Route guidance device, route guidance program, route guidance method, and route guidance system

Abstract

The system guides passengers along a suitable route that takes into consideration the needs of those inside the vehicle. [Solution] A route guidance device that generates and guides a vehicle from its starting point to its destination, which acquires passenger information based on images taken inside the vehicle, creates a response acquisition prompt to obtain information on alternative roads that the passenger should avoid based on the passenger information, outputs the created response acquisition prompt to a generation system AI to obtain a response, and generates and guides the vehicle based on the acquired response.

Description

Technical Field

[0001] The present invention relates to a route guidance device, a route guidance program, a route guidance method, and a route guidance system.

Background Art

[0002] Conventionally, when guiding a driving route from a departure place to a destination of a vehicle, a technique for searching a driving route based on the attributes of a person in the vehicle interior (for example, a pregnant woman, an infant, etc.) has been proposed (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, since the driving route is generated based on predetermined conditions and states regarding the attributes of the person in the vehicle interior and the road to be traveled, matters to be noted other than the predetermined conditions and states, and the vehicle interior situation and road conditions that change according to the passage of time It was difficult to guide the driving route in response to. As a result, there was a concern that it was impossible to guide a suitable driving route considering the person in the vehicle interior, and there was a possibility that an unpleasant driving state such as vibration or large swaying to the left and right could not be avoided.

[0005] In view of the above problems, an object of the present invention is to provide a technique capable of guiding a suitable driving route considering the person in the vehicle interior.

Means for Solving the Problems

[0006] An exemplary route guidance device of the present invention is a route guidance device that generates and guides a vehicle on a route from its starting point to its destination, and acquires passenger information about the passengers based on images taken inside the vehicle, creates a response acquisition prompt to acquire information on alternative roads that the passengers should avoid based on the passenger information, outputs the created response acquisition prompt to a Generative Artificial Intelligence (AI) to acquire a response, and generates and guides the vehicle on the route based on the acquired response. [Effects of the Invention]

[0007] According to the present invention, when generating a driving route to guide a vehicle from its starting point to its destination, prompts containing information that do not necessarily require user input are automatically created and input to the generation AI, and the driving route is generated based on the response of the generation AI. By utilizing the response of the generation AI, the driving route is not generated based on predetermined conditions and states, so it is possible to generate diverse driving routes that take into account matters other than predetermined conditions and states, as well as the in-vehicle conditions and road conditions that change over time. Therefore, it becomes possible to guide the vehicle along a suitable driving route that takes into consideration the people inside the vehicle without requiring troublesome user input. [Brief explanation of the drawing]

[0008] [Figure 1] Overall configuration diagram of the route guidance system of this embodiment [Figure 2] An explanatory diagram showing the in-vehicle conditions that affect route guidance. [Figure 3] Block diagram showing the configuration of the route guidance system in Figure 1. [Figure 4] A schematic diagram showing an example of a travel route. [Figure 5] A diagram showing an example of a road data table. [Figure 6] A diagram showing an example of a keyword data table. [Figure 7] A flowchart illustrating the route guidance process performed by the route guidance device. [Figure 8] This figure shows an example of a vehicle performance data table. [Figure 9] This figure shows an example of a vehicle characteristics data table. [Modes for carrying out the invention]

[0009] Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. However, the present invention is not limited to the embodiments described below.

[0010] <1. Route guidance system> Figure 1 is an overall configuration diagram of the route guidance system 1 of this embodiment. The route guidance system 1 is a system that generates and guides the vehicle V1 from its starting point to its destination. In this embodiment, the route guidance system 1 guides the vehicle along a route in response to the changing conditions inside the vehicle and road conditions over time.

[0011] The route guidance system 1 includes a route guidance device 10 mounted on the vehicle V1, a generation system AI 20 included in a server device, a camera 2, a display device 3, and a speaker 4 (see Figures 1 and 3). The route guidance device 10 and the generation system AI 20 are connected via a communication network N, such as a mobile communication network, enabling bidirectional communication.

[0012] The route guidance device 10 is an in-vehicle device mounted on the vehicle V1. The route guidance device 10 may be a device permanently installed in the vehicle V1, or it may be a portable device that can be carried inside or outside the vehicle V1. The route guidance device 10 may be a component of, for example, a drive recorder, a tablet terminal, a mobile terminal, etc.

[0013] The route guidance device 10 is a so-called navigation device that generates and guides a driving route from the departure point to the destination of the vehicle V1. Specifically, the route guidance device 10 guides the driving route in response to the in-vehicle situation and road conditions that change over time. FIG. 2 is an explanatory diagram showing the in-vehicle situation that affects route guidance. The route guidance device 10 acquires a photographed image of the interior of the vehicle V1 from the camera 2. The route guidance device 10 generates and guides a suitable driving route that takes into account the in-vehicle situation, that is, the people in the vehicle such as the pregnant woman P1 and the child P2 shown in FIG. 2 for example.

[0014] The generative AI 20 is included in, for example, a server device or the like outside the vehicle V1. The generative AI 20 is composed of a large language model or the like.

[0015] The generative AI 20 can be constructed by training an AI model with a large number of various types of information related to the type of person, the state of the person, the physical condition and mental state of the person, etc. As a typical method, a large language model trained by a large number of documents including documents related to the type of person, the state of the person, the physical condition and mental state of the person, etc. can be applied. In addition, for example, various types of information related to the suitability of road types for various types of people and people's states, various types of information showing the relationship between various types of people and people's states and the physical condition with respect to vibration, various types of information showing the relationship between road types and vibration, etc. By training the AI model with the learning data, the generative AI 20 can be generated.

[0016] The camera 2, the display device 3, and the speaker 4 are provided in the vehicle V1. The camera 2 photographs the interior of the vehicle V1 (pregnant woman P1, child P2, etc.). The display device 3 displays an image related to the driving route provided to the driver D1 and the like. The speaker 4 outputs a voice related to the driving route. Note that the camera 2, the display device 3, and the speaker 4 may be configured to be included in a drive recorder, a tablet terminal, a mobile terminal, or the like.

[0017] FIG. 3 is a block diagram showing the configuration of the route guidance system 1 in FIG. 1. In FIG. 2, the components necessary for explaining the features of the present embodiment are shown, and the description of general components is omitted.

[0018] <2. Route Guidance Device> The route guidance device 10 includes a communication unit 11, a storage unit 12, and a controller 13. The route guidance device 10 also includes an operation unit (not shown) for receiving operation inputs from the driver D1 or the like.

[0019] The communication unit 11 is an interface for data communication with other devices (such as the camera 2) via a communication network. The communication unit 11 includes a communication device for performing wired and wireless communications with other devices. The wireless communication device is composed of, for example, a transceiver of a 5G communication (fifth-generation mobile communication system) mobile phone network.

[0020] The storage unit 12 is composed of a volatile memory and a non-volatile memory, and stores various information necessary for content playback processing. The volatile memory is composed of, for example, a RAM (Random Access Memory). The non-volatile memory is composed of, for example, a ROM (Read Only Memory), a flash memory, or a hard disk drive. Programs and data readable by the controller 13 are stored in the non-volatile memory. At least a part of the programs and data stored in the non-volatile memory may be acquired from other computer devices connected by wire or wirelessly, or from a portable recording medium.

[0021] The storage unit 12 stores a route guidance program 121, map data 122, a road data table 123, a prompt template table 124, and a keyword data table 125. The contents of these programs, data tables, etc. stored in the storage unit 12 will be described separately. Further, the storage unit 12 stores a plurality of data tables for various processes, etc.

[0022] The controller 13 consists of a processor that performs calculations and other processing, and controls various operations in the route guidance device 10. The processor includes, for example, a CPU (Central Processing Unit). The controller 13 executes the route guidance program 121 stored in the memory unit 12 and performs the process of generating the travel route. The route guidance program 121 includes various programs that realize various functions of the route guidance device 10.

[0023] The controller 13 includes, as its functions, a destination setting unit 131, a route generation unit 132, an image acquisition unit 133, a state detection unit 134, a prompt creation unit 135, an answer acquisition unit 136, a route correction unit 137, and a guidance unit 138. In this embodiment, the functions of the controller 13 are realized by the processor executing calculation processing according to the route guidance program 121 stored in the storage unit 12.

[0024] The destination setting unit 131 receives input for the destination of vehicle V1 from the driver D1 or the like via the operation unit and sets the destination. More specifically, the destination setting unit 131 extracts destination candidates by referring to the map data 122 stored in the storage unit 12 based on the destination name entered by the driver D1 or the like, and outputs them to the display device 3. Then, the destination setting unit 131 sets the destination selected by the driver D1 or the like from among the destination candidates. The destination setting unit 131 stores information related to the destination in the storage unit 12 as needed for subsequent processing. Information related to the destination is used to generate the driving route.

[0025] The route generation unit 132 generates a travel route for vehicle V1 from its starting point to its destination. More specifically, the route generation unit 132 searches for a travel route for vehicle V1 from its starting point to its destination and extracts candidate routes. The starting point of vehicle V1 is, for example, the current location of vehicle V1. The destination of vehicle V1 is the destination set by the destination setting unit 131. The route generation unit 132 stores information related to the candidate routes in the storage unit 12 as needed for subsequent processing.

[0026] Figure 4 is a schematic diagram showing an example of a driving route. In Figure 4, circles indicate road intersections (junctions), straight lines indicate roads connecting two adjacent intersections, Sr is the current location (starting point), and Er is the destination. Specifically, the route generation unit 132 searches for a driving route from the vehicle V1's starting point Sr to the destination Er by using location information indicating the vehicle V1's current location Sr, location information indicating the destination Er set by the destination setting unit 131, and map data 122 stored in the storage unit 12. It is desirable for the route generation unit 132 to search for multiple driving routes from the vehicle V1's starting point Sr to the destination Er. The route generation unit 132 stores road data, which is information related to the driving route, in the road data table 123 of the storage unit 12.

[0027] Figure 5 shows an example of a road data table 123. In Figure 5, the dashed arrows indicate that the data records on both sides of the arrow are connected consecutively. As shown in Figure 5, the items in the road data table 123 include "Link ID", "Node Data", "Distance", "Road Width Rank Value", "Curve Blank Value", "Road Surface Rank Value", "Area Rank Value", "Road Width Weighting Value", "Curve Weighting Value", "Road Surface Weighting Value", "Area Weighting Value", and "Passage Cost".

[0028] The "Link ID" in Road Data Table 123 is identification information used to identify a dataset of road information. A "link" refers to a road between two adjacent nodes (intersections) (for example, link L11 in Figure 4). While nodes are generally intersections, points equivalent to intersections in route search processing, or points where road conditions differ significantly, are often designated as nodes. In other words, the "Link ID" is identification information that indicates a road (link) between two adjacent nodes. A Link ID is set for each link connecting two adjacent nodes, and a road information data record is constructed for each Link ID.

[0029] The "node data" in the road data table 123 is the identification code for the two nodes located at both ends of the road identified by the link ID (for example, nodes N11a and N11b in Figure 4). Various information about each node is stored in a node database, which consists of data records with the node identification code as the identification data (primary key), and is extracted and used as appropriate in various processes, for example, by using the node identification code as the key. The "distance" in the road data table 123 indicates the distance between the two nodes located at both ends of the road identified by the link ID, and represents the distance traveled when traveling along that road by vehicle or other means.

[0030] The "Road Width Rank Value," "Curve Blank Value," and "Road Surface Rank Value" in Road Data Table 123 are data indicating the degree of road width, curves, and road surface condition of the road identified by the Link ID, respectively. The Road Width Rank Value indicates the size of the road width. The Curve Blank Value is a value determined by the curvature and number of curves. The Road Surface Rank Value indicates the degree of unevenness of the road surface. In other words, this data indicates the physical and mental burden on passengers (for example, the degree of impact on motion sickness).

[0031] As described above, by including road data that indicates road conditions (road width, curves, and road surface) related to the physical and mental burden on passengers, it is possible to take into account the impact of road conditions on passengers throughout the entire travel route.

[0032] The "Area Rank Value" in Road Data Table 123 is data that indicates a value corresponding to the type of area (region) to which the link identified by the Link ID is located. The Area Rank Value is information that indicates the type related to the physical and mental burden on passengers in area (region) types such as urban areas, suburbs, and mountainous areas. In this embodiment, the Area Rank Value is a value that indicates the degree to which congestion is likely to occur.

[0033] In this embodiment, the greater the physical and mental burden on the passengers, the higher each rank value becomes, and the higher the cost of traveling on the road.

[0034] The "Road Width Weighting Value," "Curve Weighting Value," "Road Surface Weighting Value," and "Area Weighting Value" in Road Data Table 123 are data that indicate the weighting of the road width rank value, car blank value, road surface rank value, and area rank value of the link identified by the link ID, respectively. These data are rewritten according to the type and status of the passenger. These "Road Width Weighting Value," "Curve Weighting Value," "Road Surface Weighting Value," and "Area Weighting Value" are stored in each data record with the passenger type and status as the primary key, forming a database. Then, the "Road Width Weighting Value," "Curve Weighting Value," "Road Surface Weighting Value," and "Area Weighting Value" corresponding to the passenger type and status are extracted from the database and written to Road Data Table 123.

[0035] The road width rank value increases as the road width narrows. The evaluation value for a road's "road width" is expressed, for example, as (road width rank value) × (road width weighting value). The curve blank value increases as the curve radius (curve radius) decreases, as the number of curves increases, and as the distance of the curves increases. The evaluation value for a road's "curves" is expressed, for example, as (curve blank value) × (curve weighting value). The road surface rank value increases as the road surface irregularities increase. The evaluation value for a road's "road surface" is expressed, for example, as (road surface rank value) × (road surface weighting value). The area rank value increases, for example, in areas where congestion is likely to occur. The evaluation value for a road's "area" is expressed, for example, as (area rank value) × (area weighting value). Note that each weighting value indicates how much each rank value item affects the passenger depending on the type and condition of the passenger.

[0036] The "transit cost" in Road Data Table 123 is data indicating the degree of driving load experienced by a passenger when traveling along a road identified by a Link ID. The transit cost can be calculated, for example, by multiplying the road distance by an evaluation value related to "road width," an evaluation value related to "curves," an evaluation value related to "road surface," and an evaluation value related to "area." However, the method of calculating the transit cost is not limited to this method; it is also possible to calculate the transit cost using an appropriate calculation formula (set based on experiments, etc.) using the above values, or by using a data table (constructed based on experiments, etc.) with the above values ​​as parameters.

[0037] The route generation unit 132 then calculates the total cost of travel for each of the extracted candidate routes. A travel route is composed of a combination of links representing multiple roads that are traversed from the starting point to the destination. The total cost is the sum of the travel costs incurred when traversing the multiple roads (links) that make up the travel route. The route generation unit 132 then selects the travel route with the smallest total cost from among the extracted candidate routes as the route to guide the driver D1, etc. Alternatively, a predetermined number of travel routes with the smallest total cost may be presented to the user, and the route selected by the user from the presented routes may be selected as the guided travel route.

[0038] The image acquisition unit 133 acquires images of the vehicle interior (captured images) taken by the camera 2 via the communication unit 11. The route guidance device 10 has a "route guidance mode according to the conditions inside the vehicle" when guiding a suitable driving route that takes into consideration the people inside the vehicle interior. When the driver D1 or the like receives an operation to start this route guidance mode via the operation unit, the route guidance device 10 acquires captured images of the inside of the vehicle interior using the image acquisition unit 133.

[0039] The image acquisition unit 133 acquires images of the in-vehicle conditions, including people inside the vehicle, such as a pregnant woman (P1) and a child (P2) as shown in Figure 2, as information related to the in-vehicle conditions that affect route guidance (determining an appropriate route). The image acquisition unit 133 stores the acquired images (image data) in the storage unit 12 as needed for subsequent processing. This image data is used to detect the type of person (pregnant woman P1, child P2, etc.) and the state of the person that affect route guidance.

[0040] The state detection unit 134 detects the type of person and their state that may affect route guidance, based on the captured image acquired by the image acquisition unit 133. More specifically, the state detection unit 134 performs analysis on the captured image to detect the type of person, such as "pregnant woman" or "child," and their state, such as "holding drink C1." The type of person and their state detected by the state detection unit 134 correspond to the data type input to the prompt creation unit 135. The information regarding the type of person and their state detected by the state detection unit 134 is stored in the data table of the storage unit 12.

[0041] The detection of a person's species and state based on captured images may be performed, for example, using an artificial intelligence (AI) model for detecting a person's species and state. The AI ​​model takes captured images of the vehicle interior as input and outputs the person's species and state. More specifically, the AI ​​model is generated by training a large number of captured images in which the person's species and state are clearly identified as supervised training data.

[0042] The prompt creation unit 135 creates a response acquisition prompt to obtain, as dialogue (text data), the type of person affecting route guidance detected by the state detection unit 134, and the characteristics of roads to be avoided in relation to the person's state, from the generative AI. In other words, the prompt creation unit 135 creates a response acquisition prompt to obtain avoidance road information corresponding to the person included in the captured image. The prompt creation unit 135 creates a response acquisition prompt to output to the generative AI 20.

[0043] The prompt template table 124 of the memory unit 12 stores templates for response acquisition prompts. These templates consist of a standard instruction phrase for acquiring a response and parameter items to be inserted into that phrase, for example, "(parameter: person type data) + (parameter: person status data) + (standard phrase)".

[0044] "Person type data" is the type of person that affects route guidance, as detected by the state detection unit 134, and is data that represents, for example, a pregnant woman or a child. "Person status data" is the status of a person that affects route guidance, as detected by the state detection unit 134, and is data that represents, for example, a person holding a drink.

[0045] A standard phrase is a phrase used to obtain the characteristics of roads to be avoided as dialogue, such as "What kind of roads should I avoid?". Ideally, the standard phrase should be one that elicits an answer expressed in words (preferably single words) related to the roads to be avoided, such as "What kind of roads should I avoid?", which asks about the relevant road conditions or road types. A standard phrase like "What kind of roads are preferable?" is also preferable, but the subsequent return route determination process will use different (opposite) route determination conditions compared to the case of "What kind of roads should I avoid?".

[0046] The prompt creation unit 135 sets the parameter values ​​(data) corresponding to the template to create a prompt for obtaining answers. Specifically, the prompt creation unit 135 creates prompts for obtaining answers such as, for example, "What kind of road should you avoid if the passenger type is a pregnant woman?" and "What kind of road should you avoid if the passenger type is a child and the passenger status is eating or drinking?"

[0047] As described above, by creating a prompt to obtain the roads to be avoided as dialogue, it is possible to create a prompt that takes into account the influence of the person's type (e.g., pregnant woman, child) and their state (e.g., holding a drink). This makes it possible to obtain a response (dialogue) that is expected to include the influence of the person's type and state in the route guidance.

[0048] The response acquisition unit 136 outputs (applies) the response acquisition prompt created by the prompt creation unit 135 to the generation system AI 20 to acquire a response (dialogue). The generation system AI 20 takes the response acquisition prompt as an input value and outputs road characteristic information corresponding to the driving route with conditions suggested by the response acquisition prompt as a response (dialogue) to the route guidance device 10 (response acquisition unit 136).

[0049] Specifically, the response acquisition unit 136 acquires (receives) dialogue (sentences) generated by the generation AI 20 as responses, such as "Bumpy roads that cause the car to shake, or mountain roads with steep inclines, can be very strenuous for pregnant women. Avoid bumpy roads and mountain roads with steep inclines," "Getting caught in traffic jams means being in the car for a long time, which can be painful for the elderly, especially in the lower back and spine. Avoid driving on roads with frequent traffic jams," and "When a child is drinking juice, the shaking of the car can cause the juice to spill. Avoid roads with many curves or sharp turns." The response acquisition unit 136 then stores the acquired responses (dialogues) in the memory unit 12.

[0050] The route correction unit 137 extracts keywords from the responses (dialogue) obtained by the response acquisition unit 136 and corrects the road data related to the driving route based on those keywords. More specifically, the route correction unit 137 refers to the keyword data table 125 pre-stored in the memory unit 12, obtains correction values ​​(weighting value correction values) for the road data corresponding to the keywords extracted from the dialogue (sentences) generated by the generation system AI 20, and corrects the road data (each weight value) in the road data table 123.

[0051] In other words, the weighting values ​​of the roads (links) used during route selection are adjusted according to the conditions that should be avoided on those roads. As a result, the cost of traveling along the roads (links) is appropriately adjusted according to the passenger's situation, and consequently, an appropriate route is selected that suits the passenger's situation.

[0052] In this embodiment, the AI20's response to the prompt question is a line indicating road conditions to be avoided; therefore, the keywords are words that indicate the road conditions or characteristics to be avoided. Specific keywords include words such as "shaking," "uneven road," "traffic jam," "curve," and "mountain road."

[0053] Figure 6 shows an example of a keyword data table 125. As shown in Figure 6, the items in the keyword data table 125 include "Data ID", "Keyword", "Synonym", "Road width weighting correction value", "Curve weighting correction value", "Road surface weighting correction value", and "Area weighting correction value".

[0054] The "Data ID" in Keyword Data Table 125 is identification information used to identify the dataset of a "keyword". A Data ID is set for each keyword, and a data record for that keyword is constructed for each Data ID.

[0055] The "keywords" in the keyword data table 125 are words corresponding to keywords extracted by the route correction unit 137 from the dialogue (text) generated by the generation system AI 20, and are data related to words that represent the type of road or the condition of the road among the avoidance road information corresponding to the person. As described above, keywords include words such as "shaking," "uneven road," "traffic jam," "curve," and "mountain road."

[0056] The "Synonyms" section of Keyword Data Table 125 contains data related to synonyms that describe road characteristics similar to the "Keyword." For example, if the keyword is "curve," synonyms would include words such as "sharp curve," "winding road," "winding road," and "winding road."

[0057] The "Road Width Weighting Correction Value," "Curve Weighting Correction Value," "Road Surface Weighting Correction Value," and "Area Weighting Correction Value" in Keyword Data Table 125 are data representing the correction values ​​for the "Road Width Weighting Value," "Curve Weighting Value," "Road Surface Weighting Value," and "Area Weighting Value" in Road Data Table 123, respectively. These data are pre-set (stored) based on keywords, for example, through experiments.

[0058] For example, if the keyword is "narrow road," the road width weighting adjustment value will be high because road width has a significant impact on passengers. Similarly, if the keyword is "curve," the curve weighting adjustment value will be high. Also, if the keyword is "bumpy road," the road surface weighting adjustment value will be high. And if the keyword is "traffic jam," the area weighting adjustment value will be high.

[0059] The route correction unit 137 then uses the correction value to correct the road data in the road data table 123 stored by the route generation unit 132. More specifically, the "road width weighting value" in the road data table 123 is corrected to, for example, (road width weighting value (initial value: a predetermined value is set when a predetermined change occurs in the passenger's condition)) × (road width weighting correction value). The "curve weighting value" in the road data table 123 is corrected to, for example, (curve weighting value) × (curve weighting correction value). The "road surface weighting value" in the road data table 123 is corrected to, for example, (road surface weighting value) × (road surface weighting correction value). The "area weighting value" in the road data table 123 is corrected to, for example, (area weighting value) × (area weighting correction value).

[0060] In this embodiment, if the dialogue generated by the AI20 contains multiple keywords, each weighting value is corrected multiple times using the respective correction values. It is also possible to apply a method of correcting each weighting value using the largest (highest) correction value.

[0061] The route generation unit 132 then calculates the travel cost for each link based on the weighted values ​​and rank values ​​corrected by the route correction unit 137. The route generation unit 132 then determines the travel route with the smallest total travel cost as the travel route to guide driver D1, etc.

[0062] The guidance unit 138 outputs the route information determined by the route generation unit 132 to the display device 3 and the speaker 4. In other words, the route guidance information is output as an image to the driver D1 etc. via the display device 3, and as an audio output to the driver D1 etc. via the speaker 4. In this way, the route guidance device 10 (route guidance system 1) generates and guides the passengers along a route suitable for their situation from the departure point to the destination of the vehicle V1.

[0063] Furthermore, if a sudden, predetermined change in circumstances occurs (is detected), for example, if the vehicle encounters an accident or traffic congestion while driving, or if, for example, child P2 starts drinking beverage C1 while driving, the route guidance device 10 will alert the driver D1, etc., via voice output or the like, and then perform route determination processing again.

[0064] According to the above configuration, when generating a driving route to guide vehicle V1 from the starting point to the destination, prompts containing information that do not necessarily require user input are automatically created and input to the generation system AI 20, and the driving route is generated based on the response of the generation system AI 20. By using the response of the generation system AI 20, the driving route is not generated based on predetermined conditions and states, so it is possible to generate diverse driving routes that take into account matters other than predetermined conditions and states, as well as the in-vehicle conditions and road conditions that change over time. Therefore, it becomes possible to guide the vehicle along a suitable driving route that takes into consideration the people inside the vehicle without requiring troublesome user input.

[0065] <3. Example of route guidance system operation> Figure 7 is a flowchart showing the route guidance process performed by the route guidance device 10. The operation shown in this flowchart is realized by a computer program (route guidance program 121) executed by the controller 13 (the computer that constitutes the controller 13).

[0066] The computer program that implements the route guidance method according to this embodiment in a computer device is installed in a computer device such as the route guidance device 10 to realize the various functions described above. Furthermore, such a computer program is provided to the computer device via a computer-readable non-volatile recording medium. For example, optical discs on which the computer program is recorded are distributed and sold, or computer programs stored on the hard disk of a server device are distributed and sold via a network environment. Also, the computer program that implements the route guidance method according to this embodiment in a computer device may consist of only one program, or it may consist of multiple programs.

[0067] The process shown in Figure 7 is initiated, for example, when vehicle V1 starts up, camera 2 begins acquiring images (video) of the interior of the vehicle, and the driver D1 or the like initiates the "route guidance mode according to the conditions inside the vehicle".

[0068] Prior to the start of this process, the controller 13 receives input for the destination of vehicle V1 from the driver D1 or the like via the destination setting unit 131 and acquires data for the departure point and destination of vehicle V1.

[0069] In step S101, the controller 13 (image acquisition unit 133) acquires images of the interior of the vehicle from the camera 2, and then proceeds to step S102. The controller 13 (image acquisition unit 133) stores the acquired images (image data) in the storage unit 12.

[0070] In step S102, the controller 13 (state detection unit 134) detects the type of person that may affect route guidance and the person's state based on the captured image acquired in step S101, and then proceeds to step S103. More specifically, the controller 13 (state detection unit 134) performs analysis processing on the captured image to detect, for example, the type of person, such as "pregnant woman" or "child," and the person's state, such as "holding a drink."

[0071] In step S103, the controller 13 determines whether the route search initiation conditions are met. If the conditions are met, the controller proceeds to step S104; otherwise, it proceeds to step S108. Possible route search initiation conditions include changes in the type or state of a person (changes that warrant a change in the route, with the conditions set in advance based on experiments, etc.), when a destination change operation is performed, when the route is deviated from, or when a route change is preferable due to road conditions (such as congestion).

[0072] In step S104, the controller 13 (prompt creation unit 135) creates a response acquisition prompt to obtain the type of person (passenger) detected in step S102 and the road to be avoided as dialogue, and then proceeds to step S105. More specifically, the controller 13 (prompt creation unit 135) uses the prompt template table 124 of the memory unit 12 to create a response acquisition prompt to output to the generation system AI 20.

[0073] In step S105, the controller 13 (response acquisition unit 136) outputs the response acquisition prompt created in step S103 to the generation system AI 20 to acquire the response (dialogue), and then proceeds to step S106. The controller 13 (response acquisition unit 136) stores the acquired response (dialogue) in the storage unit 12.

[0074] In step S106, the controller 13 (route correction unit 137) extracts keywords from the response (dialogue) obtained in step S105, corrects the road data related to the driving route based on those keywords, and proceeds to step S107. More specifically, the controller 13 (route correction unit 137) refers to the keyword data table 125 in the memory unit 12, obtains correction values ​​for the road data corresponding to the keywords extracted from the dialogue (sentence) generated by the generation system AI 20, and corrects the road data (each weighted value) in the road data table 123 of the memory unit 12.

[0075] Furthermore, to reduce the processing load when correcting road data, the correction may be limited to road links around the origin and destination (the area where an optimal route including the origin and destination is expected to exist (for example, the area within a predetermined distance around the straight line connecting the origin and destination)).

[0076] In step S107, the controller 13 (route correction unit 137) uses the road data corrected in step S106 to perform a route search (search for the route that minimizes the total travel cost of the route) to determine the route to guide the passenger, and then proceeds to step S108.

[0077] In step S108, the controller 13 (guidance unit 138) outputs the travel route determined in step S107 to the display device 3 and speaker 4, and proceeds to step S109. The travel route is output (provided) to the driver D1, etc., via the display device 3 and speaker 4, and route guidance is provided.

[0078] In step S109, the controller 13 terminates processing if the processing termination conditions are met, and returns to step S101 if they are not met. The termination conditions include, for example, arrival at the destination, completion of vehicle movement (engine stop, etc.), termination of the "route guidance mode according to the conditions inside the vehicle", termination of the drive of the route guidance device 10, etc.

[0079] <4. Variation> Furthermore, road data may be corrected using predetermined performance-related correction values ​​for each vehicle type V1. Figure 8 shows an example of the vehicle type performance data table 126. The vehicle type performance data table 126 is stored in the server device.

[0080] As shown in Figure 8, the items in the vehicle performance data table 126 include "Vehicle ID," "Road width weighting correction value," "Curve weighting correction value," and "Road surface weighting correction value." The "Vehicle ID" is identification information used to identify the dataset of various correction values ​​for each vehicle. A vehicle ID is set for each vehicle, and data records for various correction values ​​are constructed for each vehicle ID.

[0081] The "Road Width Weighting Correction Value," "Curve Weighting Correction Value," and "Road Surface Weighting Correction Value" in Vehicle Performance Data Table 126 represent the correction values ​​for the "Road Width Weighting Value," "Curve Weighting Value," and "Road Surface Weighting Value" in Road Data Table 123, respectively. This data suggests the extent to which road width, curves, and road surface conditions affect the occupants of the vehicle (vehicle of the vehicle type in question).

[0082] The route correction unit 137 then obtains various correction values ​​corresponding to the vehicle type ID of vehicle V1 from the server device via the network. More specifically, the vehicle (route guidance device 10) transmits its own identification data (vehicle identification data or vehicle type data) to the server device and receives correction value data corresponding to the identification data from the server device. The route correction unit 137 uses the obtained correction values ​​to correct the road data in the road data table 123 stored by the route generation unit 132. With this configuration, road data related to the driving route can be updated based on the vehicle's specifications (size, etc.) and performance. Therefore, from the perspective of vehicle type, it becomes possible to guide a suitable driving route that takes into consideration the people inside the vehicle.

[0083] Alternatively, instead of obtaining the vehicle type function data table 126 from the server device, the vehicle characteristics data table 127 of vehicle V1 may be stored in the storage unit 12, and the road data in the road data table 123 of the storage unit 12 may be corrected using various correction values ​​in the vehicle characteristics data table 127. Figure 9 shows an example of the vehicle characteristics data table 127.

[0084] As shown in Figure 9, the items in the vehicle characteristics data table 127 include "road width weighting correction value," "curve weighting correction value," and "road surface weighting correction value." The "road width weighting correction value," "curve weighting correction value," and "road surface weighting correction value" in the vehicle performance data table 126 are data that represent the correction values ​​for the "road width weighting value," "curve weighting value," and "road surface weighting value" in the road data table 123, respectively.

[0085] The route correction unit 137 then corrects the road data in the road data table 123 stored by the route generation unit 132, using various correction values ​​from the vehicle characteristics data table 127, in addition to the corrections based on the passenger's condition as described above. This configuration allows for updating of road data related to the driving route based on the vehicle's characteristics. Therefore, it becomes possible to guide the vehicle along a suitable driving route that takes into account the characteristics of the vehicle itself and the people inside the vehicle.

[0086] Furthermore, to reduce the processing load, it is also effective to narrow down the keywords used to determine the correction content of the road data table 123 extracted from the responses (dialogue) of the generative AI, depending on the area currently being driven in or planned to be driven in, and the type of road (special roads such as expressways). For example, if the user has specified driving on an expressway, there are almost no curved roads that have a significant impact on passengers, so it is effective to exclude "curve" from the keywords to be extracted. In addition, in suburban areas, there are very few traffic lights and other events that cause vehicles to start and stop, so it is effective to exclude events that cause vehicles to start and stop, such as "traffic lights," from the keywords to be extracted.

[0087] <5. Things to keep in mind> The various technical features disclosed as embodiments herein can be modified in various ways without departing from the spirit of the technical creation. That is, the above embodiments are illustrative in all respects and not restrictive. The technical scope of the present invention is indicated by the claims rather than by the above descriptions of embodiments, and includes all modifications that fall within the meaning and scope equivalent to the claims. Furthermore, the multiple embodiments shown herein may be combined as appropriate to the extent possible.

[0088] Furthermore, although the above embodiment explains that various functions are implemented in software through CPU arithmetic processing according to a program, at least some of these functions may be implemented by electrical hardware resources. These hardware resources may be implemented entirely or partially by, for example, ASICs (Application Specific Integrated Circuits) or FPGAs (Field Programmable Gate Arrays). Conversely, at least some of the functions implemented by hardware resources may be implemented in software.

[0089] Furthermore, the system may include a computer program that enables a processor (computer) to implement at least some of the functions of the route guidance system 1 (route guidance device 10). Such a computer program can be stored on a computer-readable non-volatile recording medium (for example, in addition to the non-volatile memory mentioned above, it can also be provided (sold, etc.) on an optical recording medium (for example, an optical disc), a magneto-optical recording medium (for example, a magneto-optical disc), a USB memory, or an SD card, etc.), and it can also be provided from a server device via a communication line such as the Internet, a method known as download provision. [Explanation of symbols]

[0090] 1. Route guidance system 2 cameras 3 Display device 4 speakers 10 Route guidance device 11 Communications Department 12 Storage section 13 Controllers 20 Generative AI 121 Route guidance program 122 Map data 123 Road Data Table 124 Prompt Template Table 125 Keyword Data Table 126 Vehicle Performance Data Table 127 Vehicle Characteristics Data Table 131 Destination Setting Section 132 Path generation unit 133 Image acquisition unit 134 State detection unit 135 Prompt Creation Section 136 Answer acquisition part 137 Route Correction Unit 138 Information Department

Claims

1. A route guidance device that generates and guides a vehicle on its route from its starting point to its destination, Based on images taken inside the vehicle, passenger information about the occupants is obtained. Based on the aforementioned passenger information, a response prompt is created to obtain information on alternative routes that the passenger should avoid. The generated response acquisition prompt is output to the generation AI to acquire the response. Based on the acquired response, the aforementioned driving route is generated and guidance is provided. Route guidance device.

2. The response acquisition prompt includes information on the type of person and the status of the person. The route guidance device according to claim 1.

3. Keywords are extracted from the response obtained from the generation AI, and road data related to the driving route is corrected based on the keywords. The route guidance device according to claim 1.

4. The road data includes data indicating the condition of the road, which is corrected based on the keywords. The route guidance device according to claim 3.

5. The aforementioned keyword is selected according to the area rank value of the travel route. The route guidance device according to claim 4.

6. A route guidance method that generates and provides directions for a vehicle's journey from its starting point to its destination, Based on images taken inside the vehicle, passenger information about the occupants is obtained. Based on the aforementioned passenger information, a response prompt is created to obtain information on alternative routes that the passenger should avoid. The generated response acquisition prompt is output to the generation AI to acquire the response. Based on the acquired response, the aforementioned driving route is generated and guidance is provided. Route guidance method.

7. A route guidance program that generates and provides directions for a vehicle's journey from its starting point to its destination, Based on images taken inside the vehicle, passenger information about the occupants is obtained. Based on the aforementioned passenger information, a response prompt is created to obtain information on alternative routes that the passenger should avoid. The generated response acquisition prompt is output to the generation AI to acquire the response. The process of generating and guiding the vehicle along the aforementioned route based on the acquired response is as follows: Let the computer do it. Route guidance program.

8. A route guidance system that generates and provides directions for a vehicle's journey from its starting point to its destination, The system comprises a route guidance device that generates and guides the aforementioned driving route, and a generation AI that outputs information on alternative routes. The aforementioned route guidance device, Based on images taken inside the vehicle, passenger information about the occupants is obtained. Based on the aforementioned passenger information, a response prompt is created to obtain information on alternative routes that the passenger should avoid. The generated response acquisition prompt is output to the generation system AI to acquire the response. Based on the acquired response, the aforementioned driving route is generated and guided. The aforementioned generation system AI is The aforementioned response acquisition prompt is used as the input value, and the aforementioned avoidance road information is output as the response. Route guidance system.

Images