significantly reduce the processing load required to avoid collisions via virtual signals compared to the conventional method, whereas previously it was necessary to calculate the interactions between people and cars for each combination of agents (people and cars). As a result, it has become possible to make efficient, highly accurate predictions while maintaining the agent's freedom of action.
[0067] In addition, the simulation device 1 may determine whether or not a person or a vehicle is inside the collision avoidance area based on the detection results from the virtual person entry sensor, the virtual person exit sensor, the virtual vehicle entry sensor, and the virtual vehicle exit sensor. This enables the simulation device 1 to easily count the number of people or cars inside the collision avoidance area and efficiently control the virtual traffic signal.
[0068] Furthermore, the simulation device 1 may manage a list of identifiers of people and vehicles within the collision avoidance area. This allows the simulation device 1 to more accurately determine the number of people or vehicles in each area when multiple collision avoidance areas are close to or overlap each other, and to appropriately control the virtual signal.
[0069] The simulation device 1 may set a collision avoidance area wider than the area where the pedestrian flow and the vehicular flow actually intersect, by a predetermined margin for each of people and vehicles. This makes it possible to provide a realistic margin and easily set a collision avoidance area of any shape and size according to the environment, enabling highly accurate simulation.
[0070] The simulation device 1 may provide a queueing model on a route leading to a virtual stop line for people, the queueing model being made up of a plurality of cells with a prescribed capacity, and restricting the flow of people to moving to the previous cell. This enables the simulation device 1 to simplify the movement of people in the area in front of the collision avoidance area where traffic lines are restricted, and also limits the decision on whether or not to enter the collision avoidance area to the leading cell, thereby further reducing the processing load.
[0071] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments. Furthermore, the effects described in the above-described embodiments are merely a list of the most preferable effects resulting from the present invention, and the effects of the present invention are not limited to those described in the embodiments.
[0072] The simulation method by the simulation device 1 is realized by software. When realized by software, a program constituting this software is installed in an information processing device (computer). These programs may be recorded on a removable medium such as a CD-ROM and distributed to users, or may be distributed by being downloaded to the user's computer via a network. Furthermore, these programs may be provided to the user's computer as a Web service via a network without being downloaded. [Explanation of symbols]
[0073] 1 Simulation device 10 Control section 11 Setting section 12 People Flow Simulation Department 13 Vehicle Flow Simulation Section 14 Virtual signal control section 20 Memory section
Claims
1. A setting unit that sets an area where there is no actual traffic signal and where pedestrian flow and vehicular flow intersect as a fixed collision avoidance area; a people flow simulation unit including a virtual signal for people for controlling whether or not people can enter the collision avoidance area, and a virtual stop line for people where people wait to enter the collision avoidance area; a vehicle flow simulation unit including a vehicle virtual signal for controlling whether or not a vehicle can enter the collision avoidance area, and a vehicle virtual stop line where a vehicle waits before entering the collision avoidance area; a virtual signal control unit that determines whether or not there are people or vehicles inside the collision avoidance area for each period based on the prediction results for each period by the people flow simulation unit and the vehicle flow simulation unit, and controls the people virtual signal and the vehicle virtual signal in accordance with the determination result.
2. the people flow simulation unit includes a virtual person entry sensor that detects a person entering the collision avoidance area, and a virtual person exit sensor that detects a person exiting the collision avoidance area, the vehicle flow simulation unit includes a vehicle virtual entry sensor that detects a vehicle entering the collision avoidance area, and a vehicle virtual exit sensor that detects a vehicle exiting the collision avoidance area, 2. The simulation device according to claim 1, wherein the virtual signal control unit determines whether or not a person or a vehicle is present within the collision avoidance area based on detection results by the virtual person entry sensor, the virtual person exit sensor, the virtual vehicle entry sensor, and the virtual vehicle exit sensor.
3. The simulation device according to claim 2 , wherein the virtual signal control unit manages a list of identifiers of people and vehicles present within the collision avoidance area.
4. 3. The simulation device according to claim 1, wherein the setting unit sets the collision avoidance area to be wider than the intersection area by a predetermined margin for each of a person and a vehicle.
5. 3. The simulation device according to claim 1 or 2, wherein the people flow simulation unit provides a queuing model consisting of a plurality of cells with a specified capacity on a route leading to the virtual stop line for people, limiting people flow to movement to the previous cell, and limits a decision on whether or not to enter the collision avoidance area to the leading cell.
6. A simulation program for causing a computer to function as the simulation device according to claim 1 or 2.