JP2001241959A - Method for determining route search conditions in road maps - Google Patents

Abstract

(57) [Summary] [Problem] To propose a method of determining a route search condition that can optimize the route search condition. An actual travel route is collected, a difference δi of a coincidence rate with a reference route Li calculated by various route search conditions βi is taken on a horizontal axis, and a vertical axis of the collected actual travel route is taken on a vertical axis. The frequency is determined, and one or more route search conditions for searching for a route close to the actual traveling route are determined using the average value mi and the variance vi of the frequency distribution (steps S1 to S5). According to the present invention, it is possible to determine a route search condition for guiding a route on which a driver actually travels. Therefore, it is possible to understand what road each vehicle travels, and to predict what road the vehicle will travel in the future, which can be used for organizing and relaxing road traffic. .

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a route search method for collecting actual travel routes of vehicles, searching for a route to a destination on a road map in consideration of the actual travel routes. This is a method for determining a route search condition on a map.

[0002]

2. Description of the Related Art When a traffic information center on the ground receives data on a destination from a vehicle, a technology for searching for a route to the destination on a road map and providing this route data to a vehicle-mounted device is known. . Also, in the in-vehicle device,
2. Description of the Related Art Searching a route to a destination on a road map and displaying the searched route on a display screen in a vehicle has been performed.

[0003]

By the way, the routes searched by the computer as described above are based on the link traveling time base, the link distance base, the link road attribute, the number of right turns, the right turn timing, the first route found. The search is performed based on certain route search conditions in consideration of priority, link priority close to the direction of the destination, and the like. The route searched based on this route search condition does not always match the driver's feeling exactly.
For example, in a route search by a computer, an arterial road such as a national highway is often selected, and there is a case where a bypass known by a driver is rarely used. For this reason, the route is far from the actual traffic flow.

[0004] If a method of searching for a route that matches the driver's feeling is established, it is possible to accurately predict future traffic conditions, and it becomes easier to take measures to mitigate road traffic.
Therefore, according to the present invention, a method of determining a route search condition that can optimize the route search condition by collecting the actual travel route on which the vehicle actually travels is proposed.

[0005]

Means for Solving the Problems and Effects of the Invention (1) The method for determining a route search condition according to the present invention collects actual traveling routes,
This is a method for determining one or a plurality of route search conditions capable of searching for a collected actual traveling route or a route close thereto. According to the above configuration, it is possible to collect the actual traveling routes and determine a route search condition that leads the route on which the driver actually travels.

[0006] Therefore, it is possible to understand what road each vehicle travels, and in the future,
It is also possible to predict the road on which the vehicle will travel, which can be useful for organizing road traffic. The determination as to whether the searched route matches or approaches the actual traveling route may be made based on the matching rate of the route (claim 2). The route search condition searches for a route under a plurality of route search conditions,
The route search condition can be determined based on the coincidence rate between these searched routes and the actual traveling route (claim 3). As this determination method, a known method such as a linear programming method, a quadratic programming method, and a genetic algorithm can be used.

[0007] The "based on the matching rate" means, for example,
This means that the route search condition is weighted according to the magnitude of the coincidence rate with the actual traveling route, and the route search condition is determined by giving priority to the route search condition having the larger weight. When collecting the actual traveling route, a plurality of actual traveling routes are collected, and one or more route search conditions capable of searching for a route representing the actual traveling route are determined based on the collected actual traveling route distribution. (Claim 4). Since the actual traveling route differs for each vehicle, it is preferable to obtain a plurality of actual traveling routes and perform the statistical processing because the accuracy of determining the route search condition is improved.

In order to determine a route representative of the distribution of the actual traveling route, an average value, a median value, a maximum value, or the like of the matching rate distribution of the route may be employed. (2) The route search condition determination method of the present invention collects actual travel routes, and sets a route search condition capable of searching for a route that realizes route characteristics of the collected actual travel routes or route characteristics close thereto. This is a method of determining one or more (claim 8). The route characteristic is, for example, a route distance or a travel time of the route (claim 9).

In the above-described method for determining the route search condition, a route is searched under a plurality of route search conditions, and the route characteristics of the searched route match or approach the route characteristics of the actual traveling route. The route search condition can be determined at the same time (claim 10). As this determination method, a known method such as a linear programming method, a quadratic programming method, and a genetic algorithm can be used. If the route characteristic is travel time, according to this method, the route search condition that realizes the travel time of the collected actual travel route is determined, so how long each vehicle travels on the road is determined. Can be grasped, and the time required for the vehicle to travel on the road in the future can be predicted, which can be used to alleviate the traffic congestion of road traffic.

When the route characteristic is a travel distance, according to this method, a route search condition for realizing the travel distance of the collected actual traveling route is determined, so that how long each vehicle travels is determined. In addition, the distance traveled by the vehicle on the road can be predicted in the future, which can be used to improve the fuel efficiency of the entire road traffic. When collecting the actual travel route, collect a plurality of actual travel routes,
A route characteristic distribution of an actual traveling route may be obtained, and one or more route search conditions for searching for a route that realizes the route characteristic representative of the distribution or a route characteristic close thereto may be determined (claim 11). . This is because the actual travel time differs for each vehicle, and it is therefore preferable to obtain a plurality of travel times and perform statistical processing.

In order to determine a route representative of the route characteristic distribution of the actual traveling route, an average value, a median value, a maximum value, or the like of the route characteristic distribution may be employed.

[0012]

BEST MODE FOR CARRYING OUT THE INVENTION A traffic information center collects actual travel routes of a vehicle, searches for a route, and transmits the route to the vehicle.
Embodiments of the present invention will be described in detail with reference to the accompanying drawings. FIG. 1 is a schematic diagram of the system of the traffic information center A. The traffic information center A collects the actual travel route of the vehicle and road traffic information, searches for a route connecting major intersections at any time, and transmits the route through a communication line. The route data is transmitted to the optical beacon.

The traffic information center A will be described in detail.
The traffic information center A has an external memory 13 such as a magnetic disk, a controller 11, and a communication line modem 12. The controller 11 is a memory control unit that obtains necessary data from the external memory 13, a Dijkstra method or a potential method (Shibata, Tenme, Shimoura “Development of a stochastic path search algorithm” Sumitomo Electric No. 143, p. 1
65, September 1993), a route calculation processing unit for calculating an optimal route from the destination to the departure point, an SRAM and a D.
It is composed of a RAM and the like.

The external memory 13 stores various data (referred to as route network data) relating to links connecting respective major intersections and the like provided in the route information providing area. The specific link is as follows. For example, if the road map is as shown in FIG. 2 (a), the links are, as shown in FIG. 2 (b), a link L12 connecting intersections N1 and N2, and a link connecting intersections N2 and N1. Link L21,... B1, B2, and B3 indicate installation points of the optical beacon.

The route network data is stored in the form of a route network table as shown in Table 1. The route network table contains the identification number (link number) of each link, link distance, national road, prefectural road,
Road attributes such as municipal roads or highways, good or bad scenery, presence or absence of railroad crossings and tunnels, presence of curves and mountain roads, coordinates of the start and end points of the link, link direction, intersection shape, and end points of the link It includes an arc cost (described later) for one or a plurality of exit links, a pointer to the exit link, and the like.

[0016]

[Table 1]

If the words are defined here, when the next link connected to this link is seen from the end point of the link, this next link is referred to as an “exit link”, and the next link is referred to as the “exit link”. If you see the link before connecting to the link,
The link before this is called "entry link". In addition, exit,
The term “enter” focuses only on the traveling direction of the link, and has nothing to do with the direction in which the route is searched. For example, when searching for a route from the departure point, the exit links of each link are followed one after another, and when searching for the route from the destination back, the incoming links of each link are sequentially followed.

The term "connection cost" refers to the cost of turning right or left or going straight from the link to the exit link. For example, when entry is prohibited, the connection cost is infinite, and when there is a traffic light, the cost takes into account the average signal waiting time when turning right or left or going straight. The “arc cost” refers to the sum of the link cost, which is the travel time when passing through the link, and the connection cost for exiting the exit link (see FIG. 3B). Here, the arc is
As shown in FIG. 3 (a), it means from immediately after the start node of the link to immediately after the start node of the next link.

For the arc cost and the pointer to the exit link, for example, as shown in the following (1) and (2), a value for each route search condition is stored so as to be able to cope with a change in the route search condition. are doing. (1) Shortest route search condition: arc cost based on link travel time. The arc cost may be a cost at a legal speed, but in the present embodiment, a statistical cost derived from a past vehicle speed is used. When using a statistical cost, a different value (predicted value) may be adopted for each day, day of the week, and time zone. Statistically, day, day,
This is because the degree of traffic congestion and traffic regulations may change for each time zone.

Further, if traffic congestion information or the like of a road enters the traffic information center A, a change may be made in consideration of the information. For example, when a certain link is interrupted both up and down due to an accident, the arc cost of the link becomes infinite until traffic resumes. If the up lane of a certain road is congested, the up arc cost increases in accordance with the congestion. (2) Shortest distance route search condition: arc cost based on link distance. As described above, this arc cost may be changed in consideration of traffic congestion information and the like in the traffic information center A when the information is entered. The arc cost is also processed during the route search process. For example, (a) In the case of a route search condition using a road that is higher than a national road as much as possible: increase the arc cost of prefectural roads and municipal roads. (b) In the case of a route search condition that avoids a curve or a mountain road as much as possible: Increase the arc cost of a link having a curve or a mountain road. (c) When not using an expressway: The arc cost of a link whose attribute is an expressway is increased. (d) When the number of right turns is limited: When making a right turn exceeding the number of times, the arc cost to the right turn exit link is increased. (e) Restriction of right turn timing: Restriction of right turn timing means turning right at any one of the initial, middle and end timings of the route. For example, when making a right turn at the beginning of a route, if it is determined that the route is initial based on the travel distance and travel time from the departure point to the point, the arc cost to exit links other than the right turn exit link is increased, and If it is determined that the initial period has passed, the arc cost to the right turn exit link is increased. (f) When giving priority to the route found first: reduce the arc cost of the link retrieved from the pivot table (described later) first, and determine the probability that another route searched later will be selected as the optimal route. Reduce. (g) When giving priority to a link in a direction close to the direction to the destination: Reduce the arc cost of a link whose link direction is close to the direction to the destination.

The route search conditions actually employed are as described above.
(1) Combination of (2) and (a) to (g) above. Less than,
Each of the route search conditions that are combinations is β1, β2,
β3, ‥‥ (represented as βi). βi is, for example, a shortest route search condition using an expressway, or a shortest distance route search condition with a right turn restriction of three times. Next, the coincidence rate of the traveling trajectories will be described. FIG. 4 is a road map including links La to Lf, in which a solid-line route including links La, Lb, Lc, and Ld and a broken-line route including links La, Le, Lf, and Ld are depicted. The travel time of link La is ta, and the distance is d.
a, the travel time of the link Lb is represented by tb, and the distance is represented by db, etc., the coincidence rate δ of the broken-line route based on the solid-line route is
Based on travel time, (ta + td) / (ta + tb + tc)
+ Td), (da + dd) / (da +
db + dc + dd), and the difference rate is (te + tf) / (ta + tb + tc + td) based on the travel time and (de + df) / (da + db + dc + dd) based on the travel distance.

By defining the coincidence rate or the difference rate in this way, a route close to the route and a route farther from the route can be evaluated numerically with respect to a certain route. Therefore, a plurality of paths can be represented on a one-dimensional axis.
The path distance difference d is (de + df)-(db + dc). The travel time difference t is (te + tf)-(tb + tc). A plurality of routes can be represented on a one-dimensional axis by using the route distance difference d and the travel time difference t.

Next, a procedure for collecting actual traveling routes of a plurality of vehicles will be described. This collection is performed through the optical beacon B. As shown in FIG. 5, the optical beacon B includes a modem 23 connected to a communication line, a control device 24, a route memory 21,
It has a terminal 22 for transmitting and receiving data to and from the onboard navigation device. The optical beacon B, from the on-vehicle navigation device, an ID code identifying the vehicle, a point at which the vehicle communicated with the previous optical beacon, a route from the point at which the vehicle communicated with the previous optical beacon to the current point of communication, and the travel time actually taken, Receive information on the route from the departure place of the vehicle and the travel time from the departure place.

The above information (hereinafter referred to as "actual driving data"; the collected vehicle routes are referred to as "actual driving routes") is processed by the control device 24 and transmitted from the modem 23 to the traffic information center A through a communication line. Sent to If the traffic information center A collects the actual traveling route from each vehicle, the actual traveling route is stored in the external memory 13. The traffic information center A calculates an optimal route from a link corresponding to each major intersection to a link corresponding to another major intersection in the route providing area at each designated time or at a fixed time.

In this calculation, one or a plurality of route search conditions for searching for the actual traveling route are determined based on the accumulated actual traveling route data. The details of the determination will be described in detail according to a flowchart (FIG. 6). The controller 11 of the traffic information center A determines one link O as a departure place and one destination link D. Then, the actual traveling route rj (j = 1) passing through the departure point O and the destination point D
To M) are extracted (step S1). Set the route search condition βi to N
(N is a natural number of 2 or more) to search for the route Li (step S2). The route search method will be described later. Next, one route Li is selected from a plurality of routes Li searched by variously changing the route search conditions βi (this is referred to as “reference route Li”), and the horizontal axis represents the actual travel route rj and the reference route Li. The coincidence rate δj is taken, the frequency of the actual traveling route rj is taken on the vertical axis, and the frequency distribution ei is obtained (step S3). This process is performed for each reference route Li (i = 1 to N). The frequency distribution is a graph as shown in FIG. In the graph, the average value mi and the variance vi of the actual traveling route distribution are entered.

Since there are a plurality of reference routes Li, a distribution diagram as shown in FIG. 7 can be drawn by the number N of reference routes.
FIG. 8 is a graph in which each distribution ei (i = 1 to N) is simultaneously drawn. Using the average value mi and the variance vi for each route search condition βi, linearly calculate the weighting coefficient ki that minimizes {ki} (1-mi) ² + vi} (i = 1 to N, Δki = 1) (1) It is determined using a programming method, a quadratic programming method or the like (step S4). The weight coefficient ki represents the probability of realizing the actual traveling route.

In the equation (1), the average value mi and the variance vi are used, but in addition, an expression combining the mode, the median value, and the variance vi may be used. Then, the priorities of the route search conditions βi are determined in the order of the magnitude of the weight coefficient ki.
(Step S5). For example, assume that N = 2, β1 is the shortest time condition, and β2 is the shortest distance condition. As a result of calculating a weighting factor that minimizes equation (1), k1 = 0.6 and k2 =
In the case of 0.4, the shortest time condition β1 is No. 1 and the shortest distance condition β2 is No. 2.

When only one optimal route search condition is determined, β1 is determined. When two optimal route search conditions are determined, β1 and β2 are determined. In this manner, one or more route search conditions that can guide the actual traveling route can be determined. In determining the route search condition, the frequency distribution ei of the coincidence rate δj with the reference route Li has been obtained. However, instead of the coincidence rate δj, the route distance difference dj and the travel time difference tj
May be used. Also, a ratio of route distance or a ratio of travel time can be used.

FIG. 9 is a flowchart for explaining a procedure for determining one or more route search conditions using the route distance difference dj. Referring to FIG. 9, an actual travel route rj (j = 1 to M) passing through the departure point O and the destination D is extracted (step T1). The route L is changed by changing the route search condition βi N times.
Search for i (step T2). Next, for each of the reference routes Li, the horizontal axis represents the path distance difference dj between the actual traveling route rj and the reference route Li, and the vertical axis represents the frequency of the actual traveling route rj, thereby obtaining a frequency distribution ei (step T3). . The frequency distribution is a graph as shown in FIG. In the graph, the average value mi and the variance vi of the actual traveling route distribution are entered.

Since there are a plurality of reference paths Li, a distribution map as shown in FIG. 10 can be drawn for the number N of reference paths. FIG. 11 is a graph showing each distribution ei (i = 1 to N) at the same time. Average value mi and variance v for each route search condition βi
Using i, the weighting factor ki that makes Σki (mi ² + vi) (i = 1 to N, Σki = 1) (2) close to 0 is linear programming,
It is determined using a quadratic programming or the like (step T4). The weight coefficient ki represents the probability of realizing the actual traveling route.

In the equation (2), the average value mi and the variance vi are used. However, an expression combining the mode, the median value, and the variance vi may be used. Then, the priorities of the route search conditions βi are determined in the order of the magnitude of the weight coefficient ki.
(Step T5). The above determination of the route search condition is performed after specifying one set of the departure link and the destination link, and is not performed for all the set of the departure link and the destination link in the area. Absent. In fact, departure links,
By changing the destination link, the procedure for determining the route search condition in FIG. 6 can be repeated. However, the number of all origin and destination link pairs is huge, so for some typical origin and destination link pairs,
By determining the route search conditions, it is considered that one or more route search conditions for the area can be determined.

The traffic information center A performs a route search from the departure place based on one or a plurality of determined route search conditions. Here, a route search method will be briefly described. The route calculation processing unit performs an optimum route calculation using a pivot table (described later) provided in the DRAM based on the route network data read from the external memory 13, and stores the route calculation result in the SRAM. This is recorded in the form of a result table (described later).

The above-mentioned "pivot table" is a place for temporarily storing links during a route search when a route is calculated by the Dijkstra method or the potential method, and a number of links stored in the route network table. This is a first-in first-out table that stores the links required for the current calculation among the links. As shown in Table 2, the pivot table includes a column of a pivot valid flag indicating whether a link to be searched for a route is included in the pivot table, and a column of a pointer to a route network table for specifying the link. I have.

[0034]

[Table 2]

As shown in Table 3, the "result table" includes a link number and a link distance (these are copied from the route network table), a pivot registration flag, a pivot pointer, a route cost, and a Each column has a pointer.

[0036]

[Table 3]

The pivot registration flag is a flag indicating whether or not the link has been registered in the pivot table. The pivot pointer is a pointer that indicates where in the pivot table the link is registered. The route cost is an arrival cost from the departure point link to the link. More precisely, the path cost of a link is the sum of the arc costs of the links on the path from the departure value to the link (see FIG. 3B).

Next, a route search method using the route network table, the result table and the pivot table will be described. In the following calculation, a method of making a tree from the departure point to the destination and calculating the route will be described, but conversely, it is refused in advance that a method of making a tree from the destination to the departure and calculating the route may be used deep. In the former, the departure link is the calculation start link, whereas in the latter, the destination link is the calculation start link.

The route search method of this embodiment is basically
One of the links constituting the route network data of the route providing area is set as a calculation start link, and an optimum route tree from the calculation start link to all other links in the same area is obtained, and the optimum route tree is obtained. This is a procedure for acquiring a path from the calculation start link to the calculation start link using the link. The procedure for calculating the optimum route performed by the controller 11 is well known, and will be briefly described below.

First, a route network table corresponding to a day, a day of the week, and a time zone for which a route is to be obtained is obtained from the external memory 13. Then, one departure point link is set. Furthermore, a route search condition βi having a high priority based on the weight coefficient ki
The route is searched in order from. This is because the search is performed in order from the route closest to the actual traveling route of the vehicle. Next, the arc cost is corrected based on the road traffic information currently obtained. This correction is based on traffic information
This is performed based on accident information, construction information, and the like.
If this correction is made, the arc cost takes into account the actual running and the dynamic cost that changes due to the date and time, the day of the week, traffic congestion, construction, an accident, and the like.

Then, the result table is initialized.
That is, the pivot registration flag is cleared to 0, and the path cost is set to infinity (actually, the maximum value determined by the number of bits in the memory). Also initializes the pivot table. Specifically, the pivot valid flag is cleared to 0.
Next, the departure point link S is registered in the pivot table, the pivot valid flag of the row is set to 1, and the route cost of the departure point link S is set to a finite value, for example, the link cost of the departure point link S on the result table. change. This is to start the route calculation from the departure point link S.

Thereafter, the route search is started. First, refer to the pivot table to check whether there is a registered link. If there is a link, the link (link L
Is removed, the link L is deleted from the pivot table, and the pivot valid flag is set to 0. Next, the exit link P of the link L is referred to with reference to the route network table.
To explore. When one exit link P is specified, the path cost of the link L (previously set to a finite value) on the result table is referred to, and the path cost obtained by adding the arc cost to the link P to this path cost is calculated as the path cost A. And compares it with the path cost of the link P on the result table (the initial value is infinite). If the path cost A is smaller, the path cost of the link P on the result table is set to A.
Replace with And link P on the result table
By setting the link L as the entry link of the link, the link is connected.

The link P is registered in the pivot table with reference to the pivot table registration flag of the link P on the result table. This is to search for the next link based on the link P. Further, for links registered in the pivot table, if there is another exit link, the route search is continued for the other exit links. In this process, if the route cost A is larger, the further route search is terminated.

When the search is completed for the link registered in the pivot table, the route search is continued for other links registered in the pivot table. When there are no more links registered in the pivot table in this way, the tree of the optimal route is determined. On the result table, each link is traced in reverse, and the route cost and the incoming link of each route are registered in the column of the optimal route tree of the guidance table.

The optimum route tree is obtained in the same manner as described above even on the basis of the route search condition βi having the next highest priority. In this manner, the optimum route tree is obtained based on one or a plurality of route search conditions βi. In addition, Traffic Information Center A
Does the same for the other departure links as long as it calculates the optimal route from the other departure links and records the result in each guidance table.

If it is desired to predict the future in a different day, day of the week, and time zone, the date, day of the week, and time zone are changed.
Therefore, finally, a plurality of optimum route trees are obtained for each day, day of the week, time zone, each departure point link, and each route search condition βi. The traffic information center A transmits the route information in the form of a guidance table to the optical beacon B corresponding to the departure point link. Optical beacon B
If the route information is obtained from the traffic information center A, it is stored in its own route memory 21.

Next, when the vehicle transmits a route providing request and destination data to the optical beacon B, the optical beacon B
Day, day of week, time zone, route search condition βi from route information
Is selected and the route is transmitted to the vehicle through the transceiver 22. There are various possible route selection criteria when there are a plurality of route search conditions βi. For example, a route in which the route search condition βi is changed for each vehicle may be transmitted. May be changed, or may be changed for each time zone. It is preferable that the number of times of transmitting the route of the route search condition βi having a higher priority be greater than the number of times of transmitting the route of the route search condition βi having a lower priority.

Referring to FIG. 12, the configuration of vehicle-mounted navigation device 30 will be briefly described. The in-vehicle navigation device 30 has a function of acquiring route information from the optical beacon B, displaying the route information, and guiding the route information. The in-vehicle navigation device 30 includes a GPS receiver 32 as an azimuth sensor, and acquires a vehicle speed signal of an engine control unit (ECU) 34 as a vehicle speed sensor. ,
Based on the position information based on the vehicle speed signal, a vehicle position is determined based on a comparison with a road pattern stored in an on-board dedicated map memory (not shown) (so-called map matching method, see JP-A-64-53112). It has a function to detect.

The in-vehicle navigation device 30 further provides information on the links (identification numbers of the links, starting points of the links) of the traffic information center A in the route providing area where the route information from the traffic information center A is provided. And the coordinates of the end point, etc.). When the user inputs a destination on the screen using the remote control key 33, the link closest to the destination is specified from the links, The destination link and the position data of the own vehicle can be transmitted to the optical beacon B via the transceiver 31.

The embodiment of the present invention has been described above.
The embodiment of the present invention is not limited to the above embodiment. For example, in addition to the optical beacon, a radio beacon, an in-vehicle car phone, a mobile phone, a car number reading camera installed on the road, or the like may be used as a means for collecting the actual traveling route. As a communication means between the vehicle and the traffic information center, a radio wave beacon, an in-vehicle car phone, a mobile phone, or the like may be employed in addition to the optical beacon.

In the above embodiment, the actual travel route of the vehicle is collected at the traffic information center, the route is searched and transmitted to the vehicle. However, the actual travel route of the vehicle is collected at the traffic information center. The optimal route search condition βi may be obtained and sent to the vehicle, and the route may be searched by the vehicle. In addition, the vehicle is omitted, the actual travel route of the vehicle is collected at the traffic information center, the route is searched for the optimum route search condition βi, and the searched route is used for predicting future traffic conditions.
It is also conceivable to carry out this.

[Brief description of the drawings]

FIG. 1 is a system schematic diagram of a traffic information center A.

FIG. 2A is a road map including an optical beacon, and FIG.
It is a link figure corresponding to (a).

FIG. 3A is a diagram showing a series of links to be connected, and FIG. 3B is a diagram for explaining the concept of an arc cost and a path cost.

FIG. 4 illustrates links La to La for explaining the matching rate of routes.
It is a road map including Lf.

FIG. 5 is a schematic configuration diagram of an optical beacon.

FIG. 6 is a flowchart illustrating a procedure for determining one or a plurality of route search conditions that can search for a collected actual traveling route or a route close to the collected traveling route based on a matching rate of the routes.

FIG. 7 is a frequency distribution diagram in which the horizontal axis represents the coincidence rate δj between the actual traveling route rj and the reference route Li, and the vertical axis represents the frequency of the actual traveling route rj.

FIG. 8 is a graph in which respective frequency distributions ei (i = 1 to N) are simultaneously drawn.

FIG. 9 is a flowchart for explaining a procedure for determining one or a plurality of route search conditions using a route distance difference dj.

FIG. 10 is a frequency distribution diagram in which the horizontal axis represents the path distance difference dδj between the actual traveling route rj and the reference route Li, and the vertical axis represents the frequency of the actual traveling route rj.

FIG. 11 is a graph in which respective frequency distributions ei (i = 1 to N) are simultaneously drawn.

FIG. 12 is a schematic configuration diagram of a vehicle-mounted navigation device.

[Explanation of symbols]

 11 Controller 12 Modem for communication line 13 External memory 21 Path memory 22 Terminal 23 Modem 24 Controller 30 In-vehicle navigation device 31 Transceiver 32 GPS receiver 33 Remote control key 34 Engine control unit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2C032 HB22 HB24 HB25 HD16 HD23 HD26 2F029 AA02 AB07 AB13 AC02 AC04 AC06 AC08 AC14 5H180 AA01 BB02 BB04 BB05 BB13 BB15 FF05 FF12 FF13 9A001 FF03 JJ11

Claims (15)

[Claims] 1. A route search condition for a road map, wherein actual route information is collected and one or more route search conditions for searching for the collected actual route or a route close thereto are determined. Decision method. 2. The method for determining a route search condition in a road map according to claim 1, wherein the comparison between the collected actual traveling route and the searched route is performed based on a coincidence rate of the routes. 3. A method for determining a route search condition, comprising the steps of searching for a route under a plurality of route search conditions, and determining a route search condition based on a matching rate between the searched route and the actual traveling route. 3. A method for determining a route search condition in a road map according to claim 2, wherein: 4. A route in a road map, wherein a plurality of actual traveling routes are collected, and one or more route search conditions capable of searching for a route representative of the plurality of collected actual traveling routes are determined. How to determine search conditions. 5. The method for determining a route search condition in a road map according to claim 4, wherein the comparison between the collected actual traveling route and the searched route is performed based on a coincidence rate of the routes. 6. The representative of the actual traveling route means an average value, a median value, or a maximum value of the distribution rate of the coincidence rate of the route.
Method for determining route search conditions in the described road map. 7. A method for determining a route search condition, comprising the steps of searching for a route under a plurality of route search conditions, and determining a coincidence rate between the searched route and a route representing the distribution of the actual traveling route. 6. The method for determining a route search condition in a road map according to claim 5, wherein the route search condition is determined based on the following. 8. A method for collecting one or more actual travel routes and determining one or more route search conditions capable of searching for a route realizing the route characteristics of the collected actual travel routes or a route characteristic close thereto. For determining route search conditions in a changing road map. 9. The method according to claim 8, wherein the route characteristic is a route distance or a travel time of the route. 10. The method for determining a route search condition, wherein a route is searched under a plurality of route search conditions, and a route characteristic of the searched route matches a route characteristic of the actual traveling route. 10. The method for determining a route search condition in a road map according to claim 9, wherein the route search condition is determined so as to approach. 11. A method for collecting a plurality of actual traveling routes, obtaining a route characteristic distribution of the plurality of collected actual traveling routes, and searching for a route that realizes a route characteristic representative of the distribution or a route characteristic close thereto. A method for determining a route search condition in a road map, wherein one or more route search conditions that can be performed are determined. 12. The method according to claim 11, wherein the route characteristic is a route distance or a travel time of the route. 13. The method for determining a route search condition in a road map according to claim 11, wherein the representative of the route characteristic distribution of the actual traveling route is an average value, a median value, or a maximum value of the route characteristic distribution. 14. A route search condition determining method according to claim 1, wherein a route is searched under a plurality of route search conditions, and a route characteristic of the searched route is a route representative of a distribution of the actual traveling route. The method for determining a route search condition on a road map according to claim 11, wherein the route search condition is determined so as to match or approach the route characteristic. 15. The route search condition includes a link travel time base, a link distance base, a link road attribute, a right turn frequency, a right turn timing, a route found first, a link close to the direction of the destination, The method for determining a route search condition in a road map according to claim 1, wherein the route search condition is one or a plurality of route search conditions selected from the following.