JP2000266562A - On-vehicle route searching device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable an on-vehicle route searching device to get effective road information, by constituting the device in such a way that the device can predict a running route from prestored route data and the present running information even when no destination is set, and can receive the traffic information on the predicted route. SOLUTION: When the destination of a vehicle is inputted from an operating and inputting means 50, a route searching means 30 searches the minimum-cost route by reading out road information from map data 11. The decided route information is displayed on a displaying means 42 as the vehicle moves. When the vehicle is started in a state where no destination is inputted nor set, the running route of the vehicle is detected from the set of present locations detected by means of a present location detecting means 12 and stored as running information. A coincidence computing means 24 computes the degree of coincidence by using the running route data stored in a route storing means 10 and the running information, and a running route predicting means 23 selects the most coincident route. The selected route is shown to the driver of the vehicle through an output device 40 as a predicted route together with the road traffic information received by means of a VICS receiving means 13.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle route search apparatus for providing a driver with a travel route to a destination, and more particularly, to predict a route from a driving history of a driver even when the driver does not set a destination. Then, the present invention relates to a search for a traveling route including road information of a planned traveling route.

[0002]

2. Description of the Related Art Conventionally, various types of vehicle route searching devices (navigation devices) have been known, and these devices have come to be mounted on ordinary passenger cars.
In an ordinary navigation device, first, a destination point is set, so that a route from the current position to the destination point is searched for and guided.

[0003] VICS (Vehicle Information an)
d Communication System) In a navigation device using a road traffic information communication system (hereinafter referred to as a VICS navigation device), if a destination is set, road traffic information to the destination is received, and the information is received. Considering this, a route guidance that avoids congestion is provided. Further, when the destination is not set, traffic information around the vehicle position is received and provided.

[0004]

However, the most troublesome operation when using the navigation device is setting a destination. Usually, when going out for the purpose of excursions, etc., there are many cases where geographical guidance is not provided, and even if it is somewhat troublesome, the destination is set. However, if the user is familiar with destinations such as commuting, routes, and the like, the destination is often not set. Therefore, unexpected traffic congestion during commuting should be avoided because the destination is not set even though the destination can be known in advance if the destination is set by the VICS navigation device. Could not. The present invention
For such commonly used routes, it is necessary to provide effective road traffic information without setting destinations, and to be able to search for routes considering road traffic information. Aim.

[0005]

SUMMARY OF THE INVENTION The present invention provides a current position detecting means for detecting a current position of a vehicle, a route searching means for searching for a route between two points using map data, and information from a road traffic information center. In the in-vehicle route search device provided with a receiving unit for receiving, a route storage unit storing a plurality of traveling route data including a departure place, a destination, and a route therebetween stored in advance, and a destination point is not set. In the case, based on the stored traveling route data and the traveling information of the current vehicle, a route is selected, and traveling route prediction means for predicting the traveling route and received traffic information on the predicted route are output. And output means for performing the operation.

According to a second aspect of the present invention, the route search device further comprises an input unit for inputting necessary information such as a departure point, a destination point, and a route. According to a third aspect of the present invention, in the route search device, the traveling route prediction means calculates a coincidence which is an evaluation value whose value increases as the traveling information matches the stored traveling route data. It is characterized in that it further has a calculation means.

According to a fourth aspect of the present invention, the route search device further comprises a traveling route detecting means for detecting a current traveling route of the vehicle from a set of current positions obtained by the current position detecting means. .

According to a fifth aspect of the present invention, in the route search device, the travel information comprises a current position of the vehicle. According to a sixth aspect of the present invention, in the route search device according to the fifth aspect, the travel information further includes a travel route from the point at which the vehicle has started traveling to the current position of the vehicle. . According to a seventh aspect of the present invention, in the route search device according to the fifth or sixth aspect, the travel information further includes a day of the week at the departure point or a travel start time. According to an eighth aspect of the present invention, in the route searching device according to the sixth or seventh aspect, the traveling information further includes a traveling direction determined from the detected traveling route.

[0009] The invention of claim 9 is the invention of claims 5 to 8.
In the route search device according to any one of the above aspects, the route storage means stores and stores travel information detected every time the vehicle travels by the route detection means as travel route data.

[0010] According to a tenth aspect of the present invention, in the route search device according to any one of the first to ninth aspects, the route searching apparatus further includes a grouping means for grouping the traveling route data based on the similarity between the routes. And The invention according to claim 11 is the route search device according to claim 10, wherein the grouping means includes:
It is characterized by grouping based on the correlation between the routes, which is the evaluation value of the similarity between the traveling route data.

According to a twelfth aspect of the present invention, in the route search device according to the eleventh aspect, the route storage means stores, in the group of travel route data, routes having high mutual correlation as the same group. It is characterized by the following. According to a thirteenth aspect of the present invention, in the route search device according to the twelfth aspect, the route storage means ranks and stores the traveling route data classified into groups in descending order of cross-correlation. Features.

According to a fourteenth aspect of the present invention, in the route search apparatus according to the twelfth or thirteenth aspect, the grouped travel route data includes departure information including position information of departure points of all routes in the same group. It is characterized by further having point information. The invention of claim 15 is the invention of claims 12 to 14
In the route search device according to any one of the first to third aspects, the group-divided traveling route data further includes day-of-week information including days of a departure date of all routes belonging to the same group. According to a sixteenth aspect of the present invention, in the route search device according to any one of the twelfth to fifteenth aspects, the traveling route data classified into groups includes departure time information including departure times of all routes in the same group. It is characterized by having further.

According to a seventeenth aspect of the present invention, in the route search device according to any one of the thirteenth to sixteenth aspects, the traveling route predicting means uses the traveling route data sorted in a higher rank in order from the group. And performs route prediction.

According to an eighteenth aspect of the present invention, in the route search device according to any one of the fourteenth to seventeenth aspects, the coincidence calculating means includes a departure point of the traveling route data grouped with the departure point of the traveling information. It is characterized in that the higher the evaluation of the matching degree based on the information is, the larger the matching degree is calculated. The invention of claim 19 is the invention of claims 15 to 18
In the route search device according to any one of the first to third aspects, the coincidence calculating means calculates a higher degree of coincidence as the day of the week at the travel start point of the travel information and the evaluation of the grouped travel route data by the day of week information are higher. It is characterized by having more things. According to a twentieth aspect of the present invention, in the route search device according to any one of the sixteenth to nineteenth aspects, the degree of coincidence calculating means includes a travel start time of the travel information and a departure time information of the travel route data grouped. It is characterized in that the higher the evaluation by, the greater the degree of coincidence is calculated. According to a twenty-first aspect of the present invention, in any one of the twelfth to twentieth aspects,
In the route search device described in the paragraph, the coincidence calculating means includes:
It is characterized in that the higher the evaluation of the degree of coincidence between the traveling direction of the traveling information and the direction of the destination point in the departure point of the grouped traveling route data, the greater the degree of coincidence is calculated.

According to a twenty-second aspect of the present invention, in the route search device according to any one of the second to twenty-first aspects, the route storage means further stores the route information input by the input means as travel route data. It is characterized by. According to a twenty-third aspect of the present invention, in the route search device according to any one of the thirteenth to twenty-second aspects, the route storage means comprises:
According to the setting from the input means, the traveling route data is ranked higher and stored.

According to a twenty-fourth aspect of the present invention, in the route search device, the route search means sets the current position as a start point of the route search and the destination point of the predicted route as a destination point of the route search.
A route re-search means for searching for a minimum cost route by adding the received traffic information from the road traffic information center to the cost calculation of the route search is further provided. According to a twenty-fifth aspect of the present invention, in the route searching device according to the twenty-fourth aspect, the route re-search means starts a re-search for a route in response to a request signal from the input means.

[0017]

According to the route search apparatus of the present invention, the current position detecting means for detecting the current position of the vehicle and the route search for searching for a route between two points using map data. Means and a receiving means for receiving information from the road traffic information center, when the destination is not set, the departure place and the destination stored in the route storage means in advance. Based on a plurality of travel route data including a route therebetween and travel information of the current vehicle, a travel route can be selected by the travel route prediction means to predict a travel route. The receiving means for receiving the information from the information center receives the traffic information on the predicted route, and the output means can output the information. Therefore, the driver can predict the traveling route from the route data stored in advance and the traveling information of the current vehicle without setting the destination, and can receive the traffic information on the predicted route. Even if a troublesome destination is not set, valid road information can be obtained.

Further, according to the second aspect of the present invention, the route search device further includes input means for inputting necessary information such as a departure point, a destination point, and a route. Will be able to

Further, according to the third aspect of the present invention, the traveling route predicting means is a coincidence degree which is an evaluation value whose value increases as the traveling information of the current vehicle coincides with the stored traveling route data. Is further provided, so that the most suitable route determined from the stored route data and the current driving information can be selected.

Further, according to the invention of claim 4, the route search device further includes a traveling route detecting means for detecting a current traveling route of the vehicle from a set of the current positions obtained by the current position detecting means. Therefore, the traveling route of the vehicle can be detected.

Furthermore, according to the fifth aspect of the present invention, since the travel information includes the current position of the vehicle, the current position can be used as the travel information when calculating the degree of coincidence. Furthermore, according to the invention of claim 6, the traveling information further includes a traveling route to the current position of the vehicle with the point at which the vehicle starts traveling as a starting point. As a result, the travel route to the current position can be used as the travel information, and a more suitable route can be selected as the predicted route.
Furthermore, according to the invention of claim 7, since the travel information further includes the day of the week at the departure point or the travel start time, when calculating the degree of coincidence, the day of the week at the departure point is used as the travel information. Alternatively, the travel start time can be used, and a more suitable route can be selected as the predicted route.
Since the travel information further includes a travel direction determined from the detected travel route, the travel direction can be used as the travel information when calculating the degree of coincidence, and the predicted route can be used as the travel information. , A more suitable route can be selected.

Further, according to the ninth aspect of the present invention, the route storage means stores and stores travel information detected each time the vehicle travels by the route detection means as travel route data. The travel history is stored and stored as travel route data if the road has been taken. This makes it possible to provide a road that has been taken as a predicted route.

Furthermore, according to the tenth aspect of the present invention, the route search device further includes grouping means for grouping the traveling route data based on the similarity between the routes, so that the accumulated traveling route data is , The routes are grouped based on the similarity between the routes. Further, according to the eleventh aspect, the grouping means can perform grouping based on the correlation between the routes, which is the evaluation value of the similarity between the traveling route data. further,
According to the twelfth aspect of the present invention, the route storage unit can store, as the same group, routes having high mutual correlation in the traveling route data classified into groups. Further, claim 1
According to the third aspect of the present invention, the route storage means can rank and store the traveling route data in the descending order of the cross-correlation.

Further, according to the fourteenth aspect of the present invention, the grouped traveling route data further includes departure point information including position information of departure points of all routes in the same group. Furthermore, according to the invention of claim 15, the group-divided traveling route data further has day-of-week information including the day of the departure date of all routes belonging to the same group.
Further, according to the sixteenth aspect, the grouped traveling route data further includes departure time information including departure times of all routes in the same group. Further, according to the seventeenth aspect of the present invention, the traveling route predicting means can perform the route prediction by using the traveling route data classified into groups ranked higher in order.

According to the eighteenth aspect of the present invention, the coincidence calculating means calculates the starting point of the traveling information according to the sixth aspect and the starting point information of the grouped traveling route data according to the fourteenth aspect. In other words, the higher the evaluation of the degree of coincidence, the larger the degree of coincidence can be calculated, so that a more suitable path can be predicted. Further, according to the nineteenth aspect of the present invention, the degree-of-coincidence calculating means includes the day of the week at the travel start point of the travel information according to the seventh aspect of the invention and the day of the week in the travel route data classified according to the invention of the fifteenth aspect. The higher the evaluation using each piece of information, the larger the degree of coincidence can be calculated, so that a more suitable route can be predicted. Furthermore, according to the twentieth aspect, the coincidence calculating means uses the travel start time of the travel information according to the seventh aspect and the departure time information of the grouped travel route data according to the sixteenth aspect. The higher the evaluation, the larger the degree of coincidence can be calculated, so that a more suitable path can be predicted. Further, according to the twenty-first aspect of the present invention, the coincidence calculating means includes
The higher the evaluation of the degree of coincidence between the traveling direction of the traveling information according to the invention and the direction of the destination point in the departure point of the grouped traveling route data, the greater the degree of coincidence can be calculated. You can predict the route.

Further, according to the invention of claim 22, since the route storage means can further store the route information input by the input means of the invention of claim 2 as travel route data, It is possible to input a route to be passed in advance. Claim 23
According to the invention of the above, the route storage means can rank and store the traveling route data in a higher rank by the setting from the input means, so that even a route which is not used frequently can be easily set. become able to.

Further, according to the invention of claim 24, the route search means sets the current position as the starting point of the route search and the destination point of the predicted route as the destination point of the route search and receives the information from the received road traffic information center. The route re-search for searching for the minimum cost route by adding the traffic information to the cost calculation of the route search can be performed. Thereby, when traffic regulation and traffic congestion due to construction are predicted on the predicted route by the received traffic information, the route can be searched in consideration of the information. As a result, you can search for routes that avoid traffic jams and restrictions,
A route suitable for traveling can be obtained. Further, according to the twenty-fifth aspect, the route re-search means can start the re-search of the route in response to the request signal from the input means. When the user wants to guide another route, a better route can be obtained by making a request using the input means.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of a route search device according to the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a vehicle-mounted navigation device according to an embodiment of the present invention. Control device 2
0 is a traveling route detection unit 21, a traveling information storage unit 22,
Traveling route predicting means 23, grouping means 25, route searching means 3
0, search data storage means 35, search route storage means 34,
The route search control means 36 is provided. When a destination is input from the operation input unit 50, the route search unit 30 reads out the road information stored in the map data 11, performs a route search in the route search unit 31, and performs a route search in the travel cost calculation unit 32. The cost of the search route is calculated, and the route with the minimum cost is searched. Further, the road information read from the map data 11 by the route search means 34 as necessary is stored in the search data storage means 35 together with the data being searched. The route information searched and determined is stored in the searched route storage unit 34. Remembered,
The search route is overwritten on the map of the range including the current position and displayed together with the vehicle symbol on the display means 42 as the vehicle moves.

However, if the vehicle starts running without setting the destination, the travel route predicting means 2
By 3, a route is described in detail later. First, a traveling route of the vehicle is detected by a traveling route detecting unit 21 from a set of current positions detected by the current position detecting unit 12 and stored as traveling information in the traveling information storage unit 22. Next, using the travel route data and travel information stored in the route storage means 10, the degree of coincidence calculation means 24 is used.
Is calculated, and the traveling route prediction means 23 selects the most matching route, and the selected route is
The driver is guided via the output device 40 as a predicted route. Further, the road traffic information on the predicted route is VIC.
It is received by the S receiving means 13 and is guided to the driver via the output device 40. Further, the traveling information detected each time the vehicle travels is grouped by the grouping unit 25 by a grouping method described later in detail, and stored as route data by similarity to the route data stored in the route storage unit 10. It is memorized.

The route search control means 26 includes the operation input means 5
0, and a search path to the output means 40;
If there is no destination setting, the control unit performs route prediction to the traveling route prediction unit 23, displays and controls road information read from the map data 11, controls the current position detection unit 13, and controls the VICS reception unit 13 and the like. The map data 11 includes, for example, a CD-ROM, an MO, a hard disk, and an external storage device configured by the device. The path storage unit 10 may be configured by, for example, an MO, a hard disk, and a readable and writable external storage device configured by the device, or may be configured by a readable and writable internal storage device such as a RAM memory. .

The on-vehicle navigation device includes a current position detecting means 12 for knowing the current position of the vehicle. The current position detecting means 12 is, for example, a GPS (Global Positioning System) device including a receiving device, receives positioning data from a communication satellite, detects the current position of the vehicle, and obtains a route search means 30 and a route search. Notify the control means 36. In addition to the above-described method and apparatus, the current position detecting means 12 is a method and an apparatus for measuring and estimating the azimuth and the moving distance by a known in-vehicle sensor, and point code data representing the current position from a transmitting means installed outside the vehicle. May be received by a vehicle-mounted antenna and a receiver to determine the current position, and a device therefor.

The information output means 40 includes several streets,
A map such as the shortest route connecting the vehicle symbol, the current position, and the destination position is displayed. For example, it is a display device using a liquid crystal panel or the like, and includes a touch switch on a screen,
On the map displayed on the screen, the destination can be set by touching a point corresponding to the destination, and the navigation device can be operated by touching various displayed commands. The audio output means 41 is constituted by a speaker and its peripheral devices. Route guidance through speaker, VI
The information received by the CS receiving means 13 is provided.

The operation input means 50 comprises an operation panel, the above-mentioned touch panel switch, a microphone, and their peripheral devices in the on-vehicle navigation device of this embodiment. Through the microphone, an operation instruction can be given by voice to the navigation device. Since each device is publicly known, detailed description is omitted.

Next, a basic configuration of a road network which is map information will be described. The map information is composed of nodes which are intersections and links which are roads, and may be composed mainly of nodes or links. In the present embodiment, the information is configured centering on the node and includes the following information. (1) The code number of the node that defines the link (2) The traffic flag that regulates the traffic such as going straight, turning left and right (3) The link traffic flag that regulates the traffic of the link such as vehicle type regulation and time zone regulation (4) Link attribute values such as link distance, travel time or speed (5) Link attribute values such as road type and road class (6) x and y coordinates (latitude and longitude) of nodes (7) Frequency of use, road type, etc. Link attribute value such as link hierarchy value (8) Node attribute value such as hierarchy value of direction information based on frequency of use as inflow direction or outflow direction for nodes Furthermore, administrative boundaries, ports, rivers, facilities, railways, and their names And information such as coordinates. The control device 20 reads necessary map information from the map information stored in the map data 11 based on the input search start point and search end point as necessary, and displays the information for route guidance and for route search. Used as data for

As shown in FIG. 2 (a), the map information is link data (FIG. 2 (d)) which is road information based on coordinate information of latitude and longitude, a node code number, and a pointer Pn which is internal information. Is associated via the point data (FIG. 2C). Node data (Fig. 2 (b))
Is composed of node attribute values such as a node code number, latitude and longitude, the number of connected links, and a hierarchical value of direction information based on the frequency of use between connected links. The node code number is an identification number assigned to each node,
Different node codes have different node code numbers. In addition to the above, the node attribute value is a plurality of pieces of information on the attribute of the intersection, such as the width of the intersection and the presence or absence of the traffic light. The link data (FIG. 2D) includes a link number, a start node code number, an end node code number, the number of links constituting the link, a link distance, a link traffic time, a road type and a road class such as a national road and a prefectural road, and usage. It is composed of a hierarchical value based on frequency, a link traffic flag, other link attribute values, and a node code number constituting a link.

Next, route data used for route prediction will be described with reference to FIGS. Route data (Fig. 3
(C)) is a path number Kj assigned in the order stored,
Start node code number, end node code number, number of nodes that make up the route, route distance, route traffic time, departure time at which the route departed, departure point, arrival point, route traffic flag, route attribute value, node that makes the route It consists of a code number and stores a predetermined number, for example, 100. The 100 paths are sequentially stored up to the 100th path, and when the 101st path is stored, the first path is deleted. In this embodiment, any traveling may be used, and the number of routes will be described as 100 for simplicity. However, if short-distance traveling is repeated daily, the number of routes will be 100 immediately. It may be stored when the distance is longer than a certain distance. Further, the number of routes is not limited to 100. Further, instead of limiting the number, the physical memory for storing the path may be full.

As described above, the stored routes are subjected to the similarity between the routes, that is, the grouping method described in detail later.
Paths with high correlation are set to the same group, and a group-specific data table (FIG. 3B) is created. The high similarity means that, for example, when going from a house to a workplace, they pass almost the same road every day. In such a case, even if they are not exactly the same, the departure point and the destination point are the same, they are almost the same road, and the intersection where you turn left or right may be different due to traffic lights or lane change. It is well. Even in such a case, the route is to be classified into the same group as a route having high similarity. To determine the same group, in addition to the criterion that the routes are exactly the same, determine the similarity by determining the correlation between the routes and determine whether the magnitude of the correlation is the same or not. This is a criterion for determination. Next, when the degree of similarity to the other paths in the 100 paths is determined, and a group of paths having a high similarity to the other paths is set as a representative group of paths, and the path prediction is performed. Used for As a group of representative routes, among the group of routes classified into groups, the routes are ranked in descending order of similarity to other routes, and the route is predicted from the highest order in predicting the route.

The group-specific data table (FIG. 3 (b)) includes a group-specific number assigned to each group, the number of routes forming the group, the average and variance of departure points for all routes forming the group, and the departure. Mean and variance of time, probability of each departure day,
The group information includes the average and variance of arrival points, a representative route number selected from the routes forming the group, direction information, the total length of the route, the attribute value of the route, and the route number of the group. The paths classified into groups are ranked according to the degree of their correlation, and stored in the group-based ranking table (FIG. 3A) in the order of rank.

At this time, at the same time, a link list (FIG. 4A) is created for the links constituting the route data.
The link list is a list of links constituting the traveling route, and is arranged in longitude and latitude order. By referring to the link point data table (FIG. 4 (b)) and the link data table (FIG. 4 (c)) via the pointer Qn, it is determined which path includes the link. . As described above, the link is searched from the longitude and latitude, and the search is performed to find which route includes the searched link. Actually, when the vehicle deviates significantly from the route on which the vehicle originally traveled and merges with another route, it is desirable that a search is made using the middle of the route as the starting point. In such a case, in order to obtain the correlation between the routes from the links in the middle of the route, it is necessary to search for the links in the middle of the route and the links having high correlation. The first condition of a link having a high correlation is a link existing nearby. Therefore, by using the link list, for any link in the middle of the route, search for links nearby from latitude and longitude,
A route existing in the vicinity can be searched from the link data.

Next, a method for obtaining the inter-path correlation value and the cross-correlation value will be described. The accumulated route data is used to determine the similarity by obtaining a correlation value between routes and a cross-correlation value. The correlation between the stored routes is determined by the following method. For example, here are M stored route data. Each of the accumulated M route data is represented by X ₁ X ₂ ... X _M. any,
The j-th link belonging to the i-th route X _i is _defined as L _ij ,
Let N _i be the number of links that make up X _i .

[0041] defining a calculation for obtaining a cross-correlation with the route X _i and other routes as follows. Let the cross-correlation value of the path X _{i be} E
It is defined as _i by the following formula.

E _i = (Σ _k X _i * X _k ) / M (k ≠ i) * is a correlation operation (1) A correlation value between the path X _i and the path X _k is V _ik , and the following equation is used. Define.

V _ik = X _i * X _k = (Σ _j (Σ _p L _ij * L _kp ) / m _ijk ) / N _i (2) The two routes, the route X _i and the route X _k are complete When matches

V _ik = 1 (3)

[0042] In addition, the link to configure the route X _i L
_The calculation for calculating the correlation value between _ij and the link L _kp constituting the path X _k is defined by the following equation. As shown in FIG.
w is the average distance between the links, θ is the link L _ij and the link L _kp
Is the angle made by

L _ij * L _kp = a (exp (−bw)) + c (exp (−dθ)) (4) exp is used for normalization.

Where a + c = 1, 0 ≦ θ ≦ π / 2, where L _ij = L _kp , L _ij * L _kp = 1 (5) a, c is the distance between the links and the angle formed by Determined by weight. b and d are determined by the distance between the links and the allowable range of the formed angle. L _ij * L _kp is for _{checking the} degree of coincidence of the route by comparing each link. When comparing individual links, even if the links do not match, when viewed as a route, the distance between the links is short,
If you are traveling in the same direction, the destinations can generally be considered the same. In that case, the coefficient b is determined depending on how far the distance between the links is allowed. Specifically, it is determined from the allowable range of the distance between the links and the value of the calculation result that is recognized to have a correlation. Similarly, the angle θ
Is also determined.

In the equation (2), _p of Σ _p L _ij * L _kp is equal to the number of links constituting the route X _i and the number of links constituting the route X _k, and when the respective link distances do not match, that is, the link L of the route X _i
ij , the link L of the links constituting the route X _k
If there is no link that matches _ij , the correlation between the j-th link L _ij on the route X _{i and} the link on the route X _k within a circle having a radius R with the link L _ij as the center. Is used to calculate a correlation value on a link close to the link L _ij . In this case, if the link distance is large, the correlation is lost, and the calculation itself is meaningless. Therefore, the radius R is limited. Therefore, the radius R is also determined by the allowable range of the inter-link distance. The number of links on the route _Xk existing within the radius R is _defined as _mijk . The number is normalized by the number m _ijk to obtain a correlation value of the link L _ij with the link on the path X _k .

(6) (Σ _p L _ij * L _kp ) / m _ijk (6)

Next, correlation values of all the links on the path X _i with the links on the path X _k are obtained, and the sum thereof is obtained.
By dividing by the number of links N _i , the normalized expression becomes (2)
It is an expression. Equation (2), the route X _i, determine the path correlation value between the stored M number of paths, the sum is normalized, and the cross-correlation value E _i of the route X _i.

The inter-path correlation value V _ik takes a larger value as the correlation between the paths increases, and is 1 when they completely match and 0 when they do not completely match. Further, the cross-correlation value E _i takes a higher value as the route having a higher correlation with other routes out of the M routes, and the value becomes 1 when all the M routes match. . For example, commuting outward, holiday shopping forward path, always if they are identical, the cross-correlation value E _i of the pathway, appear both large. However, more of the many commuters forward a number of times, fewer shopping outward of the number of times, the cross-correlation value E _i appear larger. Furthermore, commute forward, even if slightly path is different when viewed from the entire commute forward path, because of the high path between correlation, cross-correlation value E _i will be largely appears.

Using the cross-correlation value and the inter-path correlation value defined as above, the routes are classified into groups, and among the routes in the same group, the route having the highest cross-correlation value is the representative route of the group. And The average of start points, the variance, the average of departure times, the variances, and the probabilities of each day of the departure day are obtained for all the routes constituting the group.

Next, a representative route K _q of group-specific route groups group-specific number G _i, and the group information obtained from the path constituting the group G _i, from the travel information, selects a prediction path. When selecting the path, the path K _q is as predicted path, by the following equation various factors were taken into account to compute as the value increases is determined to be appropriate, the evaluation value for the path K _q the path takes the largest value sought F _q to provide a predicted path. An equation for _{obtaining the} evaluation value _Fq will be described.

Equation 7] (path length of the sum / path K _q matching link distance) F = k1 × + k2 × (exp (-k3 × ( departure point - average starting point of the group _{G i)) + k4 × (} exp (- k5 × (departure time−average of departure times of group G _i )) + k6 × (δ × (departure day × (probability for each departure day of group G _i ))) + k7 × (sum of projection distances of the traveling route onto a straight road) / linear distance path _{K q) k1, k2, k3} , k4, k5, k6, k7 are coefficients ... (7)

The first term of the equation (7) is that the traveling route is
It means the ratio that matched the representative route. In the first term, if a match is treated as a perfect match only, the value of the first term will not increase unless the vehicle is traveling on the representative route _Kq . In this case, the perfect match is set to 1, and in the case of incomplete match, a correlation value k8 as shown in equation (4) is introduced.

Equation 8] (path length of the sum / path K _q matching link distance) =
modifying the ((path length k8 × sum of link distance) / pathway K _q).

The second term of the equation (7) is a term for evaluating the degree of coincidence of the starting position. exp is a function for normalization.
The values of k2 and k3 are changed according to the dispersion of the starting positions. If the variance is small, k3 is increased to reduce the starting position coincidence error, and k2 is increased to increase the weight occupying F. The third term of the equation (7) is considered in the same manner as the second term, and k4 and k5 are set. The fourth term of the equation (7) is
Although almost the same as the second term, the amount of change is the day of the week, not the continuous amount. Therefore, the degree of coincidence is checked based on the appearance probability of each of the days constituting the group for each day of the week. k6 is determined in the same manner as k2, but δ is a coefficient determined depending on whether the appearance probabilities for each day of the week are equal. The fifth term of equation (7) sets a straight route by connecting the departure point and the arrival point with a straight line with respect to the representative route, and sets the projection of the actual traveling locus on the straight route as an evaluation value. That is, the degree of coincidence of the traveling directions in a macro manner appears in the evaluation value. Further, the coefficient k1,
The determination of k2, k4, k6, and k7 is determined by the ratio of the weight occupied by the F factor by the respective factors in addition to the above-described elements.

With the above-described method, the degree of coincidence with the travel information and the route group classified into groups is evaluated. In the above example, the evaluation of the degree of coincidence was performed based on five factors. However, by setting the coefficient of an arbitrary term to 0, the factor can be omitted. Further, other factors may be added. Further, in the above example, the day of the week and the time are treated as different factors. However, a matrix may be introduced by taking the day of the week and the time as a set, and may be taken into consideration.

Next, the processing procedure of the route search device of this embodiment will be described with reference to the flowchart of FIG. This route search device is linked with the engine key of the vehicle,
When the engine is started to run the vehicle, the switch is automatically turned on. Step 20
At 2, it is checked whether the previous travel information has been stored. If it is determined that there is the device information stored up to the previous time, the flow proceeds to step 204 to
By calculating the cross-correlation value between the previous traveling information stored in the other and the other stored routes, the grouping means 25
They are grouped and added to the route storage means 10. In step 206, travel information is detected and collected. In step 208, it is checked whether a destination has been input. If there is an input of the destination, a route search is performed by the route search means 30 to provide a route (step 216). If there is no input of the destination, the traveling route prediction means 23 performs a route prediction (step 210). It is checked whether there is a corresponding predicted route (step 212), and if the route is not predicted, the process returns to step 206 to collect travel information,
The route is predicted again, and the process is repeated until the route is predicted from the traveling information. When the route can be predicted, the route is presented to the driver as to whether or not the route is acceptable (step 214). If the driver indicates to search for a route other than this route, the process returns to step 206 and the route is predicted again. From the driver,
If there is no reply and if OK, the process proceeds to step 218, where the traveling information is collected. In step 220, it is checked whether or not the engine key has been turned off, and the process ends. When the switch is not turned off, the road traffic information on the route presented as the predicted route and the searched route is received and presented to the driver. Here, according to step 224,
It is checked whether there is a detour request, and if there is a request for re-searching the detour route from the input means 50, step 22 is executed.
At 6, the route is searched again. In step 228, it is checked whether or not the current presentation route is sufficient. Step 2
At 14, when the predicted route is neither NO nor OK, the information is presented as it is, and if the current travel route and the predicted route are far apart, the travel information to the local point is used. It is necessary to re-estimate the route again. Here, a request from the driver is also accepted, but the degree of coincidence between the predicted route and the travel information is calculated again,
It is determined internally based on the magnitude of the obtained degree of coincidence.

Next, the procedure of the processing performed by the control device 20 regarding the method of creating the route data used for the route prediction and the route data grouping and ranking in step 204 by adding the travel information to the route data in step 204 is shown in the flowchart of FIG. It will be described according to the following. Map information configured as above,
Route data is created based on the traveling information stored each time the vehicle travels. The traveling information is detected as a route by a traveling route detecting unit 21 from a set of the current positions detected by the current position detecting unit 12 and the map data 11 by a traveling information collecting method described later. In step 302, new travel information in the travel information table is stored as a new route Kn. Here, although it means the n-th route, as described above, n is not an infinite number but a certain number. Therefore, the oldest route is deleted. Further, here, the route data is re-created every time one new route is accumulated for the sake of easy understanding. However, a predetermined number of routes may be accumulated, or a predetermined traveling information storage memory may be completely filled. At this point, the route data may be recreated. Further, the route data may be recreated by other input settings, or may be recreated under a plurality of conditions.

Hereinafter, a specific description will be given with reference to FIGS. 7 and 8, where n is 6. In the road network shown in FIG. 7, Ni attached to each intersection represents a node code number, and Li attached to each link represents a link number.
I is a point. At this time, the route Ki is a route including a departure point, an arrival point, and a transit node as shown in FIG.

Next, an inter-path correlation value _Vi6 between the stored new path K6 and the paths of the five path data Ki ( _1≤i≤5 ) stored so far is obtained by equation (2) (step 2). 304). The inter-path correlation value indicating the degree of coincidence between the paths in FIG. 8A is obtained. Here, since the correlation value between the routes is the same from both the routes, V _ij = V _ji .
Next, the path correlation value Ei of each path is obtained by the equation (1) (step 306). Next, the routes belonging to the same group as the route Ki are classified according to the magnitude of V _ij . If the correlation between the paths is high, V _ij has a large value, and if the correlation is low, V _ij has a small value. Therefore, it is determined whether or not the routes belong to the same group based on the magnitude of V _ij . The magnitude of the threshold value V _ij is empirically determined and set.

When V _{ij which} is recognized as having a correlation depending on the magnitude of V _ij and V _ij which is not recognized are classified, FIG.
This case is classified as in (b). From this, it can be understood that the route K1 has a correlation with the route K3 and the route K6, and the g1 preliminary group includes K1, K
3, K6. Similarly, since K2 has no correlation with any of the other pathways, the g2 spare group is K2, one group. Similarly, K3, K4, K
For K5, a spare group of g3, g4, g5, g6 is determined (step 308). Next, g1, g2,.
6 is selected, and those having the same route configuration are selected. Here, g1, g3, and g6 have the same path configuration. Further, although g2 is solely used, g4 and g5 have the same configuration. Extract only those with different route configurations,
In this order, the group numbers are G1, G2, and G3 (step 3).
10). Therefore, G1 is also attached to g3 and g6. Next, group information based on all the constituent routes of each group is obtained, and a group-specific data table shown in FIG. 3B is created. (Steps 312, 314, 316, 318, 320). Here, the representative route number is
Among the paths constituting the group, the path having the highest path correlation value Ei is used. However, other than that, the same route may be stored as the frequency, and the frequency having the higher frequency may be used from the frequency information.

Here, the g2 spare group consisting of one route is also 1
Although one group is treated as a group, in fact, if a group consisting of a single path is set as a group, the group of one path becomes noise, so that much time is required for path prediction or correct. Route prediction is not performed. Therefore, whether or not the spare group is a group is restricted by the number of paths constituting the group, the magnitude of the path correlation value Ei, and the like. The maximum value of the path correlation value Ei is not constant depending on the number of paths and the number of groups. The maximum value decreases as the number of groups increases. When the limit is set by the path correlation value Ei, what percentage of the E value of the group having the largest E value is
In the following cases, a method of not forming a group may be used. Further, the number of groups may be limited. In this case, if the number of groups is up to 100, a group composed of one path also appears until the first data is accumulated, but it does not matter which group is selected. Next, M groups are ranked in descending order of the path correlation value Ei of the representative path, and a group-by-group ranking table of FIG. 3A is created.

Next, the collection of travel information will be described in detail with reference to the flowchart of FIG. The travel information is collected based on the information obtained by the current position detecting means 12 (step 60).
2) In the traveling route detecting means 21, the traveling route is arranged as a series of nodes and detected as link data (step 60).
4). In step 606, it is checked whether the point is a departure point. If the point is a departure point, in step 608, information such as a departure time and a day of the week at the departure point is set in the travel information table shown in FIG. In step 610, current information other than the node code number is set. When the travel information is used as the route data, the current point is the arrival point, and the current node code number is the end point node code number.

Next, the route prediction will be described in detail with reference to the flowchart of FIG. Referring to the group-by-group ranking table of FIG. 3A, the i-th representative route data is obtained (step 402, step 404, step 40).
6). In step 408, a match evaluation value Fi is calculated according to equation (7) (step 408). Match evaluation value Fi
Are obtained for all the representative routes of ranks 1 to M,
The route having the largest value is set as the predicted route (step 414). Here, for all of the representative routes, the degree of coincidence is determined and one path having the largest value is presented as the predicted path, but a plurality of paths may be presented in descending order of the value. Is calculated,
When a value larger than the predetermined matching degree is shown, the route may be presented as a predicted route.

Next, a method of calculating the path cost will be described. The cost is an evaluation value obtained by comprehensively evaluating the path length, the required time, the cost, and the like for evaluating the shortest path.

Where αi (coefficient for factor) (9), the general formula of the method for finding the minimum cost path is

## EQU10 ## Cost = Σi (factor × αi) (i = 0, n) (10) Specifically, for example, it is obtained by the following equation.

## EQU11 ## Cost = distance.times..alpha.1 + time.times..alpha.2 + number of lanes.times..alpha.3 + right / left turn ratio.times..alpha.4 + link hierarchy value.times..alpha.5 + direction information hierarchy value.times..alpha.6 (11) where αi represents a coefficient. Is a coefficient determined by each factor, and is determined by the weight of the factor.

In this way, the search for the least cost route is performed according to the cost obtained. Next, the route search will be described in detail with reference to the flowchart of FIG. However, since a number of methods are known for searching and determining individual nodes, any method may be used, and a detailed description thereof will be omitted in this embodiment. In the route search means 30, a route search for the least cost route is performed. The route search unit 31 searches for a plurality of connection node candidates by a known method (step 502). For the connection node candidates, the link cost is calculated by the running cost calculation unit 32, and the node having the minimum link cost is determined. First, at step 504, the counter i is set to 1 to set the first candidate node. In step 506, the ith node indicated by the counter is set. Step 50
In step 8, the link cost is calculated according to equation (11). In step 510, the link cost is compared with the link cost having the minimum cost up to the (i-1) th candidate node. If it is the minimum cost, the node code number is temporarily stored as the minimum cost node (step 512). It is checked whether the node searched as a candidate is completed (step 514).
If not completed, the counter is incremented by 1 and steps 506 to 516 are repeated for the next candidate. If it is determined that the search is to be ended, the once stored minimum cost node is stored in the search route storage unit 34 as the determined search node. According to the above procedure, the nodes having the lowest cost are sequentially searched, and the shortest path obtained by the path search for extending the search branch is stored in the search path storage means 34 as a sequence of node code numbers constituting the path.

Next, the route re-search will be described with reference to the flowchart of FIG. The re-searching of the route is almost the same as the above-described searching of the route, and therefore only different portions will be described. A feature of the route re-search is that when calculating the cost, the received road traffic information is added as a factor. Equation (11) is modified as follows and step 7
08 to be used for cost calculation.

## EQU12 ## Cost = distance × α1 + time × α2 + number of lanes × α3 + right / left turn ratio × α4 + link hierarchy value × α5 + direction information hierarchy value × α6 + road traffic information × α7 (12) The route is searched again by the method. However, the road traffic information is a value indicating the degree of congestion of the link.
Costs are high. As a result, a route search taking into account the newly obtained road traffic information is executed.

In the present embodiment, the switch of the route search device and the engine key of the vehicle are linked so that
It switches on automatically, but of course you can switch it off when the engine is running, or you can switch it on again. Further, even when the engine is stopped and the vehicle is stopped, the route search device can be turned on, and the route search can be performed.

Further, the ranking table for each group is ranked according to the magnitude of the cross-correlation value. In addition to this, when an index table is created using the same starting point and the route is predicted, first, Alternatively, filtering may be performed at the departure point to increase the speed. Similarly, an index table for departure time and departure day may be created. These index tables may be used alone or in combination.

Further, the contents of the group-specific data table are not limited to this, and representative relay points may be stored. The direction information and the direction information to the relay point may be stored together, or the direction information at a fixed straight-line distance after departure may be held. Further, the route data is not limited to this, and may have the same contents as the above-described group-specific data table. Furthermore, in the above-described embodiment, the similarity between the paths is represented by the evaluation value of the cross-correlation, but other similarity evaluation methods may be used.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a configuration example of data of road information.

FIG. 3 is an explanatory diagram of a configuration example of travel route data.

FIG. 4 is an explanatory diagram of a configuration example of link data included in travel route data.

FIG. 5 is an explanatory diagram of a configuration example of travel information.

FIG. 6 is an explanatory diagram illustrating a correlation between links in the embodiment.

FIG. 7 is an explanatory diagram of an example of grouping a plurality of routes on map data.

FIG. 8 is a data configuration example for explaining route grouping using a correlation value between routes.

FIG. 9 is a flowchart showing a processing procedure of the route search device according to one embodiment of the present invention.

FIG. 10 is a flowchart showing a processing procedure for grouping new routes.

FIG. 11 is a flowchart showing a processing procedure for predicting a route.

FIG. 12 is a flowchart showing a processing procedure of a route search.

FIG. 13 is a flowchart showing a processing procedure for collecting travel information.

FIG. 14 is a flowchart illustrating a procedure of a route re-search.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Route storage means 11 ... Map data 12 ... Current position detection means 13 ... VICS reception means 20 ... Control device 21 ... Travel route detection means 22 ... Travel information storage means 23 ... Travel route prediction means 24 ... Criticality calculation means 25 ... Grouping unit 30 Route search unit 31 Route search unit 32 Travel cost calculation unit 33 Route research unit 34 Search route storage unit 35 Search data storage unit 36 Route search control unit 40 Output device 41 Voice Output means 42 ... Display means 50 ... Operation input means

Claims (25)

[Claims] 1. A current position detecting means for detecting a current position of a vehicle, a route searching means for searching a route between two points using map data, a receiving means for receiving information from a road traffic information center, A route storage device that stores a plurality of traveling route data including a departure point, a destination, and a route therebetween stored in advance, and a storage unit that stores the destination when no destination is set. Travel route prediction means for selecting a route based on the obtained travel route data and current travel information of the vehicle, and predicting a travel route; and output means for outputting received traffic information on the predicted route. And a route search device mounted on a vehicle. 2. The route search device according to claim 1, wherein said route search device further comprises input means for inputting necessary information such as a departure point, a destination point, and a route. 3. The traveling route predicting unit further includes a coincidence computing unit that computes a coincidence, which is an evaluation value whose value increases as the traveling information and the stored traveling route data coincide with each other. The route search device according to claim 1 or 2, wherein 4. The route search device according to claim 1, further comprising a traveling route detecting unit that detects a current traveling route of the vehicle from a set of current positions obtained by the current position detecting unit. The route search device according to claim 1. 5. The route search device according to claim 1, wherein the travel information comprises a current position of the vehicle. 6. The route search device according to claim 5, wherein the travel information further includes a travel route to a current position of the vehicle, starting from a point at which the vehicle starts traveling. 7. The route search device according to claim 5, wherein the travel information further includes a day of the week at the departure point or a travel start time. 8. The route search device according to claim 6, wherein the travel information further includes a travel direction determined from the detected travel route. 9. The apparatus according to claim 5, wherein said route storage means stores and stores said travel information detected each time the vehicle travels by said route detection means as travel route data. A route search device according to claim 1. 10. The apparatus according to claim 1, wherein said route search device further comprises grouping means for grouping said traveling route data based on similarity between routes.
The route search device according to the paragraph. 11. The route search device according to claim 10, wherein said grouping means performs grouping based on a correlation between routes, which is an evaluation value of the similarity between the traveling route data. 12. The route search according to claim 11, wherein the route storage means stores, in the grouped travel route data, routes having a high mutual correlation as a same group. apparatus. 13. A route according to claim 12, wherein said route storage means ranks and stores the traveling route data classified into groups in descending order of cross-correlation. Searching device. 14. The method according to claim 12, wherein said grouped travel route data further includes departure point information comprising position information of departure points of all the routes in the same group. The route search device according to the above. 15. The method according to claim 12, wherein the grouped travel route data further includes day-of-week information including a day of a departure date of all the routes in the same group. The route search device according to the paragraph. 16. The travel route data classified into groups further include departure time information comprising departure times of all the routes in the same group.
5. The route search device according to any one of 5). 17. The method according to claim 13, wherein said traveling route prediction means performs the route prediction by using the traveling route data ranked in the higher rank in order. The route search device according to the paragraph. 18. The method according to claim 18, wherein the higher the degree of matching based on the departure point of the travel information and the departure point information of the grouped travel route data, the greater the degree of coincidence. The route search device according to any one of claims 14 to 17, characterized in that: 19. The degree-of-coincidence calculating means calculates a higher degree of coincidence as the day of the week at the travel start point of the travel information and the evaluation of the grouped travel route data based on the day-of-week information are higher. The route search device according to any one of claims 15 to 18, further comprising: 20. The coincidence calculating means further comprises calculating a higher coincidence as the traveling start time of the traveling information and the evaluation of the grouped traveling route data based on the departure time information are higher. The route search device according to any one of claims 16 to 19, characterized in that: 21. The degree of coincidence calculating means calculates a higher degree of coincidence as the evaluation of the degree of coincidence between the traveling direction of the traveling information and the direction of the destination at the starting point of the grouped traveling route data is higher. The route search device according to any one of claims 12 to 20, further comprising: 22. The route search apparatus according to claim 2, wherein said route storage means further stores the route information input by said input means as travel route data. 23. The route storage device according to claim 13, wherein the route storage device ranks and stores the traveling route data in a higher rank based on the setting from the input device. Route search device. 24. The route search means, wherein the current position is set as a start point of the route search and the destination point of the predicted route is set as a destination point of the route search, and the received traffic information from the road traffic information center is searched for the route. 24. The route search device according to claim 1, further comprising a route re-search means for searching for a minimum cost route in addition to the cost calculation of (1). 25. The route search apparatus according to claim 24, wherein said route search means starts a search of a route in response to a request signal from said input means.