JPH08271275A - Route selection apparatus - Google Patents

Abstract

PURPOSE: To provide a route selection apparatus which provides one route suitable for a driver's preference out of a plurality of recommended routes. CONSTITUTION: A communication means 1 transmits position information and receives a plurality of routes and pieces of attribute information on the respective routes. In a map storage means 3, map information which displays a map containing the plurality of routes is stored, and user information which is used to identify a user is input via an input means 2. A storage means 6 stores a plurality of kinds of functions to select a route so as to be classified by the user information. Then, a route selection means 4 receives first information and second information which are pieces of information to be reflected in the selection of a route and which indicate the degree of a preference as well as the unser information. The route selection means 4 refers to the storage content of the storage means 6, it uses a first function decided according to the user information and to the kind of the first information, it converts values of respective parameters, it uses a second function decided according to the user information and to the kind of the second information, it converts the converted values, it adds all converted values, and it displays a route having a largest added value as an optimum route.

Description

Detailed Description of the Invention

[0001]

[Field of Industrial Application] A route guidance device that selects and provides one route from a plurality of recommended routes obtained by combining a starting point and a destination on a road network, and in particular,
The present invention relates to a device that learns a preference peculiar to a driver and provides a route that best suits the preference as a recommended route.

[0002]

2. Description of the Related Art A system for selecting and providing one route from a plurality of recommended routes is disclosed in, for example, Japanese Patent Application Laid-Open No. 4-17098.

This system is a system for receiving information sent from a beacon installed on a road.
As this reception information, a plurality of recommended routes from the starting point to the destination are provided.

The recommended route has been provided by determining the route that seems to be the best in light of the criteria previously set by the driver and then giving that route to the driver as route information. Then, the driver drives the vehicle so as to reach the destination according to the route information provided.

By the way, in the conventional system, the driver can set only one criterion in advance. As the parameters used for such a criterion, for example, distance, time, traveling cost, etc. can be considered.

[0006] For example, when "distance" is selected as a parameter used as a criterion, it is a criterion that the driver presets to reach the destination with the shortest distance. , The route from which the distance from the starting point to the destination is the shortest will be presented from the system side.

Similarly, when "time" is selected as the parameter used for the judgment criterion, it is a judgment criterion preset by the driver that the driver can reach the destination in the shortest time. Of these, the system will present the route that minimizes the travel time from the departure point to the destination.

As described above, according to the conventional system, the route is presented to the driver with the route that best satisfies the only criterion previously set by the driver as the optimal route information.

[0009]

However, in the prior art as described above, it is only possible to provide a route for optimizing one parameter.

[0010] However, it is an empirical fact that, in many cases, if the route provision is not performed by considering a plurality of parameters as much as possible, the user often cannot be said to have received a satisfactory route presentation. Is about to be shown.

For example, when "distance" is selected as a parameter used as a criterion, the route can be reached at the shortest distance, but the route is
It is conceivable that the route is selected by ignoring the "travel cost". That is, although the shortest distance is taken into consideration, routes such as expressways that are cheap toll are not considered at all, and certain parameters are optimized too much, while other parameters are considered at all. It also happened frequently when a route was not selected.

Although there is a correlation between the optimization of the distance, the optimization of the time, the optimization of the traveling cost, etc., this has not been taken into consideration in the prior art.

Therefore, an object of the present invention is to realize a system which considers the mutual relations of various parameters, learns the driver's preference, and provides a route satisfying the driver's preference.

[0014]

[Means for Solving the Problems] In order to solve the above problems, the following means can be considered.

That is, the communication means for performing at least the transmission of the position information of the departure point and the destination, the plurality of routes connecting the two points, and the reception of the attribute information consisting of the plurality of parameters showing the characteristics of each route, and the plurality of the communication means. Map storage means for storing map information for displaying a map including a route, input means for inputting at least user information for identifying a user, the plurality of routes and corresponding to each route Display means for displaying at least the attribute information to be displayed, route selection means for selecting one route from a plurality of routes, and storage means for storing a plurality of types of functions for route selection for each user information.

Then, the route selection means is first information and second information which are information given through the input means to be reflected in the route selection and which are information indicating a degree of preference for each of the parameters. And the user information is received, the stored contents of the storage means are referred to, and a predetermined first function is used in accordance with the types of the user information and the first information. The value is converted, and further, the converted value is converted using a second function that is predetermined according to the types of the user information and the second information,
The device is characterized in that all the converted values are added and the route having the largest added value is displayed as the optimum route on the display means.

The storage means further includes the first information and the second information given via the input means.
The user information is accepted, and a first function predetermined according to the type of the first item and a second function predetermined according to the type of the second information are set as user information. A device having a function of storing the data in association with each other is also conceivable.

Further, the route selection means receives as inputs the first neural networks provided corresponding to the respective parameters and for realizing the first function, and the outputs from the respective first neural networks. A second neural network for realizing the second function;
An adder that adds the output values of the neural network,
The first neural network has a parameter selection unit that selects a route having the largest addition value as an optimal route, and a weighting unit that gives a weighting coefficient of neurons forming the neural network. The input layer having one neuron for inputting the value of n, a plurality of intermediate layers having a plurality of neurons, and an output layer having one neuron, The neural network is configured by including an input layer having a number of parameters for receiving a value output from an output layer of the neural network, and an output layer having a plurality of neurons, and the storage means stores the first information. Predetermined according to the type,
The weighting factors for all neurons forming the first neural network and the weighting factors predetermined for all neurons forming the second neural network, which are predetermined corresponding to the second information, are set as user information. Correspondingly stored, the weighting means refers to the contents of the storage means and assigns the weighting coefficient corresponding to the input user information to the neurons forming the first and second neural networks. A device that does this is also conceivable.

[0019]

By the communication means, the position information of the starting point and the destination is transmitted, the plurality of routes connecting the two points, and the attribute information consisting of the plurality of parameters indicating the characteristics of each route are received.

The map storage means stores map information for displaying a map including the plurality of routes, and user information for identifying the user via the input means,
First information and second information, which are information indicating the degree of preference for each parameter, are input.

The storage means stores a plurality of types of functions for route selection for each user information.

The route selecting means receives the first information, the second information and the user information given through the input means.

Further, the route selection means refers to the stored contents of the storage means, and uses a first function that is predetermined according to the types of the user information and the first information, and sets each of the parameters. The value is converted, further, the converted value is converted by using a second function predetermined according to the types of the user information and the second information, and all the converted values are added. The route with the largest added value is displayed on the display means as the optimal route.

Further, the storage means is further predetermined according to the type of the first information and the second function given through the input means and predetermined according to the type of the first information. It has a function of storing and learning the obtained second function corresponding to the user information.

[0025]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a configuration example of the route guiding device of the present invention.

The route guidance device according to this embodiment includes a communication unit 1, an input unit 2, a road map storage unit 3, a route selection unit 4, which is a main part of the present invention, a display unit 5, and a driver. And a model storage means 5. Signal lines are connected to necessary parts between the respective means, and information is transmitted through the signal lines. Reference numeral 7 is a signal line to which an external signal is input. As will be described later, the route selecting means 4 is configured to be activated when a predetermined signal is input via this signal line. There is.

The components of this embodiment will be described below.

The communication means 1 is a means for transmitting information between a vehicle-to-vehicle communication means such as a beacon system installed on a road (not shown in FIG. 1) and a route guidance device. The information to be received includes a plurality of recommended routes and attribute information for each recommended route. The attribute information includes a distance from a departure place to a destination, a running time, a degree of traffic jam, which is information indicating a degree of traffic jam, a travel charge, a difficulty of driving, a landscape, and the like.

On the other hand, as the information transmitted by the route guidance device side, the position information of the departure place and the destination to receive the presentation of the recommended route, the name of the driving vehicle, the information indicating the preference unique to the driving vehicle described later, and the recommended route The route etc. selected by us can be given.

Here, a configuration example of the communication means 1 will be described with reference to FIG.

The communication means 1 comprises a transmitter 10 and a receiver 11
And is configured.

Radio waves are usually used as the information transmission medium, and the transmitter 10 is composed of, for example, a transmitting antenna and a transmitting circuit. The receiver 11 is composed of, for example, a receiving antenna and a receiving circuit. For the transmitting and receiving antennas, for example, rod antennas or microstrip type antennas are preferably used to reduce the size of the device.

When an optical beacon is used as the vehicle-to-vehicle communication means, since the information transmission medium is light, the transmitter may be configured with a light emitting element and the receiver may be configured with a light receiving element. . In this way, the transmitter 10 and the receiver 11 may be appropriately designed in accordance with the type of information transmission medium.

The transmitter 10 receives the information to be transmitted from the route selection means 4 side and transmits the information to the vehicle-to-vehicle communication means.

On the other hand, the receiver 11 receives the information sent from the vehicle-to-vehicle communication means, and sends the received information to the route selection device 4.

Next, a configuration example of the input means 2 will be described with reference to FIG.

The input means 2 is means for receiving various commands given by the driver and inputting them to the route guidance device.

The input means 2 is composed of buttons, for example, as shown in FIG. These buttons may be incorporated in the route guidance device, but may also be realized as a wireless remote controller, for example an infrared remote controller of a TV.

In the configuration example shown in FIG. 3, a desired function is selected from a "menu" button for displaying various functions of the route guiding device on the display means 5 and various displayed functions. "Cursor" button for executing the selected function, "decision" button for executing the selected function, and "enlargement" button and "reduction" button for enlarging or reducing the road map displayed on the display means 5. are doing.

The "cursor" button moves the cursor displayed on the display means 5 by pushing the button in the direction indicated by the arrow corresponding to the button. The cursor may have various shapes such as a square shape and an arrow shape.

Next, the road map storage means 3 is means for storing a so-called digital road map, and by reading the road map information stored in the means,
A map can be displayed on the display means 5. This means can be realized by a CD-ROM or a hard disk device in which digital road map information is stored.

Here, the road map information stored in the road map storage means 3 is represented by a link table as shown in FIG. 4, and is related to roads (hereinafter referred to as "links") and nodes. It will be represented by information related to a point where roads intersect each other (hereinafter, referred to as “node”) represented by a table.

In addition to this, various facilities such as buildings and parks,
There is also position information, name data, etc. of railways, water systems, etc., and these information must be stored in the road map storage means 3, but since they are not related to the essence of the present invention, detailed description thereof is omitted here. To do.

Now, as shown in FIG. 4, the node table stores the node numbers and their coordinates.
The link table stores link numbers and start and end nodes of the links. The node numbers and link numbers are set so that they do not overlap each other in order to display the map accurately.

The map database shown in FIG. 4 shows a basic structure. In addition to this data, information such as the node adjacent to the node, right turn propriety information, link length, link width, etc. It goes without saying that each table can be added to form each table.

Next, a situation in which a road map is actually displayed using the information stored in the node table and the link table shown in FIG. 4 will be slightly explained by taking the road map of FIG. 5 as an example.

The following processing may be performed by the display means 5.

The node numbers “15” and “34” stored in the node table shown in FIG. 4 are the nodes indicated by the black circles “●” in FIG. 5 from the coordinate values corresponding to them. I understand.

The link number "2" in the link table
Looking at "1", it can be seen that the road is from node 15 to node 34. If this road is not a one-way street,
There is also a road from the node 34 to the node 15, and when the link table is referred to, the link number 5
You can see that 5 is attached. In the route guidance described later, the route information is provided in the link number string. For example, the display means 5 sequentially displays the provided link number sequence in a specific color, so that the driving vehicle travels to the destination in consideration of the displayed route. .

The display means 5 refers to the stored contents of the road map storage means 3 to display a road map, and the input means 2
Has a function of displaying various menus requested by operating the “menu” button, “enter” button, and the like. For example, a display unit including a liquid crystal display, a CPU, a ROM (predetermined program Built-in), and a display control unit configured by an electronic device such as a RAM. FIG. 6 is a display example in which the road map data (node table) stored in the road map storage means 3 is read, and the coordinates of each node are connected and displayed on the display means 5.

In FIG. 6, the white circle “◯” represents the current position of the vehicle, and the line represents the road.

The current position of the vehicle can be detected by a navigation system equipped with a GPS device or the like. It is conceivable that this navigation system is configured to be built in the route guidance device or configured to be attached to the route guidance device.

In FIG. 6, roads are indicated by lines for simplification. This road map display is preferably configured to be executed at the same time when the power of the route guidance device is turned on, for example. In addition, “enlargement” included in the input unit 2
Alternatively, when the "reduce" button is pressed, the display means 5 performs display processing so that the road map is enlarged or reduced centering on the position of the vehicle.

Next, the driver model storage means 6 and the route selection means 4, which are the main parts of the present invention, will be described.

The driver model storage means 6 is means for storing the model of the driver learned by the route selection means 4.

In the following, the driver's model will be represented by synapse weights for the neurons constituting the neural network. Therefore, the driving vehicle model storage unit 6 stores all the information on the synapse weights for the neurons, peculiar to the driving vehicle.

Further, the driver model storage means 6 stores the driving vehicle models for a plurality of drivers, and the route selecting means 4 selects the required driving vehicle model for the driver to obtain synapse weight information. Read out.

The route selecting means 4 is the main part of the present invention, and will be described below.

First, a configuration example of the route selecting means 4 will be described with reference to FIG.

As shown in FIG. 7, the route selecting means 4 has C
It has a PU 40, a RAM 41, and a ROM 42, and the respective constituent elements are connected by a signal line 43.

The ROM 42 stores a control program for the route guiding device and software for realizing the neural network. Therefore, information such as the number of layers and the number of neurons in each layer, which is the configuration information of the neural network, is also stored.

Attribute information is used as input to the neural network, and route selection and driver model learning are performed by
As described above, the driver model is realized by the neural network, and is realized by the synapse weight for the neurons constituting the neural network as described above.

The RAM 41 temporarily stores the synapse weights read from the driver model storage means 6 and variables used by the control program.

The CPU 40 processes an instruction or request input via the input means 2 according to a predetermined control program. Also, although not shown,
The communication unit 1, the input unit 2, the road map storage unit 3, the display unit 5, and the driver model storage unit 6 are connected to the CPU via a signal line 43 and necessary information is transmitted.

The detailed structure of the above-mentioned neural network will be described later.

Now, the operation of the route selecting means 4 will be roughly described below.

First, when the driver's name is input through the input means 2, the driving vehicle model corresponding to the driver is read from the driver model storage means 6. If the driver's name is not registered, the process proceeds to the learning process of the driver's model.

Next, the destination (destination) and the preference are designated from the input means 2. The “preference” is information that determines how to use attribute information for selecting a route to select an optimal route for a driving vehicle. Specifically, for example, a driver who prefers a route with a short "distance" to a destination specifies a preference that "the route is" very good "if the" distance "is" very short ". It will be.
And this is equivalent to modeling the preference of a driving car.

The operation of the neural network for learning and selecting a route in response to such preference designation will be described later in detail.

Next, the route selection means 4 transmits the destination to the vehicle communication means via the communication means 1 and waits for the recommended route to be transmitted from the vehicle communication means. Then, when a plurality of recommended routes are transmitted, a route suitable for the driver's preference is selected using the read driver's model, and the selected route is displayed on the display means 5.

In this embodiment, the recommended route information transmitted by the vehicle-to-vehicle communication means includes the above-mentioned link string to the destination, the distance to the destination, the travel time, the congestion level, the fee, the difficulty of driving, It shall consist of attribute information such as landscape.

When the attribute information is not transmitted together, the traffic information such as travel time and congestion degree transmitted by the vehicle-to-vehicle communication means and the road map information stored in the road map storage means are separately provided. It is also possible to use it to create attribute information, and the processing unit for performing such processing is composed of a CPU, a ROM (which stores a predetermined program), a RAM, etc. It may be built in the selection means 4.

Next, FIG. 10 shows a configuration example of the neural network according to the present invention.

FIG. 21 shows a concrete example of the structure expressing the network.

This neural network comprises a preprocessing unit 50 and first neural networks 60-1 to 60-1.
-6, the second neural network 70, and the adder 80. These can be realized by software.

The preprocessing unit 50 multiplies and outputs the six types of attribute information such as distance, travel time, congestion level, fee, running difficulty, and scenery, and the preference setting value corresponding to the driver's preference. To do.

This preference setting value can be set uniquely to the driving vehicle, and the set value may be stored in the driving vehicle model storage means 6.

Although each type of attribute information is set to six types for the sake of explanation, it goes without saying that the attribute information may be of any type as long as it is a plurality of types.

The first neural network 60-1 to 60-1
60-6 converts and outputs the output of the preprocessing unit 50,
This is a one-input, one-output neural network. Six neural networks are provided for each attribute information.

The first neural network 60-1 to 60-1
As shown in FIG. 21, the specific configuration of 60-6 is a four-layer neural network including one input layer, two intermediate layers, and one output layer.

Then, as shown in FIG. 21, the input layer is
One neuron, each of the first and second intermediate layers, has three neurons, and the output layer has one neuron. 1 for input layer and output layer
Since the learning data is composed of two neurons, the learning data is “x” and the training data is “y”. When learning is performed by this neural network, an arbitrary function y = f (x) is represented by a synapse weight (represented by a broken line in FIG. 21). Can be stored as). This synapse weight is the driver model storage means 6
Is stored.

In addition, the second neural network 70
Is the first neural network 60-1 to 60-6.
The evaluation value for the route is determined by using the output of
This is a neural network with one input and one output.

The specific structure will be described with reference to FIG. 21. The second neural network 70 is a two-layer neural network, with 6 neurons in the input layer and 41 neurons in the output layer. It is configured with. Since this neural network has two layers, the input value is output as the output value in proportion to the synapse weight shown by the alternate long and short dash line.

A membership function described later is stored as this synapse weight.

The adder 80 adds the values output by 41 neurons included in the output layer of the second neural network 70, and outputs the added value as an evaluation value.

As shown in FIG. 21, the adder 80
May be composed of one neuron having a synapse weight of “1”.

The driver model is expressed as synapse weights of the first neural networks 60-1 to 60-6 and the second neural network 70 and stored in the driver model storage means 6.

The usage of the route guiding device and the corresponding operation will be described below.

First, in order to use the route guidance device,
It is necessary to display the menu and determine the operation item for operating the device.

The menu display is displayed as a list on the display means 5 by pressing the "menu" button provided in the input means 2.

FIG. 8 is a diagram showing a menu display example, in which "driver setting" and "route guidance" are displayed as selectable items.

Character color (color of the character "Set driving car name")
The item whose background color is reversed indicates that it is the currently selected item, and the process corresponding to the item is executed by pressing the “OK” button provided in the input means 2.

By pressing the cursor button provided with the arrow pointing vertically in the input means 2, the item in which the character color and the background color are reversed can be changed. That is, the item to be selected can be changed by operating the cursor button.

It should be noted that, instead of making the item color selectable by inverting the character color and the background color, for example, a cursor such as an arrow is displayed beside the item, and the cursor is moved by operating the cursor button. A method of selecting is also possible. By such an operation, an item is selected.

The operation when either the setting of the driving vehicle name or the route guidance is selected will be described below, focusing on the operation of the neural network.

First, a process for selecting a driving vehicle name setting item and determining a driving vehicle name will be described.

(1) Driver Name Determination Processing When the item "driver setting" is selected and the determination button is pressed, a menu as shown in FIG. 9 is displayed next. In FIG. 9, “A” to “D” are names of drivers who are already registered.

The driver presses the "cursor" button provided on the input means 2 to select his / her name and presses the "enter" button. When the driver name is determined, the CPU 40 reads the driving model for the driver from the driver model storage means 6 and stores it in the RAM 41.

If a new driving vehicle name is registered, that is, if "undefined" is selected, a description will be given later.
Newly model the driver.

Here, the driver name is set by the input means 2
I am trying to do it through, but like an electronic key,
When the identification code of the driver's name can be registered, when the electronic key is attached to a predetermined part of the vehicle and the engine is started,
It is also conceivable that the engine controller mounted on the vehicle receives the trigger signal and the driver identification code (information transmitted by the signal line 7 in FIG. 1) and sends them to the route selecting means 4. Accordingly, it is also preferable that the route selecting means 4 obtains information corresponding to the name of the driving vehicle from the driving vehicle model storage means 6 and uses the obtained information to perform the route selection.

In FIG. 1, the trigger signal and the driver identification code are input to the route selecting means 4 via the signal line 7. Although not shown, these pieces of information are connected to the signal line 43 shown in FIG. 4 and input to the CPU 40.

When the trigger signal is input, the CPU 40 collates the driver identification code with the registered driver name and reads out the model of the matching driver. If they do not match, the driver is newly modeled.

Next, the modeling of the driver will be described.

The “modeling of the driver” described so far means the first neural network 60 shown in FIG.
-1 to 60-6 and determining the synapse weight for each neuron included in the second neural network 70.

In order to determine the synapse weight, a model determination screen as shown in FIG. 11 is displayed on the display means 5, and the input means 2 is used to move the "cursor" to select the desired item. , By pressing the "OK" button, the rule sentence is completed and the model decision is prepared.

In FIG. 11, as an example, a rule sentence "If" congestion "is" very "and" heavy ", the route is" very bad "is created.

Rule texts are similarly completed for other attribute information such as distance, travel time, fee, difficulty in running, and landscape.

Although all selectable items are displayed in FIG. 11, for example, ““ distance ”is“ very ””
If it is "intense", the rule sentence "The route is very bad." Is inappropriate for Japanese, so "distance"
If is selected, it may be possible to forcibly prevent the items such as “intense”, “high”, and “good” from being displayed.

Next, after completing the rule sentences for all the attribute information, when the "decision" button provided in the input means 2 is pressed again, the CPU 40 displays the driver name registration screen as shown in FIG. It is displayed on the display means 5.

Using the registration screen of FIG. 12, while inputting the driver's name in hiragana by operating the cursor buttons, select "Register" after converting the Chinese characters (the cursor for selecting the driver's name and registration is (Not shown), when the "OK" button is pressed, the CPU 40 causes the synapse weights for the neurons included in the first neural networks 60-1 to 60-6 and the second neural network 70 shown in FIG. The driver name is stored in the driver model storage means 6
To be stored.

On the contrary, when the driver name is set, each of the first neural networks 60-1 to 60-6 and the second neural network 70 stored corresponding to the driver name is provided. The synaptic weight for the neuron is read.

The image display process and the process corresponding to the input in FIGS. 11 and 12 are executed by the CPU 40 in the ROM 42.
The device may be configured so as to perform according to a program stored in advance in.

Now, the first neural network 60
The synapse weight for each neuron included in -1 to 60-6 and the second neural network 70 is determined as follows.

First, the first neural network 60
Since -1 to 60-6 are 1-input 1-output neural networks, their input / output relationships can be expressed by the functions shown in FIG.

FIG. 13 shows three function examples, but it goes without saying that the function examples used are not limited to this.

The first neural network 60-1 to 60-1
For 60-6, one of the three functions is selected and associated and registered. This registration process is a process of creating a rule statement for each attribute information described above.
It is determined by which item is selected from the items "very", "average", and "slightly".

For example, when "very" is selected, "function 1" and "average" are selected, when "function 2" and "little" are selected, and when "function" is selected. It may be decided to adopt “function 3”.

Therefore, using the input / output relationship of each function as the teacher data, the first neural networks 60-1 to 60-60
If the function to be adopted is determined by the user's selection operation, the synapse weight stored in the ROM 42 is read out by performing learning using -6 in advance and storing the synapse weight in the ROM 42. The driver model storage means 6 may be registered in association with the name of the driving vehicle.

It should be noted that “average” may be automatically selected for the attribute information for which no rule sentence is created.

The range of the input values shown in FIG. 13 is
Although it is 0 to 1.0, in order to adopt the function shown in FIG. 13, the input value may be divided by the maximum value in each attribute information. For example, when the maximum value of “distance” is “45 (Km)” in a plurality of recommended routes, for example, when “35 (Km)” is given as “distance”, the first neural network The value entered in should be "35/45 = 0.77". The same applies to other attribute information such as "travel time".

Next, the second neural network 70
The synapse weight for the neuron included in is determined according to the concept of fuzzy reasoning. This will be described below.

“Very good” in the above rule sentence,
As shown in FIG. 14, “good”, “normal”, “bad”, and “very bad” are expressed by membership functions in fuzzy reasoning.

In FIG. 14, the horizontal axis is the input axis, and the horizontal axis corresponds to each neuron included in the output layer of the second neural network 70, as shown in FIG.

FIG. 15 shows the configuration of the second neural network 70. It has an input layer having six neurons and output layer neurons A1 to A41.
It is composed of two layers, an output layer having 41 neurons.

At this time, the membership function and the synapse weight of the second neural network 70 shown in FIG. 15 are made to correspond as follows.

That is, the output layer neurons A1, A2, A3, ..., A of the second neural network 70.
39, A40, and A41 are the membership functions (see FIG.
4) A neuron that outputs a value of "-1.0" on the input axis, a neuron that outputs a value of "-0.95", a "-"
A neuron that outputs a value of "0.9", ..., "0.
A neuron that outputs a value of "9", a neuron that outputs a value of "0.95", and a neuron that outputs a value of "1.0".

It is assumed that a rule sentence is set that "if" distance "is" very "" short "on the driver model setting screen of FIG. 11, the route is" very good ".

From among the membership functions of FIG. 14,
When the input value of “1.0” is input to the neuron in the input layer corresponding to “distance” by extracting the function “very good”, as shown in FIG. 22, “very good” is obtained. Determines the synapse weights so that the following function is generated by the value output from each neuron in the output layer.

Then, "very good", "good", "normal", "bad", and "very bad" specified in the rule sentence.
The synapse weight learned in advance so as to output the membership function corresponding to is read from the ROM 42 and registered in the driver model storage means 6 in association with the name of the driving vehicle.

In the configuration example of the second neural network 70 shown in FIG. 15, the number of output layer neurons is “41”, which is the range of values on the input axis “−1.0 to 1”. This is because ".0" is divided into "0.05". When this division unit is "0.1", the number of output layer neurons is "21", and when it is "0.01", the number of output layer neurons is "201". You can do this.

Secondly, the operation when the route guidance is selected as the menu and the operation of the apparatus will be described.

(2) Route Guidance When the item "Route Guidance" is selected on the menu screen shown in FIG. 8, the CPU 40 executes a process relating to route guidance according to a predetermined program. Then, a menu screen as shown in FIG. 16 is displayed.

The item "destination setting" is entered via the input means 2.
When you select, a white arrow is displayed as shown in the upper right of FIG. 17, move this arrow by operating the cursor button, and press the “OK” button at the destination position, the destination will be It is determined.

Then, when the "OK" button is pressed,
The CPU 40 transmits destination information to the vehicle-to-vehicle communication means via the communication means 1. At this time, the driver's preference can be set.

That is, when the item "preference setting" is selected on the screen shown in FIG. 16, the screen shown in FIG. 18 is displayed. This screen is used to determine which of the attribute information for the route should be emphasized, and whether it should be considered "important", "normal", "non-critical", or not considered at all. Do not consider "no consideration",
Operate the cursor button to select, and operate the enter button to confirm.

These pieces of information define the value of the "preference setting value" input to the preprocessing section shown in FIG.

In FIG. 18, "distance" is "normal", "travel time" is "important", "congestion level" is "normal", "fare" is "unimportant", "difficult to drive" and ""Landscape" is selected for "Landscape". These "important",
The setting values corresponding to "normal", "unimportant", and "no consideration" are input to the preprocessing unit shown in FIG. 10 and are multiplied by the route attribute information.

As the set value, for example, "important" is set to "1.
You can set "0", "normal" to "0.7", "unimportant" to "0.3", and "no consideration" to "0.0".

For example, when the “distance” is “normal” and the distance information is “45 km”, the value “45 × 0.7” is input to the input layer of the first neural network 60-1. Will be done.

Now, a method of selecting one route from a plurality of recommended routes to the destination will be described.

When the communication means 1 receives the information of a plurality of recommended routes, the CPU 40 inputs the attribute information associated with each recommended route into the preprocessing unit 50 shown in FIG.

The preprocessing section 50 outputs a value obtained by multiplying the attribute information and the set "preference setting value" to the first neural networks 60-1 to 60-6, respectively.

The first neural network 60-1 to 60-1
Reference numeral 60-6 converts the input data using the function shown in FIG. 13, and then outputs the converted value to the second neural network 70.

The second neural network 70 refers to the membership function and outputs a value corresponding to each output value of the first neural networks 60-1 to 60-6.

Then, the adder 80 sums the values of the membership function output by the second neural network 70 and outputs the sum as an evaluation value. This evaluation value is the optimum route in which the largest route reflects the driver's preference. The above process is performed for each of the plurality of recommended routes.

Finally, the CPU 40 compares the evaluation values of the recommended routes, selects the route with the largest evaluation value, and displays the route and the corresponding attribute information on the display means 5.

At this time, other recommended routes received by the communication means 1 and their attribute information are also displayed at the same time to inform the driver. This facilitates the change of the route of the driving vehicle even when the mental state changes such as selecting a route other than that determined by the device.

FIG. 19 shows an example of a screen displaying a route.

FIG. 19 shows a state in which “route 1” having the shortest “travel time” set as “important” in FIG. 18 is selected. Further, the route 2 which is another recommended route is also displayed together with its attribute information.

At this time, the driver has the same "travel time", "distance", and "congestion degree", and the "charge" is low,
When it is determined that the route 2 is good, the "cursor" button provided in the input means 2 is operated to select the route 2 and the "decision" button is pressed. This allows the driver to appropriately select a route different from the route recommended by the route selecting means 4.

However, when the driver selects a route different from the route recommended by the route selecting means 4, the route selecting means 4 judges that the driver's preference has changed, and learns the driver model. .

In learning the driver model, the recommended values recommended by the route selection means 4 (“route 1” in FIG. 19) and the evaluation values of the route selected by the driver (“route 2” in FIG. 19) are exchanged. Then, using the attribute information of the path as the “input signal” and the replaced evaluation value as the “teacher signal”, the first neural network shown in FIG. A process of changing the synapse weight for each neuron included in the networks 60-1 to 60-6 and the second neural network 70 is performed.

In the above-described embodiment, only the destination is transmitted to the vehicle-to-vehicle communication means. However, not only the destination but also the preference set by the driver is transmitted by the communication means 1 to select the route.
It is also possible to carry out at the vehicle-to-vehicle communication means installed on the road or at the central center in which a central processing unit for controlling the vehicle-to-vehicle communication means is installed. That is, the route selecting device is provided in the central center, and the route selected in consideration of the preference of each driving vehicle can be provided to the driving vehicle. As a result, the device mounted on each vehicle is miniaturized.

In this case, the communication means 1 receives the vehicle-to-vehicle communication means or the central processing unit for controlling the vehicle-to-vehicle communication means, the route selected based on the driver's preference, the attribute information and the evaluation value of the route. Then, the route selection unit 4 may display the route having the highest evaluation value on the display unit 5.

FIG. 20 shows another device configuration example according to the present invention.

In FIG. 20, the driver model storage means 6 which is a component of the route guidance device shown in FIG. 1 is a rewritable ROM in which the ROM 42 provided in the route selection means 4 of the route guidance device shown in FIG. 10 is rewritable. This is a configuration realized by that.

That is, the driver model stored in the driver model storage means 6 is stored in the rewritable ROM, thereby simplifying the configuration. With such a configuration, further downsizing of the device can be realized.

As described above, according to the present invention,
It has become possible to realize a device that provides one route suitable for the learned driver's preference from a plurality of recommended routes.

[0160]

According to the present invention, the driver's preference regarding route selection can be learned in an interactive manner with the driver.
Then, it is possible to provide one route suitable for the learned driver's preference from a plurality of recommended routes calculated for the combination of the starting point and the destination on the road network.

Further, the learned driver's preference can be stored in the storage means, and the route guidance device recognizes the driver to determine the route in consideration of the driver's unique preference. Can be provided to.

The learning is performed by appropriately combining the neural network realized by software and the fuzzy concept, and the device configuration is simplified without increasing the device configuration.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration example of a route guiding device according to the present invention.

FIG. 2 is a configuration diagram showing a configuration example of a communication unit.

FIG. 3 is a configuration diagram showing a configuration example of input means.

FIG. 4 is an explanatory diagram of a configuration example of a road map database.

FIG. 5 is an explanatory diagram showing an example of a road map.

FIG. 6 is an explanatory diagram of a display example of a road map on a display means.

FIG. 7 is a configuration diagram showing a configuration example of a route selection unit.

FIG. 8 is an explanatory diagram of a menu display example.

FIG. 9 is an explanatory diagram of an example of a driver name setting screen.

FIG. 10 is a configuration diagram showing a configuration example of a neural network.

FIG. 11 is an explanatory diagram of an example of a driver model setting screen.

FIG. 12 is an explanatory diagram of an example of a driver name registration screen.

FIG. 13 is an explanatory diagram of the input / output relationship of the first neural network.

FIG. 14 is an explanatory diagram of an example of synapse weights of a second neural network.

FIG. 15 is a configuration diagram showing a configuration example of a second neural network.

FIG. 16 is an explanatory diagram of a menu display example.

FIG. 17 is an explanatory diagram of an example of a screen in which a destination is set.

FIG. 18 is an explanatory diagram of an example of a driver preference setting screen.

FIG. 19 is an explanatory diagram of an example of a route display screen.

FIG. 20 is a configuration diagram showing another configuration example of the route guiding device according to the present invention.

FIG. 21 is a configuration diagram showing a specific configuration example of a neural network.

FIG. 22 is a configuration diagram showing a configuration example of a neural network that outputs a membership function.

[Explanation of symbols]

1 ... Communication means, 2 ... Input means, 3 ... Road map storage means,
4 ... Route selection means, 5 ... Display means, 6 ... Driver model storage means, 50 ... Preprocessing section, 60 ... First neural network, 70 ... Second neural network, 80
… Addition section

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G09B 29/10 9194-5L G06F 15/40 370C (72) Inventor Shigeo Abe Omikamachi, Hitachi City, Ibaraki Prefecture 7-1-1, Hitachi Ltd., Hitachi Research Laboratory (72) Inventor Yoshihiko Miyamoto 7-1-1, Omika-cho, Hitachi City, Ibaraki Hitachi Ltd. Hitachi Research Laboratory

Claims (7)

[Claims] 1. A communication means for at least receiving position information of a departure place and a destination, a plurality of routes connecting two points, and a plurality of parameters showing a characteristic of each route, and a plurality of communication means. Map storage means for storing map information for displaying a map including a route, input means for inputting at least user information for identifying a user, the plurality of routes and corresponding to each route Display means for displaying at least the attribute information to be displayed, and one from a plurality of routes.
Route selection means to select one route, and for each user information,
Storage means for storing a plurality of types of functions for route selection, wherein the route selection means is information given via the input means to be reflected in the route selection, and a degree of preference for each of the parameters. The first information, the second information, and the user information, which are information indicating that the user information and the user information are received, and the contents stored in the storage unit are referred to, and are determined in advance according to the types of the user information and the first information. The value of each of the parameters is converted using the first function, and the converted value is converted using the second function that is predetermined according to the types of the user information and the second information. The route selection device is characterized in that all the converted values are added and the route having the largest added value is displayed as the optimum route on the display means. 2. The storage means according to claim 1, further comprising: the first information, which is given via the input means.
The second information and the user information are accepted, and a first function predetermined according to the type of the first item and a second function predetermined according to the type of the second information are received. A route selection device having a function of storing the user information in association with the user information. 3. The route selecting means according to claim 1, wherein the route selecting means includes a first neural network for realizing the first function, which is provided corresponding to each of the parameters, and each of the first neural networks.
A second neural network for realizing the second function, which receives the output from the neural network as an input, an adder unit for adding the output values of the second neural network, and a path having the largest added value are optimized. It has a selection unit for selecting as a path and a weighting means for giving a weighting coefficient of the neurons constituting the neural network, and each of the first neural networks has one neuron for inputting the value of the parameter. The second neural network is configured to have an input layer, a plurality of intermediate layers having a plurality of neurons, and an output layer having a single neuron, and the second neural network is a value output from the output layer of the first neural network. An input layer that has as many neurons as the number of parameters that accepts An output layer that is provided, and the storage means stores the weighting coefficients for all the neurons that make up the first neural network, which are predetermined in correspondence with the type of the first information, and the second The weighting coefficient, which is predetermined corresponding to the information, for all the neurons forming the second neural network is stored in correspondence with the user information, and the weighting means stores the contents of the storage means. With reference to the above, a route selection device characterized in that a weighting coefficient corresponding to the input user information is given to the neurons constituting the first and second neural networks. 4. A mounting means for mounting a storage medium storing identification information for identifying a user, and a means for reading out the identification information stored in the storage medium according to any one of claims 1, 2 and 3. The route selection means further refers to the user information and the identification information, and when it is determined that the users are the same, acquires the information corresponding to the user from the storage means. Then, the route selecting device is characterized by activating its own means. 5. The membership function according to claim 1, wherein the function predetermined according to the types of the user information and the second information is a membership function expressing an input / output relationship by a plurality of linear patterns. A route selection device characterized by being present. 6. A configuration comprising: a parent device that selects one of a plurality of routes and provides information on the selected route; and a child device that receives the route information from the parent device. A route selection device, wherein the child device is a map storage unit that stores map information for displaying a map including the route, a destination, user information, and information to be reflected in route selection. , Input means for inputting at least first information and second information, which are information indicating the degree of preference for a plurality of parameters indicating the characteristics of the route, the selected route, and a plurality of indicating the features of the route The child side communication means for receiving the attribute information consisting of the parameters and transmitting the information input by the input means, and the display means for displaying at least the attribute information and the map information including the route. Is configured, furthermore, the parent device includes a path selection means for selecting one path from a plurality of paths,
And a parent communication unit that receives the information transmitted by the child communication unit and transmits the selected route, and a storage unit that stores a plurality of types of route selection functions for each user information. The means receives the first information, the second information, and the user information, which are the information received by the parent communication means, refers to the stored contents of the storage means, and determines the types of the user information and the first information. Accordingly, the value of each of the parameters is converted using a predetermined first function, and further, using a second function predetermined according to the types of the user information and the second information, A route selection device, wherein the converted value is converted, all converted values are added, and a route having the largest added value is sent to the parent side communication means as an optimum route. 7. The route according to claim 1, further comprising display control means, wherein the display control means displays the first information and the second information on the display means. Selection device.