JP6723967B2 - Estimating device, estimating method, and program - Google Patents

Description

The present invention relates to an estimation device, an estimation method and a program.

In order to cope with the congestion that can occur at a large-scale event or the like, grasping the flow of people and formulating control measures using a multi-agent simulator (MAS) have been conventionally performed. Highly accurate simulation is necessary for more accurate grasp of people flow and formulation of effective control measures. For this reason, it is indispensable to have a technique for determining the parameters for reproducing the observed data based on the observed data. As such a parameter, for example, a parameter indicating the number of people passing through the moving route can be cited.

Here, for example, AC traffic estimation technology in the communication field is known as a technology that solves a problem similar to the technology of estimating the number of people passing by each traveling route from the number of people observed at a plurality of observation points (for example, patents Reference 1, non-patent reference 1). The AC traffic estimation technique is a technique for estimating the amount of communication between each node pair in a communication network from the amount of outgoing traffic and the amount of incoming traffic of each node, the amount of traffic of each link, and route information.

When applying AC traffic estimation technology to the problem of estimating traffic movement, it may not be possible to observe the same person at multiple observation points in the same time zone, because it takes time to move traffic unlike communication. .. However, assuming that the traffic volume is stationary in the target time period, the problem of estimating the movement of traffic can be regarded as the same problem setting as the AC traffic estimation technique in the communication field.

Japanese Patent No. 5060513

Zhang Y. et al, "Fast Accurate Computation of Large-Scale IP Traffic Matrices from Link Loads", SIGMETRICS '03, pp.206-217, 2003.

By the way, the AC traffic estimation technique is premised on that the traffic flows only through a route designed by a designer, for example. On the other hand, in order to estimate the movement of traffic, it is also necessary to estimate a route candidate selected by a moving target (for example, a person, a car, a motorcycle, a bicycle, etc.). The estimation of cannot be modeled.

In addition, for highly accurate simulations, data that actually measures the traffic volume is required, but when measuring the traffic volume using, for example, a tally counter, there are times when the traffic volume could not be measured. It may exist. That is, there are cases where the observation value data obtained by observing the traffic volume is missing.

The present invention has been made in view of the above points, and an object of the present invention is to estimate the traffic volume by route from observation value data including a defect.

Therefore, in the embodiment of the present invention, based on the input road network data, a route generation unit that generates a plurality of routes from a departure point of a movement target to a destination, and the route generation unit generated by the route generation unit. Based on a plurality of routes and a plurality of observation points that respectively observe the traffic volume of the movement target, it is determined whether or not the movement target passing through each of the plurality of routes is observed at each of the plurality of observation points. A routing matrix generation means for generating a routing matrix to represent, the routing matrix generated by the routing matrix generation means, and a traffic amount of the moving object observed at each of the plurality of observation points in a plurality of observation periods, Route-based traffic volume estimation means for estimating the traffic volume of the movement target on each of the plurality of routes based on one observation value data.

It is possible to estimate the traffic volume by route from the observed value data including the deficiency.

It is a figure which shows an example of a structure of the number-of-people-by-routes estimation apparatus in embodiment of this invention. It is a figure which shows an example of observation value data. It is a figure which shows an example of the hardware constitutions of the number-of-people estimation device according to routes in an embodiment of the present invention. It is a figure showing an example of functional composition of the number-of-people estimation device according to an embodiment of the present invention. It is a flow chart which shows an example of the processing which the number estimating person according to route in an embodiment of the invention performs.

Embodiments of the present invention will be described below with reference to the drawings. Hereinafter, the route-by-route number estimation device 10 that estimates the number of persons by route as an example of the traffic volume by route will be described. However, the traffic volume by route is not limited to the number of people by route. The traffic volume by route may be, for example, the number of vehicles by route, the number of bikes by route, the number of bicycles by route, the number of living organisms by route, or the like. Therefore, the route-specific number estimation apparatus 10 according to the embodiment of the present invention can be similarly applied to the case of estimating the route-based traffic volume. Note that a moving object such as a person, a car, a motorcycle, a bicycle, or a living thing that moves along a route may be referred to as an “agent”.

<Configuration of the number-of-people estimation device 10 for each route>
First, the configuration of a route-specific number estimation device 10 according to an embodiment of the present invention will be described with reference to FIG. FIG. 1 is a diagram showing an example of the configuration of a route-specific number-of-people estimation device 10 according to an embodiment of the present invention.

The number-of-people-by-route estimation device 10 illustrated in FIG. 1 generates candidates for routes that an agent can pass through (hereinafter, also simply referred to as “route candidates”), and then calculates the number of people for each route (hereinafter, referred to as “route candidates”). It is also a computer or computer system that estimates "the number of people by route." A route-specific number estimation program 100 is installed in the route-based number estimation device 10 shown in FIG. The route-based number-of-people estimation program 100 may be one program or a program group composed of a plurality of programs or modules.

The route-specific number estimation apparatus 10 shown in FIG. 1 performs route candidate generation and route-specific number estimation by the processing executed by the route-specific number estimation program 100.

<Definition of variables>
Here, each variable used in the embodiment of the present invention is defined as follows.

_-Xt,i : Estimated number of people by route.

-T: An index indicating the time. t is 0≦t≦T.

-T: The last time of the time zone to be observed.

R _i : Route candidate (node sequence through which an agent can pass).

I: index of route candidate. i is 1≦i≦I.

I: the number of route candidates.

_-Yt,j : The observed value of the number of people at each observation point.

M _j : observation point (node sequence to be observed).

-J: Index of the observation point. j is 1≦j≦J.

・J: Number of observation points.

A: routing matrix.

Here, a matrix of T rows and J columns having the number of people Y _t,j at each observation point as each element is represented as observation value data Y. The observation value data Y is data obtained by measuring and totaling the traffic volume with a certain granularity in time and space. As a method of obtaining the observation value data Y, for example, as described in “Ministry of Land, Infrastructure, Transport and Tourism 2015 National Road/Street Traffic Situation Survey Overview of General Traffic Volume Survey Results,” There is a method of counting traffic by measuring the number of traffic.

An example of the observation value data Y is shown in FIG. As shown in FIG. 2, the observation value data Y is observed at the observation point M _j (j=1, 2,..., J) during the observation time period (observation period) t=0 to t=1. The observed values Y _1,j (j=1, 2,..., J) are the elements of the (1,j) component. Similarly, the observation value data Y is the observation value Y _2,j (observed at the observation point M _j (j=1, 2,..., J) during the observation time period t=1 to t= _2. Let j=1, 2,..., J) be elements of the (2, j) component. Similarly thereafter, the observation value data Y is the observation value Y observed at the observation point M _j (j=1, 2,..., J) during the observation time period t=T−1 to t=T. _{Let T,j} (j=1, 2,..., J) be the element of the (T,j) component.

There may be a missing value in the observed value Y _t,j included in the observed value data Y. In other words, a missing measurement may exist in a certain observation time zone of a certain observation point M _j . NULL may be set for the observed value Y _t,j that is a missing value, or a predetermined value that is determined in advance (for example, a predetermined code value indicating a missing value or an observed value cannot be taken). Value, etc.) may be set.

The time width of each observation time may be different. For example, the time width between the observation times t=0 to t=1 may be different from the time width between the observation times t=1 to t=2. Further, the observation time may be different for each observation point M _j . For example, during the period from t=0 to t=T, the observation point M ₁ is observed T times, while the observation point M ₂ is observed T/2 times. In other words, the observation cycle may be different for each observation point M _j . In this case, collectively for each observation point M _j where the duration of the observation are aligned (i.e., the observation period is summarized by the same observation spot M _j), also expresses the observed value data Y by a plurality of matrices good.

<Hardware Configuration of Route-Estimated Number of People Estimating Device 10>
Next, a hardware configuration of the route-based people estimation device 10 according to the embodiment of the present invention will be described with reference to FIG. FIG. 3 is a diagram showing an example of a hardware configuration of the route-based number-of-people estimation device 10 according to the embodiment of the present invention.

The route-based people estimation device 10 shown in FIG. 3 includes an input device 11, a display device 12, an external I/F 13, a RAM (Random Access Memory) 14, a ROM (Read Only Memory) 15, and a CPU (Central Processing). Unit) 16, a communication I/F 17, and an auxiliary storage device 18. These pieces of hardware are connected to each other via a bus B so that they can communicate with each other.

The input device 11 is, for example, a keyboard, a mouse, a touch panel, etc., and is used by the user to input various operations. The display device 12 is, for example, a display or the like, and displays the processing result of the route-specific number-of-people estimation device 10. In addition, the route-specific number-of-people estimation device 10 may not include at least one of the input device 11 and the display device 12.

The external I/F 13 is an interface with an external device. The external device includes a recording medium 13a and the like. The number-of-people-by-path estimation apparatus 10 can read or write the recording medium 13a or the like via the external I/F 13. The route-based number estimation program 100 and the like may be recorded in the recording medium 13a.

The recording medium 13a includes, for example, a flexible disk, a CD (Compact Disc), a DVD (Digital Versatile Disk), an SD memory card (Secure Digital memory card), and a USB (Universal Serial Bus) memory card.

The RAM 14 is a volatile semiconductor memory that temporarily holds programs and data. The ROM 15 is a non-volatile semiconductor memory that can retain programs and data even when the power is turned off. The ROM 15 stores, for example, OS settings and network settings.

The CPU 16 is an arithmetic device that reads programs and data from the ROM 15 and the auxiliary storage device 18 onto the RAM 14 and executes processing.

The communication I/F 17 is an interface for connecting the number-of-people-by-routes estimation device 10 to a communication network. The route-specific number-of-people estimation program 100 may be acquired (downloaded) from a predetermined server device or the like via the communication I/F 17.

The auxiliary storage device 18 is, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), or the like, and is a non-volatile storage device that stores programs and data. The programs and data stored in the auxiliary storage device 18 include, for example, an OS, an application program that implements various functions on the OS, a route-specific number-of-people estimation program 100, and the like.

The number-of-people-by-routes estimation device 10 according to the embodiment of the present invention has the hardware configuration shown in FIG.

<Functional Configuration of Route-Specific Number Estimating Device 10>
Next, a functional configuration of the number-of-people-by-routes estimation device 10 according to the embodiment of the present invention will be described with reference to FIG. FIG. 4 is a diagram showing an example of a functional configuration of the route-specific number-of-people estimation device 10 according to the embodiment of the present invention.

4 includes a route candidate generation unit 110, a routing matrix generation unit 120, and a route-specific number of people estimation unit 130. Each of these units is realized by the processing executed by the CPU 16 for estimating the number of people by route.

The route candidate generation unit 110 inputs the road network data G, the node sets V and U, the magnification α, and the observation point list M to generate the route candidate list R. The route candidate list R is a list of route candidates R _i . However, the route candidate R _i is a node sequence in which a link exists in the road network data G.

Note that the route candidate generation unit 110 may not input at least one of the node sets V and U, the magnification α, and the observation point list M. That is, the node sets V and U, the magnification α, and the observation point list M are arbitrary input data.

The road network data G is a directed graph representing a target road network. The road network data G is expressed as G={N, E}, where N is a set of nodes (eg, intersections) belonging to the road network and E is a set of links (eg roads) belonging to the road network. ..

The node sets V and U are a set of nodes for limiting the combination of the starting point (O) and the destination (D) (hereinafter, referred to as “OD combination”). The OD combination is a combination of a node serving as a starting point and a node serving as a destination.

Examples of the nodes included in the node sets V and U include a node indicating a landmark (for example, a node in which a person may occur or disappear). Specifically, the node set V may be, for example, a set of nodes indicating stations. Further, as the node set U, for example, a set of nodes indicating the entrance of the event venue can be cited.

The magnification α is a value for limiting the allowable range of the route candidate R _{i on} the way. The value of the magnification α is preset by, for example, the user of the route-specific people estimation device 10 or the like.

Observation place list M is a list of observation points M _j for excluding In any observation point M _j is not observed path candidate R _i.

The routing matrix generation unit 120 inputs the route candidate list R and the observation point list M and generates a routing matrix A.

The route-specific number of people estimation unit 130 inputs the routing matrix A and the observation value data Y, and estimates the route-based number of people X. Then, the route-specific number of people estimation unit 130 outputs the estimated route-based number of people X. The number X of people by route is a matrix representing the traffic volume by route. Note that the route-specific number of people estimation unit 130 may output the route-based number of people X to any preset output destination. For example, the number-of-people-by-route estimation unit 130 may output the number of people X by route to the display device 12, output (save) to the auxiliary storage device 18, the recording medium 13a, or the like, or the communication I/F 17 It may be output (transmitted) to a server device or the like on the network via. In addition, the route-specific number of people estimation unit 130 may output the route-based number of people X to another program (for example, a pedestrian flow simulation program).

<Process Performed by Route-Specific Number Estimating Device 10>
Next, the processing executed by the route-specific number-of-people estimation device 10 according to the embodiment of the present invention will be described with reference to FIG. FIG. 5 is a flowchart showing an example of a process executed by the number-of-people-by-route estimation apparatus according to the embodiment of the present invention.

Step S101: The route candidate generation unit 110 generates the route candidate list R by executing the following procedures (1) to (4) for all OD combinations. When the node sets V and U are not input, the route candidate generation unit 110 creates an OD combination from the nodes included in the road network data G. On the other hand, when the node sets V and U are input, the route candidate generation unit 110 creates an OD combination from the nodes included in the node set V and the nodes included in the node set U. Creating an OD combination from the nodes included in the node set V and the nodes included in the node set U is equivalent to selecting each edge of the complete bipartite graph of the node set V and the node set U.

The nodes included in the road network data G, the nodes included in the node set V, and the nodes included in the node set U can both be a departure point (O) or a destination (D). And That is, for the node N ₁ and the node N ₂ included in the road network data G, the OD combinations are a combination in which the node N ₁ is the starting point and the node N ₂ is the destination, and the node N ₂ is the starting point and the node N _2. There are combinations with ₁ as the destination. Similarly, for the node sets V and U, for the node N ₁ included in the node set V and the node N ₂ included in the node set U, the OD combination is the node N ₁ as the starting point and the node N ₂ as the destination. And a combination in which the node N ₂ is the starting point and the node N ₁ is the destination.

(1) The route candidate generation unit 110 includes a node indicating a departure place (O) included in an OD combination (hereinafter, referred to as “node O”) and a node indicating a destination included in the OD combination (hereinafter, referred to as “node O”). The total number of routes to and from "node D" is counted. Each enumerated route is a route candidate R _i .

The enumeration of all the paths between the node O and the node D can be performed by using, for example, Graphillion. Regarding Grafilion, for example, ERATO Minato Discrete Structure Processing System Project (Author), Shinichi Minato (edited) “Ultra-high-speed graph enumeration algorithm-a new approach to combinatorial problems developed by <numerical value of Fukasugi>” in Morikita Publishing 2015. It is disclosed.

(2) The route candidate generation unit 110 searches the shortest route between the node O and the node D by the shortest route search algorithm and calculates the distance of the searched shortest route. For this, NetworkX or the like, which is a Python library, can be used.

(3) Next, the route candidate generator 110 calculates the distance (route candidate distance d _i ) of each route candidate R _i obtained in (1) above and the shortest route obtained in (2) above. Compare with the distance (shortest distance d _min ). Then, the route candidate generation unit 110 excludes the route candidates R _i whose route candidate distances d _i are α times or more of the shortest distance d _min among the respective route candidates R _i .

(4) Next, the route candidate generation unit 110 excludes the route candidate R _i that is not observed from the route candidates R _i obtained in (3) above. Whether or not the route candidate R _i is observed can be determined using the observation point list M.

More specifically, for example, the observation point M _j included in the observation point list M is M _j =[M _j,1 , M _j,2 ,..., M _j,n ], and the route candidate R _i is R. Let _i =[R _i,1 , R _i,2 ,..., R _i,k ]. However, n is the number of nodes included in the observation point M _j , and k is the number of nodes included in the route candidate R _i . At this time, any node of R _i,1 , R _i,2 ,..., R _i,k is one of M _j,1 , M _j,2 ,..., M _j,n . If it is the same node as that node, the route candidate R _i is determined to be observed. On the other hand, any of the nodes R _i,1 , R _i,2 ,..., R _i,k has a node of M _j,1 , M _j,2 ,..., M _j,n . If neither node is the same node, it is determined that the route candidate R _i is not observed.

In particular, when the leading node R _i,1 of the route candidate R _i is the same node as any one of M _j,1 , M _j,2 ,..., M _j,n , the route candidate The departure of R _i is observed. Similarly, if the tail node R _i,k of the route candidate R _i is the same node as any one of M _j,1 , M _j,2 ,..., M _j,n , The arrival of the candidate R _i is observed. Further, when the observation point M _j is a subsequence of the path candidates R _i, passing the path candidate R _i is observed.

The list of route candidates R _i obtained in (4) above for all OD combinations is the route candidate list R. Then, the route candidate generation unit 110 outputs the obtained route candidate list R to the routing matrix generation unit 120.

Note that, as described above, at least one of the node sets V and U, the magnification α, and the observation point list M may not be input to the route candidate generation unit 110. However, depending on the size of the road network data G, the number of route candidates R _i included in the route candidate list R may become enormous. Therefore, it is preferable to limit the route candidates R _i by using at least one of the node sets V and U, the magnification α, and the observation point list M.

Step S102: The routing matrix generation unit 120 generates the routing matrix A from the route candidate list R and the observation point list M. If each element of the routing matrix A is A _j,i , the routing matrix A is generated by the following expression 1.

つまり、経路候補Ｒ_ｉが観測地点Ｍ_ｊで観測されれば、ルーチング行列の（ｊ，ｉ）要素Ａ_ｊ，ｉが１となり、そうでない場合はＡ_ｊ，ｉが０となるようにルーチング行列Ａを生成する。

That is, if it is observed by the route candidate _{R i} is the observation point _{M j, (j,} i) of the routing matrix elements _{A j, i} becomes 1, otherwise _{A j, i} becomes zero as routing matrix Generate A.

Whether or not the route candidate R _i is observed at the observation point M _j is determined as follows, for example. Let the route candidate R _i be R _i =[R _i,1 , R _i,2 ,..., R _i,k ]. However, k is the number of nodes included in the route candidate R _i . The observation points M _j included in the observation point list M are M _j =[M _j,1 , M _j,2 ,..., M _j,n ]. However, n is the number of nodes included in the observation point M _j . The observation point M _j has any of the attributes of “departure”, “arrival”, and “passage”. When the attribute is “departure”, it is assumed that the head element (head node) R _{i,1 of} the route candidate R _i is included in the observation point M _j, and is observed, otherwise it is not observed. When the attribute is “arrival”, it is assumed that the tail element R _{i,k of the} route candidate R _i is included in the observation point M _j, and is observed, and otherwise it is not observed. When the attribute is “pass”, it is assumed that at least a part of the route candidate R _i is included in the observation point M _j, and is observed, and otherwise it is not observed. As described above, the observation point list M includes the observation points M _j having at least one attribute of “departure”, “arrival”, and “passage”, and the route candidates R _i are set as the observation points according to each attribute. It is determined whether or not it is observed in M _j , and the value of the (j,i) element A _j,i of the routing matrix is set.

Each element of the routing matrix A generated in this way determines whether or not a moving object (for example, a person, a car, a motorcycle, a bicycle, etc.) passing through each of a plurality of routes is observed at each of a plurality of observation points. Will be represented. That is, A _j,i represents whether or not the moving target passing through the route candidate R _i can be observed at the observation point M _j . As a result, the routing matrix A specifies which element in the observation value data Y should be taken into consideration when estimating the number of moving objects passing through the route candidate R _i .

Step S103: The number-of-people-by-route estimation unit 130 estimates the number X of people by route from the routing matrix A and the observation value data Y by executing the following procedures (1) to (4).

(1) The route-specific number-of-people estimation unit 130 shapes the observation value data Y to generate the observation value data S. The observation value data S after shaping is a matrix of D rows and 3 columns, and the d-th row is [M _d , U _d , Y _d ′]. Here, D is the number of observation values Y _t,j obtained by removing the missing values from the observation values Y _t,j included in the observation value data Y. In other words, D is the number of elements of the observed value Y _t,j that is not a missing value among the observed values Y _t,j included in the observed value data Y. The observation value data is obtained by re-indexing the observation values Y _t,j that are not missing values in order from 1 and converting the value of each element into [M _d , U _d , Y _d ′]. It is S.

M _d is an index of the observation point of the d-th (d=1, 2,..., D) observation value Y _d ′, and 1≦M _d ≦J. U _d is information indicating the observation value observed time period _{Y d} '(the observation period), the divided time zone for each predetermined time width _{ΔT t k (k = 1,2,} ···) as , D is a list of time sections (unit time) t _i corresponding to the time zone in which the observed value Y _d ′ is observed. Hereafter, ΔT will be referred to as an observation time width.

For example, when a certain observation value Y _d ′ is obtained by observing the time zone represented by t ₂ to t ₄ , the value U _d ={t ₂ , t ₃ , t ₄ }. Similarly, when another certain observations _{Y d} 'is obtained by observing the time zone represented by _t 3 ~t _4, a value _{U d = {t 3, t} 4}.

The observation time width ΔT is preferably set to be sufficiently small. By setting the observation time width ΔT sufficiently small, it is possible to use a plurality of observation value data Y with different time widths in the observation time zone, or when one observation value data Y has different observation time zones. Also, it is possible to generate the observation value data S considering all the observation times. Specifically, for example, when there is an observation time zone having a time width of “10 minutes” and an observation time zone having a time width of “15 minutes”, ΔT is set to “5 minutes” (that is, all observation time periods). Observed value data S is generated as a common divisor of the time width of the band. That is, a unit time t ₁ , t ₂ ,..., _{T T} is obtained by dividing the entire observation period of the observation value data Y into a predetermined observation time width, and each observation period is a combination of one or more unit times. U _d is re-expressed (normalized). Then, the observation value data S is data obtained by removing the missing value from the observation value data Y and reshaping the observation value data Y so that the observation values for each normalized observation period (U _d ) and each observation point can be specified. As a result, even when the observation value data Y having a different traffic period or observation period or the observation data Y having a difference in the traffic amount observation period or the observation period is given, the difference in the observation period It is possible to allow a shift in the observation period.

(2) Next, the number-of-people-by-path estimating unit 130 generates the observation matrix H by the following Expression 2 with each element of the observation matrix H set as H _d,t .

ここで、Ａ_Ｍｄはルーチング行列ＡのＭ_ｄ行目の要素を並べた１×Ｉの行列である。また、観測期間Ｕ_ｄにｔが含まれない場合のＨ_ｄ，ｔ＝０はスカラではなく、各要素が０の１×Ｉの行列を表す。結果として、観測行列Ｈは、Ｄ×（Ｉ×Ｔ）の行列となる。

Here, A _Md is a 1×I matrix in which elements in the M _d-th row of the routing matrix A are arranged. Further, H _d,t =0 when t is not included in the observation period U _d is not a scalar and represents a 1×I matrix in which each element is 0. As a result, the observation matrix H becomes a D×(I×T) matrix.

(3) Next, the number-of-people-by-route estimation unit 130 obtains X′ that minimizes the objective function L by the following Expression 3.

ただし、人数は非負であるため、制約条件Ｘ_ｔ，ｉ´≧０を満たすように目的関数Ｌを最小化する。言い換えると、Ｙ´＝ＨＸ´を満たす（Ｉ×Ｔ）次元のベクトルＸ´を求める。

However, since the number of persons is non-negative, the objective function L is minimized so as to satisfy the constraint condition X _t,i ′≧0. In other words, the (I×T)-dimensional vector X′ that satisfies Y′=HX′ is obtained.

As a method of obtaining X′ that minimizes the above Expression 3, for example, a trust region reflective algorithm or the like may be used. The trust region reflective method algorithm is, for example, Coleman, TF and Y. Li. "A Reflective Newton Method for Minimizing a Quadratic Function Subject to Bounds on Some of the Variables," SIAM Journal on Optimization, Vol. 6, Number 4, pp. .1040-1058, 1996., etc.

(4) Next, the route-by-route number-of-people estimating unit 130 arranges the (I×T)-dimensional vector X′ obtained in (3) above, in order from the head element, in each row by I rows. Convert to column matrix X. This matrix X is the number X of people by route. The (t, i) element of X is the estimation result of the number of people who pass the route R _j at the time t.

Step S104: The route-specific number of people estimation unit 130 outputs the route-based number of people X obtained in the above step S104.

As described above, the route-based number estimation device 10 according to the embodiment of the present invention can estimate and output the route-based number X from the observed value data Y. Moreover, the route-by-route number-of-people estimation device 10 according to the embodiment of the present invention may have a missing value in the observation values included in the observation value data Y or may have different observation period lengths. Also, the number X of people by route can be estimated.

By using the route-based number X obtained by the route-based number estimating device 10 according to the embodiment of the present invention as an input to, for example, a pedestrian flow simulator, it is possible to reconstruct the temporal transition of the pedestrian flow. Further, it is considered that the number X of people by route is useful for grasping the customer flow line in marketing and the examination of guidance measures in crowd traffic security.

The present invention is not limited to the above specifically disclosed embodiments, and various modifications and changes can be made without departing from the scope of the claims.

10 Route-Specific Number Estimation Device 100 Route-Specific Number Estimation Program 110 Route Candidate Generation Unit 120 Routing Matrix Generation Unit 130 Route-Specific Number Estimation Unit

Claims (8)

Route generation means for generating a plurality of routes from the starting point of the movement target to the destination based on the input road network data,
Based on the plurality of routes generated by the route generation means and a plurality of observation points for observing the traffic volume of the movement target, the movement target passing through each of the plurality of routes is the plurality of observation points. Routing matrix generation means for generating a routing matrix representing whether or not each is observed,
Based on the routing matrix generated by the routing matrix generation means and the first observation value data indicating the traffic volume of the movement target observed at the plurality of observation points in a plurality of observation periods, respectively. A traffic amount estimating means for estimating a traffic volume of the moving object in each of the routes,
An estimation device comprising: The route-based traffic volume estimation means,
The second observation value data excluding the missing values included in the first observation value data is generated, and the second observation value data is generated in each of the plurality of routes based on the routing matrix and the second observation value data. The estimation apparatus according to claim 1, wherein the estimation unit estimates the traffic volume of the moving object. The route-based traffic volume estimation means,
Further, the common divisor of the plurality of observation periods is the observation time width,
As a unit time, the whole of the plurality of observation periods divided by the observation time width,
Each of the plurality of observation periods is composed of one or more unit times,
The traffic volume of each route per unit time is an unknown number,
It is assumed that each of the second observation value data is represented by the sum of the traffic volume of each route observable at the observation point of the second observation value data for each unit time constituting the observation period. The estimation device according to claim 2, wherein the unknown amount of traffic of each route per unit time is estimated. The route generation means,
Of the nodes included in the road network data, a departure point and a destination are selected from a first set including a predetermined first node and a second set including a predetermined second node. The estimation device according to any one of claims 1 to 3, wherein a plurality of routes from the selected starting point to the destination are generated. The route generation means,
The first distance obtained by multiplying the distance of the shortest route from the departure place to the destination by a predetermined value is compared with the second distance of each of the plurality of routes, and the first distance or more is determined. The estimation device according to claim 1, wherein a route corresponding to the second distance is excluded from the plurality of routes. The route generation means,
The estimation device according to any one of claims 1 to 5, wherein a route that is not observed at any of the plurality of observation points is excluded from the plurality of routes. Based on the input road network data, a route generation procedure for generating a plurality of routes from the starting point of the movement target to the destination,
Based on the plurality of routes generated by the route generation procedure and a plurality of observation points for observing the traffic volume of the movement target, the movement target passing through each of the plurality of routes is the plurality of observation points. A routing matrix generation procedure for generating a routing matrix representing whether or not each is observed,
Based on the routing matrix generated by the routing matrix generation procedure and the first observation value data indicating the traffic volume of the movement target observed at each of the plurality of observation points in a plurality of observation periods, A route-by-route traffic volume estimation procedure for estimating the traffic volume of the moving target on each of the routes,
An estimation method characterized by being executed by a computer. The computer program for functioning as a putative equipment according to any one of claims 1 to 6.

Images