JP6038304B2 - Traffic estimation system - Google Patents

Description

  The present invention relates to a system for estimating a traffic volume on a road, and more particularly to a system for estimating a traffic volume by measuring a traveling speed of a vehicle traveling on a road.

  Japanese Unexamined Patent Application Publication No. 2009-140007 (Patent Document 1) discloses a technique for estimating a traffic volume using a probe car. In Patent Document 1, speed data of a probe car traveling on a road section is collected, and the traffic volume of the road section is estimated using a predetermined relationship between traffic volume and speed (hereinafter referred to as road characteristics). Then, the estimated traffic volume is set to one or more of the number of occurrences or time intervals of acceleration / deceleration of the probe car traveling on the road section, the maximum acceleration absolute value, the speed variation, and the number of lane changes. Correct based on (see summary).

  In this technique, it is necessary to create a road characteristic for each road section in advance. In Patent Document 1, a representative road characteristic is created using a road type such as an expressway or a general national road (see paragraph 0048). Then, the traffic volume of a non-traffic volume observation road section (hereinafter referred to as an estimated road section) in which the traffic volume cannot be directly observed is estimated using the representative road characteristics of the same road type.

JP 2009-140007 A

However, in practice, even in a road section of the same road type, the driving environment differs depending on the road width, curve, road gradient, number of traffic lights, etc., and the road characteristics are not always the same. For this reason, the traffic volume of the estimated road section estimated using the representative road characteristics generated for each road type has a limit in estimation accuracy, and may greatly deviate from the actual traffic volume.
In addition, there is a problem that the road characteristics cannot be classified based on the road type in an area where accurate road information cannot be obtained due to a map development situation or an area where the road information itself is not supported. These challenges are often found in areas where development is ongoing and traffic is changing significantly.
Furthermore, the representative road characteristics and the road characteristics of the estimated road section are not always the same in at least the speed range of the traffic jam basin and the free basin. As the road characteristics of one estimated road section, there is a case where it is possible to estimate the accurate traffic volume by using the representative road characteristics of different representative road sections in each speed range. Therefore, it is desirable to estimate the traffic volume of the estimated road section using the representative road characteristics of the representative road section having similar road characteristics for each speed range.

  Accordingly, an object of the present invention is to provide a highly accurate traffic estimation system that estimates traffic based on representative road characteristics having high similarity and travel speed collected from a probe car.

  In order to achieve the above object, the traffic volume estimation system of the present invention determines the similarity of the driving environment between the representative road section and the estimated road section from the driving speed data obtained from the probe car, and represents the representative with high similarity. The traffic volume of the estimated road section was estimated using the road characteristics of the road section.

  The similarity of the driving environment is indirectly determined from the speed characteristics. That is, the representative road characteristic of the representative road section having the speed characteristic closest to the speed characteristic of the estimated road section is used as the road characteristic of the estimated road section.

  The speed characteristic can be obtained, for example, as a relationship between the travel speed in each road section and travel data acquired from a probe car such as a travel speed variance value, a traffic jam rate, the number of stops of a vehicle, or the number of brakes.

  According to the present invention, the travel environment of the estimated road section is estimated using the travel speed data of the probe car, and the traffic volume of the estimated road section is estimated using the representative road characteristics of the representative road section having a similar travel environment. Therefore, it is possible to accurately estimate the traffic volume of the estimated road section without depending on the difference in road characteristics that is difficult to estimate from the map information such as the road type or the map maintenance status.

  Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

It is the figure which showed the whole structure of the traffic estimation system which concerns on embodiment of this invention. An example of the format of the traffic volume data output from the traffic volume output device 17 is shown. It is an example of the data format of the travel speed of the representative road and estimated road input into the traffic volume estimation system. It is a processing flow of the road characteristic estimation apparatus 11 in the traffic volume estimation system 10 which concerns on embodiment of this invention. FIG. 5 is a supplementary explanatory diagram for explaining the road characteristic estimation processing flow of FIG. 4. It is another processing flow of the road characteristic estimation apparatus 11 in the traffic volume estimation system 10 which concerns on embodiment of this invention. FIG. 7 is a supplementary explanatory diagram for explaining the road characteristic estimation processing flow of FIG. 6. It is a processing flow of the speed characteristic calculation apparatus 12 in the traffic volume estimation system 10 which concerns on embodiment of this invention. FIG. 9 is a supplementary explanatory diagram illustrating a flow characteristic calculation process flow of FIG. 8. In the processing flow of the speed characteristic calculation device 12, it is a supplementary explanatory diagram when the speed characteristic is expressed by using a traffic jam rate in step S82. In the traffic volume estimation system 10 which concerns on embodiment of this invention, it is a processing flow of the characteristic data synthesizer 13. In the traffic volume estimation system 10 according to the embodiment of the present invention, an example of a characteristic data format including road characteristics, characteristic parameters, and speed characteristics of a road section output from the characteristic data synthesis device 13 and stored in the storage device 18. It is. In the traffic volume estimation system 10 which concerns on embodiment of this invention, it is a processing flow of the road characteristic selection apparatus 15. FIG. It is a supplementary explanatory drawing explaining the road characteristic selection processing flow of FIG.

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

  FIG. 1 is a diagram showing an overall configuration of a traffic volume estimation system according to an embodiment of the present invention.

  The traffic volume estimation system 10 includes a road characteristic estimation apparatus 11, a representative road speed characteristic calculation apparatus 12, a characteristic data synthesis apparatus 13, an estimated road speed characteristic calculation apparatus 14, a road characteristic selection apparatus 15, a traffic volume estimation apparatus 16, a traffic A quantity output device 17 and a storage device 18 are provided. For example, the road characteristic estimation device 11, the representative road speed characteristic calculation device 12, the characteristic data synthesis device 13, the estimated road speed characteristic calculation device 14, the road characteristic selection device 15, the traffic volume estimation device 16, and the traffic volume output device 17 It can be implemented on one computer. The storage device 18 can also be built in the computer. For this reason, each apparatus may be regarded as a processing unit. In this case, “apparatus” may be described as “part”.

  The road characteristic estimation device 11 collects the number of passing vehicles and the traveling speed per unit time of the vehicle detector, using an observation road section in which a vehicle detector or the like is installed and capable of steady traffic observation as a representative road section. An apparatus for generating road characteristics of a road section. The traveling speed is not limited to the data detected by the vehicle sensor, and traveling speed collected from a probe car or the like may be used.

  Here, the road characteristics will be described. The traffic volume Q (vehicles / h) of the road is the traffic density k (vehicles / km), which is the number of vehicles existing within a unit distance at a certain moment, and the average speed v (km / h) of vehicles passing through the unit distances. ) And (Equation 1).

Q = k × v (Formula 1)
Although it is difficult to directly observe the traffic volume Q or the traffic density k in all road sections, for example, using a traffic model based on the correlation between the traffic density k and the travel speed v, the traffic density can be calculated from the travel speed v. It is possible to estimate k. In the traffic model, various models have been conventionally used, for example, an underwood model shown in (Expression 2) and a Greenberg model expressed in (Expression 3).

v = vf × exp (−k / k0) (Formula 2)
exp () represents an exponential function, Vf represents a free speed at which the traffic density is zero, and k0 represents a critical density which is the traffic density when the traffic volume is maximum.

v = c × log (kj / k) (Formula 3)
Here, log is a logarithm, c is a constant, and kj is a saturation density that is a traffic density when the speed is zero. Further, when the relationship between the speed v and the traffic density k is expressed by an exponential function, for example, it is expressed by (Expression 4) where α and β are constants.

v = α × exp (β × k) (Formula 4)
Thus, the average speed v and the traffic density k can be expressed by their relational expressions and parameters. In the Underwood model of (Equation 2), these parameters are represented by a free velocity Vf and a critical velocity k0, and in the Greenberg model of (Equation 3), a constant c, a saturation density kj, and an exponential function (Equation 4). ) Are constants α and β. Here, these relational expressions are called road characteristic relational expressions, and the parameters thereof are called road characteristic parameters. The road characteristic relational expression specifying the road characteristic parameter is called a road characteristic. The road characteristic estimation device 11 estimates road characteristic parameters of the representative road section. In this embodiment, the road characteristic relational expression will be described assuming an underwood traffic model of (Equation 2). However, the present invention is not limited to the underwood traffic model, and other traffic models may be used. is there. In addition, the road characteristic and the road characteristic relational expression should show not only the relationship between the speed (average speed) v and the traffic density k, but also the relationship between the speed (average speed) v and the traffic volume Q. You can also.

The speed characteristic calculation device 12 is a device that observes a traveling speed obtained from a vehicle detector, a probe car, or the like for a predetermined period, and calculates a feature amount indicating a traveling environment of the road section. Here, the feature amount is defined as a speed characteristic. The speed characteristic is expressed using one of the following relationships.
(A) Average speed and variance for a given period (standard deviation is acceptable)
(B) Average speed and congestion rate (c) Average speed and number of stops The above relationship represents “ease of running”. This “ease of running” is defined as the speed characteristic of the road section. When travel data such as the number of brakes and steering wheel operation (operation angle) are obtained from the probe car in addition to the travel speed, “ease of travel” or speed characteristics may be calculated including the travel data. The speed characteristic is a characteristic amount obtained by statistically processing a set of travel speed data obtained from a vehicle sensor, a probe car, or the like. This feature amount reflects the tendency or property of the set of travel speed data.

  The characteristic data synthesis device 13 is a device that associates the road characteristic calculated by the road characteristic estimation device 11 with the speed characteristic calculated by the speed characteristic calculation device 12. More specifically, both characteristic data are stored in the storage device 18 in association with the unique road section ID given to each road section.

  The speed characteristic calculation device 14 is a device that acquires a traveling speed from a probe car via a communication unit and calculates a speed characteristic of an estimated road section. The calculated speed characteristic is the same as that of the speed characteristic calculation device 12. A feature amount (speed characteristic of the estimated road section) indicating the traveling environment of the estimated road section is calculated from the traveling speed observed for a predetermined period. The obtained speed characteristic of the estimated road section is output to the road characteristic selection device 15. The speed characteristic calculated by the speed characteristic calculation device 14 is a feature amount obtained by statistically processing a set of travel speed data acquired from the probe car. This feature amount reflects the tendency or property of the set of travel speed data.

  In the road characteristics and the speed characteristics described above, the average speed is used as the speed v, but a speed other than the average speed may be used as a representative value.

  The road characteristic selection device 15 selects a road section ID having a speed characteristic most similar to the speed characteristic of the estimated road section from the speed characteristics of the representative road stored in the storage device 18. The road characteristic of the selected road section ID is selected from the storage device 18 and is output to the traffic estimation apparatus 16 as the road characteristic of the estimated road section.

  The traffic estimation device 16 uses the travel speed of the estimated road section obtained from the probe car and the road characteristics of the estimated road section given from the road characteristic selection device 15 to determine the time zone during which the probe car traveled. Calculate the traffic volume of the estimated road section. The calculated traffic volume is output to the traffic volume output device 17.

  The traffic volume output device 17 outputs the traffic volume of the estimated road section obtained from the traffic volume estimation device 16 to an external application (not shown) such as a display device (not shown) or a traffic simulation.

  The traffic volume, travel speed, and travel speed of the estimated road that are input to the traffic volume estimation system 10 are input in real time via communication means (not shown), and are stored for a predetermined period of time. Stored data may be input. When handling the accumulated data of the latter, specify the representative road and area that generate the road characteristics of the road section, the road information estimated by the traffic estimation system, its section ID, or area information such as mesh code, and the traffic volume. A user interface is provided so that a time zone to be set can be set, and the user interface can be operated via the user interface. Further, when the accumulation period becomes longer, the estimation process takes time. In preparation for the case where the estimated processing time becomes longer, the estimated processing time is calculated from the past processing time, and the remaining processing time or the processing end time is displayed on the display device.

  FIG. 2 shows an example of the format of the traffic volume data output from the traffic volume output device 17. The traffic volume data includes the road ID 21, the estimated date and time 22 that is the start time of the time zone in which the traffic volume was estimated (or the representative time of the time zone), the provision date and time 23 at which this data was provided, and the estimated traffic volume at the estimated date and time. 24. The estimated traffic volume 24 is the estimated number of vehicles passing the road in a predetermined unit time with the estimated date 22 as the representative time. The unit time is set in accordance with the probe car data processing time zone. For example, when the travel speed update interval provided from the probe car is 5 minutes, the estimated time zone of the estimated traffic volume 24 is set to 5 minutes. It is also conceivable to set 30 minutes or 60 minutes as an estimated time zone with a predetermined multiple of 5 minutes, and estimate the traffic volume from the average traveling speed in the set estimated time zone.

  FIG. 3 is an example of a data format of the travel speed of the representative road and the estimated road input to the traffic volume estimation system 10. Here, assuming that the traveling speed is provided from the probe car, a part of traveling data that can be collected from the vehicle is included in addition to the traveling speed. The probe data provided from the probe car includes a road section ID 31 for identifying the road section where the probe data is collected, a vehicle ID 32 for identifying the vehicle, and an entry time 33 which is the time when the vehicle entered the road section. The outflow time 34 that has flowed out of the road section is included. The probe data further includes, as travel information, an average speed 35 when the vehicle travels on this road section, the number of times the vehicle has stopped 36, the number of times of deceleration by braking (the number of decelerations may be estimated from the number of accelerations), etc. Including.

  FIG. 4 shows a processing flow of the road characteristic estimation apparatus 11. FIG. 5 is a supplementary diagram for explaining FIG. 4 and shows the relationship between traffic density k and speed v. The road characteristic estimation device 11 observes the traffic volume and travel speed provided from an external device such as a vehicle detector or a probe car for a predetermined period in all representative road sections and stores them in an internal memory. This processing flow is executed at regular intervals, and the road characteristics are always updated based on the latest traffic volume and travel speed. Hereinafter, each step will be described in detail.

  In step S40, a loop is started in which the processing from steps S41 to S44 is repeated for the traffic volume and travel speed for a predetermined period of all the representative road sections within the predetermined period stored in the internal memory.

  In step S41, the number of vehicles Q (i) for each unit time i of each representative road section in the observed predetermined period is totaled. Step S42 calculates the average speed v (i) of each road section observed for a predetermined period. In step S43, the traffic density k (i) of each road section is calculated from Q (i) and v (i) based on the relational expression of (Expression 1). The relationship between v (i) and k (i) calculated here is shown as a point 51 on the graph with the horizontal axis representing traffic density and the vertical axis representing speed as shown in FIG. Such points representing the relationship between v (i) and k (i) are calculated for every unit time i on each representative road.

  In step S44, the road characteristic (52 in FIG. 5) based on the traffic model of (Equation 2) is estimated from the relationship between the plurality of k (i) and v (i) calculated in step S43, and the road characteristic parameter vf and k0 is output. In step S45, it is determined whether or not the processing for generating the road characteristics and its parameters has been completed for all the representative road sections. If the processing has been completed, the processing flow ends. If the processing has not been completed in all the representative road sections, the process returns to step S41.

  FIG. 6 shows another processing flow of the road characteristic estimation apparatus 11. FIG. 7 is a supplementary diagram for explaining the processing flow of FIG. 6 and shows the relationship between the traffic density k and the speed v. The processing flow of FIG. 5 shows processing on the assumption that the road characteristic of the representative road section and the estimated road section is a traffic model composed of a single curve. However, the traffic model cannot always be expressed by a single curve, and a traffic model in which a plurality of curves are combined depending on the location may be used. This processing flow executes processing for estimating the road characteristics of the road section represented by such multiple curves. Hereinafter, each step will be described in detail.

  The processing from step S40 to S43 is as described in FIG. In step S60, the process from step S61 to S63 is repeated in order to generate a plurality of road characteristics by dividing the relationship between v (i) and k (i) into discontinuities, that is, in an optimum speed range. A loop is started.

  In step S61, a speed boundary value vd (71 in FIG. 7) for generating road characteristics is set. Here, the number of road characteristics that can be expressed by combining the road characteristics of the representative and estimation target road sections, and the selection of the traffic model that serves as the reference for the road characteristics in each speed range are determined before executing this process. The relationship between v (i) and k (i) is analyzed and determined. Here, as shown in FIG. 7, it is assumed that the road characteristics of the road section are expressed by combining the road characteristics in two speed ranges of a high speed basin and a low speed basin of the traveling speed v. And each road characteristic of two speed ranges shall follow the underwood traffic model of (Formula 2). The speed boundary vd set in step S61 is corrected to an optimal value by the iterative process S60 with an appropriate speed set in advance as an initial value. In step S62, the road characteristics are estimated using the relationship between v (i) and k (i) in each velocity region (the above-described high-speed basin and low-speed basin) divided by vd. Thus, the road characteristics (72 in FIG. 7) and its characteristic parameters (vfh, k0h) and the road characteristics (73 in FIG. 7) and its characteristic parameters (vfl, k01) are obtained.

  Step S63 reproduces the traffic density k ′ (i) from the road characteristic generated in step S62 and v (i), calculates an error dk with k (i), and the road characteristic and speed boundary where dk is minimized. Holds vd.

  In step S64, it is determined whether or not the error evaluation of the road characteristics calculated for all vd within the predetermined speed range has been completed. The characteristic parameter and the speed boundary value vd are output, and this processing flow ends. If the processing has not been completed for all vd, vd is updated and the process returns to step S60. In step S45, it is determined whether or not the processing for generating the road characteristics and its parameters has been completed for all the representative road sections. If the processing has been completed, the processing flow ends. If the processing has not been completed in all the representative road sections, the process returns to step S41.

FIG. 8 shows a processing flow of the speed characteristic calculation device 12. FIG. 9 is a supplementary diagram for explaining the processing flow of FIG. 8 and shows the relationship between the variance σ and the average velocity v _av . The speed characteristic calculation device 12 observes a traveling speed provided from an external device such as a vehicle detector or a probe car for a predetermined period in all representative road sections, stores it in an internal memory, and calculates an average speed per unit time i. Calculate the variance. Here, the relationship between the average speed and the dispersion is the speed characteristic. This processing flow is executed at regular intervals, and the speed characteristics are always updated based on the latest traveling speed. Hereinafter, each step will be described in detail.

In step S80, a loop is started in which the processing from steps S81 to S83 is repeated for the traveling speeds of all road sections within a predetermined period stored in the internal memory. Step S81 calculates average speed _vav from the running speed of a predetermined period. In step S82, the running speed variance σ for a predetermined period is calculated using v _av calculated in step S81.

V _av and σ calculated in step S81 and step 82 represent the speed characteristics of this road section. For example, as shown in FIG. 9, the speed characteristics calculated in each speed range of each road section are plotted in a space where the vertical axis is v _av and the horizontal axis is σ. In this example, it is assumed that the road characteristics of a road section are expressed by two characteristics of a high speed basin and a low speed basin, and the speed characteristics of each road section and each speed range are obtained. Fast basin road characteristic _{p Rh_001} the point 91 of the road section ID001, road characteristic _{p Rl_001} the point 93 of the low speed basin, fast basin road characteristic _{p Rh_010} the point 92 of the road section ID010, the road characteristic _{p Rl_010} slow watershed point 94. If the points plotted on this space are close, this means that the speed characteristics of the road sections, that is, the traveling environment are similar.

  The road section in which the feature point is placed on the upper left in FIG. 9 is considered to be a road section that is less likely to be jammed, has a high traveling speed, and is easy to travel at a stable speed (small variation in speed). On the other hand, the road section where the feature point is placed in the lower right is considered to be a road section where traffic congestion occurs frequently at low speeds (the speed varies greatly) and is difficult to run.

  Step S83 determines whether or not the processing has been completed in each speed basin when the road characteristic of the road section is composed of a plurality of road characteristics for each speed basin. If the calculation of velocity characteristics is not completed in all velocity flow regions, the process returns to step S81. When the calculation of the speed characteristics is completed in all speed flow areas, the process proceeds to step S84, and the speed characteristics are stored in the internal memory together with the road section ID.

  In step S85, it is determined whether or not the processing for calculating the speed characteristics for all the representative road sections has been completed. When the process is completed, the process flow ends. If the processing has not been completed for all representative road sections, the road section to be calculated is updated, and the process returns to step S81.

  In this embodiment, the speed characteristic of the road section is expressed by a spatial point consisting of the relationship between the average speed and the variance. Instead of dispersion, a method of calculating a standard deviation, a traffic jam rate, a traffic jam time, a traffic jam frequency, a vehicle stop time or the number of stops from the traveling speed, and using it may be considered. In addition, when information about the brake and the accelerator is obtained from the probe car, the speed characteristic may be expressed by the relationship between the data and the average speed. The former is processing data (result) based on the distribution state of travel speed data for a road section for which speed characteristics are to be obtained. Alternatively, it is processing data (result) representing the distribution state of the traveling speed data. This processing data can be obtained by statistically processing a set of traveling speed data. Further, the processing data is processed in association with a representative speed value (for example, average speed), and changes depending on the driving environment even if the representative speed value is the same value. The latter brake and accelerator information is information related to an operation that causes a change in the traveling speed of the probe car (vehicle). This information is statistically processed in association with the representative speed value, and becomes processing data (result) that varies depending on the driving environment. As such processing data, for example, the number of brakes and the number of accelerators can be obtained. This processing data varies depending on the driving environment even if the speed representative value is the same value. Accordingly, this processing data affects the distribution state of the traveling speed data. Both the former and the latter processing data are data related to the distribution state of the traveling speed data, and are traveling data obtained based on the traveling state of the probe car or the vehicle. This processing data varies depending on the driving environment even if the speed representative value is the same value. Further, the processing data (travel data) changes based on the travel state.

  FIG. 10 is a supplementary diagram for explaining a method for calculating the traffic jam rate. This congestion rate is used to express speed characteristics instead of the variance σ in step S82 described in FIG. 100 is a change in the traveling speed v of a certain road section, 101 is an observation time T, and 102 is a reference speed vth for determining whether or not there is a traffic jam. The reference speed vth is set by using the speed value of the traveling speed of the observation time T, the tens percentile value, the regulation speed, or the like. In a road section expressed by a plurality of road characteristics divided into speed basins, the reference speed vth is not always set to a traffic jam basin or a low speed basin, and may be set to a free basin or a high speed basin. At the observation time T, the time zones t1, t2, and t3 where the traveling speed is lower than vth are traffic jam times. From this, the congestion rate ConjT of this road section is calculated by (Equation 5). j is the number of traffic time zones (j = 1, 2,...) (in this example, j = 3).

ConjT = Σt (j) / T (Formula 5)
The congestion rate of each road section obtained by (Equation 5) is plotted on the horizontal axis in FIG. 9 to express the speed characteristics. The road section in which the feature point is placed on the upper left in FIG. 9 is an easy-to-run road section that can travel at a high speed. This road section is a road section with a short traffic time (not so crowded). The road section where the feature point is placed in the lower right is a road that is difficult to run at low speed (problems in road structure such as poor road surface conditions and curves, etc. may be considered). In addition, this road section is a road section that has a long traffic jam. In general, roads with a high congestion rate are considered to have a low average speed. On the other hand, in the road section where the speed at the time of non-congestion is high, even if the speed at the time of traffic congestion is taken into consideration, the average speed is calculated to be higher. For this reason, it is not possible to see whether the road section is subject to traffic jams. Therefore, by placing a feature point in a space centered on the traffic congestion rate and speed as shown in FIG. 9, the speed characteristic of the road section is shown, and the similarity of the traffic flow in the road section can be determined.

  Here, the traffic congestion rate indicates whether the road section has a lot of traffic. Instead of this traffic congestion rate, the number of times that traffic congestion was detected in each time zone of the observation time T (the number of traffic jams), and the value obtained by dividing the traffic congestion frequency by the unit time number of the observation time zone T (the occurrence frequency of traffic jams) ) May be used.

  FIG. 11 shows a processing flow of the characteristic data synthesis device 13. This processing flow is executed at the timing when the characteristic data is updated in the road characteristic estimation device 11 or the speed characteristic calculation device 12. Hereinafter, each step will be described in detail.

In step S110, a loop is started in which the processing from steps S111 to S114 is repeated for the road characteristics and speed characteristics of all representative road sections held in the internal memory. Step S111 reads the road characteristics and the road characteristic parameters (vf and k0) held in the internal memory, and the speed boundary vd when the road section is represented by a plurality of different road characteristics divided into speed basins. In step S112, speed characteristics _vav and σ having the same ID as the road section ID read in step S111 are read. In step S113, when the road characteristics of the road section are configured by a plurality of road characteristics for each speed basin, it is determined whether or not the processing is completed in each speed basin. If the process has not been completed, the process returns to step S111. When the reading processing of all road characteristics and speed characteristics of the road section is completed, the process proceeds to step S114. In step S114, the read characteristic data is associated with the road section ID and output to the storage unit 18. In step S115, it is determined whether or not the process of associating the characteristic data for all the representative road sections has been completed. When the process is completed, the process flow ends. If the process has not been completed, the representative road section to be processed is updated, and the process returns to step S111.

  FIG. 12 shows an example of a characteristic data format including road characteristics, characteristic parameters, and speed characteristics of a representative road section output from the characteristic data synthesizing apparatus 13 and stored in the storage device 18. Here, it is assumed that the road characteristics of the representative road section stored in the storage unit 18 are composed of a plurality of (two) road characteristics that are divided into two speed basins, a high speed basin and a low speed basin. Regarding road characteristics and speed characteristics, the characteristic data is managed by two tables of characteristic data 120 of the high speed basin and characteristic data 121 of the low speed basin. The speed boundary values of the two speed basins are managed together with the road section ID (127). The characteristic data is composed of road characteristics and speed characteristics. The road characteristic includes a road section ID (122), a road characteristic (base traffic model) (123), and road characteristic parameters (αh, βh). The speed characteristic S126 includes a road section ID (125) and a speed characteristic 126. Here, since road characteristics based on different traffic models in the road sections are shown, the characteristic parameters are αh and βh based on (Equation 4). Each characteristic is associated with a road section ID (122 and 125). The low-speed basin characteristic data 121 is the same as the high-speed basin characteristic data 122.

FIG. 13 shows a processing flow of the road characteristic selection device 15. FIG. 14 is a supplementary diagram for explaining the processing flow of FIG. 13 and shows the relationship between the variance σ and the average velocity v _av . The road characteristic selection device 15 selects from the storage means 18 the road characteristics of the representative road section having speed characteristics similar to the speed characteristics of the estimated road section obtained from the speed characteristic calculation device 14. Hereinafter, each step will be described in detail.

  In step S130, a loop is started in which the processes from step S131 to S134 are repeated for the speed characteristics of all estimated road sections obtained from the speed characteristic calculation device 14.

  In step S131, the speed characteristic of the estimated road section acquired from the speed characteristic calculation device 14 and temporarily stored in the internal memory is read. If there are a plurality of speed characteristics of the estimated road section, all the speed characteristics are read. Step 132 reads the speed characteristics of the representative road section from the storage means 18. In step S133, the speed characteristic of the representative road section closest to the speed characteristic of the read estimated road section is searched.

This process will be specifically described with reference to FIG. FIG. 14 shows the velocity characteristics of each velocity basin in each road section plotted on the space of average velocity v _av and variance σ. 142 is a speed characteristic p _{rh —} 001 of the representative road section ID 001, 143 is a speed characteristic p _{rh —} 010 of the representative road section ID 010, 144 is a speed characteristic p _{rh — 101} , 145, ₁₄₆ , and 147 of the representative road section ID 101 are low speed basins of the representative road section Speed characteristics p _{rl — 001} , p _{rl — 010} , and p _{rl — 101} are shown. The speed characteristics of these representative road sections are read from the storage means 18 in step 132. 140 indicates the speed characteristic p _{eh —} 555 of the high speed basin of the estimated road section ID ₅₅₅ , and 141 indicates the speed characteristic p _{el — 555} of the low speed basin of the estimated road section. 140 and 141 are calculated using the estimated road travel speed in the speed characteristic calculation device 14. Although the processing flow of the speed characteristic calculation device 14 is not shown in the present embodiment, it is the same as the processing flow described in FIG. 8, and the input data in FIG. 8 is the travel speed of the estimated road section. The travel speed of the estimated road section is obtained from the travel data collected by the probe car, and is provided to the speed characteristic calculation device 14 via communication means or a storage medium. In step S133, the speed characteristic of the representative road section closest to 140 and 141 is searched. In FIG. 14, the speed characteristic closest to the speed characteristic 140 of the high-speed basin is 142 and the speed characteristic closest to the speed characteristic 141 of the low-speed basin is 146, and these two speed characteristics are selected.

  A step S134 sets the road characteristic (123) and its characteristic parameter (124) from the road section ID (125 in FIG. 12) of the selected speed characteristic to the road characteristic of the estimated road section. Here, for the estimated road section ID 555, the road characteristic underwood of the representative road section ID001, characteristic parameters αh = 70, βh = 20 are set as the road characteristics of the high-speed basin. Further, the road characteristic Greenberg of the representative road section ID010, characteristic parameters αl = 25, βl = 10 are set as the road characteristics of the low-speed basin. The road characteristics, parameters, and speed boundary value vp of the set estimated road section are output to the storage means 18 and stored.

  In step S135, it is determined whether or not the processing for setting the corresponding road characteristics for all estimated road sections has been completed. When the process is completed, the process flow ends. If the process has not been completed, the estimated road section to be processed is updated, and the process returns to step S131.

  When the travel speed v of the estimated road section is input, the traffic volume estimation device 16 reads the road characteristics and characteristic parameters of the estimated road section ID, and the speed boundary value vp from the storage means 18, and the input travel speed is high. Determine whether it belongs to the basin or low-speed basin, calculate the traffic density k of the estimated road section using the road characteristics of the determined speed basin, and calculate the traffic volume Q of the estimated road section based on (Equation 1) Calculate and output to the traffic output device 17 as the estimated traffic. The traffic volume output device 17 outputs the estimated traffic volume to an external device via communication or a storage medium.

  In S133, when the number of feature points in the representative road section is small, the closest speed characteristic may be selected. However, since the characteristics of the estimated road section are limited to one representative road section, the estimation accuracy is limited. Therefore, if there are many feature points in the representative road section, use a data classification method such as a clustering method to divide similar feature points, that is, neighboring feature points into several groups, And the representative feature point closest to the feature point of the estimated road section is selected. For example, 148 is one of the groups generated by the clustering method, and 149 indicates the representative feature points of the group 148. If the feature point of the estimated road section is close to 149, the estimated road section is determined to be a road section similar to the group 148. At this time, in S134, a representative value (an average value, a median value, a mode value, etc.) is calculated and set from a plurality of road characteristics of the road sections forming the group 148 for the road characteristics of the estimated road section. In this way, the road characteristic selection process for the estimated road section may be switched according to the number of feature points of the representative road section.

  Here, the travel speed of the representative road section and the estimated road section input to the traffic volume estimation system 10 is not limited to the probe car, and the same applies even when using a terminal having position detection means such as a mobile phone or a smartphone. . Further, the traffic volume estimated from the traveling speed data collected from the probe car is not limited to the same type of vehicle, and it is also possible to estimate the traffic volume including a motorcycle or the like that affects the traveling of the probe car. . That is, in the present embodiment, the vehicle that is subject to traffic is an automobile that has a prime mover and travels on the road surface. Motor vehicles include motorcycles. The trolley bus may be excluded when there is a dedicated road surface, but may be included in the car when traveling on the same road surface. Motorbikes can be targeted depending on the road segment.

  As described above, according to the present embodiment, in the traffic volume estimation system that estimates the traffic volume of the road segment from the travel speed data collected from the probe car, the representative roads in which the road characteristics (traffic model) of the estimated road segment are calculated in advance. When selecting from among the road characteristics, the speed characteristics of each road section are calculated from the travel speed data obtained from the probe car, and the road characteristics of the representative road section having similar speed characteristics are calculated as the road characteristics of the estimated road section. Since the traffic volume is calculated as follows, the estimation accuracy of the traffic volume in the road section where the traffic volume cannot be directly observed can be accurately estimated without depending on the accuracy of the road map.

  Further, when the road characteristics of the road section are expressed by different road characteristics in a plurality of speed basins, the estimated road section calculated for each speed basin and the speed characteristics of the representative road section are compared, and similar speed characteristics are obtained. Since the road characteristic of the representative road section is the road characteristic of the estimated road section and the traffic volume of the estimated road section is calculated, even if the traffic volume of the road section having multiple road characteristics cannot be observed directly, the probe car As long as travel data is obtained, it is possible to accurately estimate the traffic volume of the road section.

  In addition, this invention is not limited to an above-described Example, Various modifications are included. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations. In addition, it is possible to add, delete, and replace other configurations for a part of the configuration of the embodiment.

  Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. Each of the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by the processor. Information such as programs, tables, and files for realizing each function can be stored in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

  Further, the control lines and information lines indicate what is considered necessary for the explanation, and not all the control lines and information lines on the product are necessarily shown. Actually, it may be considered that almost all the components are connected to each other.

  DESCRIPTION OF SYMBOLS 10 ... Traffic volume estimation system, 11 ... Road characteristic estimation apparatus, 12 ... Speed characteristic calculation apparatus, 13 ... Characteristic data synthesizer, 14 ... Speed characteristic calculation apparatus, 15 ... Road characteristic selection apparatus, 16 ... Traffic volume estimation apparatus, 17 ... Traffic output device, 18 ... Storage device

Claims (8)

Based on representative road characteristics that define the relationship between the vehicle travel speed and traffic volume for the representative road section, and the travel speed data of the vehicle passing through the estimated road section for estimating the traffic volume, the traffic volume of the estimated road section In the traffic volume estimation system that estimates
For a plurality of representative road sections, a storage device that stores representative road characteristics and speed characteristics obtained by statistically processing a set of travel speed data obtained from vehicles related to the representative road characteristics;
A speed characteristic calculating device for calculating a speed characteristic of a vehicle passing through the estimated road section;
Based on the speed characteristics of the estimated road section calculated by the speed characteristic calculation apparatus and the speed characteristics of the representative road section stored in the storage device, the speed characteristics are used for estimating the traffic volume from a plurality of representative road characteristics. A road characteristic selection device for selecting a representative road characteristic,
A traffic volume estimation system for estimating the traffic volume of the estimated road section based on the representative road characteristics selected by the road characteristic selection device and the travel speed data of the estimated road section. The traffic volume estimation system according to claim 1,
The traffic volume estimation system, wherein the representative road characteristic is set using a model formula indicating a relationship between traffic volume or traffic density and travel speed. The traffic volume estimation system according to claim 2,
The speed characteristic of the representative road section and the estimated road section is represented by a spatial value of travel data that changes based on the representative value of the travel speed for a predetermined time and the travel state of the vehicle in each road section. Quantity estimation system. In the traffic estimation system according to claim 3,
A road characteristic estimation device that estimates a representative road characteristic of the representative road section from the travel speed data and traffic volume of the representative road section; a speed characteristic calculation device that calculates a speed characteristic from the travel speed data of the representative road section; A characteristic data synthesizing device that associates the representative road characteristic and the speed characteristic of the representative road section,
The storage device stores characteristic data of a representative road section in which representative road characteristics and speed characteristics are associated with each other. In the traffic estimation system according to claim 4,
The speed characteristic calculation device calculates a representative value of the traveling speed of the estimated road section using the traveling speed data of the estimated road section for a predetermined time,
As the travel data, at least one of travel speed distribution, standard deviation, congestion rate, congestion time, congestion frequency, vehicle stop count, vehicle stop time, brake operation count, brake operation time, accelerator count, or accelerator time is calculated. A traffic volume estimation system characterized by that. In the traffic estimation system according to claim 4,
The similarity between the speed characteristic of the representative road section and the speed characteristic of the estimated road section stored in the storage device is determined, and the representative road characteristic of the representative road section having a speed characteristic similar to the speed characteristic of the estimated road section is determined. A similarity determination device to select,
The similarity determination device uses a speed characteristic represented by a representative value of travel speed and a variance value of travel speed, and represents a representative road having a speed characteristic that is spatially close to the speed characteristic of the estimated road section. A traffic estimation system, wherein a section is determined as a representative road section similar to the estimated road, and a representative road characteristic of the representative road section is selected as a road characteristic of the estimated road. In the traffic estimation system according to claim 4,
The similarity between the speed characteristic of the representative road section and the speed characteristic of the estimated road section stored in the storage device is determined, and the representative road characteristic of the representative road section having a speed characteristic similar to the speed characteristic of the estimated road section is determined. A similarity determination device to select,
The similarity determination device uses a speed characteristic represented by a representative value of a traveling speed and a traffic jam rate, and represents a representative road section of a speed characteristic that is spatially close to the speed characteristic of the estimated road section. It is determined that the representative road section is similar to the estimated road, the representative road characteristic of the representative road section is selected as the road characteristic of the estimated road,
The traffic volume estimation system is characterized in that the traffic congestion rate is based on a time or number of times when a traveling speed below a predetermined speed is observed in traveling speed data for a predetermined time. In the traffic estimation system according to claim 4,
The representative road characteristics of the representative road section are composed of different representative road characteristics for each speed band having a predetermined speed as a boundary value.
The storage means stores representative road characteristics for each speed band and corresponding speed characteristics,
The speed characteristic calculation device calculates a speed characteristic of the estimated road section for each speed band,
A traffic volume estimation system, wherein a representative road characteristic of a representative road section having a speed characteristic similar to the speed characteristic of the estimated road section is selected as a road characteristic of the estimated road section in each speed band.

Images