JP2009259158A - Traffic state simulation device, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To simulate a traffic state with high precision. <P>SOLUTION: An OD traffic volume estimating part 32 estimates an OD traffic volume on the basis of probe information and statistical information, and a traffic state calculating part 38 simulates a traffic state on the basis of road network information, signal information and the estimated OD traffic volume. Also, a traffic information processing part 28 estimates a travel time of a link, a queuing length, and a traffic congestion length on the basis of the probe information and VICS (Vehicle Information and Communication System) information and calculates the target value of traffic volume of the link from estimated values. An OD traffic volume correcting part 42 corrects the OD traffic volume so as to reduce a difference between the calculation value of the traffic volume of the link calculated on the basis of the simulated traffic state and the target value of the traffic volume of the link. The traffic state calculating part 38 uses the corrected OD traffic volume to simulate the traffic state. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a traffic condition simulation apparatus and program, and more particularly, to a traffic condition simulation apparatus and program for running a vehicle model and simulating a traffic condition.

Conventionally, a road traffic condition analysis device that acquires OD information from a toll collection system on a toll road, collects traffic conditions from roadside infrastructure sensor measurement results, adjusts parameters, and analyzes traffic conditions with a simulator. Known (Patent Document 1).
JP-A-2005-84722

  However, the technique described in Patent Document 1 has a problem that the traffic state cannot be accurately simulated because it does not correspond to a general road where a bottleneck point exists due to the presence of a traffic light.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a traffic state simulation apparatus and program that can accurately simulate a traffic state.

  In order to achieve the above object, a traffic state simulation apparatus according to the present invention includes a road information storage unit storing road information indicating a road map and a road environment, and a signal for acquiring signal information indicating a cycle of change of a traffic light. Information acquisition means, OD traffic volume acquisition means for acquiring OD traffic volume indicating the departure place, departure time, and destination of each vehicle estimated based on information indicating the actual traffic state of the vehicle, and the road information Based on the signal information acquired by the signal information acquisition means and the OD traffic volume acquired by the OD traffic volume acquisition means, a vehicle model that models a vehicle is run to simulate a traffic state. And a simulation means.

  A program according to the present invention includes a computer, a road information storage unit that stores road information indicating a road map and a road environment, a signal information acquisition unit that acquires signal information indicating a change cycle of a traffic light, an actual vehicle traffic OD traffic volume acquisition means for acquiring the OD traffic volume indicating the departure place, departure time, and destination of each vehicle estimated based on the information indicating the state, and acquired by the road information and the signal information acquisition means It is a program for causing a vehicle model that models a vehicle to run based on the signal information and the OD traffic volume acquired by the OD traffic volume acquisition unit, and functioning as a simulation unit that simulates a traffic state. .

  According to the present invention, the road information indicating the road map and the road environment is stored in the storage means, and the signal information indicating the change cycle of the traffic light is acquired by the signal information acquisition means. Further, the OD traffic volume acquisition means acquires the OD traffic volume indicating the departure point, departure time, and destination of each vehicle estimated based on information indicating the actual traffic state of the vehicle. Here, as information indicating the actual traffic state of the vehicle, for example, information indicating the actual traffic volume and driving behavior of the vehicle can be used.

  Based on the road information, the signal information acquired by the signal information acquisition means, and the OD traffic volume acquired by the OD traffic volume acquisition means by the simulation means, the vehicle model that models the vehicle is run, Simulate the condition.

  As described above, by using the road information, the signal information, and the OD traffic volume, the traffic state can be accurately simulated.

  The traffic condition simulation apparatus according to the present invention acquires at least one of traffic volume per unit time, travel time, queue length, and congestion length of a vehicle in a predetermined section based on the actual traffic condition of the vehicle. Means, a calculation means for calculating at least one of traffic volume per unit time, travel time, queue length, and traffic jam length based on traffic conditions simulated in the past by the simulation means, and acquired by the acquisition means At least one of traffic volume per unit time, travel time, queue length, and traffic jam length, and at least one of traffic volume per unit time, travel time, queue length, and traffic jam length calculated by the calculation means So that at least one of the differences in traffic volume per unit time, travel time, queue length, and traffic jam length is reduced. OD traffic volume correction means for correcting the OD traffic volume acquired by the D traffic volume acquisition means, and the simulation means simulates the traffic state using the OD traffic volume corrected by the OD traffic volume correction means. be able to. By simulating the traffic state using the OD traffic amount corrected so as to eliminate the difference from the traffic amount based on the actual vehicle traffic state, travel time, queue length, or traffic jam length, traffic The state can be simulated accurately.

  The acquisition means acquires an average value of at least one predetermined period of traffic volume per unit time, travel time, queue length, and congestion length of a vehicle in a predetermined section based on an actual vehicle traffic state. be able to. As a result, the traffic state is adjusted using the OD traffic amount corrected so as to eliminate the difference with the average traffic amount, travel time, queue length, or congestion length for a predetermined period based on the actual vehicle traffic state. By simulating, it is possible to accurately simulate the traffic state for a predetermined period.

  The acquisition means can acquire a current value of at least one of the traffic volume per unit time, travel time, queue length, and congestion length of the vehicle in a predetermined section based on the actual vehicle traffic state. . By simulating traffic conditions using OD traffic, which is modified so that there is no difference from current traffic, travel time, queue length, or congestion length based on actual vehicle traffic The current traffic conditions can be simulated accurately.

  The acquisition means predicts at least one of traffic volume per unit time, travel time, queue length, and congestion length of a predetermined section at the time of prediction based on the actual traffic state of the vehicle. Thus, at least one predicted value of the traffic volume per unit time, travel time, queue length, and traffic jam length of the vehicle in the predetermined section can be acquired. As a result, the traffic using the OD traffic corrected so as to eliminate the difference from the traffic at the time of prediction predicted based on the actual traffic state of the vehicle, travel time, queue length, or traffic jam length, By simulating the state, the traffic state at the time of the prediction target can be simulated accurately.

  The simulation means can divide the OD traffic volume into the OD traffic volume for each time zone and each region and use it for the simulation of the traffic state.

  The OD traffic volume acquisition means includes travel history information indicating a travel history of each of a plurality of vehicles, statistical information on the traffic volume of each vehicle on each of a plurality of roads, and a starting point and a destination of the vehicle. The OD traffic volume can be estimated based on at least one of the statistical information regarding the OD traffic volume.

  As described above, according to the traffic state simulation apparatus and program of the present invention, the effect that the traffic state can be accurately simulated by using road information, signal information, and OD traffic volume can be obtained. .

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

  As shown in FIG. 1, the traffic state simulation device 10 according to the first embodiment collects probe car information indicating a travel history (travel position, travel time, travel speed, etc.) transmitted from the probe car 12. Based on the probe information collection management unit 14 to be managed, the probe information database 16 storing the collected probe information, and the probe information stored in the probe information database 16, signal information indicating the change period of the traffic light is estimated. The signal information estimation unit 18, the signal database 20 that stores the estimated signal information, the VICS information collection management unit 24 that collects and manages the VICS (Vehicle Information and Communication System) information from the VICS center 22, and the collected VICS VI storing information Based on the CS information database 26 and the probe information and VICS information, the average travel time, the signal queue length, and the traffic jam length are estimated for each link, and the target value in the OD traffic volume correction unit 42 described later is used. The traffic information processing unit 28 calculates an average traffic volume for each link.

  The probe information collection management unit 14 receives and collects probe car information indicating the travel history including the travel position, travel time, and travel speed of the probe car 12 from the probe car 12 and stores the probe car information in the probe information database 16.

  The signal information estimation unit 18 estimates the blue time and red time of each traffic light from the stop time and the start time of the probe car 12 obtained from the probe car information, and from each of the estimated blue time and red time of each traffic light, Signal information indicating a change period of the traffic light is estimated and stored in the signal database 20.

  The VICS information collection management unit 24 collects VICS information including road traffic information such as traffic jam information from the VICS center 22 and stores it in the VICS information database 26.

  As will be described below, the traffic information processing unit 28 estimates each of the average travel time, signal queue length, and traffic jam length for each link (predetermined section) for which traffic information can be estimated. Traffic volume is calculated. The links for which traffic information can be estimated are a part of all links on all roads to be calculated in the simulation.

  First, for each link for which traffic information can be estimated, the average travel time of the link to be estimated is estimated based on the probe car information and the VICS information regarding the link to be estimated, and the average of the link to be estimated Estimate the congestion length.

In addition, when a traffic jam occurs, it is considered that the demand is exceeded by the number of cars that stayed, rather than the traffic volume that can be gained in one signal cycle. Therefore, the average traffic volume of the link per unit time is calculated from the estimated average traffic congestion length of the link according to the following equation (1).
Traffic = Link capacity x Traffic light time ratio + Congestion length ÷ Average vehicle length
... (1)

  Here, the link capacity is the road capacity of the target link obtained from road network information stored in the road network database 36 to be described later, and the signal green time ratio is the traffic signal of the target link stored in the signal database 20. Obtained from signal information. The average vehicle length is a value obtained in advance.

  In addition, since the signal queue length of the target link can be estimated from the stop position of the probe car without traffic jams, the average link of the estimation target link can be estimated based on the probe car information about the estimation target link. The signal queue length is estimated.

  Here, it may be assumed that the arrival number follows a Poisson distribution. The probe car is not necessarily at the tail end of the signal queue, but on average, it is considered that the probe car is located at an intermediate point of the signal queue. Therefore, twice the distance from the signal position to the stop position is Estimated as queue length.

  When the traffic volume is low, the vehicle stops near the intersection, and when the traffic volume is high, there is a vehicle ahead, so the vehicle stops away from the intersection. Thus, the average traffic volume of the target link is estimated based on the estimated length of the signal queue.

For example, assuming that the vehicles arrive at the tail of the signal queue at equal intervals, the vehicle head time can be calculated based on the signal queue length. Then, when the vehicle arrives at the tail end of the signal queue for each calculated vehicle head time, the number of vehicles that stop during the time when the traffic light is in a red signal state is expressed by the following equation (2). Calculated.
Number of stops = signal red time ÷ vehicle head time × average vehicle length (2)

  Then, based on the relationship between the number of stops and the traffic volume as shown in FIG. 2, the traffic volume per unit time corresponding to the number of stops calculated by the above equation (2) is obtained, and the average of the target link is calculated. Estimated traffic volume. Based on the relationship between the vehicle head time and the traffic volume as shown in FIG. 2, the traffic volume per unit time corresponding to the calculated vehicle head time is obtained, and the average traffic volume estimated value of the target link is calculated. It is good.

Further, the average traffic volume per unit time of the link is calculated from the estimated average traffic jam length of the link according to the relational expression representing the relationship between the traffic volume and the travel time called a link performance function. For example, the traffic volume per unit time of a link can be calculated using a modified BPR function as shown in the following equation (3).
t (x) = t {1 + α (x / C) ^β } (3)

  Here, t (x) represents the link travel time when the traffic volume is x, and t represents the travel time during free travel. Further, x represents the traffic volume per unit time of the link, and C represents the traffic capacity of the link. Α and β are predetermined parameters, for example, α = 2.62 and β = 5.00.

  Also, as shown in FIG. 3, the traffic volume per unit time of the target link may be calculated from the relationship between the traffic volume of the link and the travel time.

  For each link for which traffic information can be estimated, the average traffic volume per unit time calculated by each of the above calculation methods is averaged, and traffic information used by the OD traffic volume correction unit 42 is estimated. The target value of average traffic volume per unit time of each possible link is used.

  In addition, as shown in FIG. 1, the traffic state simulation device 10 includes a statistical information database 30 that stores statistical information indicating the results of investigations such as road traffic census and person trip surveys, and a vehicle travel history obtained from probe information. OD traffic volume estimation unit 32 that estimates the wide area OD traffic volume based on the travel history information and the statistical information of the statistical information database 30, and the initial OD that stores the estimated wide area OD traffic volume as the initial value of the OD traffic volume Traffic volume storage unit 34, road network database 36 that stores road network information indicating road maps and road environments, estimated signal information, road network information, and initial value OD traffic volume storage unit 34 or modified OD traffic volume Simulate traffic conditions based on wide-area daily OD traffic volume stored in storage unit 44 A traffic information calculation unit 38, a traffic information acquisition unit 40 that calculates an average traffic volume per unit time of each link for which traffic information can be estimated, and a traffic information processing unit 28 that calculates traffic information. Based on the target value of the average traffic volume per unit time of the link and the average traffic volume per unit time of each link acquired by the traffic information acquisition unit 40, the estimated daily wide OD traffic volume An OD traffic volume correction unit 42 to be corrected, a corrected OD traffic volume storage unit 44 that stores the corrected wide-area day OD traffic volume, and a traffic information provision unit 46 that provides traffic information obtained from a simulation result are provided. .

  The statistical information database 30 includes, as statistical information, road traffic census survey results that investigate the traffic volume of road vehicles per unit time and survey results of person trip surveys that survey departure points, destinations, and means of transportation. Is stored.

  As described below, the OD traffic volume estimation unit 32 estimates the wide area daily OD traffic volume based on the probe information and the statistical information. In this embodiment, the wide area daily OD traffic volume is estimated for each zone and for each day.

  Wide area daily OD traffic volume is estimated based on road traffic census OD survey forms and person trip surveys, and is estimated based on wide-area aggregate units such as zones handled in these survey forms. Regarding the OD traffic volume between zones in the target area, the survey data is used as it is, and the traffic volume that passes through the target area, the traffic volume from the target area to the outside, and the traffic volume from the outside of the target area to the inside, Obtained by calculating the distribution from the surrounding area. Thus, the OD traffic volume estimated from the survey data is set as the initial OD traffic volume.

  The road network database 36 stores road network information that is constructed from a DRM (digital road map) or a navigation map and indicates the number of road lanes and road types as a road environment and road connection relationships as a road map. Has been.

  The traffic condition is determined by the relationship between the traffic demand by vehicles traveling in the area and the road capacity, which is the “volume” traffic volume at the passing road points. There will be a loss of profit and traffic jams will occur.

  Therefore, the traffic state calculation unit 38 runs a vehicle model obtained by modeling a vehicle and simulates the traffic state on all the calculation target roads as described below.

  First, the road network information, signal information, and wide area daily OD traffic volume are used as input values, and the road capacity, the number of road lanes and road types indicated by the road network information, and the traffic light blue time ratio obtained from the estimated signal information And decide based on. Further, the estimated OD traffic volume is used as the traffic demand.

  In addition, as shown in the following equation (4), the regional daily OD traffic volume is divided into a OD traffic volume for each small area (for example, 3rd mesh) from the wide area OD traffic volume, and the daily OD traffic volume Divide the amount of time into 1 hour or 30 minute OD traffic volume. Note that the division ratio of the regional division may be determined in advance from statistical values such as a residential population ratio, a working population ratio, and an employee population ratio. Further, the division ratio of the time zone division may be determined in advance from statistics such as changes in the observed traffic volume.

  Then, using the determined road capacity and the divided OD traffic volume, each vehicle model is virtually run on all the calculation target roads to simulate the traffic state. In the simulation of the traffic state, the route for the combination of the starting point and the destination is changed according to the traffic state based on a preset route selection model. In addition, about the simulation method of a traffic state, what is necessary is just to use a conventionally known simulation method, and abbreviate | omits detailed description.

  Based on the simulated traffic state, the traffic information acquisition unit 40 calculates the error in the OD traffic correction unit 42, and calculates the unit for each link whose traffic information can be estimated as the calculated value in the OD traffic correction unit 42. Calculate the average traffic volume per hour, and provide the traffic information based on the simulated traffic conditions and calculate the average travel time and congestion length for each link on the calculation target road. calculate. Moreover, the traffic information acquisition part 40 records the OD information of each vehicle which flows into a link, and the generation time zone in a link as information relevant to the traffic volume for every link based on the simulated traffic state.

  Here, the principle of the present embodiment will be described. The OD traffic volume is almost impossible to observe and needs to be estimated based on statistical information obtained from road traffic census and person trip surveys. However, in the statistical information, since the sampling rate is low, the estimated OD traffic volume has a larger error than the actual OD traffic volume.

  In addition, since the travel of probe cars is part of the traffic volume, if the penetration rate of probe cars is high and the ratio of the amount of collected probe information to the traffic volume can be estimated, OD can be calculated from the probe information. It is possible to estimate the traffic volume. However, the current penetration rate of probe cars is significantly lower than the penetration rate that can estimate the OD traffic volume, and the estimated OD traffic volume has a larger error than the actual OD traffic volume.

  As described above, since the estimation accuracy of the OD traffic volume used in the simulation is lowered, the accuracy of the simulated traffic state is also lowered.

  Therefore, the OD traffic correction unit 42 corrects the estimated OD traffic as described below so that an OD traffic close to the actual OD traffic can be obtained.

  First, the calculated average traffic volume per unit time of each link calculated by the traffic information acquisition unit 40 and the average traffic volume per unit time of each link calculated by the traffic information processing unit 28. Compared with the target value, the mean square of the error in the traffic volume per unit of each link for which traffic information can be estimated is obtained as an index of the degree of coincidence of the simulated traffic conditions according to the following equation (5). The obtained value is regarded as an error.

ただし、Ｌは、交通情報が推定可能なリンクの集合とする。

Here, L is a set of links for which traffic information can be estimated.

  When the error calculated by the above equation (5) is large, the OD traffic volume passing through the link for which traffic information can be estimated is corrected so that the error is reduced. If the unit for correcting the OD traffic is the wide-area daily OD traffic, the change in the OD traffic divided by region and time zone in response to the change in the wide-area daily OD traffic is based on the relationship of the above equation (4). It is obtained by the following equation (6).

  Further, when the OD traffic volume divided into regions and time zones is changed, the traffic volume change amount per unit time of the link is generated with the OD information of each vehicle flowing into the link recorded in the traffic information acquisition unit 40. The OD traffic volume obtained based on the time zone information is obtained by the following equation (7) using the probability of passing through a link for which traffic information can be estimated.

  And the following (8) Formula is obtained from the said (5) Formula and (7) Formula.

  While changing the daily OD traffic volume in the wide area, the change in the OD traffic volume divided into regions and time zones according to the change in the regional OD traffic volume in accordance with the above formula (6) is changed. ) Calculate the amount of change in the daily OD traffic volume that minimizes the error calculated by the equation. Then, using the obtained change amount of the wide-area day OD traffic volume, the wide-area daily OD traffic volume used in the previous simulation is corrected to the wide-area daily OD traffic volume with a small traffic error.

  The traffic information providing unit 46 includes a display, and displays traffic information (link travel time, congestion length, etc.) on the calculation target road obtained from the traffic state simulated by the traffic state calculation unit 38 on the display. Provide to the operator.

  Each of the probe information database 16, the signal database 20, the VICS information database 26, the statistical information database 30, and the road network database 36 is configured by an HDD. Further, the probe information collection management unit 14, the signal information estimation unit 18, the VICS information collection management unit 24, the traffic information processing unit 28, the OD traffic volume estimation unit 32, the initial OD traffic volume storage unit 34, the traffic state calculation unit 38, the traffic Each of the information acquisition unit 40, the OD traffic volume correction unit 42, and the corrected OD traffic volume storage unit 44 is configured by a CPU, a ROM, or a RAM, and includes a traffic information processing unit 28, an OD traffic volume estimation unit 32, and a traffic state calculation. Each of the unit 38, the traffic information acquisition unit 40, and the OD traffic amount correction unit 42 is realized by a program for executing a simulation processing routine described later.

  Next, the operation of the traffic state simulation device 10 according to the first embodiment will be described. First, in the traffic state simulation device 10, probe information transmitted from the probe car 12 is collected and stored in the probe information database 16, and VICS information from the VICS center 22 is collected and stored in the VICS information database 26. Remember me. Further, statistical information by road traffic census and person trip survey is input to the traffic state simulation device 10 by the operator and stored in the statistical information database 30. Further, the traffic state simulation device 10 estimates signal information indicating a change cycle of each traffic signal on all roads to be calculated based on the probe information stored in the probe information database 16, and signals the estimation result. Store in database 20.

  Then, when a simulation start instruction is input by the operator, the simulation processing routine shown in FIG.

  First, in step 100, for each link for which traffic information can be estimated based on the probe information in the probe information database 16 and the VICS information in the VICS information database 26, the average signal queue length, traffic jam length, and travel time are calculated. Estimate each. Then, in step 102, for each link for which traffic information can be estimated based on the average signal queue length, traffic jam length, and travel time in each link estimated in step 100 above, By calculating the average traffic volume and calculating the average value of these traffic volumes for each link, the target value of average traffic volume per unit time for each link for which traffic information can be estimated is calculated. .

  In step 104, on the basis of the probe information in the probe information database 16 and the statistical information in the statistical information database 30, as an initial value of the OD traffic volume used in the simulation, for each zone on all roads to be calculated, on a daily basis. Estimate wide-area daily OD traffic volume. In step 106, the wide-area daily OD traffic estimated in step 104 is set as the wide-area daily OD traffic used in the simulation.

  In the next step 108, the regional daily OD traffic volume set in step 106 is divided into time zones and regions. In step 110, road network information is acquired from the road network database 36 and signal information is acquired from the signal database 20.

  In step 112, based on the OD traffic volume divided in step 108 and the road network information and signal information acquired in step 110, the vehicle model is virtually run to simulate the traffic state. To do.

  In the next step 114, an average traffic volume per unit time is calculated for each link for which traffic information can be estimated based on the traffic state simulated in the above step 112. The target value of the average traffic volume per unit time of each link calculated is compared with the average traffic volume per unit time of each link calculated in step 114, and the above equation (5) The error is calculated according to

  In step 118, it is determined whether or not the error calculated in step 116 is equal to or less than a threshold value. If the calculated error is larger than the threshold, in step 120, as described above, the wide-area daily OD traffic volume set in step 106 is set so that the error calculated in step 116 is reduced. Is corrected, the process returns to step 106, the corrected wide-area daily OD traffic volume is set, and the traffic state is simulated again.

  If the calculated error in step 118 is less than or equal to the threshold value, it is determined that the traffic condition is simulated using the OD traffic volume that is close to the actual OD traffic volume, and the simulation is performed in step 122. Traffic information on the calculation target road obtained from the traffic condition (travel time and congestion length for the desired link among all links on all roads) is displayed on the display, providing traffic information to the operator, and simulation processing End the routine.

  As described above, according to the traffic state simulation apparatus according to the first embodiment, the travel time, queue length, and congestion length estimated from the probe information and VICS information indicating the actual traffic state of the vehicle. Based on the OD traffic volume, signal information, and road network information, using the OD traffic volume corrected so that there is no difference from the traffic volume of the link based on the traffic status, the traffic status is accurately simulated. can do.

  In addition, even if the coverage rate of infrastructure sensors is low and most of the roads cannot be observed even if VICS information and probe information are used, the traffic state of all links in the road network is calculated by using traffic simulation. can do. In addition, it is possible to provide traffic information for the entire area of the traffic information providing service target area, and efficient route guidance and the like are possible.

  Moreover, accurate traffic demand can be estimated by estimating the OD traffic volume which has a subject in data acquisition from VICS information and probe information.

  Further, by correcting the OD traffic volume so as to eliminate the error from the actual traffic state, it is possible to estimate the OD traffic volume that is close to the actual OD traffic volume.

  Next, a second embodiment will be described. In addition, since the traffic state simulation apparatus according to the second embodiment has the same configuration as that of the first embodiment, the same reference numerals are given and description thereof is omitted.

  In the second embodiment, the link traffic volume in the calculation target period is calculated, and the error is calculated in comparison with the link traffic volume calculated from the simulated traffic state. Is different.

  In the traffic information processing unit 28 of the traffic state simulation device according to the second embodiment, as described below, for each link (predetermined section) for which traffic information can be estimated, the average travel time in the calculation target period Estimate signal queue length, traffic jam length, and traffic volume. Note that the calculation target period may be determined by time or by day of the week. Further, the calculation target period may be determined according to the weather.

  First, for each link for which traffic information can be estimated, the average travel time in the calculation target period of the estimation target link is estimated based on the probe car information and the VICS information regarding the estimation target link and the calculation target engine, Estimate the average traffic jam length during the calculation target period of the target link.

  For each link for which traffic information can be estimated, the average traffic volume of the link per unit time in the calculation target period is calculated from the estimated average traffic jam length of the link according to the above equation (1).

  In addition, for each link for which traffic information can be estimated, the average signal queue length in the calculation target period of the estimation target link is estimated and estimated based on the probe car information about the estimation target link and the calculation target period. Based on the length of the signal queue, the average traffic volume in the calculation target period of the estimation target link is estimated.

  For each link for which traffic information can be estimated, the average travel time in the calculation target period of the link per unit time is calculated from the average travel time in the calculation target period of the link using the above equation (3). Calculate traffic volume.

  For each link for which traffic information can be estimated, the average traffic volume per unit time of the link is calculated by averaging the average traffic volume in the calculation target period per unit time calculated by the above calculation methods. Target value.

  The OD traffic volume correction unit 42 corrects the estimated OD traffic volume in the same manner as in the first embodiment so that an OD traffic volume close to the OD traffic volume in the calculation target period can be obtained.

  Next, the operation of the traffic state simulation device according to the second embodiment will be described. First, when a calculation target period is input to the traffic state simulation device and a simulation start instruction is input by the operator, a simulation processing routine is executed in the traffic simulation device.

  First, based on the probe information related to the calculation target period in the probe information database 16 and the VICS information related to the calculation target period in the VICS information database 26, the average signal queue length of the calculation target period in each link for which traffic information can be estimated. Estimate traffic jam length and travel time. The unit time of each link for which traffic information can be estimated based on each of the estimated average signal queue length, traffic jam length, and travel time of the calculation target period in each link for which traffic information can be estimated The average traffic volume per unit time is calculated for each link for which traffic information can be estimated by calculating the average traffic volume for each calculation period and calculating the average value of these traffic volumes for each link. Calculate the target value of the quantity.

  Then, as the initial value of the OD traffic volume used in the simulation, the regional daily OD traffic volume is estimated on a daily basis for each zone on the calculation target road, and the estimated wide area daily OD traffic volume is used in the simulation. Set as daily OD traffic volume.

  Next, the set wide area daily OD traffic volume is divided into time zones, divided into regions, road network information is acquired from the road network database 36, and signal information is acquired from the signal database 20. Based on the divided OD traffic volume and the acquired road network information and signal information, the vehicle model is virtually run to simulate the traffic state.

  Next, based on the simulated traffic conditions, the average traffic volume per unit time is calculated for each link for which traffic information can be estimated, and the average per unit time of each link calculated above is calculated. An error is calculated by comparing the target value of traffic volume with the average traffic volume per unit time of each link calculated based on the simulation result.

  Then, it is determined whether or not the calculated error is equal to or less than a threshold value. If the calculated error is larger than the threshold value, the set wide-area day OD traffic volume is corrected so that the error becomes small. And simulate the traffic condition again.

  On the other hand, when the calculated error is less than or equal to the threshold, it is determined that the traffic state simulation in the calculation target period has been performed using the OD traffic volume close to the actual calculation target period. The traffic information obtained from the traffic state (travel time of desired link and congestion length) is displayed on the display as information on the calculation target period, the traffic information is provided to the operator, and the simulation processing routine is terminated.

  In this way, there is no difference between the average travel time, queue length, and traffic volume of the link based on the traffic jam length estimated from the probe information and VICS information indicating the actual traffic state of the vehicle. By simulating the traffic state using the OD traffic volume corrected to the above, it is possible to accurately simulate the traffic state in the calculation target period.

  Next, a third embodiment will be described. In addition, since the traffic state simulation apparatus according to the third embodiment has the same configuration as that of the first embodiment, the same reference numerals are given and description thereof is omitted.

  The third embodiment is different from the first embodiment in that the current link traffic volume is calculated, and an error is calculated by comparing with the link traffic volume calculated from the simulated traffic state. ing.

  In the traffic information processing unit 28 of the traffic condition simulation apparatus according to the third embodiment, as will be described below, for each link (predetermined section) for which traffic information can be estimated, the current travel time, the signal queue length Estimate traffic congestion length and traffic volume.

  First, for each link for which traffic information can be estimated, the current travel time of the estimation target link is estimated based on the current probe car information and VICS information regarding the estimation target link, and the current of the estimation target link Estimate the congestion length.

  Further, for each link for which traffic information can be estimated, the current traffic volume of the link per unit time is calculated from the estimated congestion length of the link according to the above equation (1).

  For each link for which traffic information can be estimated, the current signal queue length of the estimated link is estimated based on the current probe car information about the estimated link, and the estimated signal queue length Based on the above, the current traffic volume of the estimation target link is estimated.

  For each link for which traffic information can be estimated, the current traffic volume per unit time of the link is calculated from the estimated current travel time using the above equation (3).

  For each link for which traffic information can be estimated, the average traffic volume per unit time for each link for which traffic information can be estimated by averaging the current traffic volume for each link calculated by the above calculation methods. The target value for traffic volume.

  As described in the first embodiment, the initial OD traffic volume storage unit 34 stores the wide-area daily OD traffic volume corrected using the average traffic volume per unit time.

  The traffic state calculation unit 38 stores the wide-area daily OD traffic volume stored in the initial OD traffic volume storage unit 34 or the modified OD traffic volume storage unit 44, the road network information stored in the road network database 36, and the signal database 20. Based on the stored signal information, a vehicle model obtained by modeling the vehicle is run to simulate the traffic state on all the calculation target roads.

  The OD traffic volume correction unit 42 corrects the estimated OD traffic volume in the same manner as in the first embodiment so that an OD traffic volume close to the current OD traffic volume can be obtained.

  Next, the operation of the traffic state simulation device according to the third embodiment will be described. First, when an instruction is input by the operator to simulate the current traffic state into the traffic state simulation device, and when a simulation start instruction is input, a simulation processing routine is executed in the traffic simulation device.

  First, based on the current probe information stored in the probe information database 16 and the current VICS information stored in the VICS information database 26, the current signal queue length and congestion length in each link for which traffic information can be estimated. , And travel time, respectively. Then, based on each of the estimated current signal queue length, congestion length, and travel time in each link, the traffic volume per unit time is calculated for each link for which traffic information can be estimated, By calculating an average value of these traffic volumes for each link, an average traffic volume target value per unit time of each link for which traffic information can be estimated is calculated.

  As an initial value of the OD traffic volume used in the simulation, an error from the target value of the average traffic volume per unit time calculated from the average link travel time, the signal queue length, and the traffic jam length becomes small. The wide area day OD traffic volume corrected in this way is set as the wide area day OD traffic volume used in the simulation.

  Next, the set wide area daily OD traffic volume is divided into time zones, divided into regions, road network information is acquired from the road network database 36, and signal information is acquired from the signal database 20. Based on the divided OD traffic volume and the acquired road network information and signal information, the vehicle model is virtually run to simulate the traffic conditions on all roads.

  Next, based on the simulated traffic conditions, the average traffic volume per unit time of each link for which traffic information can be estimated is calculated, and the average traffic per unit time of each link calculated above is calculated. An error is calculated by comparing the target value of the volume with the average traffic volume per unit time of each link calculated based on the simulation result.

  Then, it is determined whether or not the calculated error is equal to or less than a threshold value. If the calculated error is larger than the threshold value, the set wide-area day OD traffic volume is corrected so that the error becomes small. And simulate the traffic condition again.

  On the other hand, if the calculated error is less than or equal to the threshold value, it is determined that a simulation of the current traffic state has been performed using an OD traffic volume that is close to the actual current OD traffic volume. The obtained traffic information (travel time of the desired link and congestion length) is displayed on the display as current information, the traffic information is provided to the operator, and the simulation processing routine is terminated.

  In this way, the OD has been modified so that there is no difference between the current travel time, queue length, and link traffic volume based on the traffic jam length estimated from probe information and VICS information indicating the actual traffic state of the vehicle. By simulating the traffic state using the traffic volume, the current traffic state can be accurately simulated.

  In the first to third embodiments, an example of calculating the traffic volume per unit time of the link based on the travel time of the link, the signal queue length, and the traffic jam length is taken as an example. However, the present invention is not limited to this, and the traffic volume per unit time of the link may be actually observed, and the observed value may be used as the target value in the OD traffic volume correction unit. In this case, for each observable link, the observed traffic volume per unit time of the link is compared with the calculated traffic volume per unit time of the link calculated from the simulation. Should be corrected.

  Also, the case where the link travel time, the signal queue length, and the traffic jam length are estimated from the probe information and the VICS information has been described as an example, but the present invention is not limited to this, and the link travel time, the signal queue If the length and the congestion length can be actually observed, these observation values may be used.

  Next, a fourth embodiment will be described. In addition, about the part which becomes the structure similar to 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In the fourth embodiment, the future travel time of the link, the signal queue length, and the traffic jam length are predicted, and the target value of the traffic volume per unit time of the link is calculated based on these predicted values. This is different from the first embodiment.

  As shown in FIG. 5, the traffic state simulation apparatus 410 according to the fourth embodiment includes a probe information collection management unit 14, a probe information database 16, a signal information estimation unit 18, a signal database 20, and VICS information. The collection management unit 24, the VICS information database 26, the travel time up to the current time of each link for which traffic information can be estimated, the signal queue length, and the traffic jam length are estimated, and the prediction processing unit 432 described later Based on the estimated travel time, signal queue length, and traffic jam length for each link for which traffic information can be estimated, the average traffic volume is calculated as a target value for each link for which traffic information can be estimated. Traffic information processing unit 428 and actual measurement information storing travel time, signal queue length, and traffic jam length measured for each link in the past The travel time at the time of prediction for each link for which traffic information can be estimated based on the past measured values stored in the database 430, the measured information database 430, and the estimated values estimated by the traffic information processing unit 428, Prediction processing unit 432 that predicts each of signal queue length and traffic jam length, statistical information database 30, OD traffic volume estimation unit 32, initial OD traffic volume storage unit 34, road network database 36, and traffic state The calculation part 38, the traffic information acquisition part 40, the OD traffic volume correction part 42, and the traffic information provision part 46 are provided.

  For each link for which traffic information can be estimated, the traffic information processing unit 428 calculates the average travel time of the estimated link up to the current time based on the probe car information and VICS information up to the current time regarding the estimated link. And an average traffic jam length of the link to be estimated up to the current time. In addition, the traffic information processing unit 428 calculates, for each link for which traffic information can be estimated, an average signal queue of the estimation target link up to the current time, based on the probe car information up to the current time regarding the estimation target link. Estimate the length.

  For each of the travel time of the link, the signal queue length, and the traffic jam length, the prediction processing unit 432 calculates the past measured values stored in the measured information database 430 and the current time estimated by the traffic information processing unit 428. Based on the estimated value, the predicted value at the time of the prediction target is predicted, and for each link for which traffic information can be estimated, the travel time at the time of the prediction target, the signal queue length, and the congestion length are obtained. As a prediction method, a method such as autoregression, Kalman filter, or neural network may be used.

  Further, the traffic information processing unit 428 calculates the traffic volume of the link per unit time at the time of the prediction according to the above equation (1) from the predicted traffic jam length of each link for which the traffic information can be estimated. .

  Further, the traffic information processing unit 428 predicts the traffic volume at the time of the prediction target of the estimation target link based on the predicted length of the signal queue for each link for which the traffic information can be estimated. For each link for which information can be estimated, the traffic volume per unit time at the time of link prediction is calculated from the predicted travel time at the time of prediction.

  For each link for which traffic information can be estimated, the traffic information processing unit 428 averages the traffic volume per unit time at the time of the link prediction target calculated by each of the above calculation methods, and each traffic information can be estimated. The target value for the average traffic volume per unit time of the link.

  In the initial OD traffic volume storage unit 34, as in the third embodiment, the wide-area daily OD traffic volume corrected using the average traffic volume per unit time of each link for which traffic information can be estimated. Is remembered.

  Next, the operation of the traffic state simulation device according to the fourth embodiment will be described. First, when an operator inputs information indicating the prediction target time to the traffic state simulation apparatus and when a simulation start instruction is input, a simulation processing routine is executed in the traffic simulation apparatus.

  First, based on the probe information up to the current time stored in the probe information database 16 and the VICS information up to the current time stored in the VICS information database 26, the signals up to the current time in each link for which traffic information can be estimated. Estimate queue length, traffic jam length, and travel time. Further, based on the past actual measurement values stored in the actual measurement information database 430 and the estimated signal queue length, traffic jam length, and travel time up to the current time, the signal queue length, traffic jam length, and travel time are calculated. For each, a predicted value at the time of prediction of each link for which traffic information can be estimated is predicted.

  Then, based on each of the signal queue length at the time of prediction in each link for which the predicted traffic information can be estimated, the traffic jam length, and the travel time, the unit time at the time of the prediction target of each link for which the traffic information can be estimated Each traffic volume is calculated, and an average value of these traffic volumes is calculated for each link, thereby calculating an average traffic volume target value per unit time of each link for which traffic information can be estimated.

  As an initial value of the OD traffic volume used in the simulation, an error from the target value of the average traffic volume per unit time calculated from the average link travel time, the signal queue length, and the traffic jam length becomes small. The wide area day OD traffic volume corrected in this way is set as the wide area day OD traffic volume used in the simulation.

  Next, the set wide area daily OD traffic volume is divided into time zones, divided into regions, road network information is acquired from the road network database 36, and signal information is acquired from the signal database 20. Based on the divided OD traffic volume and the acquired road network information and signal information, the vehicle model is virtually run to simulate the traffic state.

  Next, based on the simulated traffic conditions, the average traffic volume per unit time of each link for which traffic information can be estimated is calculated, and the average traffic per unit time of each link calculated above is calculated. An error is calculated by comparing the target value of the volume with the average traffic volume per unit time of each link calculated based on the simulation result.

  Then, it is determined whether or not the calculated error is equal to or less than a threshold value. If the calculated error is larger than the threshold value, the set wide-area day OD traffic volume is corrected so that the error becomes small. And simulate the traffic condition again.

  On the other hand, if the calculated error is less than or equal to the threshold value, it is determined that the traffic state simulation at the time of the prediction target has been simulated using the OD traffic volume that is close to the actual OD traffic volume at the time of the prediction target. The traffic information obtained from the traffic state is displayed on the display as information at the time of the prediction target, the traffic information is provided to the operator, and the simulation processing routine is terminated.

  Thus, based on the estimated travel time, queue length, and congestion length estimated from the probe information and VICS information indicating the actual traffic state of the vehicle, and the actual measured value in the past, the travel at the time of the prediction target Estimate the time, queue length, and congestion length, and change the OD traffic volume so that there is no difference from the traffic volume of the link based on the predicted travel time, queue length, and congestion length By using and simulating the traffic state, it is possible to accurately simulate the traffic state at the time of prediction.

  In the first to fourth embodiments, an example of calculating the traffic volume per unit time of the link from each of the link travel time, the signal queue length, and the traffic jam length is an example. However, the present invention is not limited to this, and the traffic volume per unit time of the link may be calculated from at least one of the travel time of the link, the signal queue length, and the congestion length.

  Next, a fifth embodiment will be described. In addition, since the traffic state simulation apparatus according to the fifth embodiment has the same configuration as that of the first embodiment, the same reference numerals are given and description thereof is omitted.

  In the fifth embodiment, the OD traffic volume is set so that the error becomes smaller compared to the link travel time estimated from the probe information or VICS information and the link travel time calculated from the simulated traffic condition. The modification is different from the first embodiment.

  In the traffic state simulation device according to the fifth embodiment, the traffic information processing unit 28 estimates the link of the estimation target for each link whose traffic information can be estimated based on the probe car information and the VICS information regarding the link of the estimation target. The average travel time is estimated and set as the target value.

  The traffic information acquisition unit 40 calculates an average travel time for each link for which traffic information can be estimated for error calculation in the OD traffic volume correction unit 42 based on the simulated traffic state, and calculates the calculated value and To do.

  The OD traffic volume correction unit 42 corrects the estimated OD traffic volume as described below so that an OD traffic volume close to the actual OD traffic volume can be obtained.

  First, the calculated average travel time of each link calculated by the traffic information acquisition unit 40 is compared with the average travel time target value of each link calculated by the traffic information processing unit 28. Similar to equation (5), the mean square of the travel time error of each link for which traffic information can be estimated is obtained and used as the error.

  When the calculated error is large, as in the first embodiment, the OD traffic volume passing through the link on which the traffic information can be estimated is corrected so as to reduce the error.

  Note that other configurations and processes of the traffic state simulation apparatus are the same as those in the first embodiment, and thus description thereof is omitted.

  In this way, by simulating the traffic state using the OD traffic amount corrected so as to eliminate the difference from the link travel time estimated from the probe information indicating the traffic state of the actual vehicle and the VICS information, the traffic state is obtained. The state can be simulated accurately.

  In the above embodiment, the case where the travel time of the link estimated from the probe information or VICS information is set as the target value has been described as an example. However, the present invention is not limited to this. Alternatively, at least one of link travel time, signal queue length, and traffic jam length may be estimated as a target value. In this case, at least one of the link travel time, signal queue length, and traffic jam length is calculated from the simulation result to obtain a calculated value, and the OD traffic volume is set so that the error between the target value and the calculated value is reduced. You just have to fix it.

  Next, a sixth embodiment will be described. In addition, about the part which is the same structure as 1st Embodiment, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In the sixth embodiment, the correlation coefficient between links is calculated from the simulated traffic state, and the travel time of another link is calculated from the travel time of the link calculated using the correlation coefficient. The point that the time is estimated is mainly different from the first embodiment.

  Routes provided by VICS information etc. are mainly main trunk lines. When there is a general road parallel to the main trunk line, if the main trunk line is congested, the general road tends to be congested as traffic avoiding the congestion. In addition, on the road that intersects the main trunk line, when there is a lot of traffic on the main trunk line, the amount of traffic flowing in increases, so the traffic volume increases in the same tendency as the traffic volume on the main trunk line. Thus, it is conceivable that there is a high correlation between the traffic state of a certain road and the traffic state of a road parallel to or intersecting with the road.

For example, in major cities, traffic information regarding links of some main trunk lines such as L _{O1 to} L _O12 may be always available on roads as shown in FIG. Further, it is considered that L _{E1 to} L _E4 parallel to these main trunk links have a correlation in traffic state.

  Therefore, in the sixth embodiment, in the traffic state simulation device, as in the first embodiment, from the traffic state simulated using the corrected OD traffic volume, The average travel time of each link is calculated, and the correlation coefficient between all the links is calculated from the average travel time of each link.

  When the travel time of a link for which traffic information can be estimated is estimated from the collected probe information, and the travel time of a link other than the link for which traffic information can be estimated is estimated, the correlation coefficient between the links is calculated. To estimate the travel time of the link to be estimated.

For example, assuming that the travel time of the link L _x for which traffic information can be estimated is T _x and the correlation coefficient of the links L _x and L _y is K _xy , based on the travel time of the link L _x , the following ( According to the equation (9), travel times of links L _{E1 to} L _E4 other than the links for which traffic information can be estimated can be calculated.

  In this way, even if the traffic situation is not available and the travel time of the link cannot be estimated for all links, the travel time of the link that can be estimated and the correlation coefficient calculated from the simulation result are used. , The travel time of all links can be estimated.

  In the above embodiment, the case where the travel time of the link is estimated using the correlation coefficient has been described as an example. However, the present invention is not limited to this, and the link signal is calculated using the correlation coefficient. You may make it estimate queue length and congestion length. In this case, the correlation coefficient may be calculated based on the signal queue length and the traffic jam length of all links calculated from the simulated traffic state.

It is a block diagram which shows the traffic state simulation apparatus which concerns on the 1st Embodiment of this invention. It is a graph which shows the relationship between the number of stops and traffic volume, and the relationship between vehicle head time and traffic volume. It is a graph which shows the relationship between travel time and traffic. It is a flowchart which shows the content of the simulation process routine in the traffic state simulation apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the traffic state simulation apparatus which concerns on the 4th Embodiment of this invention. It is an image figure which shows the link on a road.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10,410 Traffic state simulation apparatus 12 Probe car 16 Probe information database 18 Signal information estimation part 20 Signal database 26 VICS information database 28, 428 Traffic information processing part 30 Statistical information database 32 OD traffic volume estimation part 36 Road network database 38 Traffic state Calculation unit 40 Traffic information acquisition unit 42 OD traffic volume correction unit 428 Traffic information processing unit 430 Actual measurement information database 432 Prediction processing unit

Claims (8)

Road information storage means for storing road information indicating a road map and a road environment;
Signal information acquisition means for acquiring signal information indicating the period of change of the traffic light;
OD traffic volume acquisition means for acquiring an OD traffic volume indicating the departure place, departure time, and destination of each vehicle estimated based on information indicating the actual traffic state of the vehicle;
Based on the road information, the signal information acquired by the signal information acquisition means, and the OD traffic volume acquired by the OD traffic volume acquisition means, a vehicle model that models a vehicle is run, and a traffic state A simulation means for simulating
Traffic state simulation device including An acquisition means for acquiring at least one of a traffic volume per unit time of a vehicle in a predetermined section, a travel time, a queue length, and a traffic jam length based on an actual vehicle traffic state;
Calculation means for calculating at least one of the traffic volume per unit time, travel time, queue length, and traffic jam length based on the traffic state simulated by the simulation means;
At least one of the traffic volume per unit time, travel time, queue length, and traffic jam length acquired by the acquisition means, and the traffic volume per unit time, travel time, queue calculated by the calculation means By the OD traffic volume acquisition means, the difference between the traffic volume per unit time, the travel time, the queue length, and the traffic jam length is reduced by comparing at least one of the length and the traffic jam length. OD traffic correction means for correcting the acquired OD traffic,
The traffic state simulation device according to claim 1, wherein the simulation unit simulates a traffic state using the OD traffic volume corrected by the OD traffic volume correction unit.   The acquisition unit acquires an average value of at least one predetermined period of traffic volume per unit time, travel time, queue length, and traffic jam length of the vehicle in the predetermined section based on the actual vehicle traffic state. The traffic state simulation device according to claim 2.   The acquisition means acquires at least one current value of traffic volume per unit time, travel time, queue length, and congestion length of the vehicle in the predetermined section based on the actual vehicle traffic state. 2. The traffic state simulation device according to 2.   The acquisition means predicts at least one of traffic volume per unit time, travel time, queue length, and congestion length of the predetermined section at the time of prediction based on the actual vehicle traffic state. The traffic state simulation device according to claim 2, wherein at least one predicted value of traffic volume per unit time, travel time, queue length, and traffic jam length of the vehicle in the predetermined section is acquired.   The traffic state simulation device according to any one of claims 1 to 5, wherein the simulation unit divides the OD traffic volume into OD traffic volume for each time zone and each region and uses the traffic volume for simulation of traffic conditions. .   The OD traffic volume acquisition means relates to travel history information indicating a travel history of each of a plurality of vehicles, statistical information on the traffic volume per unit time of the vehicle on each of a plurality of roads, and a starting point and a destination of the vehicle. The traffic state simulation device according to claim 1, wherein the OD traffic volume is estimated based on at least one of statistical information. Computer
Road information storage means for storing road information indicating a road map and a road environment;
A signal information acquisition means for acquiring signal information indicating a change period of the traffic light;
OD traffic volume acquisition means for acquiring OD traffic volume indicating the departure place, departure time, and destination of each vehicle estimated based on information indicating the actual vehicle traffic state, and the road information and the signal information acquisition Based on the signal information acquired by the means and the OD traffic volume acquired by the OD traffic volume acquisition means, a vehicle model that models the vehicle is run to function as a simulation means for simulating the traffic state. Program for.

Images