JP3012096B2 - Route search method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a route search method, and more particularly, to a method of pruning a route deviating from a direction of a straight line connecting a departure place and a destination, or lowering a search priority, from a departure place to a destination. The present invention relates to a route search method for performing a heuristic search for an optimum route connecting the routes.

[0002]

2. Description of the Related Art An in-vehicle navigator has a large-capacity storage device such as a CD-ROM for storing a large amount of map data, a display device, a vehicle position detecting device for detecting a current position of a vehicle, and the like. Map data corresponding to CD-R
Read from the OM, draw a map on the display screen based on the map data, fix the vehicle position mark (location cursor) at a specific position (for example, the center of the screen) on the display screen, and scroll the map according to the movement of the vehicle Display or map is fixed on the screen,
By moving and displaying the vehicle position mark, it is possible to see at a glance where the vehicle is currently traveling.

A map stored in a CD-ROM is divided into plots of appropriate longitude and latitude widths according to the scale level, and roads and the like are represented by vertices (nodes) expressed in latitude and longitude. Are drawn by sequentially connecting the nodes with straight lines. A road is formed by connecting two or more nodes, and a portion connecting the two nodes is called a link. As shown in FIG. 19, the map data is divided into four data units (4
(For each division), and one data unit corresponds to one screen. The map data includes (1) a road layer including a road list, a node table, an intersection configuration node list, and an adjacent node list; (2) a background layer for displaying roads, buildings, rivers, and the like on a map screen;
(3) It is composed of a character layer for displaying names of cities, towns and villages, names of roads, and the like.

[0004] Among them, the road layer has a configuration shown in FIG. The road list RDLT includes, for each road, the type of road (expressway, national road, and other roads), the total number of nodes forming the road, and the node table N of nodes forming the road.
It consists of data such as the position on the DTB and the width to the next node. Intersection configuration node list CRLT
Is a node table N of link other end nodes (intersection constituent nodes) connected to the respective intersections on the map.
A set of positions on the DTB. Neighbor node list NN
LT is an adjacent node that is defined on a unit boundary and shared by a plurality of units with respect to a road that spans a plurality of units (see FIG. 25). The number of adjacent nodes shown in FIG.
For the adjacent node RN defined by the unit AU ₁ (AU ₂ ) in 5 (1), the other shared unit is AU
₂ (AU ₁ ), the number of adjacent nodes is 1, and FIG.
5 (2) defined in the unit AU ₁₁ in the neighbor node RN
AU ₁₁ , AU ₂₁ , A
Number neighboring nodes since it is three U ₂₂ 3), the map mesh numbers,
It is composed of the unit code to which the adjacent node belongs in the figure and the position of the unit on the node table.
The node table NDTB is a list of all nodes on the map, and includes coordinate information (longitude and latitude) for each node, an intersection identification flag indicating whether or not the node is an intersection, and an intersection in the intersection configuration node list if the node is an intersection. A pointer to a position, if not an intersection, a pointer to the position of the road to which the node belongs on the road list; an adjacent node identification flag indicating whether the node is an adjacent node;
It is composed of a pointer or the like pointing to a position in the adjacent node list NNLT.

The in-vehicle navigator has a route guidance function for searching for an optimal route that traces the shortest distance from the departure point to the destination, displaying a guidance route on the screen and guiding the driver to travel. At this time, the driver can easily reach the destination by making the guidance route distinguishable from other roads, for example, by displaying the guidance route in a specific color and bold. A method using a horizontal search method has been proposed as a method for obtaining an optimal route connecting a starting point to a destination. This method refers to road data and refers to all intersections existing in one or a plurality of quadrants that include all quadrangular regions having a straight line connecting the departure point and the destination as a diagonal (in addition to the original intersections). , Including simple nodes that are adjacent nodes), the intersection netlist CRNL for each intersection
Is stored in the route search memory, and the shortest route from the departure point to the destination is searched for with reference to the intersection netlist.

The intersection netlist CRNL includes, for each intersection (including an intersection node and a simple node that is an adjacent node), (1) intersection sequential number (information for specifying the intersection) ( 2) Map leaf number of the map including the intersection (3) Data unit code (4) Position on node table (5) Longitude (6) Latitude More than intersection ID (7) Position on intersection configuration node list (the intersection concerned) Is the original intersection node) (8) Number of nodes constituting the intersection (same as above) (9) Position on the adjacent node list (when the intersection is an adjacent node) (10) Number of adjacent nodes (same as above) (11) Each neighbor Intersection sequential number (12) Distance to each adjacent intersection (13) Intersection sequential number immediately before the intersection (14) From the departure point Total distance to the intersection (15) Search degree or the like of the intersection are registered (where (13) to
(15) is registered when the route search is executed).

The method of creating the intersection netlist CRNL is as follows. First, as shown in FIG. 26, for each figure including a rectangular area having a straight line connecting the departure point and the destination as a diagonal line, the figure is obtained from the figure management information of the map data. Get leaf number. Then, in the map data in each of these figures leaves enter the road data for each four-divided diagram leaf (AU ₁₁ ~AU ₄₄ of FIG. 26) that includes a rectangular area of the line connecting the origin and destination diagonal (Each quadrant is identified by a leaf number and a data unit code.) The node table NDTB is searched for a node where an intersection identification flag or an adjacent node identification flag is set, and an intersection sequential number (1 ) Is prepared on the route search memory, and the intersection ID is registered ((2) to (6)). Then, from the node table, the intersection configuration node list, and the adjacent node list, the position on the intersection configuration node list, the number of intersection configuration nodes, the position on the adjacent node list, and the number of adjacent nodes are obtained, and (7) to (10) are obtained. register. Next, for each road in the road list RDLT, referring to the node table NDTB, the distance of a link between adjacent intersections (including the case where one is a simple node and an adjacent node) is determined, and one of the links is determined. The other intersection of the link is set as the adjacent intersection in the intersection net list relating to the intersection, and the intersection sequential number (adjacent intersection sequential number) registered in the intersection net list relating to the adjacent intersection and the distance of the link are registered ( (11), (12)). If the intersection netlist is created for an intersection node that also serves as an adjacent node and for an adjacent node of a simple node, an intersection netlist is created for the same adjacent node under other shared units. Since there is a case in which the intersection sequential number is found by referring to the adjacent node list NNLT, the adjacent intersection sequential number and the distance are registered in the intersection net list, Perform adjacent node connection processing. At this time, the distance is set to 0.

In this manner, the intersection netlist CR
When the NL is completed, an optimal route search process is performed by a horizontal search method based on the departure place data and the destination data.
FIG. 27 is a schematic explanatory diagram of the horizontal search method.
An intersection (including an adjacent node in a simple node) is cliffed as an intersection of straight lines, the distance between the intersections is known, the STP is the departure point (intersection), and the DSP is the destination (intersection). In the horizontal search method, the departure point is set to the 0th-order intersection (degree 0 is registered in (15) of the intersection netlist) while referring to the intersection netlist CRNL, and the departure point is adjacent to the 0th-order intersection along the road. Search for the primary intersections A _{1 to} A _4, and for each primary intersection A _{1 to} A ₄ , from the departure point via the corresponding previous intersection (the next lower intersection; here, the departure point intersection) Is accumulated and registered in each intersection net list together with the intersection sequential number for specifying the previous intersection and the search order 1 (corresponding to each intersection A _{1 to} A ₄ (( 13) to (15)).
Next, for each of the primary intersections A _{1 to} A ₄ , the secondary intersection B
_ij, and for each secondary intersection, the corresponding one before
The total distance from the departure point via the next intersection is obtained, and one corresponding to each intersection B _ij is _added to each intersection netlist.
Intersection sequential number identifying the previous intersection,
Register with search order 2.

[0009] For example, find primary intersection A ₁ 3 two-order crossroads B ₁₁ for, B _12, B _13, so as to correspond to respective intersections, B _11: from the starting point over the primary crossing point A ₁ Cumulative distance Bd ₁₁ B ₁₂ : Cumulative distance from departure point via primary intersection A ₁ Bd ₁₂ B ₁₃ : Cumulative distance Bd ₁₃ ··· (a) from departure point via primary intersection A ₁ stored with the intersection sequential number of the intersection a ₁ to. Further, the primary crossing point A ₂ 3 two-order crossroads B ₂₁ for, B _22, B ₂₃ is Motomari, each secondary intersection B
₂₁ , B ₂₂ , and B ₂₃ , B ₂₁ : the cumulative distance from the departure point via the primary intersection A ₂ Bd ₂₁ ... (B) B ₂₂ : from the departure point via the primary intersection A ₂ the total distance Bd ₂₂ B _23: stored with the corresponding intersection sequential number of the intersection a ₁ a total distance Bd ₂₃ from the starting point over the primary crossing point a _2. Similarly, for the other primary intersections A ₃ and A ₄ , adjacent secondary intersections are searched for and predetermined data is stored. Incidentally, the intersection B ₁₃ and B ₂₁ are the same intersection. In this way, intersections where data should be stored overlap,
Already, to the intersection, when the total distance of the different routes are stored, and compares the total distance Bd ₂₁ of total distance Bd ₁₃ (b) and of (a), rewrites the smaller data Remember. For example, if Bd ₁₃ > Bd ₂₁ ,
The intersection network list of the intersection B ₁₃ (= B _21),
Cumulative shown in (b) Distance Bd ₂₁ and the corresponding one before the sequential number of the intersection A ₂ is finally stored.

In the same manner, an adjacent tertiary intersection C _ij is determined for each secondary intersection B _ij , and a cumulative distance from the departure point via the corresponding previous intersection is determined for each intersection C _ij. Is stored together with the preceding intersection sequential number, and in general, the (i + 1) -th intersection is obtained for each i-th intersection with reference to the intersection netlist, and one (i + 1) -th intersection is determined by one for each (i + 1) -th intersection. Find the total distance from the departure point via the ith intersection,
If the information is stored in the intersection net list of the (i + 1) th intersection along with the intersection sequential number of the previous intersection, the vehicle finally reaches the destination (intersection) DSP. After arriving at the destination, it is stored in the intersection netlist of the (m-1) next intersection and the (m-1) next intersection stored in the intersection netlist of the destination (mth intersection). (M-2) Next intersection,... Primary intersection stored in the intersection net list of the secondary intersection, 0 stored in the intersection net list of the primary intersection
The route connecting the next intersection (departure place) in order from the departure place side to the destination side is the shortest optimal route.

However, even if the vehicle arrives at the destination along a certain route, the route search is continued while a route having a shorter cumulative distance than the cumulative distance stored in the netlist of the intersection at the destination continues. When reaching the destination on another route, if the cumulative distance on the route is shorter than the cumulative distance already stored in the intersection netlist of the destination, rewriting is performed. The route search may be ended when the distance becomes longer than the total distance registered in the intersection netlist. In addition, intersection netlist CR
The NL is prepared in the in-vehicle navigator from the road data not including the intersection net list stored in the CD-ROM as described above and prepared as a part of the road data expanded along with the road list, the intersection net list, and the like. In addition to
As part of the road data (road layer), a CD-
You may make it memorize | store in ROM. Further, even if a route search is advanced from the destination intersection toward the departure intersection, an optimal guidance route can be similarly obtained.

As described above, according to the horizontal search method, an optimal route using the shortest distance as an index in a graph theory is obtained. Therefore, by displaying the optimal guidance route in the map image on the screen together with the vehicle position mark in a specific emphasized color, it is possible to guide the driver to a desired destination.

[0013] In addition to the horizontal search method described above, the Dijkstra method (dijkstra method) is often used for searching for a guide route. Although the intersection netlist is also used in the Dijkstra method, registration of the search order is not required. FIG. 28 is a schematic explanatory diagram of the Dijkstra method. In the Dijkstra method,
While referring to the intersection netlist CRNL, first, the accumulated distance of the intersection netlist relating to the departure point intersection STP is set to zero, and then the departure point intersection STP is set as the first search branch tip node _CPtop, and the adjacent point is located adjacent to the _CPtop . one of the intersection, determine the total distance Ad ₁ from the departure point intersection STP of up to a _1, in the intersection network list according to the intersection a _1, to identify the one before the intersection (current CP _top) intersection Register with the sequential number ((13) and (14) in the intersection netlist). Then registers the intersection sequential number identifying the current adjacent intersection A ₁ in the search branch tip buffer. Similarly, for the other intersections A _{2 to} A ₄ adjacent to the CP _top , the cumulative distances Ad _{1 to} Ad ₄ from the departure point intersection STP are obtained, and the intersection immediately before in the intersection net list of each intersection is determined. The intersection sequential number is registered together with the specified intersection sequential number, and the intersection sequential number is additionally registered in the search branch leading end buffer (see FIG. 28 (1)).

Next, the exchange registered in the search
In the difference, before CP_topExcept for
Search for the one with the smallest cumulative distance,
Tana CP_topAnd In the search branch tip buffer, A₁~
A_FourIs registered, but Ad ₁~ Ad_FourInside, Ad
₁Is the minimum, then CP_top= A₁(Fig. 2
8 (2) solid line). In this case, A₁Neighborhood adjacent to
One of the differences, B₁₁A until₁Departure point ST via
Total distance Bd from P₁₁Ask for. STP-A₁Accumulation between
The distance is already at intersection A₁Distance to the intersection netlist
Release Ad₁Registered as A₁-B₁₁The distance between
A₁Intersection A in the intersection netlist ₁Adjacency from
Difference B₁₁These are registered as distances to
Distance Bd as the sum of₁₁Is found. And the adjacent intersection
Point B₁₁The intersection netlist of CP_top= A₁Identify
With the sequential number of the intersection
B in the tip buffer₁₁Intersection sequential number to identify
Register additional issues. A₁Another intersection B adjacent to₁₂, B
₁₄Similarly, the intersection netlist of each intersection
And the total distance and A₁Register the intersection sequential number of
And B in the search branch tip buffer₁₂, B₁₄At the intersection
Register the sequential number. Note that A₁Adjacent to
Difference B₁₃Is the same as the departure point intersection STP
The departure point intersection STP intersection netlist is already
Is registered as the total distance = 0, and STP-A₁
Cumulative distance Bd that is twice the distance between₁₃Always smaller
From, B₁₃Total distance to intersection netlist for
Do not register the sequential number at the intersection or intersection.
Neither is it registered in the branch tip buffer.

In this manner, when the processing for the current CP _top is completed, the accumulated distance is the smallest among the intersections registered in the search branch tip buffer, except for the ones that have been _designated as CP _top so far. And make this a new CP _top . A _1-
A ₄ , B ₁₁ , B ₁₂ , and B ₁₄ are registered, but A ₁ is excluded. Therefore, Ad _{2 to} Ad ₄ , Bd ₁₁ , Bd ₁₂ ,
If Ad ₂ was the smallest among Bd ₁₄ , CP _top
= A ₂ (see the broken line in FIG. 28 (2)). In this case, A
₂ one of the adjacent intersection adjacent to, the total distance Bd ₂₂ from the departure point intersection STP that has passed through the A ₂ to B ₂₂ obtains. Then, the intersection network list adjacent intersection B _22, registered with CP _top = A ₂ intersection sequential number, the search branch tip buffer registers the intersection sequential number of B _22. Other intersection B ₂₃ adjacent to A ₂
The intersection netlist similarly also registers the total distance and the intersection sequential number of A ₂ from the departure point intersection STP passed through the A _2, and the search branch tip buffer registers the intersection sequential number of the B ₂₃ .

The intersection B ₂₄ is the same as the departure point intersection STP, and the accumulated distance = 0 is already registered in the intersection net list of the departure point intersection STP.
And total distance Bd is twice the distance between STP-A _2
Since it is always smaller than ₂₄ , the cumulative distance and the intersection sequential number of B ₂₄ are not registered in the intersection net list, nor are they registered in the search branch tip buffer.
Further, the adjacent intersection B ₂₁ also, are the same as adjacent intersection B _12, when attempting to register a total distance Bd ₂₁ of STP-A ₂ -B _21, already total distance Bd the intersection network list B _{12 12} are registered. In this case, Bd
₂₁ compares Bd ₁₂ large and small, only towards Bd ₂₁ is small, B ₁₂ accumulated in the intersection network list (= B ₂₁₎ a distance B
Register rewrite the intersection sequential number of the d ₂₁ and B _21, and additionally registers the intersection sequential number of the B ₂₁ in the search branch tip buffer. Wakashi, if the larger of Bd _21, rewriting registration intersection sequential number of Bd ₂₁ and B ₂₁ is not, nor additional registration of B ₂₁ to search branch tip buffer.

Hereinafter, the same processing is repeated, and for the adjacent intersection at a certain CP _top , the cumulative distance and the sequential number of the intersection of the CP _top are registered, and the sequential number of the intersection of the adjacent intersection is additionally registered in the search branch tip buffer. If the adjacent intersection coincides with the destination intersection DSP at the end of the search, the search is terminated. CP _top
Since the one with the minimum total distance is always selected, the destination intersection DSP (mth-order intersection) and the intersection netlist of the destination intersection are stored together with the total distance in the (m-1) th order intersection netlist. , The (m-1) next intersection stored in the (m-1) next intersection net list, the primary intersection stored in the second intersection intersection net list , The 0-th intersection (departure intersection ST) stored in the intersection net list of the primary intersection
P) are sequentially connected in the order from the zeroth-order intersection side to the destination side, which is the shortest optimal path. As described above, according to the Dijkstra method, the shortest route can be searched more accurately than the horizontal search method described above. However, the search speed is slower than the horizontal search method.

[0018]

In both the horizontal search method and the Dijkstra method described above, even if the route search range is limited to a rectangular area having a diagonal line connecting a departure point and a destination. It takes a considerable amount of time to complete the search in order to exhaustively search all the routes included in the area, and it will be prolonged until the driver can receive the route guidance and start driving become.
To solve this problem, a heuristic search method has been proposed. In this search method, the route that deviates from the direction from the starting point to the destination is almost empirically because the optimum route is often within a certain range from the straight line connecting the starting point and the destination. In this case, unnecessary routes are not searched for by lowering the priority by weighting.

However, in order to determine whether or not the direction in which the search branch extends deviates from the direction from the starting point to the destination, the direction of the adjacent intersection viewed from a certain intersection is calculated by a trigonometric function, and If the comparison is made with the direction of the destination seen, the calculation of the trigonometric function takes a long time, and the search time may not be reduced much. In view of the above, it is an object of the present invention to provide a route search method that increases the efficiency of heuristic search and enables a significant reduction in search time.

[0020]

The above object is achieved by the present invention.
And 360 ^° all directions based on a predetermined direction
A means for dividing into sections, from a preset reference section
Means to determine the evaluation value of each section according to the degree of deviation, intersection
For each intersection in the point netlist, the intersection and the adjacent intersection
From the section to which the connecting azimuth belongs,
Register the evaluation value data consisting of the evaluation values of a number of sections
Departure place and destination are input when searching for means and routes
At the time, a section to which the azimuth connecting the departure place and the destination belongs is determined,
Means for determining a deviation between the section and the reference section, the intersection
Each of the evaluation value data registered in the netlist is
The evaluation value that comes first when shifting cyclically
True of the section to which the azimuth connecting the difference point and the adjacent intersection adjacent to the intersection belongs
Means to evaluate the optimal route using the true evaluation value
By providing means for heuristic search
Achieved.

[0021]

According to the present invention, (1) 3 based on a predetermined azimuth
The omnidirectional 60 ⁰ is divided into a plurality of sections, (2) set in advance is
Of each section according to the degree of deviation from the reference section
(3) For each intersection in the intersection netlist,
Predetermined rotation from the section to which the azimuth connecting the difference point and the adjacent intersection belongs
Evaluation in which the evaluation values of the plurality of sections are arranged in order in the direction
Register the value data and (4) find the starting point and
When a destination is entered, the direction connecting the departure point and the destination
To determine the section to be, the deviation between the section and the reference section,
(5) Each rating registered in the intersection netlist
When value data is cyclically shifted by the deviation,
Of the direction connecting the intersection and the adjacent intersection
(6) Using the true evaluation value of the section to which the
Heuristic search for the optimal route. in this way
If so, the direction connecting the starting point and the destination and the direction of the adjacent intersection
Can be obtained more quickly according to the
Route search time
Shortening becomes possible.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Overall Configuration FIG. 1 is an overall configuration diagram of an on-vehicle navigator embodying a route search method according to the present invention. In the figure, reference numeral 11 denotes a CD-ROM (map information storage unit) for storing map data. The map data includes a road layer, a background layer, a character layer, and the like. Reference numeral 12 denotes a display device (CRT) for drawing a map image or a vehicle position mark according to the current position of the vehicle, a guidance route searched for in the optimal route search, and 13 denotes a vehicle based on the traveling distance and direction of the traveling vehicle. Is a vehicle position detection unit that calculates the current position of the vehicle. The vehicle position detection unit 13
Although not shown, it has a direction sensor for detecting the traveling direction of the vehicle, a distance sensor for detecting the traveling distance, and a position calculation CPU for calculating the current position (longitude and latitude) of the vehicle based on the direction and the traveling distance. . An operation unit 14 includes a map search key, an enlargement / reduction key, a map scroll key, a route guidance mode key, and the like. A reference numeral 15 denotes a map display control device, which generates a map image around the current position of the vehicle based on map data. To generate vehicle position marks and guidance route images.

In the map display controller 15, reference numeral 15a reads out map data (for example, map data for nine screens) in a wider area than the screen display area around the current position from the CD-ROM 11 based on the vehicle position data, and temporarily stores the data in the buffer memory. And a map image drawing unit that generates a dot image map image based on the map data. 15b is a buffer memory for temporarily storing map data read from the CD-ROM 11, 15c is a guidance route drawing unit for generating a guidance route image based on guidance route data obtained by the optimal route search processing, and 15d is a map image And a video RAM for storing guidance route images. The map image drawing unit 15a rewrites the video RAM 15d at any time as the vehicle travels so that the display range of the display screen does not exceed the image range of the video RAM 15d,
The guide route drawing unit 15c also generates a guide route image according to the traveling of the vehicle and stores the guide route image in the video RAM 15d. Reference numeral 15e denotes one from the video RAM 15d so that the current position of the vehicle is located at the center of the display screen.
This is a readout control unit that reads out map images for the screen, and the readout position is instructed from the map image drawing unit 15a. 1
Reference numeral 5f denotes a vehicle position mark generating unit for displaying a vehicle position mark at the center of the display screen, which inputs vehicle direction data from the vehicle position detecting unit 13 and generates a vehicle position mark directed in the direction indicated by the data. I do.

Reference numeral 15g denotes an optimum route search unit. When a destination is input after setting the route guidance mode, the departure point and the destination are determined based on the positional relationship between the current vehicle position (departure point) and the destination. Map data of one or a plurality of quadrants adjacent to each other including a rectangular area having a straight line connecting
After creating an intersection netlist while reading it from the ROM 11 to the buffer memory 15b and storing it in the route search memory, the optimal route (shortest route) connecting the departure point and the destination is searched by the horizontal search method. The optimal route search unit 15g performs a heuristic search by pruning a route that deviates from the direction connecting the departure place and the destination based on the relationship between the direction in which the search branch extends and the direction connecting the departure place and the destination. .

Reference numeral 15h denotes a route search memory which is connected to the optimum route search unit and stores data such as an intersection net list, and 15i denotes a guide route storage unit which stores a series of nodes constituting the optimum route as guide route data. It should be noted that the route search memory 15h and the guide route storage unit 15i are configured to retain data by supplying backup power even when the power of the vehicle-mounted navigator is turned off. Reference numeral 15j denotes a synthesizing unit, which is a video RAM 15d, a vehicle position mark generating unit 1
The display device 1 combines the map image, the guide route image, and the vehicle position mark image read out from the respective display devices 5f.
2 and display the image.

The map stored in the map data CD-ROM 11 is divided into plots of appropriate longitude and latitude widths. The map data of each map is divided into four data units corresponding to one screen (each quadrant) (see FIG. 23).
In the map data, roads and the like are represented by a set of coordinates of vertices (nodes) expressed in latitude and longitude, and the drawing is performed by connecting the nodes sequentially with straight lines. A road is formed by connecting two or more nodes, and a portion connecting the two nodes is called a link. Map data includes (1) road list, node table, intersection configuration node list,
It consists of a road layer composed of a list of adjacent intersections, (2) a background layer for displaying roads, buildings, rivers, etc. on the map screen, and (3) a character layer for displaying municipal names, road names, and the like. ing. Of these, the road layer
It has the same data structure as that of FIG. 20, and includes a road list RDLT, an intersection configuration node list CRLT, a node table NDTB, an adjacent node list NNLT, and the like.

[0027] The intersection network list C of the intersection network list optimal route search unit 15g to create
The RNL is configured as shown in FIG. 2, and is an adjacent node among simple nodes in addition to an original intersection (intersection node (regardless of whether or not it is an adjacent node)). (1) Intersection sequential number (information specifying the intersection) (2) Map leaf number including the intersection (3) Data unit code (4) Location on node table (5) ) Longitude (6) Latitude or more, intersection ID (7) Location on intersection configuration node list (8) Number of intersection configuration nodes (9) Location on adjacent node list (10) Number of adjacent nodes (11) Sequential of each adjacent intersection No. (12) Distance from the intersection to each adjacent intersection (13) Direction data of each adjacent intersection viewed from the intersection (14) Intersection sea before one of the intersection Nsharu number (15) total distance from the departure point to the intersection (16) Search degree or the like of the intersection are registered (where (14) -
(16) is registered when the route search is executed).

As shown in FIG. 3, the direction data, which is the item (13) in the intersection net list, has the azimuth of 45 degrees,
North, northeast, east, southeast, south, southwest, divided west, in eight of the northwest, the intersection CP ₀ which is the subject of the intersection network list
, The position of each of the intersections CP _{1 to} CP _n is represented by an 8-bit direction data. The 8-bit direction data corresponds to the MSB, North 2SB, Northeast 3SB, East 4SB, Southeast 5SB, South 6SB, Southwest 7SB, West LSB, North. To 1 and the others to 0. For example, in the case of CP ₁ in FIG.
_When viewed from _{0, the} direction data is (10
000000), and in the case of CP _2, the direction data is (00100000) because it is located east.

FIGS. 4 to 8 are flowcharts showing the processing of the map display control device 15, FIG. 9 is an explanatory diagram of a drawing object for creating an intersection netlist, and FIG. Numeral 11 is an explanatory diagram of pruning, and will be described below with reference to these drawings.

When the power is turned on, the state immediately before the last power-off is restored.
When the power of the current vehicle-mounted navigator is turned on, the map display control device 15 reproduces the map image display state immediately before the power was turned off last time (step 101 in FIG. 4). Here, assuming that the normal guidance mode is reproduced, the map image drawing unit 15a
First, the map data is read out from the CD-ROM 11 into the buffer memory 15b, and the map data relating to a plurality of sheets around the own vehicle position is buffered from the CD-ROM 11 with reference to the map data. The data is read out to the memory 15b, a map image of an area for nine screens including the position of the own vehicle is drawn on the video RAM 15d, and the vehicle position detecting unit 1
Based on the vehicle position data input from Step 3, the readout control unit 15e cuts out a screen image of one screen centering on the vehicle position from the video RAM 15d, and synthesizes the image.
Output to The vehicle position mark generation unit 15f also generates a predetermined vehicle position mark in the direction indicated by the vehicle azimuth data detected by the vehicle position detection unit 13 and outputs it to the synthesizing unit 15j. The synthesizing unit 15j synthesizes the vehicle position mark with the map image, outputs the map to the display device 12, and displays it on the screen. Thereby, a map image around the own vehicle position is displayed on the screen together with the vehicle position mark.

If the power supply was turned off last time, and if the power supply was in the route guidance mode, the power supply was turned on, and the power supply was switched to the route guidance mode. In step 101, in addition to the processing of the map image Based on the vehicle position data, the route drawing unit 15c selects, from the guidance route data stored in the guidance route storage unit 15i, a portion that falls within the area drawn in the video RAM 15d, and
At 5d, the guidance route is drawn with a predetermined color thickly emphasized. As a result, a map image around the own vehicle position is displayed on the screen together with the vehicle position mark and the optimal guidance route connecting the departure point to the intersection.

Preparation for Route Search When the driver wants to travel to a desired destination in the route guidance mode assuming that the normal guidance mode is reproduced when the power is turned on, the driver presses the route guidance mode key on the operation unit 14 to depress the route guidance mode key. Set to the route guidance mode (YES in step 104 of FIG. 4,
105). Next, a map search key or the like is operated, and a map image including the destination is displayed on the display screen by the map image drawing unit 15a. Thereafter, the vehicle position mark is positioned at the destination using the map scroll key, and the destination is set. Is set (step 106).

When the destination is set, the optimum route search unit 1
In step 5g, the current vehicle position is automatically set as the destination (step 107). Based on the map data of the map including the departure point, it is checked whether the departure point is an intersection defined by the road data (an original intersection node or an adjacent node among simple nodes) (step 108). If the intersection is not the intersection, the nearest intersection defined by the road data is set as the departure intersection STP (step 110), and the processing after step 111 is performed. When the departure point intersection STP is determined, the optimal route search unit 15g checks whether the destination is an intersection defined by the road data (an original intersection note or an adjacent node among simple nodes) (step 111). For example, a destination intersection DSP is set (step 112), and the processing after step 201 in FIG. 5 is performed. If it is not an intersection, the nearest intersection is set as the destination intersection DSP (step 113), and the processing after step 201 is performed.

[0034] Departure intersection STP and destination intersection DS
When P is determined, the optimal route search unit 15g checks whether or not the intersection netlist has already been created in the route search memory 15h for all the rectangular areas having the straight line connecting the departure point intersection STP and the destination intersection DSP as a diagonal line ( FIG.
Step 201). If it is not done here, one or a plurality of quadrants including a rectangular area having a straight line connecting the departure point and the destination as a diagonal line are obtained by referring to the map management information, and for all these quadrants, Map data to CD-RO
After reading from M11 to the buffer memory 15b (step 202), the road data is
An intersection netlist is created in the route search memory 15h for each intersection (including an original intersection node and a simple node that is an adjacent node) (step 2).
03 and subsequent processes).

In the creation of the intersection net list, the target quadrant is BU _{11 to} BU _{33 as} shown in FIG.
First, for one BU ₁₁ of the quadrant, the intersection node where the intersection identification flag is set from the node table NDTB of the data unit, and the simple node where the intersection identification flag is set down but the adjacent node identification An adjacent node on which the flag is set is searched, and each intersection netlist in which the intersection sequential number is assigned with a serial number ascending from 1 is prepared on the route search memory, and the intersection ID is registered ((0) to (0 in FIG. 2)). 5) Steps 203 and 20
4). Then, from the node table NDTB, the intersection configuration node list CRDT, and the adjacent node list NNLT,
Obtain the position on the intersection constituent node list, the number of intersection constituent nodes, the position on the adjacent note list, and the number of adjacent nodes,
(6) to (9) of FIG. 2 are registered (step 205).
Next, for each road in the road list RDLT, referring to the node table NDTB, the distance of a link between adjacent intersections (including the case where one is a simple node and an adjacent node) is determined, and one of the links is determined. The other intersection of the link is set as the adjacent intersection in the intersection net list relating to the intersection, and the intersection sequential number (adjacent intersection sequential number) registered in the intersection net list relating to the adjacent intersection and the distance of the link are registered ( (1 in FIG. 2)
0), (11), (13), (14),. Then, the direction of each adjacent intersection viewed from the target intersection in the intersection netlist is calculated, and north, northeast, east, southeast, south, southwest,
It is determined which of the eight sections, west and northwest, is included, and 8-bit direction data in which bits corresponding to the directions of adjacent intersections are set is created and registered in the intersection netlist (FIG. 2). (12), (15), ...).
... Step 206

[0036] The same process, after performing also BU ₁₂ ~BU ₃₃ in FIG. 9 (repetition of steps 207,203～206), finally, the target made of intersection network list, the adjacent node at the intersection node And
If the intersection netlist is created for an adjacent node of a simple node, an intersection netlist may be created for the same adjacent node under another shared unit (neighbor node RN in FIG. 9). ₁ see, RN _2), by referring to the neighbor node list NNLT, locate the intersection sequential number in all other shared units,
The sequential number of the adjacent intersection and the distance (= 0) are registered in the intersection net list, and the adjacent node is connected between the divided maps (step 208).

The route searching this way, every one or more of all the 4 split view leaf comprising a square region of a straight line connecting the origin and destination and diagonal Once ready to intersection network list is complete The optimal route search unit 15g searches for an optimal route connecting the departure place and the destination by a horizontal search method based on a heuristic method using the intersection netlist (see FIG. 10). First, the optimal route search unit 15g determines that the departure point intersection S
Calculate the azimuth (measured clockwise with reference to true north) of the destination intersection DSP based on TP, and calculate the azimuth in the north, northeast, east, southeast, south, southwest, west, northwest One of the two sections is determined, and 8-bit reference direction data in which bits corresponding to the direction of the destination intersection are set is created and registered in the route search memory 15h (step 2).
09, see FIG. 11). In the case of FIG. 11, the departure point intersection ST
Since the direction of the destination intersection DSP viewed from P is northeast, the original reference direction data is (01000000),
In this embodiment, the reference direction data is set to (11100000) in order to set the range of 135 degrees around the northeast. The reference direction data is used for pruning a route deviating from the direction from the departure point to the destination during the horizontal search.

Next, the optimum route searching unit 15g first sets the search order to 0 (step 301 in FIG. 6) and determines whether or not an intersection adjacent to the 0th intersection (the 0th order is the departure point intersection) remains. Intersection netlist CRN of ground intersection STP
Check with reference to L (step 302). In step 302, the j-th intersection (j = 0, 1,
・ ・, I) is excluded. If adjacent intersections remain, of which, one adjacent intersection, for example A _4,
Neighboring intersection A in the intersection netlist related to the 0th intersection
₄ orientation indicated by the direction data (orientation of the 0th adjacent intersection A ₄ as seen from the primary intersection) is either matches the orientation (north-east) indicated by the reference direction data determined Te step 209 of FIG. 5,
Alternatively, it is checked whether the direction is the azimuth (north or east) of the next section (step 303, see FIG. 11). The direction data corresponds to (1) in FIG. 2 of the intersection netlist of the 0th intersection.
2), it is registered in association with the adjacent intersection A ₄ (15) or the like.

When the answer to the determination of step 303 is YES, the optimum route searching unit 15g starts from the 0th intersection to the adjacent intersection A _4.
In particular, the path to is not deviated from the destination direction, so that pruning is not performed. If NO, the path is deviated from the destination direction and pruned (see FIG. 10). Direction data of adjacent intersection A ₄ is (00001000), the zeroth-order crossroads (= ST
The orientation of the adjacent intersection A ₄ as seen from P) is the south. Therefore, the azimuth = northeast indicated by the reference direction data is separated from the azimuth = northeast by 2 or more in units of division, and the result in step 303 is NO. Actually, the reference direction data indicating the northeast is (110000)
Since a 0), of the direction data of the adjacent intersection A _4, also 1, so the bit positions in the reference direction data 1
All you have to do is compare them. If NO in step 303, the path of the adjacent intersection A ₄ 0th order intersection is pruning, the process returns to step 302, or other adjacent intersections to the zeroth order intersection remains 0th order intersection intersection network Look up the list. Since adjacent intersection A ₁ remains, subsequently, the orientation indicated by the direction data of the adjacent intersection A ₁ intersection network list according to the zeroth-order crossroads is either matches the orientation indicated by the reference direction data, or, next to the It is checked whether it is the direction of the section (step 303, FIG.
1). Now, the direction data of the adjacent intersection A ₁ is (0
0100000), indicating east,
YES and in step 303, in this case, subsequently, total distance D from the departure point intersection STP to the adjacent intersection A ₁
Is calculated (step 304). D is the departure intersection ST
Assuming that the cumulative distance from P to the ith intersection is d ₁ and the distance from the ith intersection to the adjacent intersection A ₁ is d ₂ ,
It is obtained by the following equation d ₁ + d ₂ → D. When i = 0 at _first, d ₁ = 0, so that D
= A d _2. d ₂ is in Figure 2 of the intersection network list CRNL according to departure point intersection STP (11), (1
4) and so on.

[0040] Next, with reference to the adjacent intersection (31) in FIG. 2 in the intersection network list CRNL according to A ₁ after entry, already per intersection A _1, the search order is (i +
1), in other words, it is checked whether the accumulated distance along a different route and the information (intersection sequential number) for specifying the previous intersection are already registered (step 305). , The adjacent intersection A
In (31) - (33) in FIG. 2 in the intersection network list CRNL according to _1, the following data (A) is currently focused 0th order intersection STP intersection sequential number of, (B) from the zero-order intersection STP the total distance to adjacent intersection a _1, (C) as the search order of the adjacent intersection a ₁ (i +
1) = 1 is registered (step 306), and thereafter, returning to step 302, referring to the intersection netlist CRNL for the departure intersection STP, the intersection adjacent to the 0th intersection of interest is still detected. A check is made as to whether or not it remains, and if it does, the same processing is repeated.

[0041] Since the adjacent intersection A ₂ remains compares the orientation indicated by the direction data of the adjacent intersection A _2, the orientation indicated by the reference direction data (step 303). Adjacent intersection A ₂
Indicates the north at (10000000), and similarly to A ₁ , YES is determined in step 303 and the total distance D from the departure point intersection STP to the adjacent intersection A _{2 is obtained.}
Look, in the intersection network list of the adjacent intersection A _2, the first 0
Intersection sequential number of next intersection and search order (i +
Register with (1) (steps 304-306). Then, the process returns to step 302. While still adjacent intersection A ₃ remains, the direction data of the adjacent intersection A ₃ is (00000
010) indicates west and is separated from the northeast indicated by the reference direction data by two or more sections.
In this case, pruning is performed without proceeding to step 304 and subsequent steps. As a result, the intersection netlist departure point intersection STP are present adjacent intersection A ₁ to A _4, of these, only A ₁ and A ₂ are the primary intersection to the corresponding intersection network list CRNL, Intersection sequential number of the previous intersection (here, departure point intersection STP), cumulative distances Ad ₁ and Ad ₂ from the departure point, search degree 1
Is registered. On the other hand, these data are in the intersection network list A ₄ and A ₃ are not registered.

When the processing is completed for all adjacent intersections included in the intersection netlist CRNL for the departure intersection STP, the optimum route searching unit 15g determines whether there is a zeroth intersection other than the departure intersection STP. (Step 307) Since there is no destination, the destination intersection D
It is determined whether the destination intersection DSP has been included in the arrival at the SP, in other words, the (i + 1) -th intersection, and if it is not included, i is incremented to 1 (step 308). Step 309). Then, one of which is the first intersection, for example by focusing on A _1,
Referring to intersection network list according to the intersection A ₁ stored in the route search memory 15i, a zero-order crossroads so far, adjacent intersection is determined whether there outside intersection is a first intersection (step 302 ). here,
Since B _11, B _12, B ₁₃ is present, one of which, for example per adjacent intersection B _11, with reference to the intersection network list of intersection A _1, obtains the azimuth indicated by the direction data of the adjacent intersection B ₁₁ Then, the orientation is compared with the orientation indicated by the reference direction data to determine whether the orientation is the same or adjacent (step 303).
Here NO and in step 303 since the orientation is south indicated by the direction data of the adjacent intersection B ₁₁ is not proceed after step 304, path extending from the intersection A ₁ to B ₁₁ is pruning, the process returns to step 302, intersection A Check if there is another intersection adjacent to ₁ .

[0043] Since the adjacent intersection B ₁₂ is present, determine the orientation indicated by the direction data of the adjacent intersection B _12, as compared with the orientation indicated by the reference direction data and compares whether the same or adjacent orientation (step 303). Now YES at step 303 since the orientation is east indicated by the direction data of the adjacent intersection B ₁₂
Next, with reference to the intersection network list according to the intersection A _1, departure point intersection STP, first intersection are currently paying attention A _1, the cumulative in the path of the adjacent intersection B ₁₂ distance D
Is calculated (step 304). Total distance A from the departure point intersection STP to the first intersection A ₁ that is currently paying attention
distance d ₂ d ₁ from the first intersection A ₁ to the adjacent intersection B ₁₂ is an item (32) in FIG. 2 in the intersection network list CRNL intersection A _1, (11), (14),
(17) from being stored as one of such, Ad ₁ + d ₂ → D by obtained is total distance D from the departure point intersection STP to the adjacent intersection B _12. Then, with reference to the intersection network list CRNL according to adjacent intersection B _12, in FIG. 2 (3
Search orders are registered in the item 3) is (i + 1) = 2 or checked (step 305), where since NO is in association with the adjacent intersection B _12, which focuses the following data (A) Current first intersection a ₁ intersection sequential number, (B) total distance from the departure point to the adjacent intersection B _12, (C) the adjacent intersection B ₁₂ searches the order of the (i +
1) = 2 the intersection B ₁₂ of intersection network list CRNL in FIG. 2 (32) Register to (34) (step 306), returns to step 302 side, next adjacent of the first intersection A ₁ If there is an intersection, the same processing is performed.

[0044] Incidentally, there are adjacent intersection B _13, adjacent intersection B as seen from the first intersection A ₁ of interest currently
The orientation of the ₁₃ examined from the direction data of the adjacent intersection B _13, partitioned or not to check which are the same or adjacent and orientation indicated by the reference direction data (step 303). Because the YES here, the case of B ₁₂ and in the same manner to obtain the total distance D from the departure point intersection STP to the adjacent intersection B _13, the intersection netlist of adjacent intersection B _13, the first intersection A ₁ It is registered together with the intersection sequential number and the search order (i + 1) (steps 304 to 306). Then, the process returns to step 302. As a result, adjacent intersections B _{11 to} B _{13 in} addition to the departure intersection are included in the intersection net list of the primary intersection A _1.
There is present, these, only B ₁₂ and B ₁₃ are the primary intersection to the corresponding intersection network list CRNL, 1
The intersection sequential number of the previous intersection, the cumulative distances Bd ₁₂ and Bd ₁₃ from the departure place, and the search order 2 are registered. On the other hand, the intersection network list B _11, these data are not registered.

Intersection A₁Intersection net squirrel for
The processing is completed for all adjacent intersections included in CRNL.
In other words, the optimal route searching unit 15 g₁Other than
It is determined whether a primary intersection exists (step 307).
Here is intersection A_TwoExists, so a new primary intersection
(Step 310), the processing after step 302
repeat. Intersection A_TwoAt the intersection netlist
Intersection B other than the departure place_{twenty one}~ B_{twenty three}Exists. this
Of which B_{twenty one}And B_{twenty two}Is not pruned in step 303,
B_{twenty three}Is at least two sections away from the target locality indicated by the reference direction data
It is pruned. B_{twenty one}In the processing of
₁₃= B_{twenty one}Therefore, in step 305, the adjacent intersection
B₁₃Is YES because the order of has already been 2. This
This is the first intersection A₁Intersection B adjacent to₁₃age
Processed (the data of (A) to (C) is stored)
In this case, the adjacent intersection B₁₃In charge of
Departure land exchange registered in the intersection netlist CRNL
Cumulative distance D '= Bd from difference point STP ₁₃And this time
The magnitude of the distance D obtained in step 304 is compared (step 3
11).

If D <D ', the adjacent intersection B
₁₃(= B₁₂2) of the intersection netlist CRNL
I-th order stored in the items (31) and (32)
Intersection A ₁Focus on the intersection sequential number of
I-th intersection A_TwoAt the intersection sequential number
At the same time, the accumulated distance D ′ is changed to D = Bd._{twenty one}Rewrite with
(Step 312), and thereafter, returns to Step 302.
If D ≧ D ′, the intersection B₁₃(= B_{twenty one})
(3) in FIG. 2 of the intersection netlist CRNL.
Return to step 302 without changing the contents of 1) and (32)
You. Intersection A_TwoWith respect to each adjacent intersection
When the process is completed and another primary intersection no longer exists,
Check whether the target intersection DSP has been reached (Step 3
07, 308), because it is not yet, increment i
2 (step 309). Next, step 302
Then, the same processing as described above is sequentially repeated.

As described above, when the direction of the route from the i-th intersection to a certain adjacent intersection is deviated by two or more sections from the direction of the destination as viewed from the departure point, the route is pruned and further advanced. Since the search does not proceed, unnecessary route search that deviates from the destination direction is omitted.

Thereafter, in the check in step 308, if the destination intersection DSP is included in all the intersections determined to be the (i + 1) -th order, and the judgment is YES, the destination intersection DSP, the destination The (m-1) -th intersection and the (m-1) -th intersection stored as the item (31) in FIG. 2 of the intersection netlist CRNL relating to the intersection DSP (assuming the m-th intersection). (M-2) next intersection stored in the intersection netlist CRNL,
... Intersection netlist CRN related to secondary intersection
The primary intersection stored in L and the zeroth intersection stored in the intersection netlist CRNL relating to the primary intersection = starting point intersection STP are sequentially connected in the order from the departure point side to the destination side to be the shortest. The route is determined (step 313).
Then, the node sequence forming the shortest route is stored as guide route data in the guide route storage unit 15i, and the route search process ends (step 314).

[0049] Even if the vehicle arrives at the destination DSP by a certain route, the route search is performed while a route having a shorter cumulative distance than the cumulative distance registered in the intersection netlist relating to the destination intersection exists. Continue, youth, destination D on other routes
When reaching the SP, if the cumulative distance on the route is shorter than the cumulative distance already registered in the intersection netlist relating to the destination intersection, rewriting is performed, and then any route is related to the destination intersection. The route search may be terminated when the distance becomes longer than the total distance registered in the intersection netlist. Further, since the guide route data is composed of only intersections (including those which are adjacent nodes among simple nodes), the intersection netlist, the road list RDLT in the road data, the node table ND
Simple route interpolation may be performed between intersections with reference to TB or the like to obtain detailed route guidance data, and then stored in the guidance route storage unit 15i. Further, a route search may be performed from the destination side to the departure place side, and the searched node sequence may be arranged in reverse order to be used as guide route data.

The route guidance Subsequently, the map display control unit 15 performs display processing of the route guidance image to the destination to the driver (step 401 of FIG. 7). That is, the map image drawing unit 15a inputs the vehicle position data from the vehicle position detection unit 13, refers to the map management information, and stores the map data relating to the nine quadrants around the vehicle position from the CD-ROM 11 into the buffer. In addition to reading the map image into the memory 15b, a map image for nine screens is drawn in the video RAM 15d. Further, the guidance route drawing unit 15c stores, based on the display area information of the video RAM 15d input from the map image drawing unit 15a, from the guidance route data stored in the guidance route storage unit 15i, the area drawn on the video RAM 15d. Select the part to enter and select the video RAM1
At 5d, the guidance route is drawn with a predetermined color thickly emphasized. Then, as the vehicle position changes as the vehicle travels, the map image drawing unit 15a reads and controls an image for one screen centering on the vehicle position from the video RAM 15d based on the vehicle position data input from the vehicle position detection unit 13. The part 15e is made to read.

If the vehicle position changes, the video RAM 15
If the range of one screen centering on the own vehicle position is likely to deviate from the map image drawn on d, the map image drawing unit 15a stores new map data around the own vehicle position from the CD-ROM 11 into the buffer memory 15b. While reading, map images for nine screens around the own vehicle position are drawn on the video RAM 15d. On the other hand, based on the vehicle direction data input from the vehicle position detection unit 13, the vehicle position mark generation unit 15f generates a vehicle position mark directed in the direction indicated by the data. The combining unit 15j combines the vehicle position mark with the center of the image read from the read control unit 15e, and outputs the map image around the vehicle position accompanied by the display of the enhanced guidance route to the display device 12 with the vehicle position mark. Is displayed. The driver can easily reach the destination by traveling along the guidance route displayed on the screen.

Thereafter, when the vehicle reaches the destination, the map display control device 15 cancels the route guidance mode and switches to the normal guidance mode (steps 402 and 403). In the normal guidance mode, normal map image display processing is performed.
Specifically, the map image drawing unit 15 a
On the basis of the vehicle position data input from Step 3, map data of the upper layer map including the vehicle position is read from the CD-ROM 11 to the buffer memory 15b and drawn on the video RAM 15d. Under the read control of the map image drawing unit 15a,
The read control unit 15e cuts out a map image of one screen centering on the vehicle position from the map images drawn on the video RAM 15d, and outputs the map image to the synthesizing unit 15j. The vehicle position mark generation unit 15f also generates a predetermined vehicle position mark and outputs it to the synthesizing unit 15j. The combining unit 15j combines the vehicle position mark with the map image that does not include the enhanced guidance route, outputs the combined map image to the display device 12, and displays the normal map image together with the vehicle position mark on the screen (step 103 in FIG. 4).

If the user wants to change the destination while traveling in the route guidance mode, he presses the route guidance mode key of the operation unit 14 (YES in step 404 in FIG. 7). Then, the map display control device 15 releases the route guidance mode and switches to the normal guidance mode (step 403).
The route guidance mode key is pressed again to set the route guidance mode (steps 104 and 105 in FIG. 4), and the destination is input (step 106). Map display control device 1
The optimum route search unit 15g of the fifth example sets the current vehicle location as the departure point, and determines the departure point intersection STP and the destination intersection DSP (steps 108, 109 or 108, 110, steps 111, 112 or 111, 113). Next, the optimal route search unit 15g stores the intersection netlist for all the rectangular areas whose diagonal is a straight line connecting the departure point intersection STP and the destination intersection DSP as the route search memory 15h.
(Step 201 in FIG. 5).
If not already created, the processing after step 202 is performed to create an intersection netlist in the same manner as described above. Conversely, if it has already been created, the process immediately proceeds to step 209 and departs. The direction of the destination viewed from the ground is obtained, the reference direction data is recreated, and the route search memory 1 is obtained.
After the registration at 5h, the route search processing of FIG. 6 is executed.
This eliminates the need to create a new intersection netlist, and saves the trouble of calculating the azimuth of an adjacent intersection viewed from the target intersection of the intersection netlist and registering the direction data. Search for the optimal route can be completed.

When the power of the on-vehicle navigator is turned off for a break or the like, the route search memory 15h and the guide route storage unit 15i are backed up by a battery so that data is retained even when the power is off. When the power is turned on, a state immediately before the power is turned off can be reproduced, and then a predetermined power-off process is performed (FIG. 8).
Steps 501 and 502).

According to this embodiment, the direction data indicating the azimuth of the adjacent intersection as viewed from the target intersection in eight sections is registered in the intersection net list in advance, and the starting point and the destination are set before starting the route search. The reference direction connecting the ground is obtained, and during the route search, for the route connecting the ith intersection to a certain adjacent intersection, the direction data and the reference direction registered corresponding to the adjacent intersections in the intersection netlist of the ith intersection are obtained. In comparison, if the segment is completely coincident or adjacent, the pruning is not performed, and in other cases, the optimum route is pruned and the optimal route is heuristically searched. Since the azimuth can be easily obtained without calculating using a trigonometric function, the time required for the search can be greatly reduced. In particular, if you can use the same intersection netlist created in the previous route search,
The effect of time reduction is great.

In the above-described embodiment, the intersection netlist is created on the map display controller 15 from the map data stored in the CD-ROM 11, but the intersection netlist is created in advance as shown in FIG. When the route search is started, the optimum route search unit 15 is included in the map data.
g stores the intersection netlist of the required area in C-ROM
11 and may be stored in the route search memory 15h via the buffer memory 15b. In addition, pruning is performed except when the azimuth of the adjacent intersection viewed from the i-th intersection is completely the same as the azimuth of the destination or is in the adjacent section. For example, as shown in FIG. A range separated by two sections (when the destination is northeast, southeast and northwest), the distance between the ith intersection and the adjacent intersection is multiplied by k times (k is a predetermined weighting factor larger than 1), and weighted to give priority. The degree may be reduced, and pruning may be performed only when three or more sections are apart.

In this case, the processing can be simplified by the following.
become. That is, in step 209 of FIG.
Finds reference direction data indicating the direction of the destination viewed from the departure point
Instead of depending on the direction of the destination from the point of departure,
Eight divisions: north, northeast, east, southeast, south, southwest, west, northwest
Search, pruning, or
Direction-specific evaluation value data (DA₁, D
A_Two, DA_Three, ..., DA₈As a combination)
I do. DA of evaluation value data by direction₁~ DA₈Is shown in FIG.
Eight, north, northeast, east, southeast, south, southwest, west, northwest
, And DA_iIs completely
The same or adjacent section where normal search is performed is 1
Pruning in the direction opposite to the direction of the destination.
Is set to 0, and the distance should be weighted.
Is an evaluation value used as the weight coefficient value. View from departure point
If the direction of the destination entered falls in the northeast section, the evaluation value for each direction
As shown in FIG.₁= 1, DA_Two= 1, DA
_Three= 1, DA_Four= K, DA_Five= 0, DA ₆= 0, DA₇= 0, DA
₈= K (where k> 1).

Then, steps 302 to 304 in FIG. 6 are modified as shown in FIG. 13, and when a certain adjacent intersection remains for the i-th intersection (YE in step 601 in FIG. 13).
S), reading the direction data of the adjacent intersection from the intersection net list of the i-th intersection, and if the value of the direction-specific evaluation value data DA corresponding to the azimuth indicated by the direction data is not 0, the data up to the i-th intersection is obtained. The cumulative distance is added DA times the distance from the i-th intersection to the adjacent intersection to obtain a total distance D from the departure point to the adjacent intersection (step 60).
2, 603), if the value of the direction-specific evaluation value data corresponding to the direction data of the adjacent intersection is 0, the process returns to step 601 to perform pruning.

Also, in contrast to this, in each intersection netlist, a difference coordinate system in the longitude direction and the latitude direction with the target intersection of the intersection netlist as the origin is provided as direction data indicating the azimuth at which each adjacent intersection is viewed from the target intersection. In step 209 of FIG. 5, the existence quadrant of the destination in the difference coordinate system having the origin as the origin is obtained in step 209 of FIG. You may do so.

Now, the difference coordinates will be described. The road represented by the road data is shown in FIG.
(1) has become as, for example, position coordinates (longitude, latitude) of intersection CP ₁ that is the creation target intersection netlist (x _1, y _1), each adjacent intersections CP for the intersection CP ₁ position coordinates of the ₂ ~ CP ₄ (longitude, latitude)
Is (x ₂ , y ₂ ) to (x ₄ , y ₄ ), the difference coordinates (X ₂ , Y ₂ ) of each adjacent intersection CP _{2 to} CP ₄ are
(X ₄ , Y ₄ ) is (X ₂ , Y ₂ ) = (x ₂ −x ₁ , y ₂ −y ₁ ) (X ₃ , Y ₄ ) as shown in the differential coordinate system XY of FIG. Y ₃ ) = (x ₃ −x ₁ , y ₃ −y ₁ ) (X ₄ , Y ₄ ) = (x ₄ −x ₁ , y ₄ −y ₁ ). That is, the difference coordinates between adjacent intersections CP ₂ ~ CP _4, each adjacent intersection viewed from the intersection CP ₁ of interest C
These are position coordinates of the relative longitude direction (X direction) and latitude direction (Y direction) of P _{2 to} CP ₄ . If the target intersection of the intersection network list is CP _4, each adjacent intersection CP _1, CP
₅ , the difference coordinates (X ₁ , Y ₁ ) of CP ₆ , (X ₅ ,
Y ₅ ) and (X ₆ , Y ₆ ) are (X ₁ , Y ₁ ) = (x ₁ −x ₄ , y ₁ −y ₄ ) (X ₅ , Y ₅ ) = (x ₅ −x ₄ , y ₅₋ y ₄ ) (X ₆ , Y ₆ ) = (x ₆ −x ₄ , y ₆ −y ₄ ). The same applies to an intersection netlist for other intersections.

Also, the existence quadrant of the destination intersection DSP is
Specifically, the longitude and latitude coordinates of the departure intersection are
(X_STP, Y_STP), The longitude and latitude coordinates of the destination intersection
(X_DSP, Y _DSP), The elephant in the differential coordinate system XY
As shown in FIG. 15, X> 0 and Y ≧ 0... First quadrant X ≦ 0 and Y> 0... Second quadrant X <0 and Y ≦ 0. ≧ 0 and Y> 0 Since the fourth quadrant, x_DSP-X_STP> 0 and y_DSP-Y_STP≧ 0 ・ ・ ・ First quadrant x_DSP-X_STP≦ 0 and y_DSP-Y_STP> 0 ... 2nd quadrant x_DSP-X_STP<0 and y_DSP-Y_STP≦ 0 ・ ・ ・ 3rd quadrant x_DSP-X_STP≧ 0 and y_DSP-Y_STP<0: The quadrant of the destination intersection DSP is obtained as the fourth quadrant. The place of FIG.
In this case, it becomes the first quadrant.

Then, in step 303 of FIG.
The ith intersection of interest and one adjacent intersection of the intersection
The point is registered in the intersection netlist of the ith intersection.
From the difference coordinates of the adjacent intersection obtained, the i-th intersection
Calculate the existence quadrant of the adjacent intersection as seen in the difference coordinate system
And the existence quadrant of the destination obtained in step 209 of FIG.
Determine whether they match, and if they match, go to step 304
The following processing is performed.
Pruning is performed (see FIG. 16A). For example, focusing on
Intersection is departure intersection, adjacent intersection is A₁That
For example, the intersection seen in the difference coordinate system with the origin intersection as the origin
A₁Is the fourth quadrant, and is the destination quadrant
Since it is different from a certain first quadrant, pruning is performed and the adjacent intersection is A
_TwoIf so, step 304 is performed without pruning because they match.
Continue to the total distance, the next intersection sequencer
Register the file number and search order. A_ThreeAnd A_FourIs the existence quadrant
Does not match the destination, so returning to step 302
Pruning.

When using the difference coordinates, FIG.
As shown in 6 (2), as viewed from the ith intersection of interest
The existence quadrant of an adjacent intersection is the destination quadrant (here,
Pruning only in the fourth quadrant on the opposite side
Pruning in the adjacent second or fourth quadrant
No, however, in step 304 of FIG.
When calculating the total distance, the ith intersection of interest and the
Distance d between adjacent intersections _TwoAbout the intersection netlist
Is multiplied by k (k> 1) and weighted
And, from the departure point to the intersection netlist of the i-th intersection
Add to the value registered as the total distance to the i-th intersection
To connect the i-th intersection with the adjacent intersection
The priority of the road can be reduced.

The search may be performed by the Dijkstra method instead of the horizontal search method. 17 and 18 are flowcharts when the optimum route search unit 15g searches for an optimum route by the Dijkstra method, FIG. 19 is an explanatory diagram of an intersection netlist used in the Dijkstra method, FIG. 20 is an explanatory diagram of evaluation value data, and FIG.
Reference numeral 1 is a schematic explanatory diagram of the Dijkstra method, which will be described below with reference to these drawings. 17 and 18 correspond to the steps from step 201 in FIG. 5 to D in FIG. It is assumed that the intersection netlist is included in the map data of the DC-ROM 11. It is assumed that the intersection netlist used here has evaluation value data registered in place of the direction data in FIG. 2 as shown in FIG. It is assumed that the item of the search order is omitted.

[0065] evaluation value data is composed of a combination of the eight data of DB ₁ ~DB _8, the direction, north, northeast, east, southeast, south, southwest, west, as represented by the northwest of the eight categories, DB ₁ is evaluation value data for the azimuth of an adjacent intersection viewed from the target intersection in the intersection netlist. Here, all the evaluation value data are weight coefficient values for the distance,
The value is any one of 1, 3, and 5. CD-RO
In the initial state stored in M11, DB ₁ assumes that the azimuth of the temporary destination viewed from the temporary departure point is north (reference section)
The adjacent intersection viewed from the target intersection in the intersection netlist
This is an evaluation value for the section to which the point orientation belongs. DB ₂ is adjacent
To the direction of the next section clockwise from the intersection direction
An evaluation value, hereinafter, similarly, DB ₃ to DB ₈ also clockwise
This is an evaluation value for the azimuth of an adjacent section one by one. Specifically, when the azimuth of the adjacent intersection is east as shown in FIG. 20 (1), DB ₁ = 3, DB ₂ = 5, DB _{2 in} the initial state.
_{3 = 5, DB 4 = 5} , DB 5 = 3, DB 6 = 1, DB 7 = 1, DB
₈ = 1.

At the time of route search, the departure point intersection S
The actual direction of the destination intersection DSP viewed from TP is north (base
By the number of segments shifted from quasi-section) in the clockwise direction, DB ₁ ~DB ₈
Is cyclically shifted clockwise. For example,
If the direction of the destination viewed from the actual departure point is east,
In the case of the adjacent intersection direction of 0 (1), as shown in FIG. 20 (2), DB ₁ = 1, DB ₂ = 1, DB ₃ = 3, DB ₄ = 5, DB ₅
= 5, DB ₆ = 5, DB ₇ = 3, DB ₈ = 1.

In the Dijkstra method, the departure point intersection STP
When the destination intersection DSP is determined, the optimal route search unit 15
g, first, it is checked whether or not the intersection netlist has already been read into the route search memory 15h for all the rectangular areas having the diagonal line connecting the departure point intersection and the destination intersection (step 701 in FIG. 17). If so, referring to the map management information, read out all the intersection netlists in the minimum one or a plurality of quadrants including the entire rectangular area from the CD-ROM 11 and read the buffer memory 15b.
(Step 702). Then, after determining whether the actual direction of the destination intersection DSP as viewed from the departure point intersection STP is any of the eight sections, it is registered in the route search memory 15h (step 7).
03, in the case of FIG. 21, the destination location is northeast), and the evaluation value data stored for each adjacent intersection of each intersection netlist by the number of sections in which the destination location deviates clockwise from north.
Shifting the DB ₁ to DB ₈ clockwise (step 70
4). As a result, the evaluation value data of each intersection netlist
DB ₁ is an evaluation value according to the azimuth of the adjacent intersection viewed from the target intersection of the intersection net list for the destination locality.

Thereafter, the optimum route search unit 15g sets the departure point intersection STP as the first search branch tip node CP _top, and sets the total distance of the intersection net list relating to the departure point intersection STP to zero (FIG. 18). Step 80 of
1, FIG. 21 (1)). And CP _top (= ST
With reference to the intersection netlist of P), it is checked whether an adjacent intersection adjacent to the CP _top remains (step 80).
2). Here, since A _{1 to} A ₄ remain,
One, for A _4, of the evaluation value data registered in correspondence with the adjacent intersection A ₄ in the intersection network list for the current CP _top, evaluation corresponding to the orientation of the adjacent intersection A ₄ as seen from the CP _top (south) The leading DB ₁ = 5 indicating the value is set as the read weight coefficient k (step 803).

From the departure point intersection STP to CP _top
The total distance d ₁ to, from the distance d ₂ Metropolitan from the CP _top to the adjacent intersection A _4, adjacent intersection A ₄ from the departure point intersection
Determining the cumulative distance Ad ₄ to at following equation _{d 1 + kd 2 = Ad 4} ( step 804). d ₁ is registered as a total distance in the intersection network list of CP _top, but here is zero. Further, d ₂ is registered as the distance to the adjacent intersection A ₄ in the intersection network list CP _top. Then, already total distance D'in the intersection network list for the intersection A ₄ it is checked whether the registered (step 805), so still, in the intersection network list of the adjacent intersection A _4, 1
Register with the intersection sequential number specifying the previous intersection (current CP _top ) (step 806,
(31) and (32) in the intersection netlist in FIG.
Then registers the intersection sequential number identifying the search branch tip buffer of this adjacent intersection A ₄ provided in the route search memory 15h (step 807). Then
Is the destination reached, in other words, the current intersection A ₄
Check if the destination matches the destination intersection (step 8)
08).

Since it is not yet, the process returns to step 802 to check whether another adjacent intersection adjacent to the current CP _top remains.
Since A ₁ , A ₃ and A ₄ remain, one of them, A ₁
, The evaluation value data DB ₁ corresponding to the azimuth (east) is read and set to k (step 803, k = 1), and the total distance Ad ₁ from the departure point intersection STP is calculated by the following equation d ₁ + kd ₂ = Ad ₁ Since the accumulated distance has not yet been registered in the intersection net list of A ₁ , it is registered in the intersection net list relating to the intersection A ₁ together with the intersection sequential number specifying CP _top (the immediately _preceding intersection). additionally registers the intersection sequential number of a ₁ in the search branch tip buffer (step 804-807). Then, it is checked whether or not the destination has been reached (step 808). Adjacent intersection A ₂
On total distance Ad ₂ calculated as k = 1 from the evaluation value data DB ₁ corresponding to the azimuth (the north) (step 803,
804), total as k = 5 the evaluation value data DB ₁ corresponding to the azimuth (west) for adjacent intersection A ₃ distance Ad ₂
The calculated, respectively, after registering with the intersection sequential number that identifies the CP _top the intersection netlist corresponding to A ₂ and A ₃ (Step 805 and 806), A
The intersection sequential number of ₂ and A ₃ is additionally registered to the search branch tip buffer (step 807).

In this way, the current CP_top= STP
If there are no remaining adjacent intersections for
In the intersection registered in the branch tip buffer,
To CP_topExcept for the ones that
Search for what is_topAnd (S
Steps 802 and 809, see the solid line in FIG. 21 (2)). Search
The branch tip buffer contains A₁~ A_FourIs registered
But Ad_ThreeAnd Ad_FourIs large and excellent because k = 5
Advertisement decreases, Ad₁And Ad_TwoIs the value of k = 1
Is small and the priority rises. Of these, Ad₁Is the smallest
If so, CP _top= A₁Becomes In this case, A₁of
A by referring to the intersection netlist₁Adjacent intersection adjacent to
One of the points, B₁₄About the updated CP_topNext to
Intersection B₁₄Value data DB for the azimuth (south) that saw
₁Is set to k = 5 based on (steps 802 and 803).
And from the departure intersection STP to CP_topCumulative distance up to
Separation d₁And CP_topIntersection B from₁₄Distance d to_Two
From, from the departure intersection to the adjacent intersection B₁₄Cumulative distance up to
Separation Bd₁₄To the following equation d₁+ Kd_Two= Bd₁₄ (Step 804). d₁Is CP_topIntersection
Registered as a cumulative distance in the netlist, d_TwoIs
CP_topIntersection B in the intersection netlist of₁₄For up to
It is registered as a distance.

[0072] and, already total distance D'in the intersection network list for the intersection B ₁₄ it is checked whether the registered (step 80
5), so still, in the intersection network list according to B _14, the immediately preceding intersection (registering with the current CP _top = intersection sequential number identifying the A ₁₎ (step 806, the intersection network list in FIG. 19 (31) and (32)). Then registers the intersection sequential number identifying the current adjacent intersection B ₁₄ on the search branch tip buffer (step 807). Next, it is checked whether the destination has been reached (step 808).
Returning to 02, it is checked whether other adjacent intersections adjacent to the current CP _top remain, and B _{11 to} B ₁₃ still remain.
For one B ₁₁ of this, after setting the k = 1 reads an evaluation value DB ₁ for orientation as seen from the CP _top (E) (step 803), via an intersection A ₁ from the departure point intersection STP Total distance B up to intersection B ₁₁
seeking d _11, since the intersection network list B ₁₁ are not yet registered total distance, in the intersection network list according to the intersection B _11, register with the intersection sequential number of CP _top, B the search branch tip buffer Additional registration of ₁₁ intersection sequential numbers (steps 804 to 8)
07). Then, it is checked whether or not the destination has been reached (step 808). Similarly, for the adjacent intersection B ₁₂ , the accumulated distance Bd ₁₂ is calculated as k = 1 from the evaluation value corresponding to the azimuth (north), and C is added to the intersection netlist corresponding to B _12.
After registering with the intersection sequential number identifying the P _top (step 803 and 806), and additionally registers the intersection sequential number in the search branch tip buffer to identify the B ₁₂ (Step 807).

[0073] direction for the adjacent intersection B ₁₃ is located in the west
Since DB ₁ = 5, the total distance D = k = 5
Seek Bd _13. STP-A between ₂ total distance d ₁ has already been registered in the intersection network list intersection A ₁ as total distance Ad ₁ (the intersection network list in FIG. 19 (3
2)), the distance d ₂ between A ₁ -B ₁₃ is registered as the distance from the intersection A ₁ in the intersection network list until adjacent intersection B _13, total distance Bd as _{d 1 + kd 2 = Bd 13} 13 Is found. However, the intersection B ₁₃ is,
Same as the departure point intersection STP, and the departure point intersection ST
Although already total distance = zero in the intersection network list P is YES in step 805 are registered, since always smaller than Bd _13, a NO at step 810. At this time, not total distance to intersection network list, the registration of intersection sequential number with respect to B _13, also not be registered in the search branch tip buffer. As a result, the total distance and the like of the intersection netlist at the intersection that has been designated as CP _top cannot be rewritten.

When the processing for the current CP _top = A ₁ is completed (NO in step 802),
Among the intersections registered in the search branch tip buffer, the one with the smallest cumulative distance is searched for, except for the ones that have been CP _top so far, and this is set as a new CP _top (step 809, FIG. 21 (2) shows a broken line). A _{1 to} A ₄ , B ₁₁ , B ₁₂ , and B ₁₄ are registered in the search edge buffer, but A ₁ is excluded. Also, Ad ₄ , Ad
_3, Bd ₁₁ is large, in the _{Ad 2, Bd 11, Bd 12} ,
If Ad ₂ is the minimum, then CP _top = A ₂ .
In this case, one of the intersections adjacent to A ₂ , B ₂₄ , is the same as the departure intersection STP, and the cumulative distance to the intersection net list and the intersection sequential number are not registered for B ₂₄ . It is not registered in the search branch tip buffer (N in steps 803 to 805 and 810).
O).

Still, CP_top= A_TwoAbout other neighbors
Difference B_{twenty one}~ B_{twenty three}(Step 802)
YES), one of these B_{twenty one}About the direction (east)
From the evaluation value data, k = 1 (step 803),
Intersection A from departure point intersection STP_TwoVia B_{twenty one}Until
Total distance D = Bd_{twenty one}(Step 804). Next
Come on, intersection B_{twenty one}Total distance already in the intersection netlist
It is checked whether D 'has been registered (step 805).
Then Bd₁₃Has been registered, so the total distance D obtained this time D =
Bd_{twenty one}(Step 810), and D is smaller
If so, B₁₃(= B_{twenty one}) To the intersection netlist
Distance Bd_{twenty one}And A_TwoRewrite intersection sequential number of
And register B in the search branch tip buffer._{twenty one}Intersection sequence
The additional registration of the char number is performed (step 811). Young
Bd_{twenty one}Is larger than Bd _{twenty one}And A_TwoRewrite
Without recording, B to search branch tip buffer_{twenty one}Without adding
Return to step 802.

[0076] Of the remaining adjacent intersection adjacent to CP _top, for B _22, evaluation value data for the orientation (north)
Since DB ₁ is 1, k = 1 and the total distance Bd ₂₂ is obtained,
And, the intersection network list B _22, it has not yet been registered total distance, the Bd ₂₂ registered with CP _top = A ₂ intersection sequential number, the B ₂₂ additionally registers the searched branch tip buffer. For B _23, orientation (West)
The cumulative distance Bd ₂₃ evaluation value data as 5 because k = 5 asked for, and, since the still total distance in the intersection network list for the intersection B ₂₃ is not registered, the Bd ₂₃ CP _top
= Register with an intersection sequential number of A _2, and additionally registers B ₂₃ to search branch tip buffer. Hereinafter, the same processing is repeated.

As a result, the direction of the route from the CP _top to a certain adjacent intersection is two directions more than the direction of the destination viewed from the departure point.
If the route is deviated by more than the classification, the route is weighted for the distance and the search priority is lowered, and the search ahead is postponed. The search is executed preferentially. That is, according to the heuristic search, the search for the destination proceeds in a narrower range than the original route search target range (see FIG. 9).

Thereafter, when the tip of the search branch reaches the destination, YES is obtained in step 808, and the CP
_{Since “top”} always selects the minimum distance of the accumulated distance, the destination intersection DSP (assumed to be the m-th intersection) and the intersection sequential number (stored in the intersection netlist of the DSP together with the accumulated distance) The (m-1) next intersection shown in (31)) of the intersection net list in FIG. 19, the (m-2) next intersection stored in the intersection net list of the (m-1) next intersection,... -The primary intersections stored in the intersection net list of the secondary intersections, and the zero-order intersections (starting point intersections STP) stored in the intersection net list corresponding to the primary intersections are sequentially set to 0.
The route connected in the order from the next departure point to the destination is determined as the shortest optimal route, and the node sequence constituting the optimal route is stored in the guidance route storage unit 15i, and the route search is completed (step 812, 813).

In the state where the intersection net list has already been read into the route search memory 15h and the evaluation value data has been shifted from the initial state, re-search for the optimum route is instructed, and a new departure point and new destination are set. If re-searching is possible in the intersection netlist of the route search memory 15h from the positional relationship (YES in step 701 in FIG. 17), the direction of the destination viewed from the new departure point is obtained and registered in the route search memory 15h ( Step 705) After shifting each evaluation value data of the intersection netlist in the clockwise direction by the number of sections in which the current destination location is shifted clockwise from the previous destination location (step 706), FIG. The process of reading the intersection netlist from the CD-ROM 11 can be omitted.

As described above, according to the Dijkstra method, the shortest route can be searched more accurately than the horizontal search method. In the above description of the Dijkstra method, the search is performed from the departure place to the destination. However, conversely, the search may be performed from the destination to the departure place. If the constituent nodes are arranged in reverse order, the route data becomes the guidance route data from the departure point to the destination. CD-ROM
The intersection netlist stored in advance in the map data is read and used. However, based on the map data not including the intersection netlist, the intersection netlist is created on the map display control device side and stored in the route search memory 15h. It may be stored.

Also, before the search is started, the evaluation value data stored in all the intersection netlists in the route search memory is shifted in a lump, but the weight coefficient is determined in step 303 in FIG. In this case, the search time can be shortened by shifting individually from the initial state. In this case, if the evaluation value data is kept shifted, the previous intersection netlist cannot be used as it is at the time of re-searching. Therefore, it is only necessary to return to the initial state after determining the weighting coefficient. In addition, the evaluation value may be set to 3 for an azimuth separated by 2 to 3 sections from the destination local position, and may be set to 5 only for an azimuth separated by 4 sections.
Further, the weight coefficient itself may be set to 5 instead of 3, and may be set to 10 instead of 5. Further, if it is set to infinity instead of 5, it becomes equivalent to pruning (step 803).
When k determined in step is infinite, the original pruning may be performed by returning to step 802).

Also, when searching by the Dijkstra method,
In the same manner as described in the above-described horizontal search method, the direction data in which the adjacent intersection is viewed from the target intersection is registered in the intersection netlist, and before the search is started, the target local position is obtained and used as the reference direction data. Medium, the direction indicated by the direction data registered corresponding to the adjacent intersection in the intersection netlist is compared with the direction indicated by the reference direction data, and the weight coefficient is determined, or the target intersection is defined as the origin in the intersection netlist. Register the difference coordinates of the adjacent intersection in the difference coordinate system to be searched, and before starting the search, find the destination existence quadrant in the difference coordinate system with the origin as the origin, and refer to the intersection netlist during the search. Calculate the existence quadrant of the adjacent intersection in the differential coordinate system viewed from the CP _top , compare it with the existence quadrant of the destination,
The weighting factor may be set to 1 when they match, the weighting factor may be set to 3 for the adjacent quadrant, and the weighting factor may be set to 5 for the opposite target quadrant (see FIG. 22).

Further, in the intersection net list, the direction data of the adjacent intersection from the target intersection is registered, and before the search is started, instead of the reference direction data, the evaluation value data for each direction according to the destination location is stored. During the search, the direction of the adjacent intersection viewed from the CP _top is obtained by referring to the direction data of the intersection net list, and the direction corresponding to the direction is determined. The weight coefficient may be determined based on the evaluation value data.

[0084]

[0085]

[0086]

[0087]

【The invention's effect】 As described above, according to the present invention, a section is defined as a reference section.
Between each intersection and its adjacent intersection
Register the heading evaluation value data in the netlist in advance.
And the section to which the direction connecting the actual departure and destination belongs
The number of shift sections in the reference section is obtained, and based on the number of shift sections.
To correct (shift) the registered evaluation value data
True evaluation value of the direction connecting the intersection of interest and the adjacent intersection
Can be obtained, and the true evaluation value can be obtained more quickly.
Is possible, and when performing a heuristic search
The route search time can be greatly reduced.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a vehicle-mounted navigator embodying a route search method of the present invention.

FIG. 2 is an explanatory diagram of an intersection net list.

FIG. 3 is an explanatory diagram of direction data.

FIG. 4 is a first flowchart showing the operation of the map display control device.

FIG. 5 is a second flowchart showing the operation of the map display control device.

FIG. 6 is a third flowchart showing the operation of the map display control device.

FIG. 7 is a fourth flowchart showing the operation of the map display control device.

FIG. 8 is a fifth flowchart showing the operation of the map display control device.

FIG. 9 is an explanatory diagram of a diagram to be created for an intersection net list.

FIG. 10 is an explanatory diagram of a route search method using a horizontal search method.

FIG. 11 is an explanatory diagram of pruning.

FIG. 12 is an explanatory diagram of a modified example of the horizontal search method.

FIG. 13 is a flowchart showing the operation of the modification.

FIG. 14 is an explanatory diagram of difference coordinates.

FIG. 15 is an explanatory diagram showing quadrant divisions in an XY differential coordinate system.

FIG. 16 is an explanatory diagram of another modification of the horizontal search method.

FIG. 17 is a first flowchart showing a route search operation by the Dijkstra method.

FIG. 18 is a second flowchart showing a route search operation by the Dijkstra method.

FIG. 19 is an explanatory diagram of an intersection netlist used in the Dijkstra method.

FIG. 20 is an explanatory diagram of evaluation value data.

FIG. 21 is an explanatory diagram of a route search method by the Dijkstra method.

FIG. 22 is a diagram illustrating a modification of the Dijkstra method.

FIG. 23 is an explanatory diagram of how to hold map data on a CD-ROM.

FIG. 24 is an explanatory diagram showing a data structure of a road layer.

FIG. 25 is an explanatory diagram of an adjacent node.

FIG. 26 is an explanatory diagram of a diagram to be created for an intersection net list.

FIG. 27 is an explanatory diagram of a route search method using a horizontal search method.

FIG. 28 is an explanatory diagram of a route search method by the Dijkstra method.

[Explanation of symbols]

 11 map information storage unit (CD-ROM) 13 vehicle position detection unit 14 operation unit 15 map display control device 15c guidance route drawing unit 15g optimal route search unit 15h route search memory 15i guidance route storage unit

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int. Cl. ⁷ , DB name) G08G 1/0969 G01C 21/00 G05D 1/02 G09B 29/00-29/10

Claims (1)

(57) [Claims] 1. The distance relationship between adjacent intersections is registered for each intersection.
Extend the search branch by referring to the recorded intersection netlist
Departure from the relationship between the direction and the direction connecting the origin and destination
Pruning and exploring routes that deviate from the direction connecting the ground and the destination
Connect the starting point to the destination by lowering the priority of the rope
Route search method for heuristic search for optimal route
Here, 360 ^° all directions based on a predetermined direction are divided into a plurality of sections.
Divide and set each according to the degree of deviation from the preset reference section.
Determine the evaluation value of the section , and for each intersection in the intersection net list,
In the specified rotation direction in order from the section to which the direction connecting the points belongs
Register evaluation value data consisting of evaluation values for multiple sections
When a departure place and a destination are input during route search,
Find the section to which the azimuth connecting the departure point and the destination belongs, and
And a deviation between the reference section and the evaluation value data registered in the intersection netlist.
When the data is cyclically shifted by the deviation
The value belongs to the direction connecting the intersection and the adjacent intersection
Heuristic search for the optimal route using the true evaluation value
Route searching method which is characterized in that search.