Route search device, control method, program, and storage medium

Abstract

To provide a route search device capable of performing route searching that appropriately takes into account individual differences among drivers. [Solution] Server device 1 receives profile information about the driver operating the mobile vehicle, as well as information about the vehicle's departure point and destination, from terminal device 2. For each of the roads constituting the multiple routes from the departure point to the destination, server device 1 calculates a driving difficulty evaluation value based on a driving evaluation model corresponding to the profile information. Then, server device 1 determines a recommended route based on the calculated driving difficulty evaluation value.

Description

Technical Field

[0004] ,

[0001] The present invention relates to path search technology.

Background Art

[0002] Conventionally, a system has been known in which data is aggregated from a plurality of vehicles to a server device, and information regarding a learning result obtained by the server device through machine learning is distributed. For example, Patent Document 1 discloses a system that collects situations such as the weather and time zone during driving of each vehicle as environmental data and learns for each situation indicated by the environmental data.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Regarding situation recognition and situation judgment during driving, there are individual differences based on the experience of the driver, etc., and the recognition and judgment may vary depending on the driver. Therefore, even in path search, the route suitable for driving may vary depending on the driver. Patent Document 1 does not mention anything about the points that should consider such individual differences of the driver.

[0005] The present invention has been made to solve the above problems, and a main object thereof is to provide a path search device capable of performing path search that preferably takes into account individual differences of the driver.

Means for Solving the Problems

[0006] The invention described in claim 1 is a route search device comprising: a first acquisition unit that acquires driver information relating to a driver operating a mobile object, and information on the departure point and destination of the mobile object; a second acquisition unit that acquires evaluation information output from a driving evaluation model by inputting captured images corresponding to each of the roads constituting a plurality of routes from the departure point to the destination into a driving evaluation model corresponding to the driver information; and a route determination unit that determines a recommended route from the plurality of routes based on the evaluation information. The invention described in claim 5 is a route search device comprising: a first acquisition unit that acquires driver information relating to a driver operating a moving object, and information on the departure point and destination of the moving object; a second acquisition unit that acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting a plurality of routes from the departure point to the destination; and a route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, wherein the driving evaluation model is a driving evaluation model in a suitable group that matches the driver information, from driving evaluation models learned for each group of judges classified according to the commonality of the profiles, based on captured images of the area around the moving object and evaluation information relating to driving evaluated by judges with profiles based on the captured images. The invention described in claim 10 is a route search device comprising: a first acquisition unit that acquires driver information relating to a driver operating a mobile object, and information on the departure point and destination of the mobile object; a second acquisition unit that acquires captured images taken on each of the roads constituting a plurality of routes from the departure point to the destination; a third acquisition unit that acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting a plurality of routes from the departure point to the destination; and a route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, wherein the third acquisition unit acquires evaluation information for each of the roads by inputting the captured images acquired by the second acquisition unit into the driving evaluation model.

[0007] The invention described in claim 14 is a control method performed by a route search device, comprising: a first acquisition step of acquiring driver information relating to a driver operating a mobile body, and information on the departure point and destination of the mobile body; a second acquisition step of acquiring evaluation information output from a driving evaluation model by inputting captured images corresponding to each of the roads constituting a plurality of routes from the departure point to the destination into a driving evaluation model corresponding to the driver information; and a route determination step of determining a recommended route from the plurality of routes based on the evaluation information. The invention described in claim 16 is a control method performed by a route search device, comprising: a first acquisition step of acquiring driver information relating to a driver operating a moving object, and information on the departure point and destination of the moving object; a second acquisition step of acquiring evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting a plurality of routes from the departure point to the destination; and a route determination step of determining a recommended route from the plurality of routes based on the evaluation information, wherein the driving evaluation model is a driving evaluation model in a suitable group that matches the driver information, from driving evaluation models learned for each group of judges classified according to the commonality of the profiles, based on captured images of the area around the moving object and evaluation information relating to driving evaluated by judges with profiles based on the captured images.

[0008] The invention described in claim 15 is a program executed by a computer, comprising: a first acquisition unit that acquires driver information relating to a driver operating a mobile object, and information on the departure point and destination of the mobile object; a second acquisition unit that acquires evaluation information output from a driving evaluation model by inputting captured images corresponding to each of the roads constituting a plurality of routes from the departure point to the destination into a driving evaluation model corresponding to the driver information; and a route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, thereby causing the computer to function as such. The invention described in claim 17 is a computer program that includes a first acquisition unit that acquires driver information relating to a driver operating a mobile object, and information on the departure point and destination of the mobile object; a second acquisition unit that acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting a plurality of routes from the departure point to the destination; and a route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, wherein the driving evaluation model is a driving evaluation model in a suitable group that matches the driver information, based on captured images of the area around the mobile object and evaluation information relating to driving evaluated by a judge with a profile based on the captured images, and learned for each group of judges classified by the commonality of the profiles. [Brief explanation of the drawing]

[0009] [Figure 1] This is a schematic configuration of the route search system. [Figure 2] This shows the block configuration of the server and terminal devices. [Figure 3] This is a correspondence table between the identification number of the driving evaluation model and each item representing the profile of the evaluator corresponding to that driving evaluation model. [Figure 4] This is an example of a flowchart illustrating the steps involved in the pathfinding process. [Figure 5] This is a schematic configuration of the route search system in a modified form. [Figure 6] This is a schematic configuration of the route search system in a modified form. [Modes for carrying out the invention]

[0010] According to a preferred embodiment of the present invention, the route search device includes: a first acquisition unit that acquires driver information relating to a driver operating a moving object, and information on the departure point and destination of the moving object; a second acquisition unit that acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting a plurality of routes from the departure point to the destination; and a route determination unit that determines a recommended route from the plurality of routes based on the evaluation information.

[0011] The above-described route search device comprises a first acquisition unit, a second acquisition unit, and a route determination unit. The first acquisition unit acquires driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle. The second acquisition unit acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting a plurality of routes from the departure point to the destination. The route determination unit determines a recommended route from the plurality of routes based on the evaluation information acquired by the second acquisition unit. In this embodiment, since the route search device performs route searching based on evaluation information of a driving evaluation model suitable for the driver of the mobile vehicle, it can perform route searching that suitably takes into account individual differences among drivers.

[0012] In one embodiment of the above-described route search device, the driving evaluation model is a driving evaluation model in a suitable group that matches the driver information, selected from driving evaluation models learned for each group of judges classified according to the commonality of their profiles, based on captured images of the surroundings of a moving object and evaluation information about driving evaluated by judges with profiles based on the captured images. Here, "suitable group" refers to a group of judges classified according to the commonality of their profiles that matches the acquired driver information. According to this embodiment, the route search device can suitably acquire evaluation information for each road necessary for route search based on a driving evaluation model that matches the driver characteristics indicated by the acquired driver information.

[0013] In another embodiment of the route search device described above, the route search device further comprises a third acquisition unit that acquires captured images taken at each of the roads constituting the plurality of routes, and the second acquisition unit inputs the captured images acquired by the third acquisition unit into the driving evaluation model to acquire evaluation information for each of the roads. In this embodiment, the route search device can acquire captured images taken at each road constituting a candidate route and suitably calculate evaluation information corresponding to each road necessary for route search.

[0014] In another embodiment of the route search device described above, the route search device further includes a storage unit that stores map data associated with evaluation information based on multiple driving evaluation models for each road, and the second acquisition unit acquires evaluation information based on the driving evaluation model corresponding to the driver information for each of the roads from the map data. In this embodiment, the route search device can suitably acquire evaluation information corresponding to each road necessary for route search by referring to the map data.

[0015] In another embodiment of the route search device described above, the route search device further comprises a transmitting unit that transmits the recommended route information to a terminal device that moves with the mobile body, and the first acquisition unit receives the driver information and the departure and destination information from the terminal device. In this embodiment, the route search device can suitably supply the terminal device that provides route guidance for the mobile body with recommended route information suitable for the driver of the mobile body.

[0016] In another embodiment of the route search device described above, the second acquisition unit transmits the driver information to a server device that stores a plurality of driving evaluation models, thereby acquiring a driving evaluation model corresponding to the driver information, and calculates evaluation information for each of the roads based on the driving evaluation model. In this embodiment, the route search device can acquire evaluation information necessary for route search and determine a recommended route based on a driving evaluation model suitable for the driver received from the server device. In a preferred example, the evaluation information may be information indicating the difficulty of driving or information indicating the quality of visibility.

[0017] According to another preferred embodiment of the present invention, there is provided a control method executed by a route search device, comprising: a first acquisition step of acquiring driver information regarding a driver who drives a moving body and information on a departure place and a destination of the moving body; a second acquisition step of acquiring evaluation information based on a driving evaluation model corresponding to the driver information for each road constituting a plurality of routes from the departure place to the destination; and a route determination step of determining a recommended route from the plurality of routes based on the evaluation information. By executing this control method, the route search device can execute a route search that preferably takes into account individual differences of the driver.

[0018] According to another preferred embodiment of the present invention, there is provided a program executed by a computer, comprising: a first acquisition unit that acquires driver information regarding a driver who drives a moving body and information on a departure place and a destination of the moving body; a second acquisition unit that acquires evaluation information based on a driving evaluation model corresponding to the driver information for each road constituting a plurality of routes from the departure place to the destination; and a route determination unit that functions the computer as a unit for determining a recommended route from the plurality of routes based on the evaluation information. By executing this program, the computer can execute a route search that preferably takes into account individual differences of the driver. Preferably, the above program is stored in a storage medium.

Example

[0019] Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.

[0020] [Outline of Route Search System] FIG. 1 shows a schematic configuration of a route search system according to this embodiment. The route search system is a system that evaluates each road by a model selected based on the driver's profile and performs route search, and includes a server device 1 and a terminal device 2.

[0021] When server device 1 receives route search request information (also called "route search request information S1") from terminal device 2, it searches for one or more recommended routes from the departure point to the destination specified in route search request information S1. Here, route search request information S1 includes profile information of the terminal device 2 user, such as gender and driving history, and server device 1 determines the recommended route taking this profile into consideration. Then, server device 1 transmits information indicating the determined recommended route (also called "recommended route information S2") to terminal device 2.

[0022] Terminal device 2 is, for example, a mobile device such as a smartphone, which performs functions such as providing route guidance to a destination and displaying a map of the area around the current location. In this embodiment, terminal device 2 receives recommended route information S2 by sending route search request information S1, which includes current location information indicating the starting point, destination information specified by user input, and user profile information, to server device 1. Then, terminal device 2 determines the route to be guided based on the received recommended route information S2 and starts route guidance. Note that terminal device 2 may store map data and other information necessary for route guidance in advance, or may receive it from server device 1 as needed.

[0023] [Device configuration] Figure 2(A) shows the schematic configuration of server device 1. As shown in Figure 2(A), server device 1 has a communication unit 11, a storage unit 12, and a control unit 15. The communication unit 11, the storage unit 12, and the control unit 15 are interconnected via a bus line.

[0024] The communication unit 11, based on the control of the control unit 15, receives route search request information S1 transmitted from the terminal device 2 and transmits recommended route information S2 to the terminal device 2, etc. The communication unit 1 is an example of a "transmission unit" in the present invention.

[0025] The storage unit 12 stores programs for controlling the operation of the server device 1 and holds information necessary for the operation of the server device 1. The storage unit 12 also stores map data 16, captured image DB 17, and driving evaluation model DB 18. The map data 16 includes road data represented by links corresponding to roads and nodes corresponding to road connections (intersections), and facility information related to facilities. The captured image DB 17 is a database that stores captured images taken on each road, associated with that road. For example, there is at least one captured image registered in the captured image DB 17 for each link.

[0026] The Driving Evaluation Model DB18 is a database of models (also called "Driving Evaluation Models") for determining driving evaluations of roads displayed in images taken on roads. A Driving Evaluation Model is a pattern classifier obtained through supervised machine learning, and is training data learned by applying, for example, a neural network (deep learning). In this embodiment, a Driving Evaluation Model is generated by training a combination of images taken on roads and evaluation values ​​of driving difficulty determined by a person who has viewed the images (also called a "Judge") as training data. Each Driving Evaluation Model registered in the Driving Evaluation Model DB18 is generated for each group of judges who share commonalities in their profiles, such as gender and driving history. The data structure of the Driving Evaluation Model DB18 will be described later. The evaluation value of driving difficulty may be a two-level evaluation value, for example, "0" indicating easy driving or "1" indicating difficult driving, or it may be an evaluation value with three or more levels. The evaluation value of driving difficulty is an example of "evaluation information" in this invention.

[0027] The control unit 15 includes a CPU, ROM, RAM, etc. (not shown) and performs various controls on each component within the server device 1. When the communication unit 11 receives route search request information S1, the control unit 15 searches for a recommended route from the starting point to the destination indicated by the route search request information S1 that minimizes the total cost of each road (link). At this time, the control unit 15 searches for a recommended route that is easy for the user of the terminal device 2 to drive by determining the cost of each road based on an evaluation value of the difficulty of driving. In this case, the control unit 15 extracts a driving evaluation model from the driving evaluation model DB 18 based on the user profile of the terminal device 2 indicated by the route search request information S1, and determines the evaluation value of the difficulty of driving for each road using the extracted driving evaluation model and the captured images registered in the captured image DB 17. The control unit 15 is an example of a computer that executes the "first acquisition unit," "second acquisition unit," "third acquisition unit," "route determination unit," and program in the present invention.

[0028] Figure 2(B) shows the schematic configuration of terminal device 2. As shown in Figure 2(B), terminal device 2 mainly consists of a communication unit 21, a storage unit 22, a sensor unit 23, an input unit 24, a control unit 25, and an output unit 26. The communication unit 21, storage unit 22, sensor unit 23, input unit 24, control unit 25, and output unit 26 are interconnected via a bus line.

[0029] Based on the control of the control unit 25, the communication unit 21 transmits route search request information S1 to the server device 1 and receives recommended route information S2 and other information necessary for route guidance transmitted from the server device 1.

[0030] The storage unit 22 stores programs executed by the control unit 25 and information necessary for the control unit 25 to perform predetermined processes. In this embodiment, the storage unit 22 stores information related to the profile generated by the control unit 25 based on inputs to the input unit 24, etc. (also referred to as "profile information"). Profile information is an example of "driver information" in the present invention.

[0031] The sensor unit 23 includes a GPS receiver 32 that generates absolute position information of the vehicle, and an autonomous positioning device 33 including a gyro sensor and a speed sensor. The sensor unit 23 supplies the generated output signals to the control unit 25.

[0032] The input unit 24 includes buttons, a touch panel, a remote controller, a voice input device, etc., for user operation. It accepts inputs such as specifying a destination for route searching and inputting the driver's profile, and supplies the generated input signals to the control unit 25. The output unit 26 includes, for example, a display or speaker that outputs based on the control of the control unit 25.

[0033] The control unit 25, which includes a CPU for executing programs, controls the entire terminal device 2. For example, the control unit 25 generates profile information based on user input from the input unit 24 and stores it in the storage unit 22. When the control unit 25 detects destination input from the input unit 24, it sends route search request information S1, which includes the input destination information, current location information output by the GPS receiver 32, etc., and the profile information stored in the storage unit 22, to the server device 1 via the communication unit 21. When the communication unit 21 receives recommended route information S2 from the server device 1, the control unit 25 outputs the recommended route to the output unit 26 and accepts input from the input unit 24 asking whether or not to set it as the guided route. After that, the control unit 25 outputs information to guide along the set route to the output unit 26.

[0034] [Extraction of driving evaluation models] When server device 1 receives route search request information S1, it extracts a driving evaluation model from the driving evaluation model DB18 based on the profile information contained in the route search request information S1. Here, each driving evaluation model registered in the driving evaluation model DB18 is generated for each group of judges who share common profile characteristics such as gender and driving history, and is associated with the classification information of the profile common to the group. Therefore, server device 1 extracts a driving evaluation model from the driving evaluation model DB18 that is associated with the classification information of the profile common to the profile indicated by the profile information.

[0035] Figure 3 is a correspondence table between the identification number (also called "Model ID") of each driving evaluation model included in the driving evaluation model DB18 and the classification information of the evaluator's profile corresponding to that driving evaluation model. The correspondence table in Figure 3 has the following items: "Gender," "Driving History," "Place of Residence," "Age," and "Model ID."

[0036] Here, the driving evaluation model with model ID "M001" is a driving evaluation model obtained by training the model with training data consisting of combinations of photographs referenced by an evaluator who is male, has less than 5 years of driving experience, resides in Tokyo, and is between 25 and 39 years old, and the driving difficulty rating determined by the evaluator based on the photographs. Similarly, the driving evaluation model with model ID "M002" is a driving evaluation model obtained by training the model with training data consisting of combinations of photographs referenced by an evaluator who is male, has less than 5 years of driving experience, resides in Tokyo, and is between 40 and 59 years old, and the driving difficulty rating determined by the evaluator based on the photographs. Furthermore, the driving evaluation model with model ID "M251" is a driving evaluation model obtained by training the model with training data consisting of combinations of photographs referenced by an evaluator who is female, has less than 5 years of driving experience, resides in Saitama Prefecture, and is between 60 and 69 years old, and the driving difficulty rating determined by the evaluator based on the photographs. Furthermore, the driving evaluation model with model ID "M342" is a model obtained by training the model with training data consisting of combinations of photographs referenced by an evaluator who is female, has more than 5 years of driving experience, resides in Kanagawa Prefecture, and is 24 years of age or younger, and the driving difficulty rating determined by that evaluator based on the photographs. Furthermore, the driving evaluation model with model ID "M445" is a model obtained by training the model with training data consisting of combinations of photographs referenced by an evaluator who is female, has more than 5 years of driving experience, resides in Hokkaido, and is 70 years of age or older, and the driving difficulty rating determined by that evaluator based on the photographs.

[0037] Thus, the driving evaluation model is generated for each group when the evaluators are grouped based on gender, driving history, place of residence, and age. Generally, the perceived difficulty of driving varies depending on the driver's gender, driving history, place of residence (i.e., the area they frequently drive in), age, etc. Taking the above into consideration, in this embodiment, evaluators are grouped based on profile items that influence the perceived difficulty of driving, and the server device 1 stores the pre-generated driving evaluation model for each group as the driving evaluation model DB18.

[0038] In the example in Figure 3, driving experience was classified into two categories: less than 5 years and 5 years or more. Residence was classified by prefecture, and age was classified into five categories. However, the classification method for each item is not limited to these. For example, driving experience may be divided into three or more categories, and residence may be classified by a broader area (e.g., 7 regions) or a more detailed area. Similarly, age may be classified into categories other than five. In addition to or instead of driving experience, the correspondence table in Figure 3 may include items related to driving history other than driving experience (e.g., number or frequency of accidents, number or frequency of sudden braking, etc.).

[0039] Server device 1 refers to the correspondence table shown in Figure 3 and extracts the driving evaluation model corresponding to the group of judges (matching group) that fits the profile indicated by the profile information from the driving evaluation model DB18. Specifically, Server device 1 refers to the correspondence table shown in Figure 3 and extracts the driving evaluation model corresponding to the model ID recorded in the record corresponding to gender, driving history, place of residence, and age included in the profile information from the driving evaluation model DB18. For example, if the gender indicated by the profile information is "male", the driving history is "3 years", the place of residence is "Tokyo", and the age is "45 years old", then the driving evaluation model with model ID "M002" recorded in the second record in Figure 3 is extracted from the driving evaluation model DB18.

[0040] [Route search process] Next, we will explain the route search process using the driving evaluation model. Figure 4 is an example of a flowchart showing the procedure for the route search process using the driving evaluation model.

[0041] First, terminal device 2 determines whether or not there is an input instructing route searching (step S101). If there is an input instructing route searching (step S101; Yes), terminal device 2 sends route search request information S1 to server device 1, which includes the destination information specified in the above input, the current location information output by the GPS receiver 32, etc., and the profile information stored in the storage unit 22 (step S102). On the other hand, if there is no input instructing route searching (step S101; No), terminal device 2 continues to monitor for the presence or absence of an input instructing route searching in step S101.

[0042] After terminal device 2 transmits route search request information S1, server device 1 receives route search request information S1 from terminal device 2 (step S201). Then, server device 1 refers to the profile information contained in route search request information S1 and extracts the applicable driving evaluation model from driving evaluation model DB18 (step S202).

[0043] Then, the server device 1 starts a route search from its current location (i.e., departure point) to the destination, as specified by the route search request information S1, with the driving difficulty evaluation value calculated from the driving evaluation model as the cost (step S203). In this case, the server device 1 extracts images corresponding to the roads that make up the route (i.e., links on the map data 16) from the captured image DB 17, and inputs the extracted captured images into the driving evaluation model to calculate the driving difficulty evaluation value corresponding to the target road (step S204). At this time, the server device 1 may set the cost for each road (link cost) to a value proportional to the calculated driving difficulty evaluation value, or it may set it to a value obtained by adding a value corresponding to the above evaluation value to the link cost calculated in a normal route search, taking into account the required time, etc. In this case, it is preferable that the driving difficulty evaluation value be set to be higher for roads that are more difficult to drive on.

[0044] The server device 1 then determines one or a predetermined number of recommended routes with the lowest total link costs for each road (step S205). In this case, for each candidate route that is a candidate for the recommended route from the departure point to the destination, the server device 1 sums up the link costs of all roads that make up the candidate route and determines one or a predetermined number of recommended routes with the lowest total link costs. This allows the server device 1 to suitably determine as a recommended route a route consisting of roads that are estimated to be easy for the driver to drive. The server device 1 then transmits the recommended route information S2 regarding the determined recommended route to the terminal device 2 that sent the route search request information S1 (step S206).

[0045] Meanwhile, after sending the route search request information S1, terminal device 2 receives the recommended route information S2 that server device 1 sent in step S206 (step S103). Then, terminal device 2 determines the recommended route selected by the user as the guided route and starts providing guidance to travel along the determined guided route (step S104).

[0046] As described above, the server device 1 in this embodiment receives profile information about the driver operating the vehicle, as well as information about the vehicle's departure point and destination, from the terminal device 2. For each of the roads constituting the multiple routes from the departure point to the destination, the server device 1 calculates a driving difficulty evaluation value based on a driving evaluation model corresponding to the profile information. Then, the server device 1 determines a recommended route based on the calculated driving difficulty evaluation value. In this way, since the server device 1 performs route searching based on a driving difficulty evaluation value calculated from a driving evaluation model suitable for the vehicle's driver, it can perform route searching that appropriately takes into account individual differences among drivers.

[0047] [Differentiation] Next, a suitable modification of the embodiment will be described. The following modifications may be applied to the above-described embodiment in any combination.

[0048] (Variation 1) Terminal device 2 may execute some or all of the route search process performed by server device 1.

[0049] Figure 5 shows an example configuration of the route search system according to this modified example. In the example in Figure 5, terminal device 2 stores map data 20 including road data and facility information, and transmits profile information "Sa" stored in storage unit 22 to server device 1. In this case, server device 1 extracts a driving evaluation model from driving evaluation model DB 18 based on the profile indicated by profile information Sa. Then, server device 1 transmits the extracted driving evaluation model information (also called "extracted model information Sb") to terminal device 2. When terminal device 2 receives input of a destination and performs route search, it determines a recommended route by executing steps S203 to S205 in Figure 4. In this case, terminal device 2 receives images of each road along the route required in step S204 from server device 1 as route image information "Sc". In this case, for example, terminal device 2 sends a request signal to server device 1 specifying the link ID of the road for which the captured image is needed, and server device 1 extracts the captured image corresponding to the link ID included in the received request signal from the captured image DB 17 and sends it as route image information Sc.

[0050] Thus, according to this modified example, terminal device 2 can suitably perform the route search process according to the embodiment in place of server device 1. In this modified example, terminal device 2 is an example of the "route search device" in the present invention, and the control unit 25 of terminal device 2 is an example of a computer that executes the "first acquisition unit," "second acquisition unit," "third acquisition unit," "route determination unit," and program in the present invention.

[0051] (Modification 2) The map data may include pre-defined driving difficulty ratings calculated based on a driving evaluation model.

[0052] Figure 6 shows the configuration of the route search system according to this modified example. In the example in Figure 6, the terminal device 2 includes information (also called "driving evaluation information Ie") in the map data 20 that associates the driving difficulty evaluation value calculated based on each driving evaluation model with each road. In this case, the driving difficulty evaluation value for each road is recorded in the map data 20 for each model ID of the driving evaluation model. In addition, in the configuration example in Figure 6, the terminal device 2 stores the correspondence table in Figure 3 in advance. When performing a route search, the terminal device 2 first identifies the model ID of the driving evaluation model corresponding to the profile information by referring to the correspondence table in Figure 3, and then obtains the driving difficulty evaluation value for each road constituting the route by referring to the map data 20 using the identified model ID and the link ID of the target road. With this configuration as well, the terminal device 2 can calculate the cost of each route based on the driving difficulty evaluation value and determine the recommended route.

[0053] Alternatively, in a configuration where the server device 1 performs route searching, the map data provided by the server device 1 may include driving evaluation information Ie. Even in this case, the server device 1 identifies the model ID of the driving evaluation model corresponding to the profile information received from the terminal device 2 by referring to the correspondence table in Figure 3, and then obtains the driving difficulty evaluation value for each road constituting the route by referring to the map data using the identified model ID and the link ID of the target road. In this configuration as well, the server device 1 can calculate the cost of each route based on the driving difficulty evaluation value and determine the recommended route.

[0054] (Variation 3) The driving evaluation model may be further classified and generated based on the vehicle's driving environment.

[0055] In this case, the correspondence table in Figure 3 includes not only profile items such as gender, driving history, place of residence, and age, but also items related to the driving environment, such as weather, road type, and time of day. The driving evaluation model is then pre-generated for each classification based on the driving environment items and stored in the driving evaluation model DB18. In this case, the driving evaluation model corresponding to each driving environment is generated by learning the combination of a captured image and the evaluation value of the driving difficulty determined by the evaluator based on the captured image, for each image taken on the road in each driving environment, as training data.

[0056] For example, when terminal device 2 receives input instructing route searching, it refers to the output of sensor unit 23, such as a raindrop sensor (not shown), and map DB 20, etc., to recognize the current driving environment of the vehicle. Then, terminal device 2 includes information indicating the recognized driving environment in route search request information S1 and transmits it to server device 1. In this case, server device 1 selects a driving evaluation model that matches the profile information and driving environment information included in route search request information S1, and uses the driving evaluation model to calculate an evaluation value of the driving difficulty required for route searching. Thus, according to this modified example, server device 1 can further consider the current driving environment of the vehicle and determine an easy-to-drive route as a recommended route.

[0057] (Modification 4) The driving evaluation model does not limit the criteria used for assessment to driving difficulty. For example, the driving evaluation model could also assess visibility.

[0058] In this case, each driving evaluation model registered in the driving evaluation model DB18 is generated by learning a combination of captured images taken on the road and the evaluation value of visibility determined by an evaluator who referred to those captured images, using this combination as training data. Then, as in the embodiment, each of these driving evaluation models is generated for each group of evaluators who share common profile characteristics such as gender and age, and is stored in the driving evaluation model DB18 in association with the classification information of the common profile. Then, in the route search process shown in Figure 4, the terminal device 2, as in the embodiment, sends route search request information S1 including profile information to the server device 1 in step S102. In this case, in step S202, the server device 1 extracts from the driving evaluation model DB18 a driving evaluation model that matches the profile indicated by the profile information contained in the route search request information S1 received from the terminal device 2. Then, in step S203, the server device 1 starts a route search using the evaluation value of visibility calculated from the driving evaluation model as the cost. In this case, the evaluation value of visibility calculated from the driving evaluation model is set so that a higher value is obtained when visibility is poor. Then, the server device 1 determines the recommended route by executing steps S204 and S205, and transmits the recommended route information S2 related to the determined recommended route to the terminal device 2 in step S206.

[0059] According to this modified version, the server device 1 can determine a route consisting of roads with good visibility as a suitable recommended route from a tourism or safety perspective. Here, a route consisting of roads with good visibility would, for example, from a tourism perspective, be a route consisting of roads with excellent scenery in a tourist area, and from a safe driving perspective, be a route consisting of safe roads where pedestrians can be easily spotted from a distance. The terminal device 2 may also receive input from the input unit 24 allowing the user to select whether to use driving difficulty or visibility as the indicator for route searching. In this case, the terminal device 2 sends information specifying which indicator to use, driving difficulty or visibility, in the route search request information S1 based on the received input to the server device 1. [Explanation of symbols]

[0060] 1 Server device 2 Terminal devices 11, 21 Communications Department 12, 22 Storage section 15, 25 Control Unit 16, 20 Map data 17 Image Database 18. Operational Evaluation Model Database 23 Sensor section 24 Input section 26 Output section

Claims

1. A first acquisition unit that acquires driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle, A second acquisition unit acquires evaluation information output from a driving evaluation model by inputting captured images corresponding to each of the roads constituting the multiple routes from the departure point to the destination into a driving evaluation model corresponding to the driver information. A route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, A pathfinding device equipped with the following features.

2. The system further includes a transmitting unit that transmits the recommended route information to a terminal device that moves together with the moving object, The route search device according to claim 1, wherein the first acquisition unit receives the driver information and the departure and destination information from the terminal device.

3. The route search device according to claim 1, wherein the second acquisition unit transmits the driver information to a server device that stores a plurality of driving evaluation models, thereby acquiring a driving evaluation model corresponding to the driver information, and calculates evaluation information for each of the roads based on the driving evaluation model.

4. The route search device according to claim 2 or 3, wherein the evaluation information is information indicating the difficulty of driving, or information indicating the quality of visibility.

5. A first acquisition unit that acquires driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle, A second acquisition unit acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting the multiple routes from the departure point to the destination, A route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, Equipped with, The driving evaluation model is a route search device in which the driving evaluation model in the adapted group that fits the driver information is learned for each group of judges classified according to the commonality of the profiles, based on captured images of the surroundings of a moving object and evaluation information regarding driving evaluated by a judge equipped with a profile based on the captured images.

6. The system further includes a memory unit that stores map data associated with evaluation information based on multiple driving evaluation models for each road. The route search device according to claim 5, wherein the second acquisition unit acquires evaluation information based on a driving evaluation model corresponding to the driver information from the map data for each of the roads.

7. The system further includes a transmitting unit that transmits the recommended route information to a terminal device that moves together with the moving object, The route search device according to claim 5 or 6, wherein the first acquisition unit receives the driver information and the departure and destination information from the terminal device.

8. The route search device according to claim 5 or 6, wherein the second acquisition unit transmits the driver information to a server device that stores a plurality of driving evaluation models, thereby acquiring a driving evaluation model corresponding to the driver information, and calculates evaluation information for each of the roads based on the driving evaluation model.

9. The route search device according to claim 7 or 8, wherein the evaluation information is information indicating the difficulty of driving, or information indicating the quality of visibility.

10. A first acquisition unit that acquires driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle, A second acquisition unit acquires images taken at each of the roads constituting the multiple routes from the departure point to the destination, A third acquisition unit acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting the multiple routes from the departure point to the destination, A route determination unit that determines a recommended route from the plurality of routes based on the evaluation information, Equipped with, The third acquisition unit is a route search device that acquires evaluation information for each of the roads by inputting the captured images acquired by the second acquisition unit into the driving evaluation model.

11. The system further includes a transmitting unit that transmits the recommended route information to a terminal device that moves together with the moving object, The route search device according to claim 10, wherein the first acquisition unit receives the driver information and the departure and destination information from the terminal device.

12. The route search device according to claim 10, wherein the third acquisition unit transmits the driver information to a server device that stores a plurality of driving evaluation models, thereby acquiring a driving evaluation model corresponding to the driver information, and calculates evaluation information for each of the roads based on the driving evaluation model.

13. The route search device according to claim 11 or 12, wherein the evaluation information is information indicating the difficulty of driving, or information indicating the quality of visibility.

14. A control method performed by a pathfinding device, A first acquisition step involves acquiring driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle. A second acquisition step involves inputting images corresponding to each of the roads constituting the multiple routes from the departure point to the destination into a driving evaluation model corresponding to the driver information, thereby acquiring evaluation information output from the driving evaluation model. A route determination step in which a recommended route is determined from the plurality of routes based on the evaluation information, A control method having

15. A program that is executed by a computer, A first acquisition unit that acquires driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle, A second acquisition unit acquires evaluation information output from a driving evaluation model by inputting captured images corresponding to each of the roads constituting the multiple routes from the departure point to the destination into a driving evaluation model corresponding to the driver information. Route determination unit that determines a recommended route from the plurality of routes based on the evaluation information. A program that causes the aforementioned computer to function.

16. A control method performed by a pathfinding device, A first acquisition step involves acquiring driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle. A second acquisition step involves acquiring evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting the multiple routes from the departure point to the destination, A route determination step in which a recommended route is determined from the plurality of routes based on the evaluation information, It has, The driving evaluation model is a driving evaluation model in a suitable group that fits the driver information, selected from driving evaluation models learned for each group of judges classified according to the commonality of the profiles, based on captured images of the surroundings of a moving object and evaluation information regarding driving evaluated by a judge with a profile based on the captured images.

17. A program that is executed by a computer, A first acquisition unit that acquires driver information relating to the driver operating the mobile vehicle, and information on the departure point and destination of the mobile vehicle, A second acquisition unit acquires evaluation information based on a driving evaluation model corresponding to the driver information for each of the roads constituting the multiple routes from the departure point to the destination, Route determination unit that determines a recommended route from the plurality of routes based on the evaluation information. The computer is made to function as follows: The driving evaluation model is a program that, based on captured images of the surroundings of a moving object and evaluation information regarding driving evaluated by a judge with a profile based on the captured images, is a driving evaluation model in a suitable group that fits the driver information, selected from driving evaluation models learned for each group of judges classified according to the commonality of the profiles.

18. A storage medium storing the program described in claim 15 or 17.

Images