JP6420749B2 - Geographic network data management apparatus and computer program - Google Patents

Description

  The present invention relates to a geographic network data management apparatus and computer program that can automatically assign attributes to nodes and links of a geographic network.

  A network structure (hereinafter referred to as NW) in which a geographical point in the geography is a node and a side or arc connecting the points is a link is referred to as a geography network.

  For example, in a simple geographic NW that handles roads, an intersection is a node and a road connecting the intersections is a link. By using this structure, route calculation becomes possible. For example, OpenStreetMap is a map service created with an NW structure, and route calculation using this structure is possible (Non-Patent Document 1).

  In addition, a system for storing attributes on the geographic network and displaying them to the user has been proposed (Non-Patent Document 2). Here, a method for displaying a barrier-free map constructed as a geographic NW on a user terminal is described.

Sittichai Choosumrong, Venkatesh Raghavan and Eugenio Realini.Implementation of dynamic cost based routing for navigation under real road conditions using FOSS4G and OpenStreetMap.International Symposium on Geoinformatics for Spatial Infrastructure Development in Earth and Allied Sciences 2010. Karimi, H., Zhang, L., Benner, J. G. Personalized Accessibility Maps (PAMs) for communities with special needs.Lecture Notes in Computer Science, 7820, pp.199? 213, 2013.

As described above, there is a system that displays attribute information stored on the geographic network, but there is no system that allocates attributes to the geographic network based on data obtained by observing the real world.
For example, when an acceleration sensor measured that there was a step on the road, the information could not be automatically given as an attribute to the corresponding link on the geographic NW.
Depending on the attribute, the configuration of the geographic network may be changed. For example, when conflicting attributes are given to one link, it is effective to create additional links that share the same start and end nodes and assign different attributes to each link.

  The present invention has been made in view of the above circumstances, and a geographic network data management apparatus capable of automatically assigning attributes to nodes and links of a geographic network and a computer for causing a computer to function as the geographic network data management apparatus The purpose is to provide a program.

The geographic network data management device according to the present invention provides, for each node included in a target area, a node table including a node ID indicating the target node, a record including the latitude and longitude of the node, and each link connecting the nodes. Per link ID that indicates the target link, a link table that includes a record including the node ID of each node, and an attribute ID that indicates the target attribute for each attribute that is set for a node or a link or part of a link, Included in the feature table having a record including an attribute, and for each node or link in which the attribute is set, the node ID of the node or the link ID of the link, the feature table record corresponding to the target attribute The attribute ID is the same as the attribute ID that is If you, the attribute start showing one end of the part, and, the the correspondence table comprising a record including the attribute endpoints of a portion of the other end, the node ID or link node link ID or node or link Read observation data including position information and attributes from the observation data storage unit, update the feature table and the correspondence table based on the observation data, and if the observation data includes an attribute start point and an attribute end point, the correspondence If there is a plurality of records in which the attribute data generation unit that sets the attribute start point and the attribute end point in the corresponding record of the table and the attribute table having the attribute IDs contradictory to each other exist in the correspondence table, the link ID A record of a new link ID that shares the start node and end node of the link of the link table Add to, said one link ID attribute is granted to contradict each other in the correspondence table, characterized in that it comprises a geographical network change unit that updates at the new link ID added.

  The computer program of the present invention is for causing a computer to function as the geographic network data management apparatus of the present invention.

  According to the present invention, an attribute can be automatically assigned to a node or link of a geographic network.

It is a block diagram of a geographic network data management apparatus. It is a figure which shows the structure of the geographic network data storage part. 3 is a diagram illustrating a configuration of an attribute data storage unit 5. FIG. It is a figure which shows an example of an expression of an attribute start point and an attribute end point. It is a figure which shows the example of the format of observation data. It is a flowchart of the observation data acquisition part 1. FIG. It is a figure which shows an example of the rule used by the contradiction determination between the attribute values of the same attribute classification. It is a figure which shows an example of the rule used by the contradiction determination between the attribute values of a different attribute type. It is a flowchart of the attribute data generation part 4. It is a flowchart of the geographic network structure change part 6. FIG. It is a figure which shows the example of a structure change of a geographic network. It is a figure which shows an example of the node change format 2. It is a figure which shows an example of the link change format.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing a configuration of a geographic network data management apparatus according to the present embodiment.
The geographic network data management device is a device that automatically assigns attributes to nodes and links of a geographic network, and includes an observation data acquisition unit 1, a geographic network data storage unit 2, an observation data storage unit 3, and attributes. A data generation unit 4, an attribute data storage unit 5, and a geographic network configuration change unit 6 are provided.

FIG. 2 is a diagram illustrating a configuration of the geographic network data storage unit 2.
The geographic network data storage unit 2 is a geographic network composed of nodes and links, and includes a node table 21 and a link table 22.

For example, the node table 21 includes a record for each point (node) included in the target area. The record includes a node ID indicating the target node, the latitude and longitude of the node, the height of the node, and the name of the node. In some cases, the height of the node and the name of the node may be omitted.
The node ID is expressed by a numeric string such as 251055431, for example. The same applies to link IDs and attribute IDs described later.

  For example, latitude is expressed as 35.6760470 and longitude is expressed as 139.7457789. For example, north latitude is positive and south latitude is negative. For example, east longitude is positive and west longitude is negative. The same applies to latitude and longitude.

For example, the node height “0” indicates the ground floor, “2” indicates the second floor, and a small number indicates the intermediate floor.
For example, the name of the node is expressed as “National Diet Building”.

For example, the link table 22 includes a record for each link (such as a road) that connects nodes. The record includes a link ID indicating a target link, a start point ID that is a node ID of one node connected to the link (referred to as a link start point), and a node ID of the other node connected to the link (referred to as a link end point). Contains the end point ID.
The link start point and link end point can be defined, for example, such that the higher latitude is the link start point and the lower latitude is the link end point.

FIG. 3 is a diagram illustrating a configuration of the attribute data storage unit 5.
The attribute data storage unit 5 indicates attributes assigned to the nodes and links of the geographic network, and includes a feature table 51 and a correspondence table 52.

  The feature table 51 includes a record for each attribute set for a node, a link, or a part of a link. The record includes an attribute ID indicating the target attribute, an attribute type, and an attribute value (attribute value). Hereinafter, a combination of an attribute type and an attribute value is referred to as an “attribute”. The feature table 51 includes only one record for the same “attribute”.

  The correspondence table 52 includes a record for each node or link for which an attribute is set. The record includes a node ID of the node or a link destination ID that is a link ID of the link, an attribute ID that is the same as the attribute ID included in the record of the feature table 51 corresponding to the target attribute, and the target attribute is a part of the link Is set to include an attribute start point indicating one end of the part and an attribute end point indicating the other end of the part.

FIG. 4 is a diagram illustrating an example of the expression of the attribute start point and the attribute end point.
The attribute start point and attribute end point are expressed by, for example, the distance from the link start point or link end point.

For example, the attribute start point “10” and the attribute end point “15” mean that the attribute starting from the 10 m point from the link start point continues to the 15 m point.
For example, the attribute start point “15” and the attribute end point “10” mean that the attribute starting from the 15 m point from the link start point continues to the 10 m point. That is, the direction of the attribute is opposite to that of the attribute start point “10” and the attribute end point “15”.

For example, the attribute start point “0” and the attribute end point “10” mean that the attribute starting from the link start point continues to the 10 m point.
Alternatively, the attribute start point and attribute end point are expressed, for example, as a ratio obtained by dividing the distance from the link start point or link end point by the link length.

For example, the attribute start point “10%” and the attribute end point “50%” mean that the attribute starting from the 10% point from the link start point continues to the 50m point.
If the attribute start point or attribute end point is a link start point, the attribute start point or attribute end point may be expressed specially as 0, and if it is a link end point, it may be expressed specially as -1.

For example, the attribute start point “0” and the attribute end point “−1” mean that the attribute starting from the link start point continues to the link end point.
For example, the attribute start point “−1” and the attribute end point “0” mean that an attribute starting from the link end point continues to the link start point. That is, the attribute direction is opposite to the attribute start point “0” and attribute end point “−1”.

Either the attribute start point or the attribute end point may be close to the link start point. When the attribute type is a vertical slope (on a slope), the lower altitude can be defined as the attribute start point and the higher altitude as the attribute end point or vice versa.
If the attribute start point and attribute end point match, the attribute start point and attribute end point of the record may be the same. Since it is not necessary to use the same two, the attribute end point may be expressed by nil, for example.
For example, the attribute start point “10” and the attribute end point “nil” mean attributes at a point 10 m from the link start point.

FIG. 5 is a diagram illustrating an example of the format of observation data.
<Observation data format 1>
The observation data indicating the attribute relating to the node includes, for example, a node ID, an attribute type, and an attribute value (FIG. 5A).
<Observation data format 2>
The observation data indicating the attribute relating to the node includes, for example, the latitude and longitude of the node, the attribute type, and the attribute value (FIG. 5B).

<Observation data format 3>
The observation data indicating the attribute regarding the whole of one link includes, for example, a link ID, an attribute type, and an attribute value (FIG. 5C).
<Observation data format 4>
The observation data indicating the partial attribute of one link includes, for example, the link ID of the link, the attribute start point indicating the one end of the link, the attribute end point indicating the other end of the link, the attribute type, and the attribute value. (FIG. 5D).

<Observation data format 5>
The observation data indicating the attribute of a part of the link spanning two or more linked links includes, for example, the attribute start point indicating one end of the part, the link ID of the link including the attribute start point (start link), the one If there is a link (intermediate link) between the start point link and the end point link, including the attribute end point indicating the other end of the part, the link ID of the link including the attribute end point (end point link), the attribute type, and the attribute value, The link ID of the intermediate link is included (FIG. 5E).

<Observation data format 6>
The observation data indicating the attribute of a part of one link or a part of a link that spans two or more links that are connected linearly is the latitude and longitude of one end (attribute start point) of the part, the part The latitude and longitude of the other end (attribute end point), attribute type, and attribute value are included (FIG. 5F).

FIG. 6 is a flowchart of the observation data acquisition unit 1.
The observation data acquisition unit 1 reads the observation data from the outside (S1), and determines whether the observation data includes latitude and longitude (S3). When the latitude and longitude are included, it is determined whether or not the position of the latitude and longitude is within a target range (for example, a target city area) (S5).

  If the observation data acquisition unit 1 does not include latitude and longitude, that is, if it includes a node ID or link ID, whether the ID is included in the node table 21 or the link table 22 of the geographic network data storage unit 2 Is determined (S7).

When the observation data acquisition unit 1 is included (S7: YES) or within the target range (S5: YES), the observation data storage unit 3 stores the observation data (S11).
If the observation data is stored in the observation data storage unit 3 or if it is out of range, the presence / absence of the next observation data is determined (S13). If there is, the process returns to step S1, and if not, the process ends. Let

The observation data acquisition unit 1 may determine the contradiction of the observation data and perform processing according to the determination result. For example, one observation data determined to have a contradiction is deleted without being stored in the observation data storage unit 3.
The observation data acquisition unit 1 determines whether the observation data is inconsistent on the same node or link. Note that the reliability of the observation data does not matter, assuming that all observation data is correct.

For example, if the attribute type is the same for the same node or link, but different attribute values are given, this is a contradiction. In addition, if the conflicting attribute types are given to the same node or link, it is contradictory.
In the former case, contradiction may or may not be allowed depending on the attribute value, and it can be determined by a rule.

FIG. 7 is a diagram illustrating an example of rules used in determining inconsistency between attribute values of the same attribute type.
For example, a rule as shown in FIG. 7 is created.

  Assume that observation data D1 is given to a certain node or link, and then observation data D2 is given to the same node or link. Here, it is assumed that both D1 and D2 are observation data indicating “slope”. At this time, it is assumed that the values of the vertical gradients of D1 and D2 are different, and the upper and lower states with respect to the reference value designated by the change are not changed. For example, when the longitudinal gradient of D1 is 3% and the longitudinal gradient of D2 is 5%, it does not exceed 8% of the reference value, so that no change with respect to the reference value occurs. In this case, it is determined that a contradiction is allowed.

Similarly, data D3 and D4 for a certain node or link are both data indicating “slope”, but it is assumed that the ascending direction is reversed between D3 and D4. In this case, since the difference in the rising direction is an unacceptable attribute value, it is determined that they are mutually contradictory.
In the latter case, there is a case where contradiction is allowed and a case where it is not allowed, and it can be determined by a rule.

FIG. 8 is a diagram illustrating an example of a rule used in inconsistency determination between attribute values of different attribute types.
Here, attribute types that contradict each other are grouped. That is, the attribute types belonging to different groups do not contradict each other.

  For example, even if the attribute type “roof” belonging to the group G1 and the attribute type “slope” belonging to the group G2 are assigned to the same node or link, there is no contradiction, but both the attribute type “slope” belonging to the group G2 It is determined that there is a contradiction when any two of "", "stairs", and "escalator" are assigned to the same node or link.

FIG. 9 is a flowchart of the attribute data generation unit 4.
The attribute data generation unit 4 reads the observation data from the observation data storage unit 3 (S21), and determines whether the observation data includes latitude and longitude (S23). When the latitude and longitude are included, the corresponding node or link is selected based on the latitude and longitude and the geographic network data storage unit 2 (S25).

The attribute data generation unit 4 updates the attribute data storage unit 5 based on the observation data if the node or link is selected (S25), or if the observation data does not include latitude and longitude (S23: NO). (S27).
Subsequently, the attribute data generation unit 4 determines whether or not there is next observation data (S29). If there is, the process returns to step S21, and if not, the process is terminated.

<Detailed processing of S25>
When observation data does not include a node ID or link ID but includes latitude and longitude, the node or link cannot be selected immediately. Here, a process of selecting a corresponding node or link based on the latitude and longitude and the geographic network data storage unit 2 will be described.

つまり、ノードの集合Ｖから、ｍに最も近いノードを選択する。 <When selecting a node>
The point shown in the observation data is m, its latitude is m ₀ , and its longitude is m ₁ (m = (m ₀ , m ₁ )).
At this time, the record (node) of the node closest to the point m is selected from the records of the node table 21. Specifically, a node v that satisfies the following conditions is selected.

That is, the node closest to m is selected from the node set V.

<When selecting a link>
The point shown in the observation data is m, its latitude is m ₀ , and its longitude is m ₁ (m = (m ₀ , m ₁ )).
At this time, for the link e = φ (v, w) where the start point is v = (v ₀ , v ₁ ) and the end point is w = (w ₀ , w ₁ ), the link e select.

<Consideration of attribute existence probability>
When a link is selected in S25, the existence probability can be considered.
A point indicated by observation data indicating the type and attribute value of an attribute (hereinafter referred to as attribute b) is μ, its latitude is μ ₀ , and its longitude is μ ₁ .

として、２変数の正規分布に従い、以下の通りとなる。

ここで、行列Ａについて、｜Ａ｜はＡの行列式、Ａ^−１はＡの逆行列である。
Σは対角行列であるため、｜Σ｜＝ρ_０ ^２・ρ_１ ^２である。 The dispersion of μ ₀ and μ ₁ is assumed to be ρ ₀ ² and ρ ₁ ² , respectively. The existence probability of the attribute b assumes a normal distribution centered on the point indicated in the observation data, and has no correlation between the latitude direction and the longitude direction (covariance is 0).
Then, the existence probability of the attribute b at a certain point x = (x ₀ , x ₁ ) is

As follows, according to the normal distribution of two variables.

Here, for the matrix A, | A | is the determinant of A, and A ⁻¹ is the inverse matrix of A.
Since Σ is a diagonal matrix, | Σ | = ρ ₀ ² · ρ ₁ ² .

である。
そのため、

である。 Also,

It is.
for that reason,

It is.

始点がｖ＝（ｖ_０，ｖ_１）、終点がｗ＝（ｗ_０，ｗ_１）であるリンクｅ＝φ（ｖ，ｗ）について、そのリンクｅのいずれかの地点の属性ｂの存在確率ａ（φ（ｖ，ｗ），ｂ）は、始点と終点を結ぶ直線の式がｘ_１＝ｇ（ｘ_０）であれば、ｆ（ｘ_０，ｘ_１，ｂ）＝ｐ（ｘ，ｂ）として、以下の式で表される。

Therefore, the existence probability of the attribute b at the point x is as follows.

Presence probability of attribute b at any point of link e for link e = φ (v, w) whose start point is v = (v ₀ , v ₁ ) and whose end point is w = (w ₀ , w ₁ ) a (φ (v, w), b) is f (x ₀ , x ₁ , b) = p (x, b) if the expression of the straight line connecting the start point and the end point is x ₁ = g (x ₀ ) ) Is expressed by the following formula.

ただし、全てのリンクについて全ての属性の存在確率を持たせるのは計算コストの観点から非現実的であり、実際には、予め閾値θ（０＜θ＜１）を設定し、存在確率がθ未満の場合は、リンクを選択せず、存在確率がθ以上の場合は、リンクを選択し、属性を付与する（当該リンクに関係するフィーチャーテーブル５１と対応テーブル５２のレコードを更新する：Ｓ２７）のが好ましい。 If the variances in the latitude and longitude directions are equal (ρ ₀ ² = ρ ₁ ² = ρ ² ), the existence probability is expressed by the following equation.

However, it is unrealistic from the viewpoint of calculation cost to have the existence probability of all attributes for all links. In practice, a threshold value θ (0 <θ <1) is set in advance, and the existence probability is θ If it is less than the link, the link is not selected, and if the existence probability is greater than or equal to θ, the link is selected and an attribute is given (the records of the feature table 51 and the correspondence table 52 related to the link are updated: S27). Is preferred.

<Update example of attribute data storage unit 5>
In step S27, the attribute data generation unit 4 searches the feature table 51 for a record including a combination of the attribute type and the attribute value in the observation data regardless of the format of the observation data. If there is no corresponding record, a record including a new attribute ID, the attribute type and attribute value is added to the feature table 51.

(Form 1)
For the observation data of this format, the attribute data generation unit 4 further adds a record including the attribute ID of the searched record or the added record and the node ID in the observation data to the correspondence table 52.
(Form 2)
For this observation data, the attribute data generation unit 4 specifies the node ID based on the latitude and longitude in the observation data, and thereafter updates the attribute data storage unit 5 in the same manner as the observation data of format 1.

(Form 3)
For the observation data in this format, the attribute data generation unit 4 further adds a record including the attribute ID of the searched record or the added record and the link ID in the observation data to the correspondence table 52.
(Form 4)
For the observation data in this format, the attribute data generation unit 4 further adds a record including the attribute ID of the searched record or the added record, the link ID in the observation data, the attribute start point, and the attribute end point to the correspondence table 52. To do.

(Form 5)
For the observation data in this format, the attribute data generation unit 4 further obtains a link to be connected to the start point link from the intermediate links in the observation data, and the position of the node that is the connection point of the start point link ( This is the attribute end point). The attribute data generation unit 4 adds a record including the attribute ID of the searched record or the added record, the link ID of the start point link in the observation data, the attribute start point, and the obtained attribute end point to the correspondence table 52.

  In addition, the attribute data generation unit 4 further obtains a link connected to the end point link from the intermediate links in the observation data, and the position of the node that is the connection point between this and the end point link (this is the attribute start point) ) The attribute data generation unit 4 adds a record including the attribute ID of the searched record or the added record, the link ID of the start point link in the observation data, the attribute end point, and the obtained attribute start point to the correspondence table 52.

  The attribute data generation unit 4 further adds a record including the link ID and the attribute ID of the searched record or the added record for each intermediate link.

(Form 6)
For this observation data, the attribute data generation unit 4 specifies a link (link ID) at least partially located on a straight line connecting the attribute start point and attribute end point based on the latitude and longitude of the attribute start point and attribute end point. .

When there is one link (link ID), the attribute data generation unit 4 updates the attribute data storage unit 5 in the same manner as the observation data of format 3.
When there are two or more links (link IDs), the attribute data generation unit 4 updates the attribute data storage unit 5 in the same manner as the observation data of the format 5. In this case, a link including the attribute start point is treated as a start link, a link including the attribute end point is treated as an end link, and other links are treated as intermediate links.

<Geographic network configuration change unit 6>
FIG. 10 is a flowchart of the geographic network configuration changing unit 6.
The geographic network configuration changing unit 6 reads a record from the correspondence table 52 (S31), and reads a record of the feature table 51 including the same attribute ID as the attribute ID in the record (S31). It is determined whether or not the attributes (attribute start point, attribute end point, attribute type, attribute value) specified by these records are inconsistent with other attributes (S33).

If it is determined that there is a contradiction, the geographic network configuration changing unit 6 updates the node table 21, the link table 22, the feature table 51, and the correspondence table 52 as necessary so as to resolve the contradiction (S35: Geographic network). Configuration change).
If it is determined that there is no contradiction, or if an update has been performed, the geographic network configuration change unit 6 determines whether there is a record in the next correspondence table 52 (S37), and if there is, returns to step S31 and does not exist If so, terminate the process.

  Further, the geographic network configuration changing unit 6 updates the node table 21, the link table 22, the feature table 51, and the correspondence table 52 as necessary even when there is no contradiction.

FIG. 11 is a diagram illustrating an example of a configuration change of a geographic network.
(Node change format 1)
For example, the geographic network configuration changing unit 6 adds a node on the link (FIG. 11A).
(Node change format 2)
For example, the geographic network configuration changing unit 6 adds an isolated node (FIG. 11B).

(Node change format 3)
For example, the geographic network configuration changing unit 6 deletes the node on the link (FIG. 11C).
(Node change format 4)
The geographic network configuration changing unit 6 deletes, for example, an isolated node (FIG. 11D).

(Link change format 1)
For example, the geographic network configuration changing unit 6 adds a link between nodes where no link exists (FIG. 11E).
(Link change format 2)
For example, the geographic network configuration changing unit 6 adds a link between nodes where the link exists (FIG. 11 (f)).

(Link change format 3)
The geographic network configuration changing unit 6 deletes, for example, a unique link between nodes (FIG. 11 (g)).
(Link change format 4)
For example, the geographic network configuration changing unit 6 deletes one link between nodes having a plurality of links (FIG. 11 (h)).

(Detailed processing of node change format 1 (add node on link))
There is a record of the node v1, a record of the node v2, and a record of the link e1 having the node v1 as the link start point and the node v2 as the link end point. The correspondence table 52 is set with the link ID of the link e1 as the correspondence destination ID. Consider a case where an attribute ID of a record including the attribute (a combination of attribute type and attribute value) f1 of the feature table 51 is set in the record. In addition, it is assumed that the attribute start point and the attribute end point are expressed as a ratio obtained by dividing the distance from the link start point by the link length.

The processing for adding a new node v3 record is shown below.
(1) A new record of the node v3 is created in the node table 21, a new node ID is set in this record, and the latitude and longitude of a point indicating any position of the link e1 are set.
(2) The node ID of the record of the node v3 is set as the end point ID of the record of the link e1 in the link table 22.
(3) Create a new link e2 record in the link table 22, set a new link ID in this record, set the node ID of the record of node v3 as the start point ID, and set the second ID as the end point ID. The node ID of the record of node v2 is set.

(4) Update the record in the attribute data storage unit 5 for the attribute f1.
(4a) When attribute f1 should be given to both link e1 and link e2 A record is newly created in correspondence table 52, the link ID of link e2 is set as the correspondence destination ID, and attribute f1 is assigned as the attribute ID. The attribute ID of the record of the included feature table 51 is set.

Note that the attribute ID of the record of the feature table 51 including the attribute f1 is already set as the attribute ID in the record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID.
Further, the attribute start point of the record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID is changed according to the length of the shortened link e1.
Further, the attribute end point of the record in the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID is updated so as to indicate the node v3.
Further, the attribute start point of the record in the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID is updated so as to indicate the node v3.
Further, the attribute end point of the record of the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID, and the attribute before the update of the record in the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID. Update to indicate end point.

(4b) When the attribute f1 should be given only to the link e1 A record is newly created in the correspondence table 52, the link ID of the link e2 is set as the correspondence destination ID, and the feature table 51 includes the attribute f1 as the attribute ID. Set the attribute ID of the record. Note that the attribute ID of the record of the feature table 51 including the attribute f1 is already set as the attribute ID in the record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID.
Further, the attribute start point and attribute end point of the record in the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID are changed according to the length of the shortened link e1.

(4c) When the attribute f1 should be given only to the link e2 A record is newly created in the correspondence table 52, the link ID of the link e2 is set as the correspondence destination ID, and the feature table 51 includes the attribute f1 as the attribute ID. Set the attribute ID of the record. Note that the attribute ID of the record of the feature table 51 including the attribute f1 is already set as the attribute ID in the record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID.

The record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID (the attribute ID of the record of the feature table 51 including the attribute f1 is already set as the attribute ID) is deleted.
In addition, the attribute start point and attribute end point of the record in the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID, and the record in the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID. Update to show previous attribute start point and attribute end point respectively.

FIG. 12 is a diagram illustrating an example of the node change format 1.
As shown in the figure, when two nodes are added on the link, the node is added on the link by the method of the node change format 1, and the method of the node change format 1 is added on the original link or the added link. Just add a node. For example, when there is an attribute in a certain part of the link, the attribute start point can be one node to be added, and the attribute end point can be the other node to be added. In this case, the attribute is set for the entire link between the two added nodes.
(Detailed processing of node change format 2 (addition and deletion of isolated nodes))

  Since it is obvious how to update the node table 21, the feature table 51, and the correspondence table 52 when adding an isolated node, a description thereof will be omitted.

(Detailed processing of node change format 3 (deletion of node on link))
There is a record for node v1, a record for node v2, a record for link e1 with node v1 as the link start point, and a node v2 as the link end point, a record with node v2, a record with node v3, and a node v2 with the link start point. There is a record of link e1 with v3 as the link end point, and the attribute ID of the record including the attribute f1 of the feature table 51 is set to the record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID. Consider a case where the attribute ID of a record including the attribute f2 of the feature table 51 is set in the record of the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID. In addition, it is assumed that the attribute start point and the attribute end point are expressed as a ratio obtained by dividing the distance from the link start point by the link length.

The processing for deleting the node v2 is shown below.
(1) One of the links to be left is arbitrarily selected from the links e1 and e2. Here, it is assumed that the link e1 is selected.
(2) In the link table 22, the end point ID of the record of the link e1 is changed from the node ID of the node v2 to the node ID of the node v3.
(3) Also, the attribute start point and attribute end point of the record of the correspondence table 52 in which the link ID of the link e1 is set as the correspondence destination ID are changed according to the length of the link e1 that has become longer.
(4) Also, a new record is created in the correspondence table 52, the link ID of the link e1 is set as the correspondence destination ID, and the attribute ID of the record of the feature table 51 including the attribute f2 is set as the attribute ID. Then, the attribute start point and attribute end point of the record of the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID is divided by the length of the link e1 that is a longer distance from the node v1 (link start point). Express as a percentage and set these to the newly created record.
(5) The node v2 record is deleted from the node table 21, and the link e2 record is deleted from the link table 22.

(Detailed processing of node change format 4 (delete isolated node))
Since it is obvious how to update the node table 21, the feature table 51, and the correspondence table 52 when deleting an isolated node, a description thereof will be omitted.

(Detailed processing of link change format 1 (adding links between nodes that do not have links))
Since it is obvious how to update the link table 22, the feature table 51, and the correspondence table 52 when adding a link between nodes where no link exists, the description is omitted.
(Detailed processing of link change format 2 (adding links between nodes with links))
There is a record of the node v1, a record of the node v2, and a record of the link e1 having the node v1 as the link start point and the node v2 as the link end point. The correspondence table 52 is set with the link ID of the link e1 as the correspondence destination ID. The attribute ID of the record including the attribute f1 of the feature table 51 is set in the first record among the three records, and the attribute ID of the record including the attribute f2 of the feature table 51 is set in the second record. Consider the case where the attribute ID of the record including the attribute f3 of the feature table 51 is set in the record 3. Also, attribute f3 is not inconsistent with attribute f1, and is inconsistent with attribute f2.

The process for adding a new link in this case is shown below.
(1) In the link table 22, a record including a start point ID indicating the node v1 (link start point), an end point ID indicating the node v2 (link end point), and a link ID of the new link e2 is created.
(2) The attribute ID of the attribute f3 is searched from the feature table 51, and a record in which this attribute ID is set is searched from the feature table 51.
(3) Since the correspondence ID of the retrieved record is the link ID of the link e1, this is changed to the link ID of the link e2.

FIG. 13 is a diagram illustrating an example of the link change format 2.
For example, attribute f4 (a set of attribute type “movement method” and attribute value “stairs”) and attribute f5 (a set of attribute type “movement method” and attribute value “slope”) are set for link e1. If it is determined that the link e1 is actually divided into a staircase and a slope, the link e2 is newly added and the attribute f5 is set.
As described above, in the present embodiment, for a conflicting attribute, the conflict can be resolved by adding a link having the attribute.

(Detailed processing of link change format 3 (deleting the only link between nodes))
Since it is obvious how to update the link table 22, the feature table 51, and the correspondence table 52 when deleting the only link between nodes, the description is omitted.
(Detailed processing of link change format 4 (one link is deleted from between nodes having multiple links))
There are a record for node v1, a record for node v2, a record for link e1 with node v1 as the link start point, a link end point with node v2, and a record with link e2 with node v1 as the link start point and node v2 as the link end point. I will do it.

Here, the attribute ID of the record of the feature table 51 including the attribute f1 is set as the attribute ID in one of the two records of the correspondence table 52 in which the link ID of the link e1 is set as the corresponding ID, and the attribute is set in the other It is assumed that the attribute ID of the record of the feature table 51 including the attribute f2 is set as the ID.
In addition, the record of the feature table 51 including the attribute f3 that does not contradict the attribute f1 but contradicts the attribute f2 as the attribute ID in the record of the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID. It is assumed that an attribute ID is set.

For example, when it is no longer necessary to assign the attribute f2 to the link e1, the link e2 is deleted.
(1) A record in which the attribute ID of the record of the feature table 51 including the attribute f2 is not set as the attribute ID is searched from the correspondence table 52 in which the link ID of the link e2 is set as the corresponding destination ID. Is read. Here, the same attribute ID as the record of the feature table 51 including the attribute f3 is read.
(2) Delete the record of link e2 from the link table 22.
(3) Delete the record of the correspondence table 52 in which the link ID of the link e2 is set as the correspondence destination ID from the correspondence table 52.
(4) A new record is created in the correspondence table 52, the link ID of the link e1 is set as the correspondence destination ID, and the attribute ID (attribute ID of the attribute f3) read in (1) is set as the attribute ID. .
When a plurality of attribute IDs are read in (1), (4) is performed for each.

  As described above, according to the present embodiment, an attribute can be automatically assigned to a node or link of a geographic network, the configuration of the geographic network can be changed according to the attribute, etc. Can be resolved.

  In the present embodiment, the geographic network configuration changing unit 6 updates the node table 21, the link table 22, the feature table 51, and the correspondence table 52 as necessary, but may be as follows. That is, the attributes set in the feature table 51 and the correspondence table 52 are set in the node table 21 and the link table 22, that is, the node table 21 with the attribute and the link table 22 with the attribute are converted into the original node table 21 and the original table. The original node table 21, the original link table 22, the feature table 51, and the correspondence table 52 are not updated, and the node table 21 with attributes and the link table 22 with attributes are updated. May be.

  A computer program for causing a computer to function as the geographic network data management apparatus of this embodiment can be recorded on a computer-readable recording medium such as a semiconductor memory, a magnetic disk, an optical disk, a magneto-optical disk, or a magnetic tape. It can be widely distributed by being transmitted through a communication network such as the Internet.

DESCRIPTION OF SYMBOLS 1 Observation data acquisition part 2 Geographic network data storage part 3 Observation data storage part 4 Attribute data generation part 5 Attribute data storage part 6 Geographic network structure change part

Claims (5)

For a node that is each point included in the target area, a node table including a node ID indicating the target node, a record including the latitude and longitude of the node,
For each link connecting the nodes, a link table including a link ID indicating the target link, a record including the node ID of each node, and
For each attribute set for a node or a link or a part of a link, an attribute ID indicating a target attribute, a feature table including a record including the attribute,
For each node or link for which the attribute is set, the node ID of the node or the link destination ID of the link, the attribute ID that is the same as the attribute ID included in the record of the feature table corresponding to the target attribute, If the attribute is set to a part of the link, a correspondence table comprising attribute start showing one end of the part, and, the record including the attribute endpoint indicating the other end of the part,
The observation data including the node ID of the node or the link ID of the link or the position information of the node or the link and the attribute is read from the observation data storage unit, and the feature table and the correspondence table are updated based on the observation data, and the observation is performed. When the data includes an attribute start point and an attribute end point, an attribute data generation unit that sets the attribute start point and the attribute end point in the corresponding record of the correspondence table ;
When there are a plurality of records having the same link ID and mutually conflicting attribute IDs in the correspondence table, a record of a new link ID that shares the link start point node and end point node of the link ID A geographic network data management comprising: a geographic network changing unit that adds one link ID to which the mutually contradictory attributes of the correspondence table are added to the table, and updates the new link ID with the added link ID apparatus. The attribute data generation unit
The geographic network data management according to claim 1, wherein a node or a link is selected based on position information included in observation data, and a record of the feature table and the correspondence table related to the node or link is updated. apparatus. The attribute data generation unit
3. The existence probability of the attribute at an arbitrary point is obtained based on position information and attributes included in the observation data, and a link at a point where the existence probability is equal to or higher than a predetermined threshold is selected. Geographic network data management device. The geographic network data management device according to claim 1, further comprising: an observation data acquisition unit that determines whether or not there is a contradiction between a plurality of observation data based on a preset rule. Computer program for causing a computer to function as geographic network data management apparatus according to any one of claims 1 to 4.

Images