JP2019040478A - Road traffic information processing apparatus and road traffic information processing method - Google Patents

Abstract

To provide a road traffic information processing technique capable of further improving an estimation accuracy of feature points such as a traffic jam generation start point and a traffic jam resolution start point on a time-space plane.SOLUTION: In a time-space plane, an inflection point extraction unit 11 extracts a deceleration inflection point and an intensification inflection point from at least a part of a plurality of trajectories representing a time series of each position of a plurality of moving objects X. The deceleration inflection point indicates that a speed has been reduced due to the moving object X entering a traffic jam. The acceleration inflection point indicates that a speed has increased due to the moving object X having escaped the traffic jam. A representative point setting unit 12 sets a representative point of a designated inflection point among the plurality of inflection points. On the basis of the representative points, a feature point estimation unit 13 estimates feature points that demarcate a free flow area and a congested flow area in the time-space plane.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a technique for processing information collected from a client as each of a plurality of mobile stations.

  Collecting probe information or FCD (floating car data) including the position of the vehicle measured in real time using the GPS function of a portable device such as a navigation device or a smartphone mounted on the vehicle, Services that provide road traffic information using probe information are being developed. Where road traffic conditions change in a complex manner due to the occurrence of a sudden event such as an accident or a fallen object, a method for predicting the travel time when a sudden event occurs using only probe information has been proposed (see Non-Patent Document 1). .

Specifically, in the time-space plane shown in FIG. 5, a sudden event occurred at time t = t ₀ and point x = x _0, and the processing of this sudden event was completed at time t = t _d . The trajectory P _i (i = 1, 2,...) Represented by the probe information of each vehicle in the case is considered. In this case, the vehicle has reached to the end of the traffic jam that passes through the incident occurs sectional x = x ₀ at time t = t ₀ later, it enters into the congestion its rate decreases. On the other hand, when the vehicle reaches the head of the traffic jam and escapes from the traffic jam, the speed increases.

In each of the plurality of trajectories P _i , a congested stretch wave L ₁ is defined by connecting a plurality of inflection points S _i (decrease inflection points) at which the speed of the vehicle decreases as a traffic jam enters. In each of the plurality of trajectories P _i , a traffic jam elimination wave L ₂ is defined by connecting a plurality of inflection points R _i (rising inflection points) at which the speed of the vehicle increases as the traffic escapes. A substantially triangular region surrounded by the traffic jam stretching wave L ₁ , the traffic jam elimination wave L _2, and the sudden event occurrence cross section is defined as a traffic jam flow region, and the outer region is defined as a free flow region.

The travel time prediction method will be described. As shown in FIG. 6, the time - in the space plane, based on the trajectory of the vehicle in accordance with the probe information present at the current time t = t _1, the traffic jam stretching wave L ₁ and at the current time t = t ₁ after extending aspects of traffic jam wave L ₂ is predicted (see dashed lines). Furthermore, the trajectory P _{j of} the vehicle is predicted based on the prediction results of the extension modes of the traffic jam stretching wave L ₁ and the traffic jam canceling wave L ₂ (see thick line). The sum of the travel time in the free flow region before entering the traffic jam, the travel time in the traffic flow region, and the travel time in the free flow region after exiting the traffic jam is the predicted travel time T = t ₂ −t ₁ (estimated arrival time). t = t ₂ ).

“Travel Time Prediction when Sudden Event Occurs Using Only Vehicle Trajectory” Sekizuka Kiichi et al. The 53rd Japan Society for Civil Engineering Planning Presentation 28-29 May 2016

However, after estimating the position of the traffic jam occurrence point (see FIG. 6 / (t ₀ , x ₀ )) on the time-space plane, the extension mode of the traffic jam stretching wave L ₁ and the jam clearing wave L ₂ is predicted. Sometimes problems arise. That is, if there are multiple lanes (or vehicle lanes) on the road, the traffic flow in each of the multiple lanes may be different, so the inflection point may fluctuate depending on the lane in which the vehicle traveled (See Fig. 6 / ○: deceleration bending point and ●: acceleration bending point). For this reason, there is a possibility that the prediction accuracy of road traffic information such as a traffic jam starting point and the required travel time T on the road is lowered.

  Therefore, an object of the present invention is to provide a road traffic information processing technique that can further improve the estimation accuracy of feature points such as a traffic jam starting point and a traffic jam eliminating starting point on a time-space plane.

  The road traffic information processing apparatus according to the present invention is such that the moving body enters a traffic jam from at least a part of a plurality of trajectories representing a time series of positions of the plurality of moving bodies on a time-space plane. A bending point extraction unit that extracts a deceleration bending point that indicates that the speed has decreased due to the above, and an acceleration bending point that indicates that the speed has increased because the moving body has escaped from a traffic jam, and the bending point extraction unit A representative point setting unit that sets a representative point of the specified bending point among the plurality of bending points extracted by the step, and the moving body is arranged on the time-space plane based on the representative point set by the representative point setting unit. And a feature point estimation unit that estimates a feature point that demarcates a boundary of a traffic flow region that is not in contact with a traffic jam and a traffic flow region in which the mobile body is in contact with the traffic jam.

  According to the road traffic information processing apparatus of the present invention, in the time-space plane, the influence of the fluctuation of the designated bending point can be reduced by setting the representative point of the designated bending point among the plurality of bending points. Further, it is possible to further improve the estimation accuracy of feature points such as a traffic jam starting point and a traffic jam eliminating starting point.

  In the road traffic information processing apparatus of the present invention, the representative point setting unit sets the center of gravity of a plurality of accelerated bending points as the designated bending point as the representative point, and the feature point estimating unit determines the representative point as the representative point. It is preferable that an intersection of a straight line parallel to the time axis and a straight line or a curve determined by a deceleration bending point on the same locus as the designated bending point is estimated as a traffic jam starting point as the feature point.

  According to the road traffic information processing apparatus having the above configuration, the representative point of the acceleration flexion point is set in view of the fact that fluctuations are likely to occur at the acceleration flexion point that is temporally or spatially close from the start point of the traffic jam. Therefore, the influence of the fluctuation is reduced. For this reason, it is possible to further improve the estimation accuracy of the traffic jam occurrence point as a feature point on the time-space plane.

  In the road traffic information processing apparatus of the present invention, the representative point setting unit determines the center of gravity of the acceleration flexion point as the designated flexion point that is spatially separated from the traffic jam starting point by a first threshold or more on the near side. And the feature point estimation unit estimates an intersection of a straight line passing through the traffic jam starting point parallel to the time axis and a negative slope straight line passing through the representative point as the traffic jam elimination starting point as the feature point. It is preferable to do.

  According to the road traffic information processing apparatus having the above configuration, the acceleration bend point that is spatially farther from the traffic jam starting point by the first threshold or more is set as the designated bend point and the representative point is set. For this reason, it is possible to avoid a situation in which the congestion elimination starting point is estimated in a situation where it is difficult to say that the congestion elimination wave (see FIG. 5) has occurred in a space that is excessively close to the origin of the congestion. It is possible to improve the estimation accuracy of the traffic jam elimination starting point.

  In the road traffic information processing apparatus of the present invention, the representative point setting unit is required that a spatial or temporal interval between a plurality of acceleration bend points whose order is continuous in time series is equal to or less than a second threshold value. It is preferable that the center of gravity of the plurality of accelerated bending points as the designated bending point is set as the representative point.

  According to the road traffic information processing apparatus having the above configuration, the acceleration inflection points having consecutive ranks in time series are excessively distant from each other, and it is difficult to say that the congestion elimination wave having a uniform change mode has occurred. Since the situation where the traffic jam elimination starting point is estimated in the situation is avoided, the estimation accuracy of the traffic jam elimination starting point as the feature point can be improved.

  In the road traffic information processing apparatus according to the present invention, the feature point estimation unit estimates the congestion elimination starting point after determining the negative slope based on an arrangement form of the acceleration bending point as the designated bending point. It is preferable.

  According to the road traffic information processing apparatus having the above configuration, it is based on a straight line suitable for dynamically expressing the congestion elimination wave according to the appearance of the time-series continuous acceleration inflection points, and therefore, the congestion as a feature point. The estimation accuracy of the cancellation start point can be improved.

1 is a configuration explanatory diagram of an information processing system according to an embodiment of the present invention. FIG. Explanatory drawing regarding the function of the information processing server which concerns on one Embodiment of this invention. Explanatory drawing regarding the estimation of the traffic jam starting point. Explanatory drawing regarding the estimation of the congestion elimination starting point. Explanatory drawing regarding the free flow area | region and traffic congestion area | region in a time-space plane. Explanatory drawing regarding the road traffic information generation in a time-space plane. Explanatory drawing regarding one output form of road traffic information. Explanatory drawing regarding the other output form of road traffic information.

(Constitution)
The road traffic information processing system as an embodiment of the present invention shown in FIG. 1 includes a road traffic information processing server 1 (road traffic information processing apparatus) and a plurality of clients 2. Each of the road traffic information processing server 1 and the plurality of clients 2 can communicate with each other via a network. The client 2 has a communication function with a moving body X having a function of moving according to a user's driving operation, such as a vehicle such as a four-wheeled vehicle, a two-wheeled vehicle, and a bicycle.

(Configuration of road traffic information processing server)
The road traffic information processing server 1 constituting the “road traffic information processing apparatus” of the present invention includes a database 10, a bending point extraction unit 11, a representative point setting unit 12, a feature point estimation unit 13, and a road traffic information generation. Part 14. The information processing server 1 may be configured by a portable terminal device such as a smartphone or a tablet.

  The database 10 stores and holds probe information including the time series of the moving object X uploaded from the client 2 to the information processing server 1. In addition, the database 10 stores and holds road traffic information and map information generated by the road traffic information generation unit 14. “Road traffic information” includes the travel cost (predicted travel time) for each link. In “Map information”, the position, shape, posture, etc. of each link that constitutes the road are not only a column of coordinate values ((latitude, longitude) or (latitude, longitude, altitude)), but also for identifying each link. It includes link identification information and data representing attributes. Each link is connected by a node. The database 10 may be configured by a database server that is separate from the information processing server 1.

  The inflection point extraction unit 11, the representative point setting unit 12, the feature point estimation unit 13, the road traffic information generation unit 14, and the necessary software and data are read from the memory, and the data is designated according to the software. An information processing unit (CPU) that executes the above arithmetic processing, and a communication device, a storage device (the memory), and the like as necessary. Details of the designated calculation process will be described later.

(Client configuration)
The client 2 is configured by a portable terminal device such as a smartphone or a tablet. “Portable” means that, for example, its size is about the size of a standard human hand, and its weight can be easily carried with one hand or in a pocket of clothes. The client 2 may be configured by a device that is larger and heavier than a portable device (for example, a device that is assembled in a moving body).

  The client 2 includes a storage device 20, a positioning device 21, an input device 22, an output device 23, and an arithmetic processing device 24.

  The storage device 20 is configured to store and hold the results of arithmetic processing performed by the arithmetic processing device 24. The storage device 20 is configured to store map information. The “map information” includes a coordinate string representing the position, shape, posture, and the like of each link constituting the road, and link identification information for identifying each link.

  The positioning device 21 measures the location (latitude and longitude) of the client 2 at regular time intervals using the GPS function. The input device 22 includes operation buttons and a microphone, and enables various operations or input settings by user operations or speech. The output device 23 includes a display device and a sound output device (speaker), and displays image content such as map information or outputs sound content. “Output” means that information is output in all forms that humans can recognize through the five senses, such as visual, auditory, and tactile sensations, such as display of the information, audio output, and vibration output. The input device 22 and the output device 23 may be configured by a touch panel display.

  The arithmetic processing unit 24 reads the software and data from the designated area of the memory as necessary, and then performs an information processing unit (CPU) that executes the designated arithmetic processing according to the software for the data and the necessity. Accordingly, it is configured by a communication device, a storage device (the memory), and the like.

(function)
The function of the information processing system having the above configuration will be described. In the client 2, the position of the client 2 or the moving body X on which the client 2 is mounted is measured in time series by the positioning device 21, and the measurement results are accumulated in the storage device 20. And the probe information including the time series of the position of the moving body X in response to an event such as a predetermined time has elapsed since the previous transmission of the probe information or the moving body X has moved a predetermined distance or more. Is transmitted from the client 2 to the road traffic information processing server 1 and stored in the database 10.

(Estimation of starting point of traffic jam)
Based on the probe information of each link (or road section) up to the current time or the time immediately before stored in the database 10, the trajectory of each vehicle on the latest time-space plane at the time is expressed (FIG. 5). reference).

The bending point extraction unit 11 travels on the first trajectory P ₁ of one vehicle on the first deceleration bending point S ₁ and the first acceleration bending point R ₁ and on the road section after the one vehicle. The second deceleration bending point S ₂ and the second acceleration bending point R _{2 on} the second locus P ₂ in the time-space plane of the other vehicle are extracted (FIG. 2 / STEP 102). For example, a point where the rate of change of the trajectory of the moving object in the time-space plane is a predetermined value or more is recognized as a bending point in the trajectory. As a result, the bending points S ₁ , R ₁ , S ₂ and R ₂ are detected as shown in FIG. The detection order of the bending point may be any of S ₁ → R ₁ → S ₂ → R ₂ and S ₁ → S ₂ → R ₁ → R ₂ .

The representative point setting unit 12 sets the center of gravity (R ₁ + R ₂ ) / 2 of the first acceleration bending point R ₁ and the second acceleration bending point R ₂ as a representative point (FIG. 2 / STEP 104). The weighted centroid γR ₁ + (1−γ) R ₂ (0 <γ <0.5 or 0.5 <γ <1) may be set as the representative point. The weight 1-γ of the subsequent second acceleration bend point R ₂ in time series may be set larger than the weight γ of the first increase bend point R ₁ . The magnitude of the weight γ is determined according to the length of the time interval between the first acceleration bending point R ₁ and the second acceleration bending point R ₂ , for example. Note that the center of gravity or the weighted center of gravity of three or more accelerated bending points detected by the bending point extraction unit 11 may be set as the representative point.

Then, the feature point estimation unit 13 makes a straight line connecting the first deceleration bending point S ₁ and the second deceleration bending point S ₂ (see the alternate long and short dash line) and the representative point (R ₁ + R ₂ ) / 2 is estimated as an estimated traffic jam starting point Q ₁ = (t ₀ , x ₀ ) (see FIG. 2 / STEP 106 and FIG. 3 / □). ). Thereby, the sudden event occurrence cross section x = x ₀ is estimated (see FIG. 5).

It should be noted that the bending point extraction unit 11 extracts three or more acceleration bending points, sets the centroids or weighted centroids of some acceleration bending points as representative points, and sets the remaining acceleration bending points and the representative points. The intersection of the approximate straight line or approximate curve representing the arrangement form and the approximate straight line or approximate curve representing the arrangement form of the three or more deceleration bending points may be estimated as the estimated congestion occurrence point Q ₁ .

(Estimation of traffic congestion elimination starting point)
It is determined whether or not the first acceleration bending point R ₁ has been extracted by the bending point extraction unit 11 (FIG. 2 / STEP 202). The first acceleration flexion point R ₁ is a time series in which the spatial component of the acceleration flexion point is spatially separated from the traffic jam occurrence point x ₀ by the first threshold value α or more in the time-space plane. This is the first speed-up bending point. When the determination result is affirmative (FIG. 2 / STEP 202... YES), the index j indicating the number of times of acceleration bending point extraction is set to “1” (FIG. 2 / STEP 204).

The representative point setting unit 12 sets the centroids Σ _{k =} 1 to j R _k / j of the first to j-th accelerated bending points R _j (j = 1, 2,...) As representative points (FIG. 2 / (STEP 206). In the state of j = 1, the first acceleration bending point R ₁ is set as a representative point as it is.

Then, the feature point estimation unit 13 uses the intersection of the straight line having the negative slope (congestion elimination wave velocity) Cr passing through the representative point and the sudden event occurrence cross section x = x ₀ as the congestion elimination start point Q as the feature point. ₂ (j) is estimated (see FIG. 2 / STEP 208, FIG. 4). In the state of j = 1, the congestion elimination wave velocity Cr may be a constant value, but the acceleration bending point R ₀ extracted by the bending point extraction unit 11 is connected immediately before the first acceleration bending point R _1. The slope of the straight line may be calculated as the congestion elimination wave velocity Cr (see FIG. 4 / dashed line).

  Subsequently, the inflection point extraction unit 11 determines whether or not the (j + 1) th acceleration inflection point has been extracted (FIG. 2 / STEP 210). The j + 1 acceleration bending point is an acceleration bending point whose spatial or temporal interval is within the second threshold value β from the jth acceleration bending point in the time-space plane among the acceleration bending points.

When the determination result is negative (FIG. 2 / STEP 210... NO), the definition of the j-th acceleration bending point is canceled and the acceleration bending point is detected again (FIG. 2 / STEP 202). ). When the determination result is affirmative (FIG. 2 / STEP 210... YES), the index j is increased by “1” (FIG. 2 / STEP 212), and the representative point setting process by the representative point setting unit 12 (FIG. 2 / STEP 206). And the process of the feature point estimation process (FIG. 2 / STEP208) etc. by the feature point estimation part 13 is repeated. In the state of j = 2, the center of gravity of the first acceleration bending point R ₁ and the second acceleration bending point R ₂ is set as the representative point. Further, although the congestion elimination wave velocity Cr may be a constant value, the inclination of the straight line connecting the first acceleration bending point R ₁ and the second acceleration bending point R ₂ may be calculated as the congestion elimination wave velocity Cr. Good (see FIG. 4 / two-dot chain line).

(Generation of road traffic information)
Then, the road traffic information generation unit 14 generates road traffic information based on the feature point estimation result (FIG. 2 / STEP 214). Based on the estimation results of the traffic jam starting point Q ₁ = (t ₀ , x ₀ ) and the traffic jam canceling point Q ₂ = (t _d , x ₀ ), the sudden event occurrence cross section x = x ₀ (t = t _{0 to} t _d ) Is estimated (see FIG. 6). An approximate straight line or an approximate curve representing the arrangement of the inflection points S ₁ , S ₂ ,... Used to estimate the traffic jam starting point Q ₁ is estimated as the traffic jam stretching wave L ₁ and used to estimate the traffic jam canceling point Q _2. An approximate straight line or an approximate curve representing the arrangement of the raised bending points R ₁ , R ₂ ,... Is estimated as the congestion elimination wave L ₂ (see FIG. 6).

The road traffic information is transmitted from the road traffic information processing server 1 to the client 2 and stored in the storage device 20. In the client 2, when the destination point is set through the input device 22, the road processing information is used by the arithmetic processing unit 24 to search for a recommended route from the current point or the departure point to the destination point. When predicting the travel time, the slope of the trajectory in each of the traffic flow region defined by the sudden event occurrence cross section, the traffic jam stretching wave L ₁ and the traffic jam canceling wave L ₂ and the free flow region outside the traffic flow region is considered (FIG. 6). reference).

  In response to this, for example, as shown in FIG. 7A, when a traffic jam occurs in the recommended route at the normal time, in addition to an alternative recommended route that avoids this, the location where the traffic jam occurs Is displayed on the display device constituting the output device 23. Also, a voice guidance such as “I will guide you on an alternative route because traffic congestion has spread on the usual route” is output from the voice output device constituting the output device 23.

  In addition, as shown in FIG. 7B, although traffic congestion has occurred in a location with a recommended route, it is expected that the traffic congestion is resolved when the mobile object X passes through the location. The guidance is displayed as an image. The voice output device constituting the output device 23 outputs a voice guidance such as “There is an accident traffic jam on the guide route, but it is expected to be resolved.

  When the destination point is set through the input device 22 in the client 2, the road traffic information generated by the road traffic information generation unit 14 is used by the road traffic information processing server 1, and the recommended route to the destination point is determined. It may be searched. In addition, information indicating the recommended route may be transmitted from the road traffic information server 1 to the client 2 together with the road traffic information on the recommended route, and may be output through the output device 23 (see FIGS. 7A and 7B).

(Function and effect)
According to the road traffic information processing system of the present invention and the road traffic information processing server 1 that constitutes the road traffic information processing system, the influence of the fluctuation of the specified bending point is reduced by setting the representative point of the specified bending point among the plurality of bending points. Therefore, the accuracy of estimating feature points such as a traffic jam starting point and a traffic jam canceling point can be further improved in the time-space plane.

Specifically, in view of the fact that fluctuations are likely to occur at the acceleration bending points R ₁ , R ₂ ,. As a result, the influence of the fluctuation is reduced (see FIG. 2 / STEP 104). Accordingly, the estimation accuracy of the traffic jam occurrence point Q ₁ as a feature point is further improved in the time-space plane (see FIG. 3).

Further, the acceleration bending point that is spatially farther than the first threshold α from the traffic jam starting point is set as the designated bending point (first acceleration bending point R ₁ ), and a representative point is set (FIG. 2). / STEP 202... YES →. For this reason, the situation where the traffic jam elimination start point is estimated in a situation where it is difficult to say that the traffic jam elimination wave has occurred is too close to the traffic jam origination point. The estimation accuracy of the starting point is improved (see FIG. 4).

Further, the center of gravity of the acceleration flexion point is represented by the requirement that the spatial and temporal intervals of the acceleration flexion points R ₁ , R ₂ ,. It is set as a point (see FIG. 2 / STEP210... YES →... STEP206). Acceleration bend points R ₁ , R ₂ ,... Are excessively dissociated, and it is avoided that a congestion elimination starting point is estimated in a situation where it is difficult to say that a congestion elimination wave showing a uniform change mode has occurred. Therefore, the estimation accuracy of the congestion elimination starting point Q ₂ as the feature point is improved (see FIG. 4).

DESCRIPTION OF SYMBOLS 1 ... Road traffic information processing server, 10 ... Database, 11 ... Bending point extraction part, 12 ... Representative point setting part, 13 ... Feature point estimation part, 14 ... Road traffic information generation part, 2 ... Client, 20 ... Storage device, 21: positioning device, 22: input device, 23: output device, 24: arithmetic processing device, X: moving body.

Claims (6)

On the time-space plane, a deceleration bend indicating that the speed has decreased due to the moving body entering a traffic jam from at least some of the plurality of loci representing the time series of the respective positions of the plurality of moving bodies. A bend point extraction unit for extracting a point and an acceleration bend point indicating that the speed has increased due to the mobile body exiting a traffic jam;
A representative point setting unit for setting a representative point of a specified bending point among a plurality of bending points extracted by the bending point extraction unit;
Based on the representative point set by the representative point setting unit, a boundary between a free flow area where the moving body is not in contact with traffic jam and a congestion flow area where the moving body is in contact with traffic jam is defined on the time-space plane. A road traffic information processing apparatus comprising: a feature point estimation unit that estimates a feature point to be performed. The road traffic information processing apparatus according to claim 1,
The representative point setting unit sets, as the representative point, the center of gravity of a plurality of accelerated bending points as the designated bending point,
The feature point estimator uses an intersection of a straight line passing through the representative point and parallel to the time axis, and a straight line or a curve determined by a deceleration bend point on the same locus as the specified bend point, as a congestion occurrence point as the feature point. A road traffic information processing apparatus characterized by estimating as follows. The road traffic information processing apparatus according to claim 1 or 2,
The representative point setting unit sets, as the representative point, the center of gravity of the accelerating bending point as the specified bending point that is spatially more than the first threshold away from the traffic jam starting point by a first threshold,
The feature point estimation unit estimates an intersection of a straight line passing through the traffic jam starting point parallel to the time axis and a negative slope straight line passing through the representative point as the traffic jam elimination starting point as the feature point. Road traffic information processing device. The road traffic information processing apparatus according to claim 3,
The representative point setting unit requires that a spatial or temporal interval between a plurality of accelerated bending points whose order is continuous in time series is equal to or less than a second threshold, and the plurality of the specified bending points are A road traffic information processing apparatus characterized in that the center of gravity of an acceleration bending point is set as the representative point. The road traffic information processing apparatus according to claim 3 or 4,
The road traffic information processing apparatus, wherein the feature point estimation unit estimates the negative inclination based on an arrangement form of an acceleration bending point as the designated bending point, and then estimates the congestion elimination starting point . On the time-space plane, a deceleration bend indicating that the speed has decreased due to the moving body entering a traffic jam from at least some of the plurality of loci representing the time series of the respective positions of the plurality of moving bodies. An inflection point extracting step for extracting a point and an acceleration inflection point representing that the speed has increased due to the mobile body getting out of a traffic jam;
A representative point setting step of setting a representative point of a specified bending point among a plurality of bending points extracted in the bending point extraction step;
Based on the representative point set in the representative point setting step, a boundary between a free flow area where the moving body is not in contact with traffic jam and a congestion flow area where the moving body is in contact with traffic jam is defined on the time-space plane. And a feature point estimation step of estimating a feature point to be performed.