Information processing device, information processing method, and program

Abstract

To predict a traffic flow using a highly accurate traffic flow theory model parameter set capable of reproducing traffic conditions similar to the current traffic situation.SOLUTION: An information processing device includes: a configuration unit that sets parameter sets for traffic flow theory models used in traffic flow simulations applying traffic flow theory models; a simulation unit that executes traffic flow simulations for each parameter set; and a determination unit that selects traffic flow simulation data from the traffic flow simulation results that is similar to the actual measured traffic flow measurement data, and determines the parameter set corresponding to the selected similar traffic flow simulation data as the parameter set to be used for traffic flow prediction.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, and a program for predicting traffic congestion. [Background technology]

[0002] Currently, traffic congestion predictions are based on the knowledge and experience of experts. Specifically, experts such as traffic congestion forecasters use past congestion trends and patterns to predict the extent and location of future congestion. However, in order to train experts, it is necessary to gain the experience necessary for congestion prediction. Therefore, a method using traffic flow simulation has been proposed.

[0003] As a related technique, Patent Document 1 discloses a traffic situation prediction system that verifies and predicts traffic congestion. According to the traffic situation prediction system of Patent Document 1, first, current traffic-related data for a target road is used to calculate model parameters for each vehicle to be set in a traffic flow simulation that can reproduce the current traffic situation. Then, the traffic situation prediction system of Patent Document 1 sets the model parameters and congestion events at required locations on the target road, and executes a traffic flow simulation until a set time is reached to predict the traffic situation.

[0004] Specifically, the traffic situation prediction system in Patent Document 1 discloses that the average speed and number of vehicles are calculated from traffic volume, density, etc., and a traffic flow simulation is used to make the vehicles travel at the average speed for a certain period of time, and optimization calculations are performed using vehicle parameters such as accelerator, acceleration, and brake as functions to calculate model parameters (initial value parameters) for each vehicle that can reproduce the current traffic situation. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-067180 Summary of the Invention [Problem to be solved by the invention]

[0006] However, the traffic condition prediction system of Patent Document 1 does not clearly disclose a method for calculating the model parameters of each vehicle in the traffic flow simulation. Furthermore, Patent Document 1 performs traffic flow simulations under the assumption that all vehicles maintain the measured average speed. However, in real traffic flows, there are speed differences between vehicles, and vehicle behavior fluctuates over time and is complex. Therefore, the degree of reproduction of current traffic conditions and the accuracy of future predictions are thought to be low, making it impossible to make accurate congestion predictions.

[0007] An example of an objective of the present disclosure is to predict traffic flow using a parameter set of a highly accurate theoretical traffic flow model that can reproduce traffic conditions similar to the current traffic conditions. [Means for solving the problem]

[0008] In order to achieve the above object, an information processing device according to one aspect of the present disclosure includes: a setting means for setting a parameter set of a traffic flow theoretical model used in a traffic flow simulation to which the traffic flow theoretical model is applied; a simulation means for executing the traffic flow simulation for each of the parameter sets; a determining means for selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from among the traffic flow simulation data resulting from the execution of the traffic flow simulation, and determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in the traffic flow prediction; It is characterized by:

[0009] In order to achieve the above object, an information processing method according to one aspect of the present disclosure includes: The information processing device A parameter set of the traffic flow theory model to be used in the traffic flow simulation to which the traffic flow theory model is applied is set, Executing the traffic flow simulation for each of the parameter sets; Selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that are the execution results of the traffic flow simulation; determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; It is characterized by:

[0010] Furthermore, in order to achieve the above object, a program according to one aspect of the present disclosure includes: On the computer, A parameter set of the traffic flow theory model to be used in the traffic flow simulation to which the traffic flow theory model is applied is set, Executing the traffic flow simulation for each of the parameter sets; Selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that are the execution results of the traffic flow simulation; determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; The present invention is characterized in that it causes the processing to be executed. [Effects of the Invention]

[0011] As described above, according to the present disclosure, traffic flow prediction can be performed using a parameter set of a highly accurate theoretical traffic flow model that can reproduce traffic conditions similar to the current traffic conditions. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating an example of an information processing device. [Figure 2] FIG. 2 is a diagram illustrating an example of a system including an information processing device. [Figure 3] FIG. 3 is a diagram for explaining a cellular automaton. [Figure 4] FIG. 4 is a diagram illustrating an example of a bottleneck. [Figure 5] FIG. 5 is a diagram for explaining the determination of the traffic flow parameter set. [Figure 6] FIG. 6 is a diagram for explaining the particle filter processing. [Figure 7] FIG. 7 is a diagram illustrating the similarity. [Figure 8] FIG. 8 is a diagram for explaining an example of displaying traffic flow (average speed) measurement data and the posterior probability of each traffic flow parameter. [Figure 9] FIG. 9 is a diagram for explaining an example of displaying the prediction result and the correct answer value. [Figure 10] FIG. 10 is a diagram illustrating the operation of the information processing device. [Figure 11] FIG. 11 is a diagram illustrating an example of a computer that realizes the information processing device according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] (Embodiment) The configuration of an information processing device in an embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram illustrating an example of an information processing device.

[0014] [Device configuration] 1 is a device (traffic flow simulation device) that predicts traffic flow using highly accurate parameters that can reproduce traffic conditions similar to the current traffic conditions through data assimilation using a traffic flow theory model proposed in traffic engineering. Also, as shown in FIG. 1, the information processing device 10 has a setting unit (setting means) 11, a simulation unit (simulation means) 12, and a determination unit (determination means) 13.

[0015] The setting unit 11 sets a parameter set of the traffic flow theoretical model to be used in a traffic flow simulation to which the traffic flow theoretical model is applied. The simulation unit 12 executes a traffic flow simulation for each parameter set.

[0016] The determination unit 13 selects traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that is the execution result of the traffic flow simulation, and determines the parameter set corresponding to the selected similar traffic flow simulation data as the parameter set to be used in the traffic flow prediction.

[0017] In this way, in the embodiment, a parameter set of a highly accurate theoretical traffic flow model that can reproduce traffic conditions similar to the current traffic conditions can be determined, thereby enabling highly accurate traffic flow prediction.

[0018] [System Configuration] The configuration of the information processing device 10 in this embodiment will be described in more detail with reference to Fig. 2. Fig. 2 is a diagram showing an example of a system having an information processing device. The system 100 has the information processing device 10, a storage device 20, an output device 30, and a network 40. The information processing device 10, the storage device 20, and the output device 30 are connected to each other so as to be able to communicate with each other via the network 40 or the like.

[0019] The information processing device 10 is, for example, an information processing device such as a CPU (Central Processing Unit), a programmable device such as an FPGA (Field-Programmable Gate Array), a GPU (Graphics Processing Unit), or a circuit equipped with any one or more of these, a server computer, a personal computer, or a mobile terminal.

[0020] 2, the simulation unit 12 is provided inside the information processing device 10, but it may be provided outside the information processing device 10. For example, the simulation unit 12 may be provided in a server computer or the like that is provided separately from the information processing device 10.

[0021] The storage device 20 is a database, a server computer, a circuit having a memory, etc. The storage device 20 stores, for example, at least setting data, traffic flow measurement data, etc. In the example of Fig. 2, the storage device 20 is provided outside the information processing device 10, but it may also be provided inside the information processing device 10.

[0022] The output device 30 acquires output information, which will be described later, and outputs generated images, sounds, and the like based on the output information. The output device 30 is, for example, an image display device using a liquid crystal, an organic EL (Electro Luminescence), or a CRT (Cathode Ray Tube). Furthermore, the image display device may also include an audio output device such as a speaker. The output device 30 may also be a printing device such as a printer.

[0023] Network 40 is a general network constructed using communication lines such as the Internet, a LAN (Local Area Network), a dedicated line, a telephone line, an in-house network, a mobile communication network, Bluetooth (registered trademark), or Wi-Fi (Wireless Fidelity) (registered trademark).

[0024] The information processing device 10 will now be described in detail. As shown in FIG. 2, the information processing device 10 in the embodiment includes a setting unit 11, a simulation unit 12, a determination unit 13, and an output information generation unit .

[0025] ●About the settings section The setting unit 11 first acquires setting data necessary for the traffic flow simulation from the setting data 21 stored in the storage device 20. Next, the setting unit 11 sets the acquired setting data in the simulation unit 12.

[0026] The setting data includes at least a road parameter set, a traffic flow parameter set, simulation condition information, inflow time series information, etc. However, the setting data is not limited to the road parameter set, the traffic flow parameter set, the simulation condition information, and the inflow time series information.

[0027] The road parameter set is a parameter set used when constructing a simulation model (road model) of a target road in a traffic flow simulation to which a traffic flow theory model is applied.

[0028] An example of a traffic flow theory model is the S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. Specifically, the S-NFS model describes the behavior of each vehicle using probabilistic cellular automata (CA).

[0029] Let us now explain cellular automata. Figure 3 is a diagram used to explain cellular automata. In a cellular automata model (CA model), space-time and speed are expressed as discrete values. In the example of the CA model in Figure 3, one lane on a road is represented by a row of cells, and the vehicle speed is represented by the number of cells (arrow) that the vehicle (circle) will move forward in the next time step (t, t+1, t+2, t+3, t+4).

[0030] In addition, the space is discretely divided (cells), and each cell is assigned a "0" or a "1" (binarization). A "0" indicates that no vehicle is present, and a "1" indicates that a vehicle is present. In other words, the "1" is moved. Vehicle movement is based on predetermined rules. For example, a rule could be that if there is an empty cell in the direction of the vehicle's movement, the target vehicle is moved to that cell.

[0031] For details of the S-NFS model, see Reference 1 (Satoshi Sakai, Katsuhiro Nishinari, Shinji Iida, “A new stochastic cellular automaton model on traffic flow and its jamming phase transition”, J. Phys. A: Math. Gen. 39 (2006) 15327-15339, [searched on May 14, 2024], Internet <URL: https: / / doi.org / 10.48550 / arXiv.cond-mat / 0611455).

[0032] However, traffic flow theory models based on the CA model are not limited to the S-NFS model. Other models include the Nishinari-Fukui-Schadschneider model (Reference 2: Katsuhiro Nishinari, Minoru Fukui, & Andreas Schadschneider, “A Stochastic Cellular Automaton Model for Traffic Flow with Multiple Metastable States”, J. Phys. A: Math. Gen. 37 (2004) 3101-3110, [Retrieved May 21, 2024], Internet <URL: https: / / iopscience.iop.org / article / 10.1088 / 0305-4470 / 37 / 9 / 003>) and the Nagel-Schrechenberg model (Reference 3: Kai Nagekl & Michael Schreckenberg, “A Cellular Automaton Model for Freeway Traffic”, J. Phys. I France. 2 (1992) 2221-2229, [Retrieved May 21, 2024], Internet < URL : https: / / jp1.journaldephysique.org / en / articles / jp1 / abs / 1992 / 12 / jp1v2p2221 / jp1v2p2221.html >).

[0033] The traffic flow parameter set is a set of parameters for the traffic flow theory model that is set when running a traffic flow simulation. Specifically, in the S-NFS model, the speed of each vehicle is determined based on the following (1) to (5).

[0034] (1) Maximum speed Vmax of the road (2) Speed ​​and distance of the vehicle ahead: Avoiding rear-end collisions (3) Random braking probability p: One-stage deceleration (random braking effect) (4) Slow start probability q: Inter-vehicle distance at the previous step (time delay of inertia and reaction, slow start effect) (5) Probability of visibility r: Speed ​​(future outlook) using the distance between vehicles S ahead. For example, if probability r, S = 2, and if probability 1-r, S = 1.

[0035] Therefore, as a traffic flow parameter set to be adjusted to reproduce the measured traffic flow, at least the maximum speed V BN Parameters such as the probability of random braking p, the probability of slow start q, and the probability of visibility r are used.

[0036] A bottleneck is a section where vehicle speed decreases or congestion occurs, such as an uphill section, a sag section, a tunnel, or a toll booth. Figure 4 is a diagram for explaining an example of a bottleneck. In the example of Figure 4, the section from 8.4 to 8.6 km (kilometers) on the target road is the bottleneck (hatched area).

[0037] As an example (parameter set example), the traffic flow parameter set is as follows: 1 cell: 10 m (meters), time step width: 1.8 seconds, speed resolution: 20 km / h (kilometers per hour) (= 10 m / 1.8 seconds = 50 / 9 m / seconds), and the maximum speed within the bottleneck V BN Possible settings are 20, 40, and 60 km / h (3 combinations), the random braking probability p is in the range of 0.05 to 0.6 in increments of 0.05 (12 combinations), the slow start probability q is in the range of 0.1 to 0.8 in increments of 0.1 (8 combinations), and the visibility probability r is in the range of 0.75 to 0.99 in increments of 0.03 (9 combinations). With the above settings, the traffic flow parameter sets will be 2592 (= 3 x 12 x 8 x 9). However, the settings for the traffic flow parameter sets are not limited to the above-mentioned settings, such as grid search.

[0038] The simulation condition information is information that indicates the simulation conditions used in the traffic flow simulation, such as the simulation end time and initial conditions.

[0039] The inflow volume time series information is information that represents the inflow volume of a target road in a time series. For example, the inflow volume time series information is a value (time series data) of the number of vehicles that flow into the start point (0 [km] point) of the target road per unit time measured by a traffic counter.

[0040] ●About the Simulation Department The simulation unit 12 executes a traffic flow simulation for each of a plurality of different traffic flow parameter sets during a set period. The set period is a pre-set period from the present time, such as 10 to 60 minutes in the past. However, the set period is not limited to the above period. In the case of the above-mentioned example of the traffic flow parameter set, the simulation unit 12 executes a maximum of 2592 traffic flow simulations.

[0041] Furthermore, the simulation unit 12 performs a traffic flow simulation for a predetermined period from the present time onward, using a traffic flow parameter set to be used in the traffic flow prediction determined by the determination unit 13 described below, to make a traffic flow prediction.

[0042] ●About the decision section The determination unit 13 calculates the posterior probability distribution for each parameter set, or the maximum posterior probability, or the expected value of the posterior probability distribution, or all of these, based on the similarity between the traffic flow measurement data and the results of traffic flow simulations (traffic flow simulation data) performed for each of multiple different traffic flow parameter sets during a set period, and determines the traffic flow parameter sets to be used in traffic flow prediction based on one or more of these.

[0043] Traffic flow measurement data is actually measured data such as the average speed of vehicles, the number of vehicles passing per unit time (flow rate), etc. In addition, traffic flow measurement data is data measured by on-board devices such as optical fiber sensing, surveillance cameras, loop coils, ultrasonic sensors, and other traffic counters, ETC2.0 probe information, etc.

[0044] Figure 5 is a diagram for explaining the determination of traffic flow parameter sets. Figure 5A shows the change in average speed based on traffic flow measurement data. Figure 5B, C, and D show the change in average speed based on traffic flow simulation data. In each of Figures 5A to 5D, the vertical axis represents the flow of time, and the horizontal axis represents the direction of traffic flow and the distance from the starting point. The average vehicle speed is also shown color-coded from 30 to 60 km / h. In other words, it shows the average vehicle speed for each 1-minute, 1-km section over the past 30 minutes.

[0045] Also, Figure 5B is a diagram obtained by setting the traffic flow parameter set to probabilities (p, q, r) = (0.10, 0.8, 0.75). Figure 5C is a diagram obtained by setting the traffic flow parameter set to probabilities (p, q, r) = (0.55, 0.6, 0.84). Figure 5D is a diagram obtained by setting the traffic flow parameter set to probabilities (p, q, r) = (0.35, 0.2, 0.99).

[0046] In the example of Figure 5, since the similarity between A in Figure 5 and D in Figure 5 is high at each time and location, it can be determined that the traffic flow parameter set (0.35, 0.2, 0.99) in D in Figure 5 is optimal. In this way, traffic flow simulation data that has a low error rate with respect to the average speed based on traffic flow measurement data, in other words, similar traffic flow simulation data, is selected for as many sections as possible.

[0047] The reason for using the error rate of average speed as the similarity is that, for example, if an error rate of up to 10% (percent) is allowed, if the average speed in the traffic flow measurement data is 100 km / h, then the average speed in the traffic flow simulation data can tolerate an error of up to ±10 km / h. In contrast, if the average speed in the traffic flow measurement data is 30 km / h, then the average speed in the traffic flow simulation data can tolerate an error of only ±3 km / h. For these reasons, the error rate of average speed is considered to be suitable as an index of the similarity (degree of agreement) of congestion (low-speed events). In other words, to focus on low-speed events, the error rate of average speed is used, and a greater penalty is imposed on the degree of disagreement in low-speed areas than in high-speed areas. Note that, in addition to average speed, the average flow rate or average density may also be used as a similarity measure, or these may be combined.

[0048] A method for determining (estimating) the traffic flow parameter set will now be described in detail. As a method for determining (estimating) the traffic flow parameter set, for example, a process using a particle filter can be considered. Specifically, the determination unit 13 executes particle filter processing using the similarity between the traffic flow measurement data and the traffic flow simulation data, and determines (estimates) the traffic flow parameter set based on the frequency distribution of the particles.

[0049] The particle filter process is performed by filtering the traffic flow simulation data obtained for each of the multiple different traffic flow parameter sets at each time t (multiple times t1 to t N (N is a positive integer greater than or equal to 2).

[0050] Fig. 6 is a diagram for explaining particle filter processing when there is one parameter set to be determined (estimated). In particle filter processing, first, as shown in Fig. 6, particles for the traffic flow parameter set are placed at sampled positions in the parameter space.

[0051] For example, in the example parameter set mentioned above, the maximum speed within the bottleneck, V BNThe parameter space is a four-dimensional real number space that is a set of four parameters: the random braking probability p, the slow start occurrence probability q, and the visibility probability r. From this four-dimensional space, specific values ​​are extracted at specified intervals within the range specified for each parameter, as in the example parameter set described above.

[0052] In the example parameter set described above, 2593 parameter sets are sampled by grid search, so 2593 parameter sets are extracted from the above four-dimensional real number space. The particles of the particle filter are the values ​​corresponding to the parameters extracted from the four-dimensional real number space, for example (V BN, It is placed at a specific location in the four-dimensional real space: p,q,r) = (20,0.05,0.1,0.75).

[0053] Next, as shown in Figure 6, the similarity between the traffic flow measurement data and the traffic flow simulation data is set for the particles of the traffic flow parameter set at the target time t. Note that the higher the traffic flow parameter set, the higher the similarity (weight) is set. In the example of Figure 6, the larger the black circle, the higher the similarity.

[0054] Next, particle filtering involves resampling according to particle weights. In other words, while maintaining the total number of particles, the number of particles placed at points in the parameter space where particles with large weights exist is increased. Conversely, the number of particles placed at positions where particles with small weights exist is reduced. As a result, many particles are placed in traffic flow parameter sets that are highly similar to the traffic flow measurement data (that can reproduce the traffic flow measurement data well), and particles with low similarity eventually disappear.

[0055] Traffic flow simulation data from time t1 to t N When particle filter processing is completed for the traffic flow parameter set, many particles are placed in the traffic flow parameter set that has a high similarity to the traffic flow measurement data, as shown in the particle filter processing results in Figure 6.

[0056] Thereafter, the determination unit 13 calculates statistical information (posterior probability distribution, or maximum posterior probability, or expected value of posterior probability distribution, or all of them) for each traffic flow parameter set based on the frequency distribution of particles of the traffic flow parameter set representing the similarity as shown in FIG. 6. In the case of the above-mentioned parameter set example, as a result of particle filter processing, a joint distribution is obtained in a four-dimensional real number space, which is the parameter space. The value of the joint distribution is high for a set of model parameters that has a high degree of reproducibility of the measured traffic flow. By calculating the marginal distribution for each traffic flow parameter from the joint distribution, the posterior probability distribution of each traffic flow parameter is obtained, and statistical information is obtained.

[0057] Then, the determination unit 13 compares the statistical information (posterior probability distribution, or maximum posterior probability, or expected value of posterior probability distribution, or one or more of them) calculated for each traffic flow parameter set, and determines (estimates) the traffic flow parameter set corresponding to the most similar traffic flow parameter set based on the comparison result, and uses it for traffic flow prediction. For example, the value with the maximum posterior probability is determined (estimated) as the most similar traffic flow parameter optimal value. In addition to the value with the maximum posterior probability, the expected value of the posterior probability distribution can be used.

[0058] The calculation of the similarity will now be described. The determination unit 13 first acquires the execution results (traffic flow simulation data) of the traffic flow simulation executed for each of a plurality of different traffic flow parameter sets during the set period. The determination unit 13 also acquires the traffic flow measurement data during the set period from the traffic flow measurement data 22 in the storage device 20.

[0059] Next, the determination unit 13 calculates the average vehicle speed at each time point and in each section based on the traffic flow measurement data for the set period, and the average vehicle speed at each time point and in each section based on the traffic flow simulation data for the set period.

[0060] Next, the determination unit 13 calculates the error rate between the average speed based on the traffic flow measurement data and the average speed based on the traffic flow simulation data at each time point and in each section.

[0061] Next, the determination unit 13 calculates the similarity based on the error rate of the average vehicle speed at each time point and in each section.

[0062] Fig. 7 is a diagram for explaining the similarity. The example in Fig. 7 shows the average speed in each of a plurality of sections at a point in time (time) t in the traffic flow measurement data and the traffic flow simulation data.

[0063] The section indicates m sections (m is a positive integer greater than or equal to 1) set on the target road. The average speed indicates the average speed in the mth section at time t. For example, the average speed could be the average speed per minute in the mth section at time t.

[0064] The error rate can be expressed as in Equation 1. The n-th (n: 1 to N (the number of all traffic flow parameter sets)) traffic flow parameter set is represented as θn.

[0065]

number

[0066] Similarity can be expressed as likelihood as shown in Equation 2. The reason for multiplying each interval in Equation 2 is that the likelihood does not decrease if the error rate of any one interval is low in the sum. Note that the hyperparameter σ in Equation 2 can be, for example, 20 to 30%. However, the value of the hyperparameter σ is not limited to 20 to 30%.

[0067]

number

[0068] In addition, in Equation 2, if the similarity (degree of agreement) is not high in all sections where congestion (low-speed events) occurs, the similarity will decrease. Furthermore, if even one section where congestion occurs does not agree, the similarity will decrease.

[0069] However, in particle filter processing, we want to treat the similarity (weight) as a positive real number, so we convert Equation 2 into Equation 3 (taking the inverse square of the logarithmic likelihood).

[0070]

number

[0071] In addition, the similarity (weight) is normalized and expressed as in Equation 4. A high weight is set for similar traffic flow parameter sets, including locations where low-speed events occur. By normalizing the weights, it becomes possible to treat the frequency distribution of the weights as a posterior probability distribution. Therefore, as described above, the optimal value of the traffic flow parameters can be determined from the frequency distribution of the weights.

[0072]

number

[0073] The output information generation unit 14 generates output information for outputting to the output device 30, for example, the road model shown in FIG. 4, traffic flow (average speed) measurement data for a set period shown in FIG. 8A, the posterior probability (frequency distribution of particles) of each traffic flow parameter based on the result of particle filter processing shown in FIG. 8B, their statistical information (posterior probability distribution, maximum posterior probability, expected value of the posterior probability distribution, or one or more of them), traffic flow simulation data for the traffic flow prediction shown in the prediction result in FIG. 9, traffic flow prediction accuracy, etc. Thereafter, the output device 30 acquires the output information output by the output information generation unit 14 and outputs it to the output device 30 based on the output information. Note that the similarity and traffic flow simulation data for the set period may also be output.

[0074] Fig. 8 is a diagram for explaining an example of displaying traffic flow (average speed) measurement data and the posterior probability of each traffic flow parameter. Fig. 9 is a diagram for explaining an example of displaying prediction results and correct values.

[0075] The average speed of A in Figure 8 is calculated by the traffic flow model parameter (V BN, This is the traffic flow data when p, q, r) = (40, 0.36, 0.12, 0.98). Also, Fig. 8B shows the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility for the above-mentioned traffic flow model parameters.

[0076] The dashed line in B of Figure 8 represents the correct value of the traffic flow model parameters mentioned above. In the example of Figure 8, it can be confirmed that the value that maximizes the posterior probability is close to the correct value. Furthermore, the traffic flow model parameter value that maximizes the posterior probability was applied, and the prediction results for the future average speed up to 60 minutes into the future are shown in Figure 9.

[0077] The information indicating the traffic flow prediction accuracy is, for example, information indicating the relationship between the predicted value of the average speed and the actual measured value of the average speed (results of regression analysis, correlation coefficient, mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), mean percentage error (MPE) and so on). For example, the prediction accuracy of the prediction result shown in FIG. 9 relative to the correct value is a correlation coefficient of 0.91, mean absolute error of 7.0 [km / h], root mean squared error of 9.8 [km / h] and mean percentage error of 18 [%].

[0078] [Device operation] Next, the operation of the information processing device in the embodiment will be described with reference to FIG. 10. FIG. 10 is a diagram for explaining the operation of the information processing device. In the following description, the diagram will be referenced as appropriate. Furthermore, in the embodiment, an information processing method is implemented by operating the information processing device. Therefore, the description of the information processing method in the embodiment will be replaced with the description of the operation of the information processing device below.

[0079] As shown in FIG. 10, first, the setting unit 11 acquires the setting data necessary for the traffic flow simulation from the setting data 21 stored in the storage device 20, and sets the acquired setting data in the simulation unit 12 (step A1).

[0080] Next, the simulation unit 12 executes a traffic flow simulation for each of a plurality of different traffic flow parameter sets during the set period (step A2).

[0081] Next, the determination unit 13 acquires traffic flow measurement data and execution results (traffic flow simulation data) of traffic flow simulations executed for each of a plurality of different traffic flow parameter sets during the set period (step A3).

[0082] Next, the determination unit 13 calculates statistical information (posterior probability distribution for each parameter set, or maximum posterior probability, or expected value of the posterior probability distribution, or all of them) based on the similarity between the traffic flow measurement data and the traffic flow simulation data for the set period, and determines (estimates) the traffic flow parameter sets to be used in the traffic flow prediction based on one or more of the statistical information (step A4).

[0083] Specifically, in step A4, the determination unit 13 executes particle filtering using the similarity between the traffic flow measurement data and the traffic flow simulation data, and determines (estimates) a traffic flow parameter set based on the frequency distribution of the particles.

[0084] Next, the simulation unit 12 performs a traffic flow simulation for a predetermined period from the present time onward, using the traffic flow parameter set to be used in the traffic flow prediction determined by the determination unit 13, to perform traffic flow prediction (step A5).

[0085] Next, the output information generation unit 14 generates output information for outputting to the output device 30, for example, the road model shown in Fig. 4, the traffic flow (average speed) measurement data for the set period in Fig. 8A, the posterior probability (frequency distribution of particles) of each traffic flow parameter based on the result of particle filter processing in Fig. 8B, their statistical information (posterior probability distribution, or maximum posterior probability, or expected value of the posterior probability distribution, or one or more of them), the traffic flow simulation data of the traffic flow prediction shown in the prediction result in Fig. 9, the traffic flow prediction accuracy, etc., and outputs the output information to the output device 30 (step A6). Note that the similarity and the traffic flow simulation data for the set period may also be output.

[0086] Thereafter, the output device 30 acquires the output information output by the output information generating unit 14, and outputs to the output device 30 based on the output information.

[0087] [Effects of the embodiment] As described above, according to the embodiment, a parameter set of a highly accurate theoretical traffic flow model that can reproduce traffic conditions similar to the current traffic conditions can be determined, thereby enabling highly accurate traffic flow prediction.

[0088] In addition, in the past, training experts required a great deal of time to gain the experience necessary for traffic congestion prediction, but according to the embodiment, traffic flow prediction (traffic congestion prediction) can be performed even by non-experts.

[0089] Furthermore, currently, traffic flow predictions (traffic congestion predictions) are made based on the results of an automatic search to see if trends and patterns in past data match, which requires a huge amount of past data as training data. It is also difficult to respond to rare or unexpected events (accidents, rapidly developing congestion, etc.). According to the embodiment, traffic flow simulations are performed using a traffic flow parameter set that can reproduce the most recent road conditions, enabling highly accurate traffic flow predictions (traffic congestion predictions) even when rare or unexpected events occur.

[0090] Furthermore, according to the embodiment, congestion can be predicted in advance and reduced through traffic control, thereby reducing the significant economic losses caused by congestion due to traffic concentration, the increased risk of accidents, and the burden on the environment caused by exhaust gases.

[0091] Furthermore, according to the embodiment, particle filtering, which is one of the methods for analyzing time-series data, is used, so that traffic flow prediction can be performed in real time with high accuracy.

[0092] In this way, traffic flow predictions can be made that can handle small amounts of data, rare and unexpected events, and in real time, and the results of traffic flow simulations can be combined (data assimilation).

[0093] The technology of this embodiment can be applied to congestion control through patrol car road pricing and traffic accident risk management (improving the management effectiveness of road operators).

[0094] It will also be effective in reducing economic losses, environmental burdens, driver stress, and improving services. Furthermore, it will play an important role in traffic control and autonomous vehicle control in the period when autonomous vehicles will be used together.

[0095] [program] The program in the embodiment may be any program that causes a computer to execute steps A1 to A6 shown in Fig. 10. By installing and executing this program on a computer, the information processing device and information processing method in the embodiment can be realized. In this case, the processor of the computer functions as a setting unit 11, a simulation unit 12, a determination unit 13, and an output information generation unit 14 and performs processing.

[0096] The program in the embodiment may be executed by a computer system constructed by a plurality of computers, in which case, for example, each computer may function as one of the setting unit 11, the simulation unit 12, the determination unit 13, and the output information generation unit 14.

[0097] [Physical configuration] A computer that realizes an information processing device by executing a program in the embodiment will now be described with reference to Fig. 11. Fig. 11 is a diagram illustrating an example of a computer that realizes an information processing device in the embodiment.

[0098] 11, the computer 110 includes a CPU (Central Processing Unit) 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These components are connected to each other via a bus 121 so as to be able to communicate data with each other. Note that the computer 110 may include a GPU or an FPGA in addition to or instead of the CPU 111.

[0099] The CPU 111 loads a program in the embodiment, which is composed of a group of codes and stored in the storage device 113, into the main memory 112 and executes each code in a predetermined order to perform various calculations. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory).

[0100] The program in the embodiment is provided in a state stored in a computer-readable recording medium 120. The program in the embodiment may be distributed over the Internet connected via the communication interface 117.

[0101] Specific examples of the storage device 113 include a hard disk drive and a semiconductor storage device such as a flash memory. The input interface 114 mediates data transmission between the CPU 111 and input devices 118 such as a keyboard and a mouse. The display controller 115 is connected to a display device 119 and controls the display on the display device 119.

[0102] The data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, reads programs from the recording medium 120, and writes processing results from the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and other computers.

[0103] Specific examples of the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as flexible disks, or optical recording media such as CD-ROMs (Compact Disk Read Only Memory).

[0104] Note that the information processing device 10 in the embodiment can be realized not by a computer on which a program is installed, but by hardware corresponding to each unit, for example, an electronic circuit. Furthermore, the information processing device 10 may be partially realized by a program and the remaining unit by hardware. In the embodiment, the computer is not limited to the computer shown in FIG. 9.

[0105] [Note] The following supplementary notes are further provided with respect to the above-described embodiments. Some or all of the above-described embodiments can be expressed by (Supplementary Note 1) to (Supplementary Note 15) described below, but are not limited to the following descriptions.

[0106] (Appendix 1) a setting unit for setting a parameter set of a traffic flow theoretical model used in a traffic flow simulation to which the traffic flow theoretical model is applied; a simulation unit that executes the traffic flow simulation for each of the parameter sets; a determination unit that selects traffic flow simulation data similar to the actually measured traffic flow measurement data from traffic flow simulation data that is the execution result of the traffic flow simulation, and determines a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; An information processing device having the above.

[0107] (Appendix 2) the simulation means further executes the traffic flow simulation for a predetermined period from the present time onward, using the determined parameter set to be used in the traffic flow prediction, to perform traffic flow prediction. 10. The information processing device according to claim 1.

[0108] (Appendix 3) The traffic flow theory model is an S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. 10. The information processing device according to claim 1.

[0109] (Appendix 4) When the S-NFS model is used as the traffic flow theory model, the parameter set is the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility. 4. The information processing device according to claim 3.

[0110] (Appendix 5) the determination unit calculates a posterior probability distribution, a maximum posterior probability, an expected value of the posterior probability distribution, or all of them for each of the parameter sets based on the similarity between the traffic flow simulation data and the traffic flow measurement data, and determines a parameter set to be used in the traffic flow prediction based on one or more of these. 10. The information processing device according to claim 1.

[0111] (Appendix 6) The information processing device A parameter set of the traffic flow theory model to be used in the traffic flow simulation to which the traffic flow theory model is applied is set, Executing the traffic flow simulation for each of the parameter sets; Selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that are the execution results of the traffic flow simulation; determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; Information processing methods.

[0112] (Appendix 7) The information processing device, Furthermore, the traffic flow simulation is executed for a predetermined period from the present time onward using the determined parameter set to be used in the traffic flow prediction, thereby predicting traffic flow. 1. The information processing method described in Appendix 6.

[0113] (Appendix 8) The traffic flow theory model is an S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. 1. The information processing method described in Appendix 6.

[0114] (Appendix 9) When the S-NFS model is used as the traffic flow theory model, the parameter set is the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility. 9. The information processing device according to claim 8.

[0115] (Appendix 10) In the determination, a posterior probability distribution, a maximum posterior probability, an expected value of the posterior probability distribution, or all of them are calculated for each parameter set based on the similarity between the traffic flow simulation data and the traffic flow measurement data, and the parameter set to be used in the traffic flow prediction is determined based on one or more of these. The information processing method according to claim 6.

[0116] (Appendix 11) On the computer, A parameter set of the traffic flow theory model to be used in the traffic flow simulation to which the traffic flow theory model is applied is set, Executing the traffic flow simulation for each of the parameter sets; Selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that are the execution results of the traffic flow simulation; determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; A program that executes a process.

[0117] (Appendix 12) The computer, Furthermore, the traffic flow simulation is executed for a predetermined period from the present time onward using the determined parameter set to be used in the traffic flow prediction, thereby predicting traffic flow. 12. The program according to claim 11, which causes the processing to be executed.

[0118] (Appendix 13) The traffic flow theory model is an S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. 12. The program described in Appendix 11.

[0119] (Appendix 14) When the S-NFS model is used as the traffic flow theory model, the parameter set is the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility. 13. The program described in Appendix 13.

[0120] (Appendix 15) In the determination, a posterior probability distribution, a maximum posterior probability, an expected value of the posterior probability distribution, or all of them are calculated for each parameter set based on the similarity between the traffic flow simulation data and the traffic flow measurement data, and the parameter set to be used in the traffic flow prediction is determined based on one or more of these. 12. The program described in Appendix 11.

[0121] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. [Industrial Applicability]

[0122] According to the above description, traffic flow prediction can be performed using a parameter set of a highly accurate traffic flow theoretical model that can reproduce traffic conditions similar to the current traffic conditions. This is also useful in fields where traffic flow prediction is required. [Explanation of symbols]

[0123] 10. Information processing equipment 11 Setting section 12 Simulation Section 13 Decision Section 14 Output information generation unit 20 Storage device 21 Setting data 22 Traffic flow measurement data 30 Output Devices 100 systems 110 Computer 111 CPU 112 main memory 113 Storage device 114 Input Interface 115 Display Controller 116 Data Reader / Writer 117 Communication Interface 118 Input Devices 119 Display Device 120 Recording Media 121 Bus

Claims

1. a setting means for setting a parameter set of a traffic flow theoretical model used in a traffic flow simulation to which the traffic flow theoretical model is applied; a simulation means for executing the traffic flow simulation for each of the parameter sets; a determining means for selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from among the traffic flow simulation data resulting from the execution of the traffic flow simulation, and determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in the traffic flow prediction; An information processing device having the above.

2. the simulation means further executes the traffic flow simulation for a predetermined period from the present time onward, using the determined parameter set to be used in the traffic flow prediction, to perform traffic flow prediction. The information processing device according to claim 1 .

3. The traffic flow theory model is an S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. The information processing device according to claim 1 .

4. When the S-NFS model is used as the traffic flow theory model, the parameter set is the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility. The information processing device according to claim 3 .

5. the determining means calculates a posterior probability distribution, a maximum posterior probability, an expected value of the posterior probability distribution, or all of them for each of the parameter sets based on the similarity between the traffic flow simulation data and the traffic flow measurement data, and determines the parameter sets to be used in the traffic flow prediction based on one or more of these. The information processing device according to claim 1 .

6. The information processing device A parameter set of the traffic flow theory model to be used in the traffic flow simulation to which the traffic flow theory model is applied is set, Executing the traffic flow simulation for each of the parameter sets; Selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that are the execution results of the traffic flow simulation; determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; Information processing methods.

7. The information processing device, Furthermore, the traffic flow simulation is executed for a predetermined period from the present time onward using the determined parameter set to be used in the traffic flow prediction, thereby predicting traffic flow. The information processing method according to claim 6.

8. The traffic flow theory model is an S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. The information processing method according to claim 6.

9. When the S-NFS model is used as the traffic flow theory model, the parameter set is the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility. The information processing method according to claim 8.

10. In the determination, a posterior probability distribution, a maximum posterior probability, an expected value of the posterior probability distribution, or all of them are calculated for each parameter set based on the similarity between the traffic flow simulation data and the traffic flow measurement data, and the parameter set to be used in the traffic flow prediction is determined based on one or more of these. The information processing method according to claim 6.

11. On the computer, A parameter set of the traffic flow theory model to be used in the traffic flow simulation to which the traffic flow theory model is applied is set, Executing the traffic flow simulation for each of the parameter sets; Selecting traffic flow simulation data similar to the actually measured traffic flow measurement data from the traffic flow simulation data that are the execution results of the traffic flow simulation; determining a parameter set corresponding to the selected similar traffic flow simulation data as a parameter set to be used in traffic flow prediction; A program that executes a process.

12. The computer, Furthermore, the traffic flow simulation is executed for a predetermined period from the present time onward using the determined parameter set to be used in the traffic flow prediction, thereby predicting traffic flow. The program according to claim 11, which causes processing to be executed.

13. The traffic flow theory model is an S-NFS (Stochastic Nishinari-Fukui-Schadschneider) model. The program according to claim 11.

14. When the S-NFS model is used as the traffic flow theory model, the parameter set is the maximum speed within the bottleneck, the probability of random braking, the probability of slow start, and the probability of visibility. The program according to claim 13.

15. In the determination, a posterior probability distribution, a maximum posterior probability, an expected value of the posterior probability distribution, or all of them are calculated for each parameter set based on the similarity between the traffic flow simulation data and the traffic flow measurement data, and the parameter set to be used in the traffic flow prediction is determined based on one or more of these. The program according to claim 11.

Images