JP2001165684A - Route selecting method and system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a route selecting system capable of finding the optimum route by estimating more accurate travel time including waiting time due to signals when the time required for a route is estimated. SOLUTION: An optimum route searching device 5 judges whether there is a signal at a crossing in an area of search based on signal flags included in route selection data in a second memory device 4. When the searching device 5 judges that there is a signal at the crossing in the area of search, a signal cost computing device 6 computes a signal cost relating to waiting time at the signal. The searching device 5 computes a passage cost relating to time for passing the crossing in the area of search based on the signal cost computed by the computing device 6. When the time required for a route is estimated, more accurate travel time is estimated including waiting time due to signals and thus the optimum route can be found.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a route selection system, and more particularly, to a system for automatically selecting an optimal route between a designated departure point and a destination on a map.

[0002]

2. Description of the Related Art In recent years, with the development of electronic technology, navigation systems for guiding and guiding vehicles have rapidly become widespread. Some conventional navigation systems include a route selection system for automatically selecting an optimal route (for example, a shortest distance route or a minimum time route) from a departure point to a destination.

[0003] In a conventional route selection system, as a method for obtaining an optimal route for guiding and guiding a vehicle,
For example, there is an automatic vehicle guidance method disclosed in Japanese Patent Application Laid-Open No. 59-105113. In this method, an optimal route between a departure point and a destination is obtained from data representing the connection state of roads using the Dijkstra method or the like, which is one of the optimal route determination methods. The theory of the Dijkstra method is described in, for example, A. V. "Data structures and algorithms" written by Aiho et al. And translated into Japanese by Yoshio Ohno
179-183 of (Baifukan Co., Ltd., issued in 1990)
It is shown on the page.

As another method for obtaining an optimum route, for example, Japanese Patent Application No. 2-338001 (Japanese Unexamined Patent Application Publication No.
No. 204881). In this method, an optimal route is searched for by simultaneously executing two of a departure point search starting from the departure point and a destination search starting from the destination. The search time is reduced by such a bidirectional search.

As another method of obtaining an optimum route, for example, Japanese Patent Application No. 3-151416 (Japanese Unexamined Patent Application Publication No.
There is a recommended route guidance device disclosed in U.S. Pat. This publication describes that a right turn or a left turn at an intersection is added to an evaluation value used when obtaining an optimum route.

[0006]

When a vehicle actually travels on the optimum route obtained by any of the above methods, it is due to signal waiting that the travel time of the vehicle is greatly affected. Stop and acceleration / deceleration. However, in the conventional method, the optimum route is searched on the road network data without considering the influence of signal waiting. As a result, when the vehicle actually travels on the optimal route obtained by the conventional method, there is a problem that the vehicle cannot arrive at the destination at the expected time due to an influence of a stop or the like due to a signal waiting.

Therefore, an object of the present invention is to provide a route selection method and system for searching for an optimum route in consideration of the influence of signal waiting.

[0008]

A first aspect of the present invention is a method for selecting an optimal route on a map, comprising: road network data representing intersections and roads; Route selection data including a traffic light flag indicating whether or not there is a traffic light at each intersection is stored, and based on the route selection data, an optimal route between the designated departure point and destination is determined. A first step of determining whether there is a traffic light at an intersection to be searched based on a traffic light flag included in the route selection data during execution of a route search process to be searched; and a determination result of the first step And a second step of calculating a passing cost when passing through the intersection to be searched, based on the second step, and an optimum route is selected based on the passing cost calculated in the second step.

In the first aspect, route selection data including a traffic light flag indicating the presence or absence of a traffic light at an intersection is stored in advance. In the second step, while the route search process is being executed based on the route selection data, the passing cost when passing through the intersection to be searched is calculated based on the presence or absence of the traffic light. Therefore, the second step is
Different passing costs can be calculated according to the presence or absence of a traffic light at an intersection. As a result, it is possible to obtain the optimum route including the effect of the waiting time or acceleration / deceleration caused by the traffic light.

[0010] The second aspect is subordinate to the first aspect,
In the second step, when it is determined that there is a traffic light at the intersection to be searched, the traffic light cost related to the waiting time at the traffic light is calculated, and based on the calculated traffic light cost, the traffic light passes through the intersection to be searched. Calculate the cost. In the second aspect, the traffic light cost is calculated only when there is a traffic light at the intersection to be searched. As a result, in the route search processing, it is possible to obtain the optimum route including the influence of the waiting time and the like caused by the traffic light.

[0011] The third aspect is subordinate to the first aspect,
The route selection data includes road type information that specifies the type of each road, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, the traffic cost related to the waiting time at the traffic light. Is calculated based on the road type information of the road constituting the intersection, and the passing cost of the intersection to be searched is calculated based on the calculated traffic signal cost.

In the third aspect, the traffic signal cost is calculated based on the type of the road constituting the intersection. Therefore, different traffic signal costs can be calculated depending on whether the road that enters the intersection is a priority road. As a result, in the route search processing, it is possible to obtain an optimum route that more accurately reflects the influence of the waiting time and the like caused by the traffic light.

[0013] The fourth aspect is subordinate to the first aspect,
The route selection data further includes width information for specifying a road width of each road, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, relating to a waiting time at the traffic light. Is calculated based on the width information of the road constituting the intersection, and the passing cost of the intersection to be searched is calculated based on the calculated traffic signal cost.

The fifth aspect is dependent on the first aspect,
The route selection data further includes lane number information for specifying the number of lanes on each road, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, a waiting time at the traffic light. Is calculated based on the number of lanes of the roads constituting the intersection, and the passing cost of the intersection to be searched is calculated based on the calculated traffic light cost.

In the fourth or fifth aspect, the traffic signal cost is calculated based on the width or the number of lanes of the road constituting the intersection. Therefore, different traffic signal costs can be calculated depending on whether the road that enters the intersection is a priority road. According to the fourth and fifth aspects of the present invention, it is possible to obtain a difference in width or a difference in the number of lanes between roads constituting an intersection, and calculate different traffic signal costs according to the priority of a road entering the intersection. . As a result, in the route search processing, it is possible to obtain an optimum route that more accurately reflects the influence of the waiting time and the like caused by the traffic light.

[0016] The sixth aspect is subordinate to the first aspect,
The route selection data includes a predetermined reference waiting time cost and a ratio of the signal waiting time to the reference waiting time cost on the road entering the intersection where the traffic light is present. If it is determined that there is a traffic light, the traffic light cost related to the waiting time at the traffic light is calculated based on the reference waiting time cost and the ratio of the traffic waiting time on the road approaching the intersection. Based on the traffic signal cost obtained, the passing cost of the intersection to be searched is calculated. In the sixth aspect, the traffic signal cost is calculated based on the reference waiting time cost and the ratio of the signal waiting time for each road entering the intersection.
Therefore, the calculated traffic light cost differs for each road entering the intersection. Thus, the influence of the waiting time and the like caused by the traffic signal on the road approaching the intersection can be accurately reflected in the route search.

The seventh aspect is subordinate to the sixth aspect,
The reference waiting time cost is an average value of the waiting time of the traffic light at all the intersections where the traffic light exists. According to the seventh aspect, since the reference waiting time is selected based on the actual waiting time of the traffic light, the traffic light cost can be calculated more accurately.

[0018] The eighth aspect is subordinate to the sixth aspect,
The route selection data includes the ratio of the signal waiting time to the reference waiting time cost for some of the roads entering the intersection where the traffic light is present. Even if there is no signal waiting time ratio for the traffic light, the traffic light cost at the intersection is calculated based on the signal waiting time ratio of the other roads entering the intersection.

In the eighth aspect, there are a plurality of roads that enter most intersections where there is a traffic light. The route selection data does not have the signal waiting time ratio for all of the approach roads, but has the signal waiting time ratio for some of the approach roads. Therefore, the size of the route selection data can be reduced. As a result,
It is possible to create preferable route selection data for applications where the capacity of the storage device is limited.

The ninth aspect is dependent on the first aspect,
In the road network data, a road network is represented by a combination of nodes and links, some intersections are represented by a plurality of nodes and at least one link, and the data for route selection is road network data. Includes an intersection flag indicating whether the link exists inside or outside the intersection, and the second step is to determine whether there is a traffic light at the intersection to be searched, Is calculated based on the intersection flag, and the passing cost of the intersection to be searched is calculated based on the calculated traffic signal cost.

In the ninth aspect, in the road network data, some intersections are represented by a plurality of nodes and at least one link. In addition, the intersection flag
Each link constituting the road network data is classified into a link existing inside the intersection and a link existing outside the intersection. By using this intersection flag, it is possible to avoid calculating the signal cost a plurality of times at an intersection composed of a plurality of nodes.

The tenth aspect is dependent on the ninth aspect, and the second step is that a link indicating a road entering the intersection to be searched exists inside or outside the intersection. Is determined based on the intersection flag, and only when it is determined that a link representing a road entering the intersection to be searched exists outside the intersection, the traffic signal cost at the traffic signal at the intersection is calculated.

The eleventh aspect is dependent on the ninth aspect, and the second step is to determine whether an unsearched link connected to a node constituting the intersection to be searched exists inside the intersection. Is determined based on the intersection flag, and only when it is determined that the unsearched link exists outside the intersection, the traffic signal cost at the traffic signal at the intersection is calculated.

In the tenth or eleventh aspect, the traffic signal cost is calculated only when the intersection flag of the approach link to the intersection to be searched indicates outside or inside the intersection. This eliminates the need to calculate the traffic light cost multiple times at an intersection composed of a plurality of nodes.

The twelfth aspect is dependent on the ninth aspect, and includes, as a route search process, a departure point side search process for expanding the route search starting from the designated departure point, and a departure point search process starting from the designated destination. In the case where the destination-side search processing for expanding the route search is being executed, the second step is that, in the destination-side search processing, an unsearched link connected to a node forming an intersection to be searched is It is determined based on the intersection flag whether it exists inside or outside the intersection, and only when it is determined that an unsearched link exists outside the intersection, the traffic signal cost at the traffic signal at the intersection Is calculated.

The thirteenth aspect is dependent on the ninth aspect, and includes, as a route search process, a departure place side search process for expanding the route search from the designated departure point as a starting point, and a departure point side search process as a starting point at the designated destination. In the case where the destination-side search processing for expanding the route search is being executed, the second step is that, in the destination-side search processing, the link that has reached the node forming the intersection to be searched is located inside the intersection. Whether it exists or exists outside is determined based on the intersection flag, and only when it is determined that the reached link exists outside the intersection, the traffic signal cost at the traffic light at the intersection is calculated.

In the twelfth or thirteenth aspect, in the departure point side search and the destination side search, an unsearched link is:
Only when it is determined that the traffic light exists outside or inside the intersection, the traffic signal cost at the traffic light at the intersection is calculated. As a result, at the intersection constituted by a plurality of nodes, the traffic signal cost is calculated only once. That is, at such an intersection, the traffic signal cost is not repeatedly calculated or the traffic signal cost is never calculated. Furthermore, when the traffic cost at the intersection composed of a plurality of nodes is calculated in the departure point side search and the destination side search by the above traffic light cost calculation method, the traffic light costs are calculated at the same node. Can be. As a result, an accurate passing cost can be calculated.

The fourteenth aspect is dependent on the first aspect, and the route selection data includes a system table in which a combination of a plurality of intersections having traffic signals interlocking with each other is recorded as a system, and the second step includes: If it is determined that there is a traffic light at the intersection to be searched, the traffic cost related to the waiting time at the traffic light is calculated based on the system table, and based on the calculated traffic cost, the traffic cost of the search target intersection is calculated. Calculate the transit cost.

Now, assume a road condition in which a traffic light is turned on in conjunction with the traffic light. Under such road conditions, after the first traffic light turns on blue, the vehicle can run smoothly without stopping at several traffic lights thereafter. Therefore, when a vehicle passes through a plurality of intersections having interlocking traffic signals, it is not practical to calculate the traffic signal cost at each intersection. Therefore, in the fourteenth aspect, the system table records combinations of a plurality of intersections where traffic signals interlock with each other. In the second step, a traffic signal cost at each intersection is calculated based on the system table. As a result, it is possible to calculate a more favorable traffic signal cost according to the interlocking of the traffic signals. As a result, in the section where the traffic lights are linked, it is possible to calculate the traffic light cost only when the first traffic light passes through an intersection, and to obtain an optimal route that matches the actual road conditions.

The fifteenth aspect is dependent on the fourteenth aspect, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, a system of an intersection that has passed in the past;
The system of the intersection to be searched is compared with the system based on the system table, and based on the result of the comparison, the traffic signal cost related to the waiting time at the traffic signal at the intersection to be searched is calculated, and the calculated signal cost , The passing cost of the intersection to be searched is calculated.

The sixteenth aspect is dependent on the fifteenth aspect, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, Whether it belongs to the same line with the intersection of
When the system at the previous intersection is different from the system at the intersection to be searched, the traffic cost related to the waiting time at the traffic light at the intersection is calculated, and the calculated traffic light is determined based on the system table. Based on the cost, the passing cost of the intersection to be searched is calculated.

A seventeenth invention is according to the fifteenth invention, and the second step is, if it is determined in the first step that a traffic signal is present at the intersection to be searched for, at the intersection that passed two times before. And whether or not the intersection to be searched belongs to the same system based on the system table.
If the system at the previous intersection is different from the system at the intersection to be searched, calculate the traffic light cost related to the waiting time at the traffic light at the intersection, and, based on the calculated traffic light cost, determine the intersection to be searched. Is calculated.

In the fifteenth to seventeenth aspects, it is determined whether or not the system of the intersection that passed in the past (one or two previous times) is the same as the system of the intersection to be searched. The traffic signal cost is calculated based on this determination result. This makes it possible to calculate a more favorable traffic signal cost according to the interlocking of the traffic signals.

It is assumed that a vehicle enters a system linked with a traffic light by turning right from a certain road. In such a situation, at the first intersection immediately after a right turn, the traffic light is often lit red. Thus, in a seventeenth aspect, it is determined whether or not the system at the intersection immediately before and the system at the intersection to be searched are the same. The traffic signal cost is calculated based on this determination result. In this way, it is possible to calculate a more preferable traffic light cost according to the lighting timing of the traffic light on the actual road.

The eighteenth aspect is dependent on the fifteenth aspect, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, the intersection that has passed in the past and the intersection to be searched. It is determined based on the system table whether or not the device belongs to the same system, and immediately after it is determined that the device belongs to the same system a predetermined number of times immediately, the waiting at the traffic light at the intersection to be searched is performed. The traffic cost related to time is calculated, and the passing cost of the intersection to be searched is calculated based on the calculated traffic cost.

The nineteenth aspect is subordinate to the fifteenth aspect, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, the intersection that has passed in the past and the intersection to be searched. Judge whether they belong to the same strain,
Immediately after determining that they belong to the same system continuously for a predetermined time, calculate the signal cost related to the waiting time at the signal at the intersection to be searched, based on the calculated signal cost, Calculate the passing cost of the intersection to be searched.

It is difficult for a vehicle to pass through a section in which interlocking traffic lights are installed without waiting for a signal. Therefore, an eighteenth aspect is a method that immediately follows that it is determined that they belong to the same system continuously for a predetermined number of times, and the nineteenth invention is that they belong to the same system continuously for a predetermined time. The traffic signal cost is calculated at the intersection to be searched immediately after it is determined that the traffic signal has been detected. Therefore,
Even if the vehicle travels in a section with an interlocking traffic light, after traveling to a certain extent in the section,
The traffic signal cost is calculated. This makes it possible to calculate a more favorable traffic light cost according to the actual lighting timing of the traffic light.

The twentieth aspect is dependent on the fourteenth aspect, and the system table records, for each time zone, a combination of a plurality of intersections having traffic signals interlocking with each other as a system. The step, when it is determined that there is a traffic light at the intersection to be searched, predicts the time of passing through the intersection to be searched, and to which system the intersection to be searched belongs at the predicted time, in the system table. Based on the result of the comparison, the system of the determined intersection to be searched is compared with the system of the intersection that has been passed in the past. The traffic cost of the intersection to be searched is calculated based on the calculated traffic cost.

In an actual road network, a combination of a plurality of traffic lights may change depending on a time zone. Therefore, in the twentieth aspect, a combination of a plurality of intersections having traffic signals that are interlocked with each other is recorded in the system table for each time zone. In the second step, the traffic signal cost is calculated with reference to the system table. As a result, it is possible to calculate a more favorable signal cost according to the actual situation.

The twenty-first aspect is a system for selecting an optimum route on a map, comprising: road network data representing intersections and roads; and a traffic signal flag indicating whether or not each intersection has a traffic light. A route selection data storage device that stores route selection data including a route selection data stored in the route selection data storage device, and an optimal route between the designated departure point and destination. An optimal route search device that executes a route search process to be searched, and a traffic light cost calculation device that operates when the optimal route search device is searching for the optimal route. Whether there is a traffic light at the intersection of
The decision is made based on the traffic signal flag included in the route selection data in the route selection data storage device, and the traffic signal cost calculation device, when the optimum route search device determines that there is a traffic light at the intersection to be searched, The traffic light cost related to the waiting time at the traffic light is calculated, and the optimum route searching device calculates the passing cost related to the time of passing through the intersection to be searched based on the traffic light cost calculated by the traffic light cost calculating device. .

The twenty-second aspect is dependent on the twenty-first aspect, and the route selection data includes a predetermined reference waiting time cost and a signal waiting time corresponding to the reference waiting time cost for a road entering an intersection with a traffic light. The traffic cost calculation device further includes a time ratio, and when the optimum route searching device determines that the traffic light is present at the intersection to be searched, the traffic light cost related to the waiting time at the traffic light is determined by the standard waiting time. It is calculated based on the time cost and the ratio of the signal signal waiting time at the intersection.

The twenty-third aspect is dependent on the twenty-second aspect, and the route selection data includes, for roads entering an intersection where a traffic light is present, for some of the roads, the signal waiting time with respect to the reference waiting time cost. Including the ratio, the traffic light cost calculation device calculates the traffic light cost at the intersection for the road entering the intersection to be searched, even if the ratio of the signal waiting time does not exist, for other roads entering the intersection. Is calculated based on the ratio of the signal waiting time.

The twenty-fourth aspect is dependent on the twenty-first aspect, wherein the road network data represents a road network by a combination of nodes and links, and some of the intersections include a plurality of nodes and at least one Expressed in combination with the node, the route selection data further includes an intersection flag indicating whether each link constituting the road network data exists inside or outside the intersection, and calculates traffic signal cost. The apparatus calculates a traffic light cost related to a waiting time at the traffic light at the intersection to be searched based on the intersection flag.

The twenty-fifth aspect is dependent on the twenty-first aspect, wherein the route selection data further includes a system table in which a combination of a plurality of intersections having traffic signals interlocking with each other is recorded as a system, and Calculates the traffic light cost related to the waiting time at the traffic light at the intersection to be searched based on the system table.

The twenty-sixth aspect is dependent on the twenty-fifth aspect, and the traffic light cost calculation device has passed two times before when the optimum route search device determines that there is a traffic light at the intersection to be searched. Whether or not the intersection and the intersection to be searched belong to the same system is determined based on the system table. If the system at the intersection before the second intersection is different from the system at the intersection to be searched, the search is performed. The traffic light cost related to the waiting time at the traffic light at the target intersection is calculated.

The twenty-seventh aspect is dependent on the twenty-sixth aspect, and the system table records, for each time zone, a combination of a plurality of intersections having traffic signals interlocking with each other as a system. The apparatus predicts the time of passing through the intersection to be searched, determines which system the intersection belongs to at the predicted time based on the system table, and determines the system of the intersection that has passed in the past and the system to be searched. And the system at the intersection, and based on the result of the comparison,
The traffic light cost at the traffic light at the intersection to be searched is calculated.

A twenty-sixth aspect is a recording medium on which a program for realizing, on a computer, an environment for selecting an optimal route on a map is recorded, wherein road network data representing intersections and roads are stored in advance. Route selection data including a traffic light flag indicating whether or not there is a traffic light at each intersection is stored, and based on the route selection data, an optimal route between the designated departure point and destination is determined. A first step of determining whether there is a traffic light at an intersection to be searched based on a traffic light flag included in the route selection data during execution of a route search process to be searched; and a determination result of the first step And a second step of calculating a passing cost when passing through the intersection to be searched based on the second step, and an optimum route is selected based on the passing cost calculated in the second step.

The twenty-ninth aspect is dependent on the twenty-eighth aspect, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, a traffic light cost related to a waiting time at the traffic light. Based on the calculated traffic signal cost, the passing cost of the intersection to be searched is calculated.

The thirtieth aspect is dependent on the twenty-eighth aspect, and the route selection data includes a predetermined reference waiting time cost and a signal waiting time for a road entering an intersection where a traffic light is present. The second step includes, when it is determined that there is a traffic light at the intersection to be searched, the traffic cost related to the waiting time at the traffic light, the reference waiting time cost, and the approach to the intersection. The traffic cost is calculated based on the ratio of the signal signal waiting time on the road to be searched, and the passing cost of the intersection to be searched is calculated based on the calculated signal cost.

The thirty-first aspect is dependent on the twenty-eighth aspect. In the road network data, a road network is represented by a combination of nodes and links. Route selection data includes an intersection flag indicating whether each link representing the road network data exists inside or outside the intersection, and the second step includes: If it is determined that there is a traffic light at the intersection to be searched, a traffic signal cost related to the waiting time at the traffic light at the intersection is calculated based on the intersection flag, and based on the calculated traffic light cost, the intersection to be searched is determined. Is calculated.

The thirty-second aspect is according to the twenty-eighth aspect, wherein the route selection data includes a system table in which a combination of a plurality of intersections having interlocking traffic signals is recorded as a system, and the second step includes: If it is determined that there is a traffic light at the intersection to be searched, based on the system table, calculate the traffic light cost related to the waiting time at the traffic light of the intersection, based on the calculated traffic light cost,
Calculate the passing cost of the intersection to be searched.

[0052]

FIG. 1 is a block diagram of a route selection system according to one embodiment of the present invention. In FIG.
The route selection system includes a first storage device 1, a position detection device 2, an input device 3, a second storage device 4 corresponding to a route selection data storage device in a claim, an optimum route search device 5, It includes a traffic signal cost calculation device 6, a third storage device 7, an output data generation device 8, and an output device 9.

The first storage device 1 includes a recording medium typified by a CD-ROM, and stores position detection data describing a detailed road network used for detecting the current position of the vehicle in the recording medium. I remember.

The position detecting device 2 obtains the approximate position and trajectory of the vehicle by using the speed of the vehicle, the turning angle, and / or the radio wave (GPS) from the satellite. Further, the position detection device 2 refers to the position detection data stored in the first storage device 1 to obtain an accurate current position of the vehicle.

The input device 3 operates in response to a user's operation, and inputs a destination to the optimum route searching device 5 (inputs a departure place if necessary. Enter time, or estimated departure and arrival times).

The second storage device 4 includes a recording medium typified by a CD-ROM, and stores path selection data used for selecting an optimum path in the recording medium.

The optimum route search device 5 is provided in the second storage device 4
The optimal route from the departure place to the destination is obtained using the route selection data stored in the. As the departure place, the current position of the vehicle detected by the position detection device 2 or information of the departure place input by the input device 3 is used. As the destination, information on the destination input by the input device 3 is used. In addition, the current time or the time at which the user is scheduled to depart (or arrive) is input to the optimal route search device 5 by the input device 3 as necessary. Further, during the search for the optimum route, the optimum route searching device 5 causes the traffic signal cost calculating device 6 to calculate the traffic signal cost necessary for calculating the passing cost when passing through each intersection. The details of the passing cost and the traffic signal cost will be described later, and the description thereof will be omitted here.

The traffic signal cost calculating device 6 calculates the traffic signal cost while the optimum route searching device 5 is searching for the optimum route. Information necessary for this calculation (information such as the presence or absence of a traffic light) is provided by the optimum route search device 5. The traffic light cost calculation device 6 calculates the traffic light cost based on the provided information, and outputs the calculated cost to the optimum route search device 5.

The third storage device 7 includes a recording medium such as a CD-ROM, and stores roads, land, and the like to be represented on a map.
It stores map display data including the shapes of ports, rivers, and parks. The output data generation device 8 uses the map display data stored in the third storage device 7 to output the current position of the vehicle obtained by the position detection device 2 and a map around the current position or the user's request. Image data for displaying a map of the specified range.

The output data generating device 8 generates image data indicating the optimum route if the range to be displayed includes the optimum route obtained by the optimum route searching device 5. Further, the output data generation device 8 provides guidance information for guiding and guiding the vehicle based on the current position of the vehicle and the obtained optimum route (for example, a voice such as "Please turn right at an intersection 100 m ahead"). Information and / or textual information).

The output device 9 includes a display and / or a speaker, and outputs the image data and the guidance information generated by the output data generation device 8 to the outside and presents them to the user. More specifically, the display displays image data and character information included in the guidance information, and the speaker outputs audio information included in the guidance information by voice.

Next, the route selection data in the second storage device 4 of FIG. 1 will be described with reference to FIGS.

In FIG. 2, the route selection data includes road network data expressing the connection status of the road network. The road network data does not faithfully represent the actual road shape (see, for example, FIG. 2 (A)), but instead of a single line based on the center line of the actual road to reduce the amount of data. This is vectorized data (see FIG. 2B). The intersection of two or more vectors represents an intersection in the road network,
Generally, it is called a node. In addition, each node:
A node identifier for uniquely specifying each is assigned. A vector having one of the two nodes as a start point and the other as an end point represents a road from one intersection to the next intersection, and is generally called a link. Further, each link is assigned a link identifier for uniquely specifying each link.

The data for route selection is further shown in FIG.
And a node table as shown in FIG. In the node table, a traffic signal flag indicating the presence or absence of a traffic signal is set for each node identifier. This makes it possible to determine whether or not a traffic light is installed at an actual intersection represented by each node.

By the way, there are roads which are independent of each other in the traveling direction. As an example of such a road, there is a road in which the upper and lower lines are separated by a median strip (see a portion with a dot) as shown in FIG. As another example, there is one in which the upper and lower lines are slightly apart, such as a side road of an elevated road. Such vertical lines of the road are represented by different links as shown in FIG. The intersection of the roads separated by the median strip in FIG. 3A is represented by a vector in which the respective vectors representing the vertical lines intersect in a substantially # -shape as shown in FIG. 3B. As a result, the intersection which was originally one is shown in FIG.
As shown in (B), a plurality of nodes (M1, M2, M3, M
4). Therefore, the links around one intersection represented by a plurality of nodes are classified into links inside the intersection and links outside the intersection. The intra-intersection link means a link formed inside one intersection represented by a plurality of nodes. In FIG. 3B, the link L2
(Refer to the thick solid line portion) is an example of the link in the intersection. In addition, the link outside the intersection is 1 represented by a plurality of nodes.
A link formed outside an intersection.

The route selection data includes a node table and a link table as shown in FIG. 3C for a case where one intersection is represented by a plurality of nodes. The node table is the same as that described with reference to FIG. 2C, and a detailed description thereof will not be repeated. In the link table, an intersection flag indicating whether the link is inside the intersection or outside the intersection is set for each link identifier. Thereby, it is possible to determine whether each link is an intra-intersection link or an inter-intersection link.

Hereinafter, a method of calculating a passing cost when the above-described route selection data is used will be described. In the present embodiment, the passing cost means an evaluation value representing a time required when the vehicle passes through the intersection, and is generally a sum of the entry / exit cost and the traffic light cost. Here, the entry / exit cost is an evaluation value indicating the traveling time of the vehicle from entering the node to exiting when the vehicle goes straight through the node or turns right or left at the node. Such entry / exit costs are well-known technologies (see Japanese Patent Application Laid-Open No. 4-372985).
Detailed description is omitted.

The traffic light cost is an evaluation value representing the waiting time at the traffic light installed at the intersection, and is one of the features of the present invention. Therefore, in the following, a method of calculating the traffic signal cost will be described in detail with reference to FIGS.

First, FIG. 4A is a schematic diagram showing a part of the road network data included in the route selection data. In FIG. 4A, two intersections are represented by two nodes N and M. That is, in FIG. 4A, one node represents one intersection. One end of the links L1, L2, R1 and R2 is connected to the node N. The node M has at least a link L2
Is connected to one end of the link L3.
FIG. 4B shows a node table and a link table of the road network data shown in FIG. 4A. FIG.
In (B), a signal flag and a signal waiting ratio are set in the node table for each node identifier. The traffic light flag is as described with reference to FIG. The node table in FIG. 4B indicates that there is a traffic light at the intersection represented by only the node N,
Furthermore, it shows that there is no traffic light at the intersection represented by only the node M. Also, the ratio of signal waiting time is
It is set for each node for which the traffic light flag is set to “present”.

Hereinafter, the ratio of the signal waiting time will be described in detail. First, the waiting time of a traffic signal means the time during which the traffic signal facing the vehicle is lit in red. The waiting time varies for each traffic light installed on the road network,
The average value of the waiting time of each traffic light can be obtained by actual measurement or the like. In the present system, this average is chosen as a preferred example of the reference latency cost. However, the reference waiting time cost may be selected in any appropriate value. As shown in FIG. 4B, the reference waiting time cost is stored in the second storage device 4 together with the node table and the link table as a part of the route selection data. The signal waiting ratio is a value indicating the ratio of the signal waiting time when entering a certain node from a certain link to the reference waiting time cost. FIG. 4 (B)
Describes a numerical value of 0.7 as a ratio of the signal waiting time of the node N. When the ratio of the node N is 0.7, the traffic light cost when entering the node N from the links L1 and L2 is 7 times the reference waiting time cost.
It is equivalent to a certain time.

In FIG. 4B, an intersection flag is set in the link table for each link identifier. The link table is as described with reference to FIG. In the link table of FIG. 4B, since the nodes N and M represent different intersections, the intersection flag of the link L2 having the nodes N and M as one end and the other end indicates that the link is outside the intersection. . Also, the intersection flags of the other links shown are
Indicates a link outside the intersection.

In the course of the route search based on the Dijkstra's algorithm, the traffic signal costs required for calculating the passing costs of the nodes N and M shown in FIG. 4 are calculated as follows.

First, as shown by a white arrow in FIG. 5A, a traffic signal cost when entering the node N from the link L1, and a link to the node N as shown by a black arrow in FIG. The traffic signal cost when entering from R1 is
It is calculated by the following equation (1). Traffic light cost = standard wait time cost x signal wait time ratio ... (1)

Now, when entering the node N from the link L1, the signal waiting ratio is 0.7 as described in the node table of FIG. Also, let S be the reference waiting time cost. By substituting these numerical values into the above equation (1), the traffic light cost represented by the following equation (2) can be calculated. Signal cost = 0.7 × S ... (2)

When the vehicle enters the node N from the link R1, the signal waiting ratio becomes 1.3. The reason will be described below. As shown in FIG. 6A, a situation is assumed where traffic signals TS1 and TS2 are installed at the intersection of two roads α and β. FIG. 6 (B) and FIG. 6 (C)
Indicates a lighting pattern of the traffic signals TS1 and TS2.
As can be seen with reference to FIGS. 6A and 6C, when the traffic light TS1 lights red, the traffic light TS2 lights a color other than red. Now, it is assumed that the signal waiting ratio of the traffic signal TS1 when entering the intersection from the road α side is 0.7. 0.7 as the ratio of the signal waiting time is a ratio when the reference waiting time cost S is set to 1. for that reason,
The residual of the ratio (= 1−the ratio of the signal waiting time of the traffic signal TS1) is added to the ratio of the signal waiting time of the traffic signal TS2. Therefore, the signal waiting ratio of the signal TS2 when entering the intersection from the road β side is 1 + the residual of the ratio (= 1−the ratio of the signal waiting time of the signal TS1) =
1.3. As is clear from the above description, FIG.
When entering the node N of (A) from the link R1, the signal waiting ratio becomes 1.3. Therefore, the traffic light cost when entering the node N from the link R1 can be calculated by the following equation (3). Traffic light cost = (1+ (1-0.7)) × S = 1.3 × S (3)

From the above, the ratio of the signal waiting time when entering the node N from R1 and R2 can be calculated based on the ratio of the signal waiting time when entering the node N from L1 and L2. . In other words, R
The ratio of the signal waiting time when entering the node N from 1 and R2 means that the ratio does not have to be described in the node table of FIG. Thereby, the route selection system of FIG. 1 has a technical effect that the traffic signal cost can be calculated while the data amount stored in the second storage device 4 is reduced. More specifically, the route selection system is typically mounted in a small space such as a vehicle. Therefore, the storage capacity of each of the data storage devices 1, 4, and 7 is naturally limited. It is a great advantage for the route selection system having such a limitation that the amount of stored data is small.

Referring back to FIG. As shown by the black arrows in FIG. 5B, the traffic signal cost when entering the node M from the link L2 is zero. Because Figure 4
This is because, in the node table of (B), the signal flag of the node M indicates that there is no signal.

With reference to FIGS. 4 to 6, the method of calculating the traffic signal cost when one intersection is represented by one node has been described. It should be noted here that in the above traffic light cost calculation, the links L1 to L3 and R1 and R2 are all links outside the intersection, and therefore the description of the intersection flag when calculating the traffic light cost (details will be described later) is omitted. That is the point.

The above description has been made on the assumption that the traffic signal cost is calculated simply based on the reference waiting time cost and the signal waiting ratio. However, for example, at the node N, the road type (national road, prefectural road, etc.) of the links (L1, L2, R1, R2) constituting the intersection is checked during the route search, and the traffic signal cost of the main road is reduced. It may be calculated as follows. Further, the road width of the link forming the intersection may be examined during the route search, and the calculation may be performed so that the traffic signal cost of a road having a wide road width is reduced. Thus, different traffic signal costs can be calculated depending on whether the road entering the intersection is a priority road. As a result, it is possible to obtain an optimum route that more accurately reflects the influence of the waiting time and the like caused by the traffic light.

Further, the road width of the link forming the intersection may be examined during the route search, and the calculation may be performed so that the traffic signal cost on a road having a wider road is smaller.

Further, the road width and the number of lanes of the links constituting the intersection are checked during the route search, and the greater the difference between them, the lower the traffic signal cost on a road with a wide road or a large number of lanes. It may be calculated as follows.

FIGS. 7 and 8 are diagrams for explaining a method of calculating the traffic signal cost when one actual intersection is represented by a plurality of nodes. First, FIG.
(A) is a schematic diagram showing a part of road network data included in route selection data. In FIG. 7A, one actual intersection is represented by four nodes N1 to N4. One end of links L5, L6, R1 and R2 is connected to node N1. The links as shown are also connected to the other nodes N2 to N4.

FIG. 7B shows a node table and a link table in which information of the road network data shown in FIG. 7A is described. In the node table of FIG. 7B, the ratio of the traffic light flag and the signal waiting time is set for each node identifier. In the case of FIG. 7B, since there is a traffic light at one intersection represented by a plurality of nodes, the traffic light flags of the nodes N1 to N4 are set to “present”. Further, for the node N1, the link L
“0.8” is set as the ratio of the signal waiting time when entering the node N1 from L5 and L6. Here, as shown in FIG. 7A, the vehicle cannot enter the node N1 from the link L6, but the route search may be executed from the destination side in the bidirectional search search (see FIG. 9). It is possible to note that the ratio of the signal waiting time of the node N1 is set even when the node N1 enters from the link L6. Other nodes N2 to N
Similarly, the ratio of the signal waiting time is set for 4 as well.

In the link table shown in FIG. 7B, an intersection flag is set for each link identifier (for details, see the description of FIG. 3C). In the case of FIG. 7B, of the links shown in FIG.
Only the intersection flags of L5, R2, and R5 are set to “inside”, and it is described that the links L2, L5, R2, and R5 are intra-intersection links.

In the course of the route search based on the Dijkstra method, there is a point to be noted when calculating the traffic signal cost at one intersection represented by the nodes N1 to N4 shown in FIG. As described with reference to FIGS. 4 and 5, the traffic light cost is calculated for each node whose traffic light flag is set to “present” in principle. However, when one intersection is represented by a plurality of nodes, the traffic signal cost must not be calculated for each of the plurality of nodes. Therefore, in the departure point route search starting from the departure point based on the Dijkstra method, a rule is defined that the traffic signal cost is calculated only when the incoming link to any of the nodes N1 to N4 is a link outside the intersection. Under such a rule, the nodes N1 to N4
The traffic signal cost when entering the intersection represented by is calculated as follows.

First, as shown in FIG. 8A, it is assumed that the vehicle passes from the node L1 to the node N2 and exits to the link R3 (left turn) or the link L2 (other than left turn). here,
The link L1 which is an approach link is a link outside the intersection (FIG. 7).
(See (B)). Therefore, at the node N2, the traffic signal cost (see the above equation (1)) is calculated. Here, the reference waiting time cost is S as described above. Also,
When entering the node N2 from the link L1, the signal waiting ratio is “0.8” as shown in FIG. 7B. By substituting these into the above equation (1), the traffic light cost in the case of exiting from L1 through the node N2 to the link R3 or the link L2 can be calculated by the following equation (5). Traffic light cost = 0.8 × S ... (5)

Next, in cases other than the left turn, as shown in FIG. 8B, the link L2 is used as an incoming link, passes through the node N3, and goes to the link L3 (straight) or to the link R5 (right turn or U-turn). Escape. At this time, the entrance link L2
Is an intra-intersection link, so that the traffic light cost is not calculated when passing through the node N3 (that is, 0).

In the case of a right turn or a U-turn, FIG.
As shown in (C), link R5 is an incoming link,
After passing through the node N4, the vehicle escapes to the link R6 (right turn) or the link L5 (U-turn). Also at this time, the traffic link cost is not calculated (that is, 0) when passing through the node N4 because the entry link R5 is a link within an intersection.

Further, in the case of a U-turn, the link L5 is set as the incoming link, passes through the node N1, and escapes to the link L6. Also at this time, the traffic link cost is not calculated (that is, 0) even when passing through the node N1, because the entrance link L5 is a link within an intersection.

As a result, as shown in FIG. 8D, when the vehicle enters from the link L1, the traffic signal cost is calculated only at the node N2 which is the first node, and the intersection represented by the plurality of nodes is calculated. Until the node escapes, the traffic signal cost is not calculated in other nodes. Therefore, even at an intersection constituted by a plurality of nodes, the traffic signal cost is calculated only once.

[0091] The traffic signal cost may be calculated only once for an intersection composed of a plurality of nodes only when exiting from the intersection. For this purpose, for example, in a departure point route search starting from the departure point, a rule is defined that the traffic signal cost is calculated only when the escape link is a link outside the intersection. Under this rule, when entering from link L1, when turning left node N2
Then, go straight on at node N3, and turn right at node N3.
At 4, the traffic light cost is calculated at the node N1 at the time of the U-turn.

It should be noted here that one feature of the present invention is that when calculating the passing cost of an intersection represented by a plurality of nodes, the traffic signal cost is calculated only once. In order to make this point concrete, in the above, in the route search starting from the departure point based on the Dijkstra method, the traffic signal cost is calculated only when the incoming link to any of the nodes N1 to N4 is a link outside the intersection. And the rule that the traffic signal cost is calculated only when the exit link is a link outside the intersection. However, if the traffic light cost is calculated only once,
A method other than these two rules may be used.

Next, a description will be given of the calculation of the traffic light cost at the time of the bidirectional search, in which the departure point is not only the starting point of the route search but also the destination is the starting point of the route search. FIG.
Is a conceptual diagram of a bidirectional search. As shown in FIG. 9, in the bidirectional search, a route search starting from the departure point SP (hereinafter, referred to as a route search)
A route search is performed in two directions, that is, a route search starting from the destination DP and a route search starting from the destination DP (hereinafter, referred to as the destination search), and a route that passes through a connection portion of the search range is optimal. Required as a route. As a result, the search range becomes about half that of the one-way route search only on the departure point side, and thus a conventionally known method for reducing the search time (see Japanese Patent Application Laid-Open No. Hei 4-204881). .

However, when there is an intersection represented by a plurality of nodes, and the positions for calculating the traffic signal cost are different from each other in the departure point side search and the destination side search, at the intersection, There is an inconvenience that the traffic signal cost is not calculated once or the traffic signal cost is calculated twice. As a result, at the connection point between the departure point side search and the destination side search, there is a possibility that the correct route cost is not calculated and the optimum route is not selected.

Therefore, a method of calculating the traffic signal cost at the intersection shown in FIG. 7A when the destination side search is performed in the above bidirectional search will be described with reference to FIG.

In the search for the departure place side, in the route that enters the intersection of FIG. 7A from the link L1, FIG.
As shown in (1), the traffic signal cost was calculated at the node N2. However, in the destination side search, as indicated by an arrow in FIG. 10, the direction in which the search spreads is opposite to the direction in which the search starts, so that the search spreads from the exit link to the approach link. For this reason, in the destination side search, for example, when searching for a route from the link L1 to the link L3, in order to calculate the traffic light cost at the same node N2 as in the case of the departure side side search, the link to be searched next is an intersection If there is an external link, the traffic signal cost is calculated (see FIG. 10A).

Next, the links L1 to L
A more detailed calculation method of the traffic light cost when searching for a route to No. 3 will be described with reference to FIGS. First, in FIG. 10B, the link L3 is searched, and the search range extends to the node N3. Next, the link L3 is set as an exit link, passes through the node N3, and
A determination is made as to whether or not to calculate the traffic light cost for the pattern in which 2 is the incoming link. In this judgment, the link L3 is a searched link, and the link L3
2 is an unsearched link, and the link L2 is an intra-intersection link. Therefore, the signal cost is not added at the node N3. Next, in FIG.
When passing through the node N2, the link L2 is a searched link and the link L1 is an unsearched link. Here, since the link L1 is a link outside the intersection, the traffic signal cost is calculated. As a result, as shown in FIG. 10A, the traffic signal cost can be calculated at the same node as the departure point side search.

If the traffic signal cost is calculated when exiting the intersection in the departure point side search, the traffic signal cost may be calculated only when the searched link is a link outside the intersection, for example. .

Now, there is a case where a plurality of traffic lights are turned on in conjunction with each other. As a specific example, when the first traffic light on the route to be driven and the next traffic light turn on blue at the same time, or when a certain time elapses after the first traffic light turns on blue, the next traffic light turns blue. May turn on. When a plurality of traffic lights are linked, the vehicle can travel smoothly on the road. Therefore, it is possible to calculate a more accurate passing cost if the traffic cost is not calculated for each of the plurality of traffic lights. Therefore, in the present route selection system, the traffic signal costs of a plurality of interlocking traffic signals are calculated as follows.

FIG. 11A is a schematic diagram showing road network data representing a road network (system) in which a plurality of traffic lights are linked. FIG. 11B is a diagram showing a node table and a system table accompanying the road network data of FIG. 11A.

In FIG. 11A, seven nodes N1 to N
Six intersections are represented by 6 and M1.
The nodes N1 to N6 and M1 are connected by links as shown. In FIG. 11A, all the intersections are represented by one node for simplicity of description. As a result, the links L1 to L7 and the links R1 and R2 are off-intersection links.

In FIG. 11B, since the node table has already been described in detail, the description thereof is omitted here. As can be seen from this node table, nodes N1 to N6 and M1 in FIG.
Each has a traffic light. In the system table, time zone information and a node identifier are set for each system number. First, in the present embodiment, the system means a route on which a plurality of traffic lights whose lighting timings are linked are installed. The system number is a number for uniquely specifying each system. The time zone information is information indicating a time zone in which traffic lights of the same system are linked. Further, the node identifier in the system table is information for specifying a combination of nodes belonging to the same system. Referring to the system table in which the above information is set, for example, the traffic signals at the nodes N1 to N6 belonging to the second system are interlocked from 5:00 in the morning to 11:00 in the night. Also, for example, the node N1 belonging to the first system
And M1 are interlocked from midnight to 5:00 in the morning. As can be seen from the above two examples, in the system table, a time zone in which a plurality of traffic lights installed in each system are turned on in conjunction with each other and nodes belonging to the system are collectively recorded.

When calculating the traffic signal costs of a plurality of interlocking traffic signals, the route selection data includes road network data, a node table, and a system table shown in FIG. However, since one intersection may be represented by a plurality of nodes, a link table may be required.

Next, an example of a procedure for calculating the traffic signal cost of the searched route based on the route selection data of FIG. 11 will be described. Now, at around 12:00 noon, FIG.
When it is predicted that the vehicle will pass through the road network shown in (A), the traffic signal cost of the route from the link R1 to the link L7 is calculated by the procedure shown in FIG. However, in the following description of the calculation method, for convenience, the case of the above-described departure point side search is limited.

In FIG. 12, first, when passing through the node M1 and exiting to the link R2 with the link R1 as an ingress link, the system to which the previously searched node belongs, and the node previously searched for It is assumed that the system to which No. belongs is number 0. In the following, the notation of the past system [x1, x2] is used, where x1 indicates the system number of the immediately preceding node, and x2 indicates the system number of the previous node. For example, when the numbers of the systems to which the nodes before the previous node and the node before the previous node belong are 0, respectively, if the numbers of the past nodes [0,
0].

At about 12:00 noon, it is known from the system table (see FIG. 11B) that the node M1 to be searched this time belongs to the fourth system. Node M1
Since the system number “4” is different from “0” as the system number two before, the signal cost is calculated at the node M1.

Next, when the link R2 is set as the incoming link and passes through the node N1 and escapes to the link L2, the past system is [4, 0]. The node N1 to be searched this time belongs to the second system. Since the current system number “2” is different from the previous system number “0”, the signal cost is also calculated at the node N1.

Next, in the case where the link L2 is set as the incoming link and passes through the node N2 and exits to the link L3, the past system is [2, 4]. The node N2 to be searched this time belongs to the second system at around 12:00 in the daytime. However, since the current system number “2” is different from the previous system number “4”, the signal cost is also calculated at the node N2.

Here, in the case where the link L3 is set as the incoming link and passes through the node N3 and escapes to the link L4, the past system is [2, 2]. Now, the node N3 belongs to the second system, and the system two immediately before is also the second system.
In this way, if the current system is the same as the two previous systems,
It is determined that the traffic signal cost is not calculated. Thereafter, even when the nodes N4, N5, and N6 are to be searched, the same applies to the case of the node N3, so that no signal cost is calculated. As is clear from the above, when the search target node belongs to the same system as the node that has passed in the past, the traffic signal cost is not calculated at the search target node. Therefore, when the traffic light passes through the interlocking route, the waiting time of the traffic light is not reflected, and it can be shown that the vehicle smoothly proceeds on the route. As a result, it is possible to execute an accurate route search according to the actual traffic situation.

In the above description, the system number of the node to be searched is compared with the system number of the immediately preceding node. However, only the system number of the immediately preceding node is stored, and the system number of the node to be searched and the 1
The system number of the previous node may be compared. However, just comparing with the previous system number, at the first signal immediately after the vehicle turns right and enters a different system,
It is not possible to express a state in which a signal wait occurs. In other words, at the node where signal waiting should occur,
There is a disadvantage that the traffic signal cost is not calculated.

The traffic signal cost calculating device 6 counts how many nodes of the same system have passed, and each time the count value reaches a predetermined number of nodes (for example, 5), the traffic signal cost is calculated. You may make it calculate. Accordingly, it can be expressed that if the vehicle travels for a certain distance, a signal wait occurs even while traveling on the same system. Further, the traffic light cost calculating device 6 may calculate the traffic light cost each time the vehicle travels on a route of the same system for a predetermined travel time instead of passing through a predetermined number of nodes. . The above-described predetermined number of nodes and travel time may be changed depending on the time zone in which the nodes pass.

FIGS. 13 to 18 are flowcharts showing the operation of the route selection system shown in FIG. Hereinafter, FIG.
The route selection operation in the route selection system of FIG. 1 will be described with reference to FIGS. In the following, a route that has the shortest travel time from the current position of the vehicle to the destination input by the user is searched only from the departure point as a starting point, and at the intersection composed of a plurality of nodes, the first node The operation in the case where the traffic light cost is calculated will be described. Hereinafter, the route that provides the shortest travel time from the current position of the vehicle to the destination is referred to as an optimal route.

First, the optimum route searching device 5 acquires information indicating the position coordinates (longitude, latitude) of the destination output from the input device 3 in response to the operation of the user (FIG. 13; step S101). ). At the same time, the optimum route searching device 5
Also obtains time information for departure to the destination. Next, the optimum route searching device 5 acquires from the position detecting device 2 information indicating the coordinates (longitude and latitude) of the current position of the vehicle and the traveling direction of the vehicle (step S102). Next, the optimum route search device 5 executes a route selection process of searching for and selecting an optimum route on the road network data (steps S103 to S1).
04).

In the above route selection processing, the optimum route searching device 5 converts the road network data representing the range where the optimum route may pass (usually a rectangular range including the departure point and the destination) into the second data. Is read from the storage device 4 and stored in an internal memory (not shown) (step S103). Here, it should be noted that the rectangular area including the departure place and the destination is very wide. Therefore, usually, step S103
In, a plurality of road network data representing a relatively small range is read out, and a map of a rectangular range including a departure point and a destination is formed by the plurality of road network data.

Next, based on the coordinates of the current position and the traveling direction, the optimum route search device 5 investigates and specifies the link on which the vehicle is currently traveling from the road network data stored in the internal memory. The optimum route search device 5 uses the specified link as a start point of the route search process, and searches for a route with the shortest travel time to a node near the coordinates of the destination (step S104). The route search process in step S104 is executed by using a shortest route search method such as the conventionally known Dijkstra method, and the detailed processing procedure is shown in the flowcharts of FIGS.

In FIG. 14, the optimum route searching device 5
First, initialization is performed (step S201). That is,
The departure point is set as a reference node (a reference node corresponds to an intersection to be searched in the claims), the destination is set as a destination node, and it is set that there is no reaching link to the reference node. Next, the optimum route searching device 5 determines whether the reference node is a target node (Step S2).
02). Initially, the origin is the reference node, so the reference node cannot be the destination node. Therefore, the optimum route searching device 5 investigates links that can travel from the departure place and selects any one of them (step S20).
3). Next, the optimum route searching device 5 performs step S203.
The arrival cost from the departure point is calculated for the node located at the tip of the link selected in (hereinafter, called the arrival node) (step S204). Step S20
4 will be described later in detail.

Next, the optimum route searching device 5 executes step S
It is determined whether the arrival cost calculated in 204 is smaller than the currently recorded minimum arrival cost (step S205). More specifically, since there are several routes from the departure place to the destination node, the optimum route search device 5 passes through a certain route from the departure place at the time of step S205 and is currently set. There are cases in which the arrival cost to reach a certain arrival node is calculated many times. The smallest of the calculated arrival costs is recorded in the internal memory in step S206 described later. In step S205, the minimum arrival cost is compared with the arrival cost calculated in step S204. Here, it should be noted that immediately after the start of the route search process, the arrival cost to each destination node is calculated for the first time.
ES. Therefore, the optimum route searching device 5 calculates the arrival cost to the arrival node calculated in step S204,
It is recorded in the internal memory as the current minimum, and the minimum reaching cost is rewritten. Further, the optimum route search device 5
Is the link from the departure point to the destination node (reach link)
Is recorded in the internal memory (step S206).

Next, the optimum route searching device 5 determines whether or not there remains any link that has not been selected in step S203 among the links that can travel from the reference node (current departure point) (FIG. 15; Step S207). If there remains any unselected link, the optimum route searching device 5 returns to the operation of step S203 again. That is,
The optimum route searching device 5 selects one of the unselected links, calculates the arrival cost, and further executes a recording process as needed (steps S203 to S20).
7).

The optimum route searching device 5 executes the processing in the above step S2
The processes from 03 to S207 are repeated, all the links that can proceed from the departure point are selected, and when the calculation of the arrival cost to each arrival node is completed, it is determined as No in step S207. Next, the optimum route searching device 5 searches for and sets the next reference node (step S208). Here, the next reference node is not yet set as a reference node,
Also, it means a node having the minimum arrival cost among the nodes whose arrival costs have already been calculated. Immediately after the start of the route search process, the reaching node located at the end of each link that can travel from the departure point is a candidate for selection, and the node having the lowest reaching cost among these reaching nodes is the next reference node. Is set as Next, the optimum route searching device 5 determines whether or not the reference node has been set in step S208 (step S209). If the reference node is not set for some reason, the optimum route searching device 5 determines that the route search has failed (step S
210), and returns to the main routine of FIG. After that, the optimum route searching device 5 outputs, via the output data generating device 8,
The fact that the route search has failed is displayed on the output device 9 (step S105), and the operation ends.

On the other hand, if the reference node has been set in step S209, the optimum route searching device 5
Returning to the operation of 202, the processes of steps S203 to S209 are executed with the newly set reference node as a search target. Hereinafter, the operation when an arbitrary node located between the departure place and the destination is set as the reference node will be described in more detail.

First, it is determined that the current reference node is not the target node (step S202), a link that can travel through the reference node from the links reached by the reference node is examined, and one of them is selected (step S202). Step S203). Here, each node existing on the road network data is set as a reference node when the shortest route from the departure place to the node is determined. The arrival link of the reference node means a link that is located on the shortest path from the departure place to the reference node and that is directly connected to the reference node. In other words, the destination link is an incoming link to the reference node. Next, the optimal route searching device 5 calculates the cost of reaching the destination node located at the tip of the destination link selected in step S203 (step S204). The detailed processing procedure of the cost calculation processing in step S204 is shown in the flowchart of FIG.

In FIG. 16, first, the optimum route searching device 5 determines a traffic signal cost by using the traffic signal cost calculating device 6 (step S301). In step S301, the optimum route searching device 5 extracts each piece of information necessary for calculating the traffic signal cost from the route selection data developed in the internal memory, and
Give to. The detailed processing procedure of the traffic light cost determination processing in step S301 is shown in the flowcharts of FIGS.

Referring to FIG. 17, the traffic light cost calculating device 6
First, whether or not there is a traffic signal at the reference node is checked based on the traffic signal flag set for the reference node (step S401). Traffic light cost calculator 6
When it is determined that there is no traffic signal at the reference node, the traffic cost is set to 0 (step S402), and step S3
Finish 01. If it is determined that there is a traffic light at the reference node, the traffic light cost calculation device 6 next determines whether the arrival link is an intra-intersection link based on the intersection flag set for the arrival link (step S).
403). If the arriving link is an intra-intersection link, the traffic light cost is set to 0 in step S402, and step S301 ends.

In step S403, if the arriving link is a link outside the intersection, the traffic signal cost calculation device 6 determines which system the reference node belongs to at the expected time based on the time expected to arrive at the reference node. Is checked based on the system table. Further, the traffic signal cost calculation device 6 sets the system to which the reference node belongs as the reference system (step S404). Then, the route to the reference node reaches the reference node is compared with the two routes before the route, and it is determined whether or not they are the same (FIG. 18; step S405). If the two previous systems are the same as the system of the reference node, the traffic light cost is set to 0 (step S4).
06), the process moves to step S409. In addition, when the system of the immediately preceding system and the system of the reference node are different, the traffic signal cost calculation device 6 acquires the ratio of the signal waiting time on the arrival link from the node table, and further calculates the reference waiting time cost from the route selection data. Obtained (step S407) (already described with reference to FIGS. 4 and 5), the traffic light cost at the reference node is calculated based on the obtained ratio of the signal waiting time and the reference waiting time cost (step S40).
8).

In step S409, the previous system number is set to the previous system, and the system number of the current reference node is set to the previous system. Then, the traffic light cost determination processing in step S301 is completed, and FIG.
The process returns to the cost calculation processing of No. 6.

Referring back to FIG. After the traffic signal cost calculation device 6 calculates the traffic signal cost in step S301,
The optimum route searching device 5 acquires an entry / exit cost required for turning right or left or moving straight from the reference node to a traveling link (a link from the reference node to the destination node) (step S302). Next, the optimum route searching device 5 determines the minimum arrival cost from the departure point to the current reference node (already calculated and recorded in the internal memory) and the entry / exit cost of the reference node (in the above-described step S302). Then, the traffic cost at the reference node (calculated at step S301) and the passing cost of the traveling link are added to calculate the arrival cost at the destination node (step S303). Here, the sum of the entry / exit cost and the traffic light cost corresponds to the passing cost in the claims. It should be noted, however, that the feature of the present application is that the signal cost is included in the passing cost of the reference node, and that the passing cost is an added value of the entry / exit cost and the signal cost. There is no point. After the above step S303, the optimal route searching device 5 returns to the route searching process of FIG.

Referring again to FIG. The optimum route search device 5 determines whether the arrival cost calculated in step S204 is smaller than the minimum value of the currently recorded arrival cost (step S205). If the calculated arrival cost is the minimum cost so far, the optimum route searching device 5 rewrites the arrival cost to the destination node stored in the internal memory as the arrival cost calculated in step S204. Further, the optimum route search device 5
Rewrites the reaching link to the reaching node to a link from the current reference node to the reaching node (step S206). On the other hand, if the calculated arrival cost is not the minimum cost thus far, the optimum route searching device 5 proceeds directly to the processing of step S207 in FIG. 15 without rewriting the arrival cost and the arrival link of the arrival node. Next, the optimum route searching device 5 determines whether or not an unselected link remains among the links that can proceed from the current reference node (FIG. 15; step S207). If there are remaining links, the processing of steps S203 to S207 is repeated for the remaining links. When the calculation of the reaching cost to each reaching node is completed for all the links, the optimum route searching device 5 has the lowest reaching cost among the nodes that are not the reference nodes but have already reached the reaching costs. The node is set as the next reference node (step S208). Next, the optimum route searching device 5 determines whether or not the reference node has been set in step S208 (step S20).
9) If it has been set, the process returns to the operation of step S202, and the processes of steps S203 to S209 are executed for the newly set reference node.

If the newly set reference node is the target node, the optimum route searching device 5 executes a route configuration process in step S211. This route configuration process is a process of determining the shortest route from the departure point to the destination, going back from the destination side. Thereafter, the optimum route searching device 5 returns to the main routine of FIG. 13, displays the shortest route search configured on the output device 9 via the output data generating device 8 (step S105), and ends the operation.

In the above embodiment, it has been described that the Dijkstra method is used in the optimum route searching device 5,
Other route search methods may be used. In addition, the search is performed based on the node.
As disclosed in Japanese Patent Application Laid-Open Publication No. H10-209, a search based on a link may be performed. In addition, the route selection data is read from the second storage device 4 before the search is performed. However, the route selection data may be read while searching, or may be searched without limiting the range in advance.

[0130] The above processing procedure may be realized by a program. CD-R
By recording and distributing it on a recording medium represented by OM, it can be easily implemented by another independent computer device. Here, examples of the computer device include a mobile phone, a personal digital assistant (so-called PDA), and a portable navigation device. Therefore, in the above embodiment,
The optimum route search device 5 can search for not only the optimum route of the vehicle but also the optimum route of the user of each computer device.

The recording medium for recording the program is not limited to a CD-ROM, but is typically a magnetic disk, a magneto-optical disk, or an optical disk.

[Brief description of the drawings]

FIG. 1 is a block diagram of a route selection system according to an embodiment of the present invention.

FIG. 2 is a diagram for explaining an example of a general road network expression form in route selection data in a second storage device 4 of FIG. 1;

FIG. 3 is a diagram for explaining an example of an expression form of a road network in which vertical lines are separated from each other in route selection data in the second storage device 4 of FIG. 1;

FIG. 4 is a diagram for explaining how to calculate a traffic signal cost when passing through a certain intersection represented by one node as in FIG. 2;

FIG. 5 is a diagram for explaining how to calculate a traffic light cost when passing through a certain intersection represented by one node as in FIG. 2;

FIG. 6 is a diagram for explaining a method of calculating an incoming link for which the ratio of the signal waiting time is not set in the node table.

FIG. 7 is a diagram for explaining how to calculate a traffic light cost when passing through a certain intersection represented by a plurality of nodes as shown in FIG. 3;

FIG. 8 is a diagram for explaining how to calculate a traffic light cost when passing through another intersection represented by a plurality of nodes as shown in FIG. 3;

FIG. 9 is a diagram for explaining an outline of a bidirectional search which is a known technique.

10 is a diagram for explaining how to determine the presence or absence of a traffic light cost at the time of destination side search starting from the destination in the bidirectional search shown in FIG. 9;

FIG. 11 is a road network (grid) in which a plurality of traffic lights are linked.
FIG. 7 is a diagram for explaining how to determine whether there is a traffic signal cost when passing through.

FIG. 12 is a diagram for explaining how to calculate a traffic signal cost when passing through a system as shown in FIG. 11;

FIG. 13 is a flowchart showing an operation of the route selection system of FIG. 1;

FIG. 14 is a first half of a flowchart showing a detailed operation of subroutine step S104 in FIG. 13;

FIG. 15 is a latter half of the flowchart showing the detailed operation of subroutine step S104 in FIG. 13;

FIG. 16 is a flowchart showing a detailed operation of a subroutine step S204 in FIG. 14;

FIG. 17 is a first half of a flowchart showing a detailed operation of a subroutine step S301 of FIG. 16;

FIG. 18 is a latter half of a flowchart showing the detailed operation of subroutine step S301 in FIG. 16;

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 first storage device 2 position detection device 3 input device 4 second storage device 5 optimal route search device 6 traffic signal cost calculation device 7 third storage device 8 output data generation device 9 output device

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Yoshiki Ueyama 1006 Kazuma Kadoma, Kazuma-shi, Osaka Matsushita Electric Industrial Co., Ltd. F-term (reference) 2C032 HB05 HB22 HC31 HD07 HD16 HD21 2F029 AA02 AB07 AB13 AC02 AC08 AC14 AC18 5H180 AA01 BB13 FF05 FF25 FF27 FF32

Claims (32)

[Claims] 1. A method for selecting an optimal route on a map, comprising: in advance, road network data representing an intersection and a road; and a traffic signal flag indicating whether a traffic signal is present at each intersection. Route selection data is stored, and is included in the route selection data during a route search process for searching for an optimal route between the designated departure point and destination based on the route selection data. A first step of determining whether or not there is a traffic light at the intersection to be searched based on a traffic signal flag to be searched; and a passing when passing through the intersection to be searched, based on the determination result of the first step. A second step of calculating a cost, wherein an optimal route is selected based on the passing cost calculated in the second step. 2. The second step, when it is determined that there is a traffic light at the intersection to be searched, calculates a traffic light cost related to a waiting time at the traffic light, and based on the calculated traffic cost. The route selection method according to claim 1, further comprising calculating a passing cost of the intersection to be searched. 3. The route selection data further includes road type information for specifying a type of each road, and the second step includes, when it is determined that there is a traffic light at the intersection to be searched, Calculating a traffic light cost related to the waiting time at the intersection based on road type information of a road constituting the intersection, and calculating a passing cost of the intersection to be searched based on the calculated traffic light cost. 2. The route selection method according to 1. 4. The route selection data further includes width information for specifying a road width of each road, and the second step includes: when it is determined that there is a traffic light at the intersection to be searched, the traffic light is used. The cost related to the waiting time of the intersection is calculated based on the width information of the road constituting the intersection, and the passing cost of the intersection to be searched is calculated based on the calculated traffic light cost. Route selection method. 5. The route selection data further includes lane number information for specifying the number of lanes on each road, and the second step includes: when it is determined that there is a traffic light at the intersection to be searched, A cost related to a waiting time at a traffic light is calculated based on information on the number of lanes of roads constituting the intersection, and a passing cost of the intersection to be searched is calculated based on the calculated traffic light cost. 2. The route selection method according to 1. 6. The route selection data includes a predetermined reference waiting time cost and a ratio of the signal waiting time to the reference waiting time cost on a road entering an intersection where a traffic light is present, the second step. If it is determined that there is a traffic light at the intersection to be searched, the traffic light cost related to the waiting time at the traffic light, the reference waiting time cost, the ratio of the signal waiting time on the road to enter the intersection The route selection method according to claim 1, wherein the calculation is performed based on the traffic signal cost, and the passing cost of the intersection to be searched is calculated based on the calculated traffic signal cost. 7. The route selection method according to claim 6, wherein the reference waiting time cost is an average value of waiting times of the traffic lights at all intersections where the traffic lights are present. 8. The route selection data includes a ratio of a signal waiting time to the reference waiting time cost for a part of roads entering an intersection where a traffic light is present, and the second step includes: Even if there is no signal waiting time ratio on the road entering the intersection to be searched, the signal cost at the intersection is calculated based on the signal waiting time ratio of other roads entering the intersection. The route selection method according to claim 6, wherein the calculation is performed. 9. The road network data, wherein a road network is represented by a combination of nodes and links, and some intersections are represented by a plurality of nodes and at least one link. Data includes an intersection flag indicating whether each link representing the road network data exists inside or outside the intersection, and the second step includes a traffic light at the intersection to be searched. If it is determined that, the traffic light cost related to the waiting time at the traffic light at the intersection is calculated based on the intersection flag, and the passing cost of the intersection to be searched is calculated based on the calculated traffic light cost. The route selection method according to claim 1. 10. The second step, wherein a link representing a road entering the intersection to be searched is:
Whether it is inside or outside the intersection,
Judgment is made based on the intersection flag, and a link representing a road entering the intersection to be searched is
The route selection method according to claim 9, wherein the traffic light cost at the traffic light at the intersection is calculated only when it is determined that the traffic light exists outside the intersection. 11. The second step is to determine whether an unsearched link connected to a node configuring the intersection to be searched exists inside the intersection or outside the intersection. 10. The route selection method according to claim 9, wherein a traffic signal cost at a traffic light at the intersection is calculated only when it is determined that the unsearched link exists outside the intersection. 12. The departure point side search processing for expanding the route search starting from the designated departure point, and the destination side search processing for expanding the path search starting from the specified destination point as the route search processing. Is executed, the second step is, in the destination side search processing, whether an unsearched link connected to a node configuring the intersection to be searched exists inside the intersection. It is determined based on the intersection flag whether it exists outside, and only when it is determined that the unsearched link exists outside the intersection, the traffic signal cost at the traffic signal at the intersection is calculated. Item 10. The route selection method according to Item 9. 13. The departure point side search processing for expanding the route search starting from the designated departure point as the route search processing, and the destination side search processing for expanding the path search starting from the indicated destination point as the starting point. Is executed, the second step is that, in the destination side search processing, a link reaching a node forming the intersection to be searched exists inside the intersection or outside the intersection. The traffic light cost at the traffic light of the intersection is calculated only when it is determined that the reached link exists outside the intersection, based on the intersection flag. Route selection method. 14. The route selection data includes a system table in which a combination of a plurality of intersections having traffic signals interlocked with each other is recorded as a system, and the second step includes the step of determining that there is a traffic signal at the intersection to be searched. If it is determined, based on the system table, a traffic signal cost related to a waiting time at the traffic signal is calculated, and based on the calculated traffic signal cost, a passing cost of the intersection to be searched is calculated. 2. The route selection method according to 1. 15. The second step, when it is determined that there is a traffic light at the intersection to be searched, a system of an intersection that has passed in the past and a system of the intersection to be searched are stored in the system table. Based on the result of the comparison, calculate the traffic light cost related to the waiting time at the traffic light at the intersection to be searched, and, based on the calculated traffic light cost, the passing cost of the intersection to be searched. The route selection method according to claim 14, wherein 16. The second step is, if it is determined that there is a traffic light at the intersection to be searched, whether the intersection passed immediately before and the intersection to be searched belong to the same system. Is determined based on the system table, and when the system at the previous intersection is different from the system of the intersection to be searched and the system at the intersection, a traffic signal cost related to a waiting time at a traffic signal at the intersection is calculated. The route selection method according to claim 15, further comprising calculating a passing cost of the intersection to be searched based on the calculated traffic light cost. 17. The second step, when it is determined that there is a traffic light at the intersection to be searched, whether or not the intersection that passed two previous times and the intersection to be searched belong to the same system Is determined based on the system table, and when the system at the intersection before the second and the system at the intersection to be searched are different, the traffic signal cost related to the waiting time at the traffic signal at the intersection is calculated. The route selection method according to claim 15, further comprising calculating a passing cost of the intersection to be searched based on the calculated traffic light cost. 18. The second step, when it is determined that there is a traffic light at the intersection to be searched, it is determined whether or not the previously passed intersection and the intersection to be searched belong to the same system. Is determined based on the system table, and immediately after it is determined that the signal belongs to the same system continuously for a predetermined number of times, the signal cost related to the waiting time at the signal at the intersection to be searched is determined. The route selection method according to claim 15, wherein the calculating unit calculates the passing cost of the intersection to be searched based on the calculated traffic signal cost. 19. The second step, when it is determined that there is a traffic light at the intersection to be searched, it is determined whether or not the previously passed intersection and the intersection to be searched belong to the same system. Is determined based on the system table, and immediately after it is determined that the vehicle belongs to the same system for a predetermined time continuously, the traffic signal cost related to the waiting time at the traffic signal at the intersection to be searched is calculated. The route selection method according to claim 15, further comprising calculating a passing cost of the intersection to be searched based on the calculated traffic light cost. 20. The system table includes, for each time zone,
A combination of a plurality of intersections having traffic lights interlocked with each other is recorded as a system, and the second step is, when it is determined that there is a traffic light at the intersection to be searched, a time for passing through the intersection to be searched. It is determined on the basis of the system table which system the intersection of the search target belongs to at the predicted time based on the system table, and the determined intersection system of the search target and the intersection that has passed in the past Based on the result of the comparison, calculate the traffic light cost related to the waiting time at the traffic light at the intersection to be searched, and, based on the calculated traffic light cost, calculate the traffic cost of the search target intersection. The route selection method according to claim 14, wherein a passage cost is calculated. 21. A system for selecting an optimum route on a map, comprising: road network data representing an intersection and a road; and a traffic signal flag indicating whether or not each intersection has a traffic signal. And a route for searching for an optimal route between the designated departure point and the destination based on the route selection data stored in the route selection data storage device. An optimal route search device that performs a search process, and a traffic light cost calculation device that operates when the optimal route search device is searching for an optimal route, wherein the optimal route search device includes:
Whether or not there is a traffic light at the intersection to be searched for is determined based on the traffic light flag included in the route selection data in the route selection data storage device, and the traffic light cost calculation device, by the optimal route search device, When it is determined that there is a traffic light at the intersection of the search target, calculate the traffic light cost related to the waiting time at the traffic light, the optimal route search device, the traffic light cost calculated by the traffic light cost calculation device A route selection system that calculates a passing cost related to a time required to pass through the intersection to be searched based on the search target. 22. The route selection data further includes a predetermined reference waiting time cost and a ratio of a signal waiting time to the reference waiting time cost for a road entering an intersection having a traffic light, wherein the traffic light cost calculation is performed. The device, when it is determined by the optimal route search device that there is a traffic light at the intersection to be searched, the traffic light cost related to the waiting time at the traffic light, the reference waiting time cost, the signal at the intersection, 22. The route selection system according to claim 21, wherein the route selection system calculates based on a waiting time ratio. 23. The route selection data includes a ratio of a signal waiting time to the reference waiting time cost for a part of roads entering an intersection having a traffic light, and the traffic light cost calculating device includes: Regarding the road entering the intersection to be searched, even if there is no signal waiting time ratio, the traffic signal cost at the intersection is calculated based on the signal waiting time ratio of the other roads entering the intersection. 23. The route selection system according to claim 22, which calculates. 24. The road network data represents a road network by a combination of nodes and links, and some intersections are represented by a combination of a plurality of nodes and at least one node. The selection data further includes an intersection flag indicating whether each link constituting the road network data exists inside the intersection or outside the intersection, and the traffic signal cost calculation device includes an intersection flag of the search target intersection. 22. The route selection system according to claim 21, wherein a traffic signal cost related to a waiting time at the traffic signal is calculated based on the intersection flag. 25. The route selection data further includes a system table in which a combination of a plurality of intersections having traffic signals interlocked with each other is recorded as a system, wherein the traffic signal cost calculation device is a traffic signal at the intersection to be searched. 22. The route selection system according to claim 21, wherein a traffic signal cost related to the waiting time is calculated based on the system table. 26. The traffic signal cost calculation device, when the optimum route search device determines that there is a traffic light at the intersection to be searched, the intersection that has passed two times before and the intersection to be searched are It is determined whether or not they belong to the same system based on the system table. If the system at the two-preceding intersection is different from the system at the intersection to be searched, the traffic signal at the intersection to be searched is determined. Calculate the traffic light costs related to the waiting time at
The route selection method according to claim 25. 27. The system table includes, for each time zone,
A combination of a plurality of intersections where there are traffic lights interlocking with each other is recorded as a system, and the traffic light cost calculation device predicts a time of passing through the intersection to be searched, and at the predicted time, It is determined based on the system table in which the system for each time zone is recorded, the system of the intersection that has passed in the past, and the system of the intersection to be searched and the system of the intersection, based on the result of the comparison 27. The route selection system according to claim 26, further comprising calculating a traffic light cost related to a waiting time at a traffic light at the intersection to be searched. 28. A recording medium on which a program for realizing an environment for selecting an optimum route on a map on a computer device is recorded, wherein road network data representing intersections and roads are stored in advance, and Route selection data including a traffic light flag indicating whether or not there is a traffic light is stored, and a route for searching for an optimal route between the designated departure point and destination based on the route selection data. A first step of determining whether or not there is a traffic light at the intersection to be searched based on a traffic light flag included in the route selection data during execution of the search processing; based on a determination result of the first step; Calculating a passing cost when passing through the intersection to be searched, and selecting an optimal route based on the passing cost calculated in the second step. The recording medium recording a program. 29. In the second step, when it is determined that there is a traffic light at the intersection to be searched, the traffic light cost related to the waiting time at the traffic light at the intersection is calculated. 29. The recording medium according to claim 28, wherein a pass-through cost of the intersection to be searched is calculated based on the search cost. 30. The route selection data includes a predetermined reference waiting time cost and a ratio of a signal waiting time to the reference waiting time cost for a road entering an intersection with a traffic light, the second step. When it is determined that there is a traffic light at the intersection to be searched, the traffic light cost related to the waiting time at the traffic light, the reference waiting time cost, based on the ratio of the signal waiting time at the intersection. 29. The recording medium according to claim 28, wherein the calculating unit calculates the passing cost of the intersection to be searched based on the calculated traffic signal cost. 31. The road network data, wherein a road network is represented by a combination of nodes and links, and some intersections are represented by a plurality of nodes and at least one plurality of nodes. The route selection data includes an intersection flag indicating whether each link representing the road network data is present inside or outside the intersection, and the second step is to provide a traffic signal at the intersection to be searched. If it is determined that there is, the traffic light cost related to the waiting time at the traffic light at the intersection is calculated based on the intersection flag, and the passing cost of the intersection to be searched is calculated based on the calculated traffic light cost. 29. The recording medium according to claim 28, which calculates. 32. The route selection data includes a system table that records, as a system, a combination of a plurality of intersections having traffic signals that are interlocked with each other, and the second step includes the step of: If determined, the traffic cost related to the waiting time at the traffic light of the intersection is calculated based on the system table, and the passing cost of the intersection to be searched is calculated based on the calculated traffic cost. A recording medium according to claim 28.