EP2143096A1

EP2143096A1 - Method for creating a directory of road sections, method for determining all road sections within a search area and computer programme

Info

Publication number: EP2143096A1
Application number: EP08716773A
Authority: EP
Inventors: Andreas Vogel; Guido Mueller
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2007-04-03
Filing date: 2008-02-08
Publication date: 2010-01-13
Also published as: DE102007016002A1; US20090234877A1; JP2010523951A; WO2008119585A2; US9214097B2

Description

description

title

A method for creating a directory of road sections, methods for determining all road sections within a search area and computer program

The invention relates to a method for creating a directory of road sections of a digital road map for a navigation system.

The invention further relates to a method for determining all road sections within a search area.

The invention further relates to a computer program for carrying out a method for creating a directory of road sections and a method for determining all road sections within a search area.

Navigation systems for vehicles require a digital map of the road network for the basic functions of location, destination selection, route calculation and route guidance. This digital road map is usually on a digital storage medium, such as a CD, DVD or SD card. On the storage medium, the data representing the digital road map is uninhabited in files. The data is present in a linear sequence in the file.

The navigation system needs this data, for example, to display the map of a road network on a screen or to project the current position on a nearest road of the road network. The current position can be determined, for example, by means of GPS (Global Positioning System) or Galileo and / or further inertial sensors, such as, for example, odometers, yaw rate sensors or accelerometers. In many applications of the navigation system, large parts of the entire road network stored on the storage medium play no or only a minor role. For example, a specific destination route may be limited to a regional area even though the road network of an entire country is stored. It is also conceivable that the user an enlarged section of his current

Wants to show position.

If all roads within a two-dimensional search area are searched for in a linearly stored digital road map, the entire road map must be searched, which entails considerable expenditure of time. In addition, the takes

Information content of digital road maps continuously increasing; On the one hand, individual roads are divided into ever shorter sections of road, on the other hand added additional information, such as geographical locations of gas stations or so-called points of interest (POI). The time required to search the roadmap can result in a delayed presentation that is for a

User is disturbing.

In order to avoid the full search of the road map in such cases, the roads contained in the digital road maps are structured.

It is known to combine adjacent streets in an information unit, a so-called cluster. In doing so, it orients itself to urban areas, city boundaries, county boundaries or similar geographical boundaries. Neighboring clusters are grouped into hierarchically higher clusters. This process continues for higher levels. The result is a relatively well-balanced, hierarchical tree. So that the tree does not become too large, roads are laid down linearly in geometric sorting on the lowest level. Typical numbers at this level are about 1000 streets.

In another known structuring the digitized road network is covered with a uniform grid, for example, from a variety of equal size

Rectangles, the so-called tiles. All roads that lie in a tile are stored together. An index is created on the storage medium, which assigns geographic coordinates to the tiles so that the index can be accessed quickly from a coordinate on the tile. However, it is disadvantageous that the Tiles are very differently covered with streets; some tiles have a lot of streets, others have little or no roads. This makes the storage ineffective and the time required for a search very dependent on the location of the search area.

The invention therefore provides a method according to claim 1.

According to the invention, it is provided that the road sections are arranged in an ordered N-tree without the additional structural information required in the prior art being required. The method according to the invention uses in each case a road section of the road map as the starting point for the division of the road map, ie as "spatial divider" of the road map Shapepoints are already available in the map, there is no need for additional data for the division of the road map, which reduces the required storage space and relieves the search in the tree according to the invention to consider additional information, such as the POI, for a given search time.

The method according to the invention can be applied to an entire road map as well as in combination with an already known method. In a combination with known methods, the method according to the invention is advantageously used at the lowest level, where a plurality of roads, typically the

Order of 1000 roads, stored in linear files. With the aid of the method according to the invention, the search at this level can be limited from the search of the approximately 1000 streets to the search of approximately 10 to 20 streets, whereby a considerable saving of time is achieved. Thus, the tree according to the invention is compatible with the previously used data structures.

Appropriately, the assignment of the road sections to the sub-levels then takes place in which part level they are completely. Can not a street section clearly one - A -

are assigned to the part levels, it is assigned to that part level in which the road section that generates the partitioning.

In one embodiment, the method according to the invention is characterized by M = 3. The list of road sections is arranged in this case in a tri-tree. Each road section divides the road maps into three sub-levels. Advantageously, the three sub-levels can be arranged as follows:

First, a straight line is defined by the start and end points of a road segment. Then two are parallel to this first straight line

Straight lines formed and arranged to each other, that the road section is completely enclosed, that is, including possibly existing shapepoints, between these two parallel lines. In this case, it may be advisable to arrange the two parallel straight lines only at such a distance from each other that just all the points of the road section are enclosed between them. Of the

Road section has a clear beginning and end point

Orientation, for example, from the beginning to the end point. If the

Road section is considered in the direction of its orientation, lies one of the three

Partial levels clearly on the left side of the road, another part level on the far right of the road section and the third part level includes the road section itself, that is, it is located in the "middle".

Another embodiment of the method according to the invention is characterized by N = 5. The list of road sections is structured in this case a so-called 5-tree. Advantageously, the five sub-levels can be arranged as follows:

First, a rectangle is placed over the considered road section. For example, the start and end points of the road section may form the endpoints of a diagonal of the rectangle, and the rectangle may be chosen to be large enough to have all shapepoints within the rectangle. It may be advisable to choose the rectangle as small as possible. In a second step, the sides of the rectangle are supplemented to half-lines, each with the same mathematical sense of direction, for example clockwise or in the Counterclockwise, have. The rectangle and the four half-lines clearly partition the road map into five sub-levels. In terms of the orientation of the road section, the sub-levels can be described as "left", "center" (rectangle), "right", "before" and "afterwards".

The search within the inventively generated N-tree can be performed very quickly if pointers are stored in each node to directly following nodes. In this way, the addresses of the direct successors are immediately available in each node, so that it is possible to enter a sub-tree of the node without a time delay.

As a further invention, a method according to claim 7 is presented.

The search duration for a road section is proportional to the number of road sections stored in the list in a conventional linear list. In contrast, in an N-tree generated according to the present invention, the search time increases only with log (N) when N is the number of road sections stored in the N-tree or the node of the N-tree. Especially with a large number of stored road sections, the advantage of the method according to the invention is clearly evident.

The N-tree is expediently generated offline. Thus, a lot of time can be invested in the balance of the N-tree, which can significantly reduce the search time online, that is in a navigation system.

Preferably, the inventive method for determining all road sections within a search area is characterized by a rectangular search area. A rectangular search area needs very little data to uniquely define it. In addition, it is very easy to process, which reduces the search time compared to more complicated geometric search areas.

As an invention, a computer program with program code according to claims 9 and 10 is presented. The invention will be explained in more detail with reference to the embodiments illustrated in the following figures. Show it:

FIG. 1 shows a detail of a first digital road map;

FIG. 2 shows a detail of a second digital road map,

Figure 3 - section of a third digital road map and

Figure 4 - a 3-tree (Tribaum) to the dargestellen in Figure 3 section.

FIG. 1 shows a section of a first digital road map 1.

The digital road map 1 contains a road section 2. The road section 2 has a starting point 3a, subsequently two shapepoints 3b, 3c and one

Endpoint 3d on. The two shapepoints 3b, 3c give one of a straight line

Course deviating, that is curvy, course of the road section 2 again. In the digital road map 1, the road section 2 stores the geographical coordinates such as the longitude and the latitude. Other information may be stored, such as altitude above

Sea level or data on petrol stations or POIs located on the road section 2.

The road section 2 is assigned an orientation. The orientation is indicated by the arrow 4, which is directed from the starting point 3 a to the end point 3d.

In Figure 1, two mutually parallel straight lines 5a, 5b are shown. The two straight lines 5a, 5b include a middle part plane 6a between them. The road section 2 runs completely in the middle part plane 6a.

Seen from the road section 2, in the direction of the orientation, a left part plane 6b is located to the left of the road section 2 or the middle part plane 6a. A right part plane 6c lies to the right of the road section 2 or the middle part plane 6a. This oriented at the Orientierang of the road section 2 arrangement of the sub-levels 6a, 6b, 6c is unique.

The three sub-levels 6a, 6b, 6c form a clear decomposition and thus one

Partitioning the digital road map 1.

The partitioning of the digital road map 1 shown in Figure 1 in three sub-levels 6a, 6b, 6c through the road section 2 forms the basis for the classification of other road sections and the construction of a tri-tree.

FIG. 2 shows the decomposition of a second digital road map 7 by a road section 8 in 5 sub-levels 9a, 9b, 9c, 9d, 9e. The orientation of the road section 8 is represented by an arrow 10.

The road section 8 has a starting point I Ia, an end point I Ib and two shapepoints 11c, Hd. The entire road section with start, end and shapepoints I Ia, I Ib, 11c, Hd lies within a rectangle that represents one of the sub-levels 9a, 9b, 9c, 9d, 9e. The sides of the rectangle have been supplemented to four half-lines 12a, 12b, 12c, 12d. Addition of sides of a rectangle to half-lines

12a, 12b, 12c, 12d is clockwise.

Through the rectangle and the half-lines 12a, 12b, 12c, 12d, the digital road map 7 is divided into the 5 sub-levels 9a, 9b, 9c, 9d, 9e. Depending on the orientation of the road section 8, the sub-levels 9a, 9b, 9c, 9d, 9e can be in the

"Middle" part plane 9a, the "right" part plane 9b, the "left" part plane 9d, the part plane 9c "before" and the part plane 9e "behind" arrange.

FIG. 3 shows a section of a third digital road map 13.

The digital road map 13 includes 10 road sections 14 a to j whose orientation is indicated by an arrow 15, respectively. For the sake of clarity, the arrow 15 is shown only for the road section 14f. In addition, a search area 16 is shown in FIG. The search area 16 at least partially detects the road sections 14a, 14b, 14c.

In the following it will first be explained how a directory of the road sections 14a to j is created by means of the method according to the invention.

The directory will receive the structure of a tri-tree 17 as shown in FIG.

For the tri-tree 17 is the example of the road section 14e following

Search tree condition defined: By the road section 14e of the same named node, the road map 13 is partitioned into three sub-levels, of which exactly one contains the road section 14e. The sub-levels are arranged depending on the orientation of the road section 14e to each other. All other road sections 14a to d can be exactly one of the sub-levels and thus one

Assign subtree of the node. In the left part of the plane remain the road sections 14a to d, in the middle part of the level of the road section 14f and in the right part of the level, the road sections 14g to j.

In the next step, the left part level is split further. This will be one of

Road sections 14a to d, which lie in this sub-level selected. In this example, this is the road section 14c.

The road section 14c divides the left sub-level now again in three sub-levels, one of which is on the left with respect to the orientation of the road 14c, another in the

Middle and the third right.

The not yet selected road sections 14a, b, d are now assigned to these sub-levels. The road section 14a lies in the left partial plane of the road section 14c, the middle part plane is empty and in the right part plane are the

Road sections 14b, c.

Next, the road section 14d is selected as the next node. Now only the road section 14b remains. This is relative to the road section 14d in the left part plane. There are no road sections in the middle and right part levels. Thus, with respect to the road section 14e, the root of the tri-tree, all road sections 14a to d are inserted in the left sub-tree.

For the middle and right subtree, the nodes are set accordingly. This results in the tri-tree 17 of Figure 4.

In the following, the inventive method for determining all road sections within a search area using the example of the road map 13, the tri-state

Tree 17 and the search area 16 explained.

The entry for the search in the tri-tree 17 is always about the root, in this

Fall the road section 14e.

With respect to the road section 14e, the search area 16 is completely in the left

Tele level. Since the search area 16 has no intersection with the middle sub-level, the road section 14e does not come into consideration as a possible search result

(Process step f)).

In method step g), only the left part plane is determined in the following example.

After step h), a subtree 18 is now to be determined, which is assigned to the left partial plane. The subtree 18 includes the road sections 14a to d.

Subtree 18 is noted after method step i).

After process step j), the process steps b) to i) are to be repeated, wherein the selection is limited to the subtrees noted in step i) until all the noted subtrees have been selected exactly once.

In the present example, only the subtree 18 has been noted, so that only this can be selected. The root node of this subtree 18 is the node with the road section 14c. For the road section 14c, the search area 16 is both in the left, right and middle part levels. After method step f), the road section 14c comes into consideration as a possible search result and is noted.

There are now subtrees 19a, b to be determined, which are attached to the node of the road section 14c. The one subtree 19a consists only of the node with the

Road section 14a. The second sub-tree 19b consists of the nodes to the road sections 14d as the root node of this sub-tree 19b and the road section 14b.

With respect to the road section 14a, the search area 16 is located in both the left, right, and middle parts planes. Since the node to the road section 14a is a leaf of the tri-tree 17, no further subtrees need to be noted. Only the road section 14a is to be noted as a possible search result.

The examination of the subtree with the road section 14d as the root causes the search area 16 to lie completely in the left partial plane. Thus, the road section 14b is obtained. With respect to the road section 14b, which corresponds to a leaf of the tri-tree 17, the search area 16 lies in the middle part plane. Thus, the road section 14b comes as a search result in question and is noted.

Now all subtrees 18, 19a and b that have been recorded in the meantime have been processed.

In the last method step, of the possible possible search results that are marked, the road sections 14a, b, c are eliminated, those with which the search area 16 actually has no overlap. It is present no road section

14a, b, c excrete. Thus, the search leads to the search result: road sections 14a, b, c.

Claims

claims

Method for creating a directory of road sections (2, 8, 14a to j) of a digital road map (1, 7) for a navigation system, in which the

Directory receives the data structure of an ordered N-tree (17), in which the road sections (2, 8, 14a-j) are stored in nodes, with the steps:

a) assigning a geographical location to each of the road sections (2, 8, 14a to j);

b) assigning an orientation to each of the road sections (2, 8, 14a-j);

c) Defining a search tree condition:

(i) through the road section (2, 8, 14a-j) of the node, the road map is partitioned into N sub-planes (6a, b, c), of which exactly one contains the road section (2, 8, 14a-j);

(ii) the part planes are dependent on the orientation of the road section

(2, 8, 14a to j) arranged to each other;

(iii) all road sections (2, 8, 14a-j) can be assigned to exactly one of the sub-levels and thus to exactly one sub-tree of the node;

d) selecting one of the road sections (2, 8, 14a to j);

e) searching the selected road section in the tree (17); the search ends at a node [end node] with maximally N-I subtrees, none of which contains the selected road section (2, 8, 14a-j);

f) attaching the selected road section (2, 8, 14a to j) as a leaf to the end node according to the search tree condition; g) repeating steps d) to f) until each road section (2, 8, 14a to j) has been selected exactly once;

h) Save the directory to a disk.

2. The method according to claim 1, characterized in that the

Road sections from step (iii) are assigned to the part plane in which they lie completely, otherwise the part plane in which the road section from step (i) is located.

3. The method according to claim 1 or 2, characterized by N = 3.

4. The method according to claim 3, characterized in that the three sub-levels are formed as follows:

a) Defining a first straight line through a start and an end point of the

Road section (2, 8, 14a to j);

b) forming two straight lines parallel to each other and parallel to each other, between them completely enclosing the road section (2, 8, 14a to j);

c) arranging the three sub-levels depending on the orientation of the road section (2, 8, 14a-j) ("left", "center", "right").

5. The method according to claim 1 or 2, characterized by N = 5.

6. The method according to claim 5, characterized in that the five sub-levels are formed as follows:

a) forming a rectangle including the road section (2, 8, 14a-j);

b) Along the sides of the rectangle are formed half-lines having the same orientation with respect to the orientation of the road section (2, 8, 14a-j); c) arranging the five sub - levels depending on the orientation of the

Road section (2, 8, 14a to j) ("left", "center", "right", "before", "after").

7. The method according to any one of the preceding claims, characterized in that are stored in each node pointers that refer to subsequent nodes.

8. A method for determining all road sections (2, 8, 14a to j) of a directory generated by one of the methods according to claims 1 to 7 within a search area, comprising the steps of:

a) defining the search area (16) as a sub-area of the road map;

b) selecting one of the subtrees starting with the tree (17);

c) determining a root node of the selected subtree;

d) determining the sub-levels of the road section (2, 8, 14a-j) stored in the root node;

e) determining with which of the particular sub-levels the search area overlaps;

f) memorizing the road section (2, 8, 14a to j) as a possible search result, if the search area has an overlap with that part plane in which the

Road section lies;

g) determining those remaining sub-levels for the further search with which the search area overlaps;

h) determining the subtrees associated with the part planes determined in step g);

i) memorize the subtrees determined in step h); j) repeating steps b) to i) until all the memorized subtrees have been selected exactly once, the selection being limited to the subtrees noted in step i);

k) leaving out the road sections (2, 8, 14a to j) marked in step f) with which the search area has no overlap.

9. The method according to claim 8, characterized by a rectangular search area.

10. A computer program with program code for carrying out all the method steps according to one of claims 1 to 9, when the program is executed in a computer.

11. Computer program with program code on a machine readable

Carrier is stored for carrying out the method according to one of claims 1 to 9, when the program is executed in a computer.