Route training device, route training method and computer program

Abstract

In order to provide a technology relating to education for improving a skill to derive a vehicle's travel route to a destination, a provision unit of this route education device gives a question including information relating to a departure point and a destination and design to think about a vehicle's travel route from a departure point to a destination. A reception unit receives an answer travel route that is an answer to the question. A retrieval unit uses a route retrieval model and the given question to retrieve a recommended route from the departure point to the destination included in the question. The route retrieval model is a model that is generated by an expert selected on the basis of the amount of experience performing machine learning on actual travel routes traveled by the vehicle and outputs the recommended route to the destination. An evaluation unit compares the recommended route and the answer travel route, and evaluates the answer using a comparison result. An output unit outputs, as an evaluation result, information indicating evaluation for the answer.

Description

[Technical field]

[0001] The present invention relates to a training technique for improving skills in deriving a driving route for a vehicle to reach a destination. [Background technology]

[0002] The route that a vehicle travels to a destination is not necessarily the shortest route to the destination. In other words, various factors such as road congestion, weather, number of pedestrians, road width, and the presence or absence of road construction are involved in the travel of a vehicle. For this reason, depending on various factors, there may be a route that takes less time to reach the destination than a route that is the shortest distance to the destination. In many cases, it is considered better to take less time to reach the destination, and in such cases, the shortest route may not be the optimal route because it is not the route that takes the shortest time to reach the destination.

[0003] Patent Document 1 (JP Patent Publication No. 2015-224958) relates to a navigation device, and discloses technology that can guide people to their destination even in large amusement parks, apartment buildings, or supermarkets where there is no road information. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2015-224958 A Summary of the Invention [Problem to be solved by the invention]

[0005] Although various navigation technologies that guide a vehicle's driving route to a destination have been proposed, it is sometimes preferable for a driver to be able to determine the vehicle's driving route himself / herself without relying solely on a navigation device. For example, it is considered convenient for a driver who drives a vehicle professionally to have the skills to determine whether or not to drive the vehicle on a driving route suggested by a navigation device and to instantly determine a driving route to a destination.

[0006] For example, as mentioned above, it is often considered better to have a shorter travel time to the destination. In particular, for emergency vehicles involved in rescue activities, shortening the time required from receiving a call for dispatch to arriving at the scene is important for improving the survival rate. For this reason, it is useful for engineers to instantly calculate the vehicle's travel route that is expected to arrive at the destination in the shortest time.

[0007] A primary object of the present invention is to provide a training technique for improving the skill of deriving a vehicle driving route to a destination. [Means for solving the problem]

[0008] In order to achieve the above object, a route training device according to the present invention includes, as one aspect thereof, A question provider that provides a question including information on a departure point and a destination, the question asking about a vehicle's driving route from the departure point to the destination; A receiving unit that receives an answer route that is an answer to the question; a route search model that generates a recommended route to a destination by machine learning the actual driving routes of vehicles driven by experts selected based on the amount of experience, and outputs the recommended route to the destination, and a search unit that searches for a recommended route from the starting point to the destination included in the problem, using the problem; an evaluation unit that compares the searched recommended route with the answer travel route and evaluates the answer using a comparison result; an output unit that outputs information representing an evaluation of the answer as an evaluation result; Equipped with.

[0009] Furthermore, one aspect of the route training method according to the present invention is to By computer, A question is presented that includes information on a departure point and a destination and asks the user to think about a vehicle's driving route from the departure point to the destination; receiving an answer route to the question; searching for a recommended route from the starting point to the destination included in the problem, using a route search model that outputs a recommended route to the destination, and the problem, and the route search model is generated by machine learning the actual driving routes that have been driven by an expert selected based on the amount of experience in a vehicle; comparing the searched recommended route with the answer travel route, and evaluating the answer using the comparison result; Information representing an evaluation of the answer is output as an evaluation result.

[0010] Furthermore, a program storage medium according to the present invention comprises, as one aspect thereof: A process of setting a question including information on a departure point and a destination, the question being about a vehicle's travel route from the departure point to the destination; receiving an answer route to the question; A process of searching for a recommended route from the starting point to the destination included in the problem, using a route search model that is generated by machine learning of actual driving routes traveled by vehicles by experts selected based on the amount of experience and outputs a recommended route to the destination, and the problem; A process of comparing the searched recommended route with the answer travel route and evaluating the answer using a comparison result; A process of outputting information representing an evaluation of the answer as an evaluation result; The computer program for causing a computer to execute the above is stored. Effect of the Invention

[0011] According to the present invention, it is possible to provide a training technique for improving the skill of deriving a driving route for a vehicle to reach a destination. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating a configuration of a route training system in a first embodiment. [Diagram 2] FIG. 13 is a diagram illustrating an example of a display of an evaluation result on a terminal device. [Diagram 3] FIG. 11 is a diagram illustrating another example of how the evaluation results are displayed on the terminal device. [Figure 4] FIG. 11 is a diagram illustrating another example of how the evaluation results are displayed on the terminal device. [Diagram 5] FIG. 11 is a diagram illustrating another example of how the evaluation results are displayed on the terminal device. [Figure 6] FIG. 11 is a diagram illustrating another example of how the evaluation results are displayed on the terminal device. [Figure 7] FIG. 11 is a diagram illustrating another example of how the evaluation results are displayed on the terminal device. [Figure 8] FIG. 13 is a diagram showing an example of a screen for allowing an engine operator to input information on the amount of experience. [Figure 9] 5 is a flowchart illustrating an example of an operation related to route training in the route training device of the first embodiment. [Figure 10] FIG. 11 is a diagram illustrating the configuration of a route training system in a second embodiment. [Figure 11] FIG. 13 is a block diagram showing the configuration of a route training device according to another embodiment. [Figure 12] 12 is a flowchart illustrating an example of an operation related to route education of the route education device illustrated in FIG. 11. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0014] First Embodiment FIG. 1 is a block diagram showing the configuration of a route education system including a route education device according to a first embodiment of the present invention. This route education system 1 is a system that implements route education to educate people with little experience in driving vehicles in order to improve their skills in deriving a driving route until the vehicle reaches a destination. In the first embodiment, the vehicle is assumed to be an emergency vehicle, and the education recipients are inexperienced engine personnel and people aiming to become engine personnel. An emergency vehicle is a vehicle that is designated and permitted to have priority passage on a road or priority use of a road in the event of an emergency business. Emergency vehicles include ambulances, fire engines, police vehicles, and vehicles used for emergency work in public utilities such as electricity, gas, and water, but here, ambulances and fire engines are particularly assumed. Of course, the route education system 1 can be applied to emergency vehicles other than ambulances and fire engines.

[0015] The route education system 1 comprises a route education device 2 and a terminal device 4. The route education device 2 is a computer device (e.g., a server) and can be connected to the terminal device 4 via an information and communication network including a wireless communication network. In the example of Fig. 1, the route education device 2 is connected to one terminal device 4, but multiple terminal devices 4 can be connected.

[0016] Here, the configuration of the terminal device 4 will be described. The terminal device 4 is an information device having a communication function and a display control function for controlling a display device to display information. The type of the terminal device 4 is not limited as long as it is an information device having a communication function and a display control function, and specific examples include a smartphone, a tablet, and a personal computer (PC). When the terminal device 4 is a smartphone or a tablet, the display device and the input device (such as a touch panel) are integrated with the terminal device 4, but when the terminal device 4 is a personal computer, the display device and the input device (such as a keyboard or a mouse) may be external.

[0017] The terminal device 4 is a device through which a trainee of route training (training to improve the skill of deriving a driving route until a vehicle reaches a destination) receives questions and answers the questions, and includes a control device 40, a display device 41, a storage device 42, and an input device 44. The display device 41 is a device that displays information using characters, images, videos, etc. The input device 44 is a device through which a person operating the terminal device 4 (hereinafter also referred to as a user or a trainee) inputs information into the terminal device 4, and is, for example, a touch panel, a keyboard, a mouse, etc. In the example of FIG. 1, the display device 41 and the input device 44 are built into the terminal device 4, but one or both of the display device 41 and the input device 44 may be external.

[0018] The storage device 42 includes a storage medium for storing data and a computer program (hereinafter also referred to as a program) 43. There are multiple types of storage devices, such as a magnetic disk device and a semiconductor memory element, and there are multiple types of semiconductor memory elements, such as a RAM (Random Access Memory) and a ROM (Read Only Memory). The type of storage device 42 included in the terminal device 4 is not limited to one. A computer device is often provided with multiple types of storage devices. Here, the type and number of storage devices 42 included in the terminal device 4 are not limited, and a description thereof will be omitted. Also, when the terminal device 4 includes multiple types of storage devices 42, they will be collectively referred to as the storage device 42.

[0019] In the first embodiment, an application program relating to route education (hereinafter also referred to as a route education application) is stored as one of the programs 43 in the storage device 42.

[0020] The control device 40 of the terminal device 4 is configured with a processor such as a CPU (Central Processing Unit). The control device 40 can have various functions based on a program 43 by reading and executing the program 43 stored in the storage device 42. Here, the control device 40 has a communication unit 45, a reception unit 46, and a presentation unit 47 as functional units based on the route training application.

[0021] The communication unit 45 communicates information with the route training device 2. Specifically, for example, the communication unit 45 receives questions to be answered by the trainee from the route training device 2 in order to implement route training. The communication unit 45 also transmits the answers given by the trainee to the received questions to the route training device 2. Furthermore, the communication unit 45 receives an evaluation result from the route training device 2, which is information indicating an evaluation of the answers given by the trainee.

[0022] The presentation unit 47 causes the display device 41 to display the questions received from the route training device 2 and the evaluation results regarding the answers to the questions in a predetermined display format.

[0023] The reception unit 46 receives information input to the terminal device 4 by the user of the terminal device 4. Here, one piece of information received by the reception unit 46 is an answer given by the user (trainee) to a question displayed on the display device 41 by the presentation unit 47. That is, in the first embodiment, a question output from the route education device 2 to the terminal device 4 is a question in which information on a starting point and a destination is given and the trainee is asked to answer a vehicle travel route from the starting point to the destination. For example, a map on which the starting point and the destination are written is displayed on the display device 41 by the presentation unit 47 together with a question sentence asking such a question. When information on a vehicle travel route from the starting point to the destination thought up by the trainee is input as an answer using the input device 44 using the displayed map, the reception unit 46 receives the input information on the travel route as an answer travel route. The information on the answer travel route received in this way is transmitted to the route education device 2 by the communication unit 45.

[0024] The method by which the trainee answers the question may be, for example, a method in which the trainee answers by drawing the answer driving route on a map, or a method in which the trainee selects the answer driving route from a plurality of driving routes presented as options.

[0025] Next, the route education device 2 will be described. The route education device 2 transmits questions (asks questions) to the terminal device 4, evaluates answers received from the terminal device 4, and transmits the evaluation results to the terminal device 4. In the first embodiment, the route education device 2 uses AI (Artificial Intelligence) technology to evaluate the answer travel route answered to the question transmitted to the terminal device 4. That is, the route education device 2 uses a route search model based on AI technology to search for a recommended route from the departure point to the destination included in the question transmitted to the terminal device 4, and evaluates the answer travel route by comparing the recommended route with the answer travel route.

[0026] The route search model is a model generated by machine learning the actual driving routes of experts. An expert is, for example, an engine worker who is selected based on the amount of experience, such as the number of times that the expert has been dispatched in the area of ​​responsibility that he / she is responsible for as part of his / her job, being a threshold value (e.g., 100 times) or more. The route search model is generated by accumulating the actual driving routes traveled by such experts when they are actually dispatched, and by machine learning the actual driving routes.

[0027] This route search model takes as input data search condition information including information representing the departure point (hereinafter also referred to as departure point information) and information representing the destination (hereinafter also referred to as destination information), and outputs a recommended route from the departure point to the destination.

[0028] An example of machine learning for generating a route search model is intent learning. Intention learning is a technique that uses inverse reinforcement learning as a core technology and machine-learns an objective function that maximizes actions using data that indicates when (state) and what decision was made (action). When generating a route search model using such intent learning, not only the actual driving route but also information that indicates the surrounding conditions when actually driving, for example, the conditions in the assigned area (hereinafter also referred to as situation information) is used to generate the model. Examples of situation information include date and time information, the presence or absence of an event, the location of the event, the scale of the event (number of participants, etc.), road traffic conditions such as congestion, the presence or absence of a disaster and the disaster situation, weather, and the implementation status of construction work.

[0029] Unlike ordinary vehicles, emergency vehicles are special vehicles that are allowed to drive with priority (driving on shoulders, driving the wrong way on one-way roads, etc.). It can be said that such priority driving information is reflected in the actual driving route. As a result, when generating a route search model, it is possible to generate a model that reflects information about shoulders, even if information related to priority driving, such as the existence of shoulders wide enough for emergency vehicles, is not input.

[0030] Here, the configuration of the route education device 2 will be described. As shown in FIG. 1, the route education device 2 includes a control device 20 and a storage device 22. The storage device 22 includes a storage medium for storing data and a computer program (program) 23. There are multiple types of storage devices, such as a magnetic disk device and a semiconductor memory device, and there are multiple types of semiconductor memory devices, such as a RAM (Random Access Memory) and a ROM (Read Only Memory). The type of storage device 22 included in the route education device 2 is not limited to one. A computer device is often equipped with multiple types of storage devices. Here, the type and number of storage devices 22 included in the route education device 2 are not limited, and the description thereof will be omitted. In addition, when the route education device 2 includes multiple types of storage devices 22, they will be collectively referred to as the storage device 22. The route education device 2 may be connected to an external storage device (database) 6, and the route education device 2 may write data and programs to the storage device 6 and read data and programs from the storage device 6 as appropriate. Furthermore, in the following description, when it is stated that "read from memory device 22" or "store in memory device 22", this also includes the possibility of reading from memory device 6 or storing in memory device 6, and a description of memory device 6 will be omitted.

[0031] In the first embodiment, the storage device 22 stores a computer program for executing processing related to route education as one of the programs 23. The storage device 22 also stores the above-mentioned route search model as one of the programs 23. Furthermore, the storage device 22 stores a past travel route, situation information at the time of travel, and information indicating the experience of the engineer who traveled the route, in association with each other as past data. The storage device 22 also stores information on the experience used when selecting an expert (such as information that the number of times the engineer has been dispatched in the area of ​​responsibility is 100 or more), information on the number of data points of the past travel route used to generate the route search model, and the like.

[0032] The control device 20 is configured with a processor such as a CPU. The control device 20 can have various functions based on a program 23 by reading and executing the program 23 stored in the storage device 22. Here, the control device 20 has a providing unit 24, a receiving unit 25, a searching unit 26, an evaluating unit 27, and an output unit 28 as functional units related to the process of route education.

[0033] The providing unit 24 issues (transmits) to the terminal device 4 a problem including departure point information and destination information, which asks for a vehicle's driving route from the departure point to the destination. The timing at which the providing unit 24 issues a problem to the terminal device 4 is, for example, when a notification of the start of training or a command requesting the next problem is received from the terminal device 4. The problem set by the providing unit 24 may be related to an area imaginary for the problem, but here, for example, the problem is a problem of asking for a driving route in an actual area, such as an area in which the training subject is in charge of work. In addition, in the first embodiment, the problem set by the providing unit 24 includes hypothetical situation information that indicates a hypothetical situation in an area including the departure point and the destination. The hypothetical situation information is the same type of information as the situation information used to generate a route search model. For example, the hypothetical situation information includes information on date and time, the presence or absence of an event, the location of the event, the scale of the event (the number of participants, etc.), road traffic conditions such as congestion, the presence or absence of a disaster and the disaster situation, weather, and the implementation status of construction work.

[0034] The providing unit 24 is capable of posing a plurality of different questions with various changes to such hypothetical situation information, starting points, and destinations. The questions transmitted by the providing unit 24 to the terminal device 4 are associated with question identification information for identifying the question. Note that the method of generating the plurality of different questions is not limited, and a description thereof will be omitted.

[0035] The receiving unit 25 receives from the terminal device 4 information on the answer travel route, which is the answer to the question posed by the providing unit 24. The terminal device 4 associates this answer travel route with question identification information related to the question corresponding to the answer travel route.

[0036] The search unit 26 searches for a recommended route using the departure point information, destination information, and assumed situation information contained in the question posed by the provision unit 24, and the route search model stored in the storage device 22.

[0037] In addition, when it is assumed that one of the routes from the departure point to the destination is a route that passes through a railroad crossing, information on train operation information may be included as the assumed situation information. Then, using the train operation information and information on the departure time as another assumed situation information, it may be determined whether or not the train will pass through the railroad crossing at the time it is estimated that the train will pass through the railroad crossing, and the result of this determination may be used in the search process for the recommended route. In addition, there are traffic lights that detect emergency vehicles and control the lighting of red and green lights so as not to impede the passage of the emergency vehicle. Information on the locations of such traffic lights may also be used in the search process for the recommended route.

[0038] The recommended route searched by the search unit 26 is a route searched using a route search model generated by machine learning the actual travel routes of the skilled person, and is therefore not necessarily a simple shortest route from the departure point to the destination. For example, if the shortest route with the shortest travel distance includes a railroad crossing, there is a risk of waiting for a train to pass at the railroad crossing, and the arrival at the destination may be delayed due to the train passing. In contrast, the route search model based on the actual travel routes of the skilled person outputs, as a recommended route, a detour route with a longer travel distance than the shortest route, but a route that can use a grade separation without a railroad crossing. This recommended route is a route that can avoid delays caused by waiting for a train to pass at a railroad crossing.

[0039] Incidentally, the intention learning mentioned as one of the machine learning methods for generating a route search model is a so-called explainable AI (Artificial Intelligence). For this reason, when a route search model is generated by intention learning, the route search model can also output an explanation of the grounds for recommending a recommended route (hereinafter also referred to as recommendation grounds information). Thus, when using a route search model generated by intention learning, the search unit 26 may also output recommendation grounds information for the recommended route.

[0040] Furthermore, when the performance data (information including the performance travel route and the situation information) is stored in the storage device 22, the search unit 26 may also search for the following data from the performance travel routes stored in the storage device 22 when searching for a recommended route. That is, the search unit 26 searches for a performance travel route that has a similar or the same combination of the starting point and the destination included in the question as a reference route. More preferably, the search unit 26 searches for a performance travel route of performance data that includes not only the combination of the starting point and the destination but also situation information (e.g., time period, day of the week, weather, etc.) that is similar to the situation of the assumed situation information included in the question as a reference route from the performance data in the storage device 22. In this case, the search unit 26 reads out, for example, situation information associated with the reference route (performance travel route) from the storage device 22.

[0041] Furthermore, the route search model may be configured to output not only the recommended route but also the estimated time required to reach the destination when traveling along the recommended route, and in this case, the search unit 26 may also output information on the required time.

[0042] The evaluation unit 27 compares the recommended route output by the search unit 26 with the answer travel route received by the receiving unit 25, and evaluates the answer (answer travel route) using the comparison result. Furthermore, in the first embodiment, the evaluation unit 27 outputs an explanation for the evaluation. The evaluation method for evaluating the answer travel route by referring to the recommended route and the explanation output method for outputting an explanation for the evaluation are not limited here, but for example, AI technology may be used. When AI technology is used, a model (hereinafter also referred to as an evaluation explanation model) generated by machine learning the recommended route, the answer travel route, the comparison result between these routes, and the evaluation and explanation is used. This evaluation explanation model is generated in advance and stored in the storage device 22.

[0043] The output unit 28 returns (outputs) information representing the evaluation by the evaluation unit 27 to the terminal device 4 as an evaluation result. In addition, the output unit 28 also outputs an explanation of the evaluation by the evaluation unit 27 to the terminal device 4. At this time, the output unit 28 may also output the recommended route used in the evaluation to the terminal device 4. In addition, when the search unit 26 also outputs recommended basis information of the route, the output unit 28 also outputs the recommended basis information to the terminal device 4. Furthermore, when the search unit 26 also searches for a reference route, the output unit 28 also outputs the reference route to the terminal device 4. Furthermore, when the search unit 26 reads out not only the reference route but also situation information associated with the reference route from the storage device 22, the output unit 28 outputs the situation information read out by the search unit 26 and performance data including the reference route to the terminal device 4. Furthermore, when information on the required time is output by the search unit 26, the output unit 28 may also output the information on the required time to the terminal device 4.

[0044] Furthermore, when transmitting the recommended route to the terminal device 4, the output unit 28 may also transmit information on the number of data points of actual driving routes used to generate the route search model and the amount of experience used to select an expert.

[0045] The communication unit 45 of the terminal device 4 receives the information including the evaluation result transmitted by the output unit 28, and the presentation unit 47 of the terminal device 4 displays the information including the evaluation result.

[0046] FIG. 2 shows an example of a display showing the evaluation result on the display device 41 of the terminal device 4. In the example of FIG. 2, an evaluation rank, which is the evaluation result, and its explanation are shown, and information showing the answer driving route and the recommended route is superimposed on the map as reference information, and information showing the starting point and the destination is also superimposed on the map. In addition, when the number of data of the actual driving route used to generate the route search model and information on the amount of experience used to select an expert are transmitted from the output unit 28 to the terminal device 4, the information may be displayed on the display device 41 as in the part surrounded by the dotted line A in FIG. 2. It is assumed that such information can make the trainee who has seen the recommended route trust the recommended route, leading to an increase in the reliability of the evaluation using the recommended route. Note that the evaluation result may simply be, for example, either a correct answer or an incorrect answer, instead of an evaluation rank.

[0047] The display mode of the evaluation result is not limited to the example of Fig. 2. For example, as shown in Fig. 3, in addition to the illustration of the recommended route, characters and symbols explaining the recommended route may also be displayed as reference information. Furthermore, when route recommendation basis information is also output from the route education device 2 to the terminal device 4, an explanation based on the recommended basis information may also be displayed on the display device 41 of the terminal device 4 as shown in Fig. 4. Furthermore, the explanation of the recommended route and the explanation of the basis of the recommended route may be configured to be selectable between display and non-display. It is considered that displaying the recommended basis information will give the education recipient who has viewed the recommended route a sense of satisfaction with the evaluation result.

[0048] Furthermore, when the actual travel route as the reference route searched by the search unit 26 is transmitted to the terminal device 4, the terminal device 4 may display the reference route on the display device 41 together with the recommended route, as shown in FIG. 5. Furthermore, when not only the reference route but also situation information when an emergency vehicle actually traveled along the reference route is transmitted to the terminal device 4 by the output unit 28, the terminal device 4 may display the situation information on the display device 41, as shown in FIG. 6. In addition, the configuration may be such that the display or non-display of the situation information can be selected. Furthermore, information representing the experience of the engineer who traveled along the reference route may be displayed. Such information representing the experience of the engineer is included in the performance data, and is transmitted from the route training device 2 to the terminal device 4 in association with the reference route. It is considered that displaying the reference route also leads to the trainee who has viewed the recommended route gaining trust in the recommended route and the evaluation results.

[0049] Furthermore, if information on the estimated time required to travel the recommended route is also output from the route education device 2 to the terminal device 4, the information on the required time may also be associated with the recommended route and displayed on the display device 41.

[0050] Incidentally, it is considered that the training subjects include beginners who have almost no experience (for example, the number of dispatches) to those who have experience to be called mid-level, and it is assumed that there is a difference in the amount of experience possessed by the training subjects. Therefore, the display mode of the reference information regarding the evaluation, etc. on the display device 41 may be different according to the amount of experience of the training subjects. For example, when the amount of experience (number of dispatches) of the training subjects is less than a threshold, as shown in FIG. 7, not only the recommended route is displayed on the display device 41, but also explanations and photos of points to be careful of on and around the recommended route, and explanations of the reasons for the recommended route are displayed. On the other hand, when the amount of experience (number of dispatches) of the training subjects is equal to or greater than the threshold, as shown in FIG. 2, the explanations of points to be careful of on and around the recommended route are omitted on the display device 41, and a simple display is provided. Note that even if the amount of experience (number of dispatches) of the training subjects is equal to or greater than the threshold, explanations and photos of points to be careful of on and around the recommended route may be popped up on the display device 41 according to the request of the training subjects. Furthermore, the experience of the training recipients may be divided into three or more ranks, and the display mode of the reference information regarding the evaluation results on the display device 41 may differ for each rank.

[0051] In this way, when the display mode on the display device 41 is made different depending on the experience amount of the trainee, a screen such as that shown in Fig. 8 may be displayed on the display device 41 to have the trainee input the experience amount of the trainee into the terminal device 4. The trainee manually inputs information indicating the experience amount (hereinafter also referred to as experience amount information) into the terminal device 4 according to this screen display, so that the terminal device 4 can acquire the experience amount information of the trainee. Furthermore, the terminal device 4 may acquire the experience amount information of the trainee as follows.

[0052] For example, the route training device 2 is capable of reading out information included in the crew member data from a database in which the crew member data is stored. The crew member data is data in which crew member identification information for identifying the crew member is associated with the experience level information of the crew member. The terminal device 4 has a function of acquiring crew member identification information of the crew member who is the training subject. Specifically, for example, the terminal device 4 acquires the crew member identification information manually input by the training subject by displaying a screen on the display device 41 that allows the training subject to input his / her crew member identification information into the terminal device 4. Alternatively, for example, the terminal device 4 may be equipped with a camera, which may photograph the face of the training subject crew member, and may acquire the crew member identification information of the training subject from the photographed image of the face using face recognition technology. The terminal device 4 may also acquire the crew member identification information of the training subject using biometric authentication other than face recognition (authentication using fingerprints, iris, veins, voice, ear acoustics, behavior, etc.). In this case, the terminal device 4 is provided with an acquisition device for acquiring the biometric information of the training subject for biometric authentication other than face recognition. Furthermore, for example, a two-dimensional code including the crew member identification information is written on the ID card carried by the crew member. The terminal device 4 may take a picture of the two-dimensional code with a camera and obtain the crew member identification information from the captured image using a two-dimensional code reading application. After obtaining the crew member identification information in this way, the terminal device 4 transmits the obtained crew member identification information to the route training device 2. As a result, the route training device 2 reads out the experience amount information of the crew member data corresponding to the crew member identification information from the database storing the crew member data, and returns it to the terminal device 4. As a result, the terminal device 4 can obtain the experience amount information of the training subject. When biometric authentication is performed as described above, data in which the biometric information of the crew member used in the biometric authentication and the crew member identification information are associated is stored in the storage device 42 of the terminal device 4 or the storage device 22 of the route training device 2. In addition, when the training subject is a person other than a crew member, for example, if the terminal device 4 is unable to obtain the crew member identification information, it may determine that the training subject is not a crew member and obtain information indicating no experience (zero number of deployments) as the experience amount information of the training subject.

[0053] Furthermore, when the evaluation result is displayed on the display device 41 of the terminal device 4, for example, images or videos showing the surrounding scenery seen from the driver's viewpoint when the vehicle travels along the recommended route, the answered travel route, or the reference route may also be displayed on the display device 41. Furthermore, when the evaluation result is displayed on the display device 41 of the terminal device 4, if there are matters to be careful of on the answered travel route (for example, in the case of heavy rain, it is better to avoid passing through the area because of the risk of flooding), an explanation of the matters may also be displayed on the display device 41.

[0054] The various display modes for showing the evaluation results on the display device 41 of the terminal device 4 as described above are realized by the route training application and the control device 40 that executes the application.

[0055] The route education system 1 in the first embodiment is configured as described above. Next, an example of the operation of a process related to route education in the route education device 2 in the first embodiment will be described with reference to Fig. 9. Fig. 9 is a flowchart illustrating an example of the operation of a process related to route education in the route education device 2.

[0056] For example, when the route education device 2 detects that it is a predetermined question-setting timing (for example, the timing when the route education device 2 receives a notification from the terminal device 4 informing the user of the start of education), the providing unit 24 transmits (sets) a question to the terminal device 4 (step 101 in FIG. 9). As described above, the question includes departure point information and destination information, and asks the user to consider a vehicle driving route from the departure point to the destination. This question also includes, for example, hypothetical situation information that indicates hypothetical conditions in the area including the departure point and the destination.

[0057] Thereafter, when the receiving unit 25 receives an answer travel route from the terminal device 4, which is an answer to the question, (step 102), the searching unit 26 searches for a recommended route from the departure point to the destination based on the search condition information, using the search condition information (starting point information and destination information) included in the question, the assumed situation information, and the route search model (step 103). Note that the trigger for the searching unit 26 to start searching for a recommended route may be when a question is sent (asked) to the terminal device 4 by the providing unit 24, rather than when the answer travel route is received, and is not limited to this example.

[0058] Thereafter, the evaluation unit 27 compares the recommended route output by the search unit 26 with the answer travel route received by the receiving unit 25, and evaluates the answer (answer travel route) using the comparison result (step 104). Then, the output unit 28 returns (outputs) the evaluation result by the evaluation unit 27 to the terminal device 4 (step 105).

[0059] The route education system 1 and the route education device 2 in the first embodiment evaluate the answer (answered driving route) given by the trainee using a recommended route output from a route search model based on the past driving routes of the expert. This allows the trainee to learn the way experts think when deriving a driving route. In other words, the route education system 1 and the route education device 2 can provide the trainee with education to improve the skill of deriving a driving route for a vehicle to reach a destination.

[0060] That is, when considering the driving route of a vehicle from a departure point to a destination, the time required from the departure point to the destination is one of the matters to be considered. In particular, when an emergency vehicle heads to a site where a dispatch request has been made, the time required may affect the survival rate, and the time required is an important matter. Such a time required to reach a destination is related to many factors such as the time of day, day of the week, and weather related to human movement, and the road width and the presence or absence of railroad crossings related to the vehicle's driving speed. It is considered that an expert acquires the relationship between such many factors related to the time required to reach a destination and the time required to reach a destination through experience, and drives along a route based on the experience. By using a recommended route output from a route search model that machine-learns the actual driving route based on such an expert's tacit knowledge, the route education system 1 and the route education device 2 can educate the trainee to improve the skill of deriving the driving route of the vehicle to the destination.

[0061] In addition, because emergency vehicles are given priority and can travel different routes from general vehicles, it is believed that experience in driving emergency vehicles is more effective in determining a route for directing an emergency vehicle to a destination than in determining a route for a general vehicle. For this reason, it is believed that it is more effective to use a recommended route output from a route search model based on the actual driving routes of an expert in training to improve the skill of deriving a driving route for an emergency vehicle.

[0062] <Second embodiment> A second embodiment of the present invention will be described below. In the description of the second embodiment, the same reference numerals are used to designate the same components as those in the route training system 1 of the first embodiment, and a duplicate description of the common components will be omitted.

[0063] FIG. 10 is a block diagram showing the configuration of the route education system 1 in the second embodiment. In the second embodiment, the route education device 2 further includes a learning unit 29 as a functional unit in the control device 20 in addition to the configuration in the first embodiment. The learning unit 29 has a function of re-learning the route search model. In the route education system 1 in the second embodiment, in order to obtain data used for the re-learning, an answer travel route that is an answer given by an expert to a question is obtained as an expert-answered route. That is, in the second embodiment, it is assumed that not only the trainee but also the expert will answer the question for considering a travel route posed by the terminal device 4 (i.e., the question posed to the terminal device 4 by the providing unit 24 of the route education device 2).

[0064] In the second embodiment, for example, before a trainee or an expert (hereinafter also referred to as an answerer) answers a question about a driving route, the terminal device 4 acquires the experience amount information of the answerer by the acquisition method as described in the first embodiment. This acquired experience amount information is associated with the answer driving route transmitted from the terminal device 4 to the route training device 2. By using this experience amount information, it is possible to determine whether the person who answered the answer driving route is a trainee or an expert. The learning unit 29 acquires, as an expert answered route, an answer driving route associated with experience amount information representing the experience amount selected as an expert as described in the first embodiment.

[0065] The learning unit 29 stores (accumulates) the acquired expert-answered route in the storage device 22. Then, at a predetermined update timing, the learning unit 29 updates the route search model stored in the storage device 22 by re-learning the route search model using an expert-answered route newly acquired after the previous update. The update timing may be every time a predetermined period (e.g., three months) elapses, or every time the number of expert-answered routes accumulated after the previous update reaches a predetermined number, and may be set appropriately here.

[0066] Other than the above, the configuration of route training system 1 in the second embodiment is similar to the configuration of route training system 1 in the first embodiment.

[0067] The route education system 1 of the second embodiment has the same configuration as the first embodiment, and can therefore achieve the same effects as the first embodiment. In the second embodiment, the route education device 2 further has a function of acquiring and storing routes answered by an expert from the terminal device 4, and updating the route search model using the stored routes answered by an expert. As a result, the route search model is updated, and the route education system 1 can increase the reliability of the recommended route output from the route education device 2, and therefore the reliability of the evaluation results can also be increased.

[0068] <Other embodiments> The present invention is not limited to the first and second embodiments, and may be implemented in various ways. For example, the route search model may be generated so that a plurality of recommended routes can be output with priorities. In this case, the search unit 26 may use such a route search model to search for a plurality of recommended routes with priorities. In this way, when the search unit 26 searches for information on a plurality of recommended routes and priorities, the evaluation unit 27 uses the information on the plurality of recommended routes and priorities to evaluate the answer travel route. This evaluation method is not particularly limited, but it is possible to use, for example, AI technology. In this case, an evaluation and commentary model generated by machine learning a plurality of recommended routes and their priorities, an answer travel route, a comparison result of those routes, and the evaluation and commentary thereof is used to evaluate the answer travel route.

[0069] In addition, when the search unit 26 searches for multiple recommended routes and priority information, when the evaluation results are displayed on the display device 41 of the terminal device 4, the multiple recommended routes may be displayed in order of priority as reference information.

[0070] Furthermore, the route search model may be generated by weighting the actual driving route used in generating the model according to the amount of experience of the expert who has traveled the route. Furthermore, the route search model may be generated by machine learning using not only the actual driving routes of the expert, but also data of the comparison result between the actual driving routes of the expert and the actual driving routes of the expert (inexperienced person) whose amount of experience is less than a predetermined numerical value.

[0071] Furthermore, in the first and second embodiments, the configuration of the route training system is described by taking an emergency vehicle as an example of a vehicle. Instead of the emergency vehicle, the route training system 1 in the first and second embodiments can be applied to route training for a vehicle in which the experience of an expert is utilized for route training, such as a taxi or a delivery vehicle. When applied to a taxi or a delivery vehicle, the information on the amount of experience is, for example, information indicating the number of years of experience as a driver. In this way, the information on the amount of experience is not limited to the number of times an emergency vehicle is dispatched.

[0072] Furthermore, in the first and second embodiments, the route training device 2 and the terminal device 4 operated by the training recipient are separate entities. Alternatively, the terminal device 4 may also have the functions of the route training device 2. In this case, the acceptance unit 46 of the terminal device 4 functions as a receiving unit of the route training device 2. In addition, the presentation unit 47 of the terminal device 4 also functions as an output unit of the route training device 2.

[0073] FIG. 11 is a block diagram showing the basic configuration of the route education device according to the present invention. The route education device 50 is, for example, a computer device, and includes a providing unit 51, a receiving unit 52, a searching unit 53, an evaluating unit 54, and an output unit 55 as functional units by the operation of a computer based on a computer program. The providing unit 51 asks a question including information on a starting point and a destination, and asks the user to think about a vehicle's driving route from the starting point to the destination. The receiving unit 52 receives an answer driving route that is an answer to the question. The searching unit 53 searches for a recommended route from the starting point to the destination included in the question, using a route search model and the asked question. The route search model is a model that outputs a recommended route to the destination, which is generated by machine learning the actual driving routes traveled by a skilled person selected based on the amount of experience. The evaluating unit 54 compares the searched recommended route with the answer driving route, and evaluates the answer using the comparison result. The output unit 55 outputs information indicating the evaluation of the answer as an evaluation result.

[0074] Next, an example of the operation of the route education device 50 related to route education will be described with reference to Fig. 12. For example, the providing unit 51 of the route education device 50 poses a question for considering a vehicle driving route from a departure point to a destination (step 201 in Fig. 12). After that, when the receiving unit 52 receives an answer driving route that is an answer to the posed question (step 202), the searching unit 53 searches for a recommended route using a route search model and the question (step 203). Note that the searching unit 53 may not start a search when the answer driving route is received, but may start a search when a question is posed, for example, and the timing when the searching unit 53 starts a search is not limited to this example.

[0075] Thereafter, the evaluation unit 54 compares the searched recommended route with the answer travel route, and evaluates the answer using the comparison result (step 204). Thereafter, the output unit 55 outputs information representing the evaluation of the answer as the evaluation result (step 205).

[0076] The route education device 50 as described above evaluates the answer using a recommended route output from a route search model based on the actual driving routes of experts. In this way, the route education device 50 can provide education to the trainee to improve the skill of deriving a driving route for a vehicle to reach a destination.

[0077] The present invention has been described above by taking the above-mentioned embodiment as an exemplary example. However, the present invention is not limited to the above-mentioned embodiment. That is, the present invention can be applied in various aspects that can be understood by a person skilled in the art within the scope of the present invention.

[0078] This application claims priority based on Japanese Patent Application No. 2021-155259, filed on September 24, 2021, the disclosure of which is incorporated herein in its entirety. [Explanation of symbols]

[0079] 1. Route Education System 2,50 Route training device 4 Terminal Equipment 24,51 Providing Department 25,52 Receiver 26,53 Search section 27,54 Evaluation section 28,55 Output section

Claims

1. A providing means for providing a question including information on a departure point and a destination, the question prompting a user to think about a vehicle's travel route from the departure point to the destination; A receiving means for receiving an answer route which is an answer to the question; a route search model that generates a recommended route to a destination by machine learning the actual driving routes of vehicles driven by experts selected based on the amount of experience, and a search means that searches for a recommended route from the starting point to the destination included in the problem, using the problem; an evaluation means for comparing the searched recommended route with the answer travel route and evaluating the answer using a comparison result; an output means for outputting information representing an evaluation of the answer as an evaluation result; A route training device comprising:

2. the route search model is a model generated by machine learning including, in addition to the past travel route, situation information that represents a surrounding situation when the vehicle traveled the past travel route, The problem includes hypothetical situation information representing a hypothetical situation of an area including the starting point and the destination; The search means further uses the assumption situation information included in the problem to search for a recommended route.

2. The route training device according to claim 1.

3. The route search model is a model capable of outputting an explanation of the reason for recommending a recommended route, The search means outputs an explanation of the reason why the recommended route is recommended, The output means outputs the searched recommended route and an explanation of the reason why the recommended route is recommended.

3. A route training device according to claim 1 or 2.

4. The vehicle further includes a learning means for acquiring an answer driving route given by an expert to a given question as an expert-answered route, and updating the route search model by re-learning the route search model at predetermined timings using the acquired expert-answered route.

4. A route training device according to claim 1.

5. The vehicle is an emergency vehicle that is allowed to travel with priority.

5. A route training device according to claim 1.

6. By computer, A question is presented that includes information on a departure point and a destination and asks the user to think about a vehicle's driving route from the departure point to the destination; receiving an answer route to the question; searching for a recommended route from the starting point to the destination included in the problem, using a route search model that outputs a recommended route to the destination, and the problem, and the route search model is generated by machine learning the actual driving routes that have been driven by an expert selected based on the amount of experience in a vehicle; comparing the searched recommended route with the answer travel route, and evaluating the answer using the comparison result; Outputting information representing an evaluation of the answer as an evaluation result. Route teaching method.

7. A process of setting a question including information on a departure point and a destination, the question being about a vehicle's travel route from the departure point to the destination; receiving an answer route to the question; A process of searching for a recommended route from the starting point to the destination included in the problem, using a route search model that is generated by machine learning of actual driving routes traveled by vehicles by experts selected based on the amount of experience and outputs a recommended route to the destination, and the problem; A process of comparing the searched recommended route with the answer travel route and evaluating the answer using a comparison result; A process of outputting information representing an evaluation of the answer as an evaluation result; A computer program that causes a computer to execute the following:

Images