Information processor and generator

Abstract

To provide an information processor capable of estimating the degree of difficulty of vehicle travel for each road segment.SOLUTION: An information processor includes: information acquisition means for acquiring traffic volume forecast information indicating the traffic volume forecasted for a period of time determined based on a future first period of time of a road segment for which the degree of travel difficulty is to be estimated, and snow depth forecast information which is information for predicting the snow depth in an area including the road segment in the first period of time; and estimation means for estimating the degree of travel difficulty for a vehicle to travel along the road segment during the first period of time based on the traffic volume forecast information and the snow depth forecast information.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information processing device, and more particularly to an information processing device that processes information to provide information that contributes to driving assistance for a vehicle. [Background technology]

[0002] Various information for assisting vehicle driving is provided using information devices such as navigation systems. For example, a driving assistance system has been proposed that provides driving assistance based on information on the degree of congestion set based on probe information related to the location where the tires have slipped and the type of tires (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2009-123153 A Summary of the Invention [Problem to be solved by the invention]

[0004] When driving a vehicle, information on the depth of snow on the road is important information for judging the difficulty of driving the vehicle. Conventionally, to obtain information on snow depth, predicted values ​​of snow depth are calculated based on information such as precipitation, temperature, amount of solar radiation, and snow depth gauges. For example, the Japan Meteorological Business Support Center (JMBSC) calculates predicted values ​​of snow depth every hour for each 5 km square mesh (grid (3 minutes latitude, 3.75 minutes longitude)).

[0005] However, even in areas with the same snow depth, the rate at which snow melts varies between roads with heavy traffic and roads with light traffic, and the degree of difficulty in driving also differs. Therefore, there was a problem in that it was not possible to estimate the degree of difficulty in driving for vehicles based only on the predicted snow depth for each area as in the past.

[0006] The present invention has been made in consideration of the above-mentioned points, and has as its object to provide an information processing device capable of estimating a degree of driving difficulty for a vehicle for each road section. [Means for solving the problem]

[0007] The invention described in claim 1 is an information processing device comprising: an information acquisition means for acquiring traffic volume prediction information indicating the traffic volume predicted for a period determined based on a first future time period for a road section whose driving difficulty level is to be estimated, and snow depth prediction information which is information predicting the snow depth at the first time period in an area including the road section; and an estimation means for estimating the driving difficulty level when a vehicle travels on the road section at the first time period based on the traffic volume prediction information and the snow depth prediction information.

[0008] The invention described in claim 6 is a generation device comprising: an acquisition means for acquiring a dataset in which snow depth information indicating past snow depth information in an area including a sample road section, traffic volume information indicating past traffic volume in the sample road section, and evaluation information regarding driving difficulty when a vehicle travels on the sample road section are associated; and a generation unit for generating an estimation model based on a plurality of the datasets, the estimation model estimating the degree of driving difficulty when a vehicle travels on a road section to be estimated at a first time in the future by inputting information regarding the expected traffic volume of the road section to be estimated in a period determined based on the first time, and the snow depth in the area including the road section to be estimated at the first time. [Brief description of the drawings]

[0009] [Figure 1] 1 is a block diagram showing a configuration of an information processing device according to an embodiment of the present invention. [Diagram 2] FIG. 13 is a diagram showing inputs and outputs when using a driving difficulty level estimation model. [Diagram 3] FIG. 13 is a diagram showing an example of an image display of an estimation result. [Figure 4]FIG. 11 is a diagram showing an example of a driving difficulty level for each road section when a route is set. [Diagram 5] FIG. 13 is a diagram showing an example of a timeline display of driving difficulty levels. [Figure 6] 1 is a block diagram showing a configuration of a model generating device according to an embodiment of the present invention. [Figure 7] FIG. 13 is a diagram showing inputs when learning a driving difficulty level estimation model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will now be described in detail. In the following description of each embodiment and in the accompanying drawings, the same reference numerals are used to designate substantially the same or equivalent parts.

[0011] 1 is a block diagram showing a configuration of an information processing device 100 according to an embodiment of the present invention. The information processing device 100 is a device that estimates a degree of driving difficulty for each road section of a road on which a moving body such as an automobile (hereinafter referred to as a vehicle) travels. In this embodiment, the information processing device 100 is mounted on the vehicle and outputs the estimation result to a display screen of a navigation device or the like.

[0012] The information processing device 100 includes a communication unit 11, a traffic volume prediction information acquisition unit 12, a snow depth prediction information acquisition unit 13, a road characteristic information acquisition unit 14, a road shape information acquisition unit 15, a vehicle information acquisition unit 16, a target period setting unit 17, an estimation unit 18 and an output control unit 19.

[0013] The communication unit 11 is a communication device that transmits and receives data via a wireless communication line, and is configured, for example, by a NIC (Network Interface Card).

[0014] The traffic volume prediction information acquisition unit 12, the snow depth prediction information acquisition unit 13, the road characteristic information acquisition unit 14, the road shape information acquisition unit 15 and the vehicle information acquisition unit 16 are each composed of a processing control unit consisting of, for example, a central processing unit (CPU (Central Processing Unit)).

[0015] The traffic volume prediction information acquisition unit 12 acquires traffic volume prediction information that indicates the future predicted traffic volume of the road section that is the target of the estimation of the traveling difficulty level (hereinafter referred to as the target road section). The traffic volume prediction information is information that indicates the future predicted traffic volume per unit time that is calculated based on statistical data that indicates the change in the past traffic volume for each road section.

[0016] In this embodiment, the "expected traffic volume" refers to the number of vehicles expected to travel in a predetermined period (e.g., one hour) immediately prior to the scheduled travel time of the vehicle. For roads with multiple lanes (e.g., two lanes on each side), the number of vehicles traveling per unit time on one lane is calculated as the expected traffic volume.

[0017] The expected traffic volume used to estimate the degree of driving difficulty is calculated, for example, using actual measured values ​​of past traffic volume in the target road section under similar conditions such as snowfall, day of the week, and time of day. Furthermore, if there is no statistical data for the target road section, the expected traffic volume may be calculated using a value estimated from the actual measured values ​​of surrounding roads for which there is statistical data. For example, if the surrounding roads for which there is statistical data are national roads and the target road section is a city road, the expected traffic volume may be calculated by multiplying the traffic volume of the surrounding road by a predetermined value (e.g., 0.6) and assuming that the actual measured value of the target road section is the value. Furthermore, if the surrounding roads have two lanes on each side and the target road section has one lane on each side, the expected traffic volume may be calculated by multiplying the traffic volume of the surrounding road by 0.5 and assuming that the actual measured value of the target road section is the value.

[0018] In this embodiment, the expected traffic volume is calculated, for example, in an external server device (not shown). The traffic volume prediction information acquisition unit 12 acquires traffic volume prediction information from the server device via the communication unit 11.

[0019] The snow depth prediction information acquisition unit 13 acquires snow depth prediction information, which is information predicting the snow depth per unit time in a predetermined area (e.g., an area of ​​5 km square) that includes the target road section. The snow depth prediction information is information on the snow depth predicted based on the amount of precipitation, temperature, amount of solar radiation, etc. In this embodiment, the snow depth prediction information acquisition unit 13 acquires snow depth prediction information announced by the Japan Meteorological Agency, etc., via the communication unit 11.

[0020] The road characteristic information acquisition unit 14 acquires road characteristic information, which is information indicating road characteristics related to the ease of snow melting or snow accumulation in the target road section. In this embodiment, the road characteristic information is information indicating whether the road section includes road parts with predetermined attributes such as bridges, overpasses, tunnel entrances, and unpaved roads. For example, bridges and overpasses are far from the ground and snow is difficult to melt, so snow is likely to accumulate on them. In addition, tunnel entrances tend to be in the shade, and the temperature of the road surface is low, so snow is likely to accumulate on them. For this reason, the road characteristic information acquisition unit 14 acquires information indicating whether the target road section includes these as road characteristic information related to the amount of snow accumulation. In this embodiment, the road characteristic information acquisition unit 14 acquires road characteristic information based on map information stored in the navigation device or the information processing device 100.

[0021] The road shape information acquisition unit 15 acquires information on the road shape related to the ease of travel during snowfall, such as the presence and degree of curves, gradients, etc., as road shape information. In this embodiment, the road shape information acquisition unit 15 acquires the road shape information based on map information stored in a navigation device or the information processing device 100.

[0022] The vehicle information acquisition unit 16 acquires information on the vehicle's drive system, such as FF (Front engine Front drive), FR (Front engine Rear drive), 4WD (4 Wheel Drive), etc., as vehicle information. The vehicle information acquisition unit 16 also acquires information on the type of tires, such as normal tires, studless tires, etc., as vehicle information. In this embodiment, the vehicle information acquisition unit 16 acquires the vehicle information by receiving this information input in advance from the user.

[0023] The various information acquired by the traffic volume prediction information acquisition unit 12, the snow depth prediction information acquisition unit 13, the road characteristic information acquisition unit 14, the road shape information acquisition unit 15 and the vehicle information acquisition unit 16 is stored in a memory unit (not shown) consisting of an SSD (Solid State Drive), RAM (Random Access Memory) or the like, and is read out when the estimation unit 18 estimates the degree of driving difficulty.

[0024] The target period setting unit 17 sets a period (hereinafter referred to as the target period) for which the estimation unit 18 estimates the degree of driving difficulty. The target period setting unit 17 receives a designation from the user of a future time (first time period) that serves as a reference for setting the target period, and sets the target period based on this. The designation of the first time period is performed based on an input by the user using an input means such as a touch panel.

[0025] The estimation unit 18 estimates the degree of driving difficulty when a vehicle travels through a target road section. The estimation unit 18 estimates the degree of driving difficulty using a driving difficulty estimation model 181, which is a trained model.

[0026] FIG. 2 is a diagram showing input and output of the driving difficulty level estimation model 181 when estimating a driving difficulty level.

[0027] Traffic volume prediction information and snow depth prediction information are input to the driving difficulty level estimation model 181 as main information necessary for estimating the driving difficulty level. Here, the traffic volume prediction information and snow depth prediction information corresponding to the target period set by the target period setting unit 17 are input. These pieces of information are essential for estimating the driving difficulty level in this embodiment, and are always input to the driving difficulty level estimation model 181 when estimating the driving difficulty level.

[0028] In addition, road characteristic information, road shape information, and vehicle information are input to the driving difficulty estimation model 181 as optional information that contributes to estimating the driving difficulty. These pieces of information are input to improve the accuracy of the estimation of the driving difficulty. For example, even with the same amount of snowfall, a curved section has a higher driving difficulty than a straight section. Therefore, by inputting road shape information indicating whether the section is a curve or a straight section as optional information, it is possible to estimate the driving difficulty with high accuracy.

[0029] As a result of calculation using the main information and optional information as inputs, the driving difficulty level estimation model 181 outputs a driving difficulty level for each road section.

[0030] 1 again, the output control unit 19 performs output control for presenting to the user the driving difficulty level estimated by the estimation unit 18. In this embodiment, the output control unit 19 outputs the estimated driving difficulty level for each road section to the navigation device, and causes it to be displayed on the display screen.

[0031] 3 is a diagram showing an example of a display screen that displays the driving difficulty level. A road section shown by a dot pattern in the figure (e.g., road section RL1) indicates that the driving difficulty level is low, i.e., the road section is easy to drive on.

[0032] Road sections marked with diagonal lines (e.g., road sections RL2, RL3, and RL4 in Fig. 3) are medium-level road sections with a medium level of difficulty and require careful driving. They also indicate that it is recommended to take precautions such as putting chains on tires before driving on the road sections.

[0033] Road sections that are filled in black (for example, RL5) indicate that the road section has a high level of difficulty and is dangerous to drive on.

[0034] "TB" in the figure represents a time bar. By advancing the time bar, the driving difficulty level at a desired time in the future (for example, several hours later) can be displayed on the screen.

[0035] By displaying the estimated future driving difficulty level for each road section in this way, the user can know in advance which road sections are difficult to drive on. Therefore, the user can select a route to the destination taking into account the driving difficulty level for each road section, for example by taking a detour to avoid road sections with high driving difficulty levels as necessary.

[0036] FIG. 4 is a diagram showing an example of how the travel difficulty level is displayed when a route is set by a navigation device.

[0037] When a vehicle's driving route is set, the driving difficulty level for each road section is displayed based on the estimated arrival time at a switching point (e.g., an intersection) of the road section on the driving route and the driving difficulty level for each time period estimated for each road section. "S" indicates the start point of the route, and "G" indicates the end point of the route.

[0038] S1 to S5 indicate consecutive road sections within the route. Sections S1 and S3 are road sections with a low level of driving difficulty and are easy to drive on. Sections S2 and S4 are road sections with a medium level of driving difficulty and require careful driving. Section S5 is a road section with a high level of driving difficulty and is dangerous to drive on.

[0039] Fig. 5 is a diagram showing an example of a display in which the driving difficulty level of the route shown in Fig. 4 is displayed as a timeline. The time required for the route is calculated based on the probe information obtained in real time and the information on the driving difficulty level estimated by the estimation unit 18, and is updated sequentially. For example, since the vehicle is scheduled to be traveling in section S1 at 2 p.m., the time periods before and after this time period are displayed as time periods with a low driving difficulty level. In addition, since the vehicle is scheduled to be traveling in section S5 at 4 p.m., the time periods before and after this time period are displayed as time periods with a high driving difficulty level.

[0040] As shown in Fig. 4 or 5, by displaying the driving difficulty level for each road section of a preset driving route or displaying the timeline thereof, the user can grasp the driving difficulty level for each position on the driving route or for each time period in the itinerary. Therefore, when two or more users travel to a destination by vehicle, the user can determine in advance the sections for which each person will be responsible for driving according to the driving difficulty level, such as a driver who is confident in driving ability being in charge of driving road sections or time periods with high driving difficulty.

[0041] In this way, according to the information processing device 100 of this embodiment, the traffic volume forecast information for each road section and the snow depth forecast information for the area including that road section are used as the main information, and the degree of driving difficulty for vehicles for each road section can be estimated and presented visually to the user.

[0042] Next, a processing operation during learning (that is, generation) of the traveling difficulty level estimation model 181 will be described.

[0043] When generating the driving difficulty estimation model 181, a sample road section is set in advance, and a vehicle actually travels on the sample road section to obtain evaluation information on the driving difficulty when traveling on the sample road section. Then, the driving difficulty estimation model 181 is generated based on a data set consisting of various information such as snow depth and traffic volume on the sample road section during travel and the above evaluation information.

[0044] FIG. 6 is a block diagram showing a configuration of a model generating device 200 that generates the traveling difficulty level estimation model 181.

[0045] The model generation device 200 includes a communication unit 21, a past traffic volume information acquisition unit 22, a past snow depth information acquisition unit 23, a road characteristic information acquisition unit 24, a road shape information acquisition unit 25, a vehicle information acquisition unit 26, a memory unit 27, a correct data acquisition unit 28, and a model generation unit 29.

[0046] The communication unit 21 is a communication device that transmits and receives data via a wireless communication line, and is configured, for example, by a NIC.

[0047] The past traffic volume information acquisition unit 22, the past snow depth information acquisition unit 23, the road characteristic information acquisition unit 24, the road shape information acquisition unit 25 and the vehicle information acquisition unit 26 are constituted by a processing control unit consisting of, for example, a central processing unit (CPU).

[0048] The past traffic volume information acquisition unit 22 acquires past traffic volume information indicating the past traffic volume of a preset sample road section. Here, the "past traffic volume" refers to the number of vehicles traveling per lane in a predetermined period of time (e.g., one hour) in the past based on the time when the vehicle traveled on the sample road section, similar to the expected traffic volume at the time of use. The past traffic volume information acquisition unit 22 acquires the past traffic volume information by acquiring, for example, traffic volume information published by the Ministry of Land, Infrastructure, Transport and Tourism via the communication unit 11.

[0049] The past snow depth information acquisition unit 23 acquires information on the past snow depth for each unit time in a predetermined area including the sample road section. The past snow depth information acquisition unit 23 acquires, via the communication unit 21, information on the past snow depth for each area announced by, for example, the Japan Meteorological Agency, as the past snow depth information.

[0050] The road characteristic information acquisition unit 24 acquires road characteristic information, which is information indicating road characteristics (bridges, overpasses, tunnel entrances and exits, unpaved roads, etc.) related to snow melt or snowfall in the sample road section. The road characteristic information acquisition unit 24 acquires the road characteristic information based on map information stored in, for example, a navigation device or the model generating device 200.

[0051] The road shape information acquisition unit 25 acquires road shape information, which is information on road shapes (curves, gradients, etc.) related to the ease of travelling in a sample road section during snowfall. The road shape information acquisition unit 25 acquires the road shape information based on map information stored in, for example, a navigation device or the model generating device 200.

[0052] The vehicle information acquisition unit 26 acquires, as vehicle information, information on the drive system (FF, FR, 4WD, etc.) of the vehicle that actually traveled the sample road section and information on the type of tires (normal tires, studless tires, etc.).

[0053] The storage unit 27 is composed of storage devices such as a hard disk, a flash memory, an SSD (Solid State Drive), etc. Various pieces of information acquired by the past traffic volume information acquisition unit 22, the past snow depth information acquisition unit 23, the road characteristic information acquisition unit 24, the road shape information acquisition unit 25, and the vehicle information acquisition unit 26 are stored in the storage unit 27 and used by the model generation unit 29 to generate a driving difficulty estimation model 181.

[0054] The correct data acquisition unit 28 acquires, as correct data, evaluation information on the driving difficulty level obtained by a vehicle actually traveling through a sample road section. Specifically, a plurality of evaluators who satisfy a predetermined standard for driving history evaluate the driving difficulty level based on driving data (driving log) such as video footage from an on-board camera of a vehicle that has traveled through a sample road section and vehicle speed, thereby generating the evaluation information. For example, the evaluators evaluate the driving difficulty level on a five-level scale (e.g., easy to drive=1, slightly difficult to drive=2, etc.) and generate the evaluation information. The correct data acquisition unit 28 normalizes the evaluation information obtained by the evaluators to fall within the range of 0 to 1, and acquires this as correct data.

[0055] The method of acquiring the correct answer data is not limited to this. For example, when the vehicle travels on the sample road section, the driver who actually drives the vehicle may be asked to act as an evaluator and answer the driving difficulty level as an impression when driving on the sample road section, and this may be acquired as the correct answer data. In addition, the driving tendency in normal driving and the driving tendency in snowy conditions in the driving log of the same driver may be compared, and if there is a large difference between them, the difficulty level may be calculated as high, and if there is a small difference, the difficulty level may be calculated as low, and the driving difficulty level expressed as a class value may be acquired as the correct answer data. In addition, the driving difficulty level may be calculated based on the driving data (driving log) by taking into account the number of slips per unit time of the vehicle that traveled on the sample road section.

[0056] The model generation unit 29 generates a driving difficulty estimation model 181 by performing machine learning based on various information acquired by the past traffic volume information acquisition unit 22, the past snow depth information acquisition unit 23, the road characteristics information acquisition unit 24, the road shape information acquisition unit 25, and the vehicle information acquisition unit 26, and the correct data acquired by the correct data acquisition unit 28.

[0057] FIG. 7 is a diagram showing input data when the traveling difficulty level estimation model 181 is trained.

[0058] Past traffic volume information, past snow depth information, road characteristic information, road shape information, and vehicle information are input to the driving difficulty estimation model 181 as input data during learning. The past traffic volume information and past snow depth information are main information during the learning stage, and the road characteristic information, road shape information, and vehicle information are optional information during learning. By inputting this optional information and performing machine learning, it becomes possible to estimate a driving difficulty level with high accuracy that takes into account the optional information when the driving difficulty estimation model 181 is used (during estimation).

[0059] Moreover, the travel difficulty level estimation model 181 is supplied with the evaluation information obtained by the correct answer data acquisition unit 28 as correct answer data.

[0060] When the driving difficulty estimation model 181 is trained, a plurality of sample road sections with a wide variety of conditions of snow depth and traffic volume are set. In addition, many data sets, each of which is a set of past snow depth information, past traffic volume information, and driving difficulty evaluation (correct answer data), are input. As a result, the driving difficulty estimation model 181 learns the tendency of driving difficulty corresponding to a combination of snow depth and traffic volume, so that when it is used, it is possible to estimate the driving difficulty under such conditions even for roads other than the sample road sections by inputting snow depth prediction information and traffic volume prediction information.

[0061] As described above, the information processing device of this embodiment obtains traffic volume forecast information for a future target period for a road section to be estimated and snow depth forecast information for an area including the road section, and estimates the driving difficulty level of a vehicle based on this information. With this configuration, it becomes possible to estimate the driving difficulty level of a vehicle for each road section.

[0062] The present invention is not limited to the above embodiment. For example, in the above embodiment, in addition to the main information, traffic volume prediction information and snow depth prediction information, road characteristic information, road shape information, and vehicle information are input as optional information to the driving difficulty estimation model 181 to estimate the driving difficulty level. However, it is not necessarily required to input these optional information when estimating the driving difficulty level, and it is also possible to estimate the driving difficulty level based only on the traffic volume prediction information and snow depth prediction information. Also, only a part of the multiple optional information may be used. Also, information that contributes to the estimation of the driving difficulty level other than the above information may be used as optional information. In that case, by including the information in the data set during learning and performing learning, a driving difficulty estimation model that takes the information into account can be generated.

[0063] In the above embodiment, an example has been described in which the estimated result of the driving difficulty level is displayed on the screen of the navigation device, etc. However, the manner in which the estimated result of the driving difficulty level is used is not limited to this, and for example, the estimated result of the driving difficulty level may be used when setting a route in the navigation device. For example, the navigation device can set a route to a destination by preferentially adopting road sections with low driving difficulty levels by searching for a route while setting a high cost for road sections with high driving difficulty levels.

[0064] In the above embodiment, the information processing device 100 has the communication unit 11, the traffic volume prediction information acquisition unit 12, the snow depth prediction information acquisition unit 13, the road characteristic information acquisition unit 14, the road shape information acquisition unit 15, the vehicle information acquisition unit 16, the target period setting unit 17, the estimation unit 18, and the output control unit 19. However, one device does not necessarily have all of these functions, and the functions may be distributed among multiple devices. For example, a server device may acquire traffic volume prediction information, snow depth prediction information, road characteristic information, road shape information, and vehicle information, and the information processing device may be configured to estimate the degree of driving difficulty based on the information acquired by the server device.

[0065] In the above embodiment, the information processing device 100 is mounted on a vehicle and outputs the estimation result to a display screen of a navigation device or the like. In this case, the information processing device 100 may be a device integrated with the navigation device. Alternatively, the information processing device 100 may be a mobile terminal such as a smartphone. On the other hand, the information processing device 100 may be, for example, a server device located outside the vehicle. In this case, the information processing device 100 as a server device may be configured to communicate with a terminal device such as an in-vehicle navigation device, a mobile terminal, and a smartphone via the communication unit 11, receive information specifying a road section to be estimated for the degree of driving difficulty and a future time (first time period) from the terminal device, and transmit the estimation result of the degree of driving difficulty to the terminal device. [Explanation of symbols]

[0066] 100 Information processing device 11 Communications Department 12 Traffic volume forecast information acquisition unit 13 Snow depth forecast information acquisition section 14 Road characteristic information acquisition section 15 Road shape information acquisition unit 16 Vehicle information acquisition unit 17 Target Period Setting Department 18 Estimation part 19 Output control section 181 Driving Difficulty Estimation Model 200 Model Generator 21 Communications Department 22 Past traffic information acquisition unit 23 Past snow depth information acquisition unit 24 Road characteristic information acquisition section 25 Road shape information acquisition unit 26 Vehicle Information Acquisition Department 27 Memory section 28 Correct Data Acquisition Section 29 Model Generation Unit

Claims

1. an information acquisition means for acquiring traffic volume prediction information indicating a traffic volume predicted for a period determined based on a first future time period for a road section for which a degree of driving difficulty is to be estimated, and snow depth prediction information which is information predicting the snow depth for the first future time period in an area including the road section; an estimation means for estimating a degree of driving difficulty when a vehicle travels on the road section during the first period based on the traffic volume prediction information and the snow depth prediction information; An information processing device comprising:

2. The information processing device according to claim 1, characterized in that the estimation means includes a trained model generated by machine learning of a plurality of data sets in which snow depth information indicating past snow depth information in an area including a sample road section, traffic volume information indicating past traffic volume in the sample road section, and evaluation information regarding the difficulty of driving when a vehicle travels on the sample road section are associated with each other.

3. the information acquisition means acquires road characteristic information indicating road characteristics related to the likelihood of snow melting or snow accumulation in the road section; 2. The information processing apparatus according to claim 1, wherein the estimating means estimates the degree of driving difficulty based on the traffic volume prediction information, the snow depth prediction information, and the road characteristic information.

4. the information acquisition means acquires road shape information indicating a shape related to the ease of travelling in the road section during snowfall; 2. The information processing apparatus according to claim 1, wherein the estimating means estimates the degree of driving difficulty based on the traffic volume prediction information, the snow depth prediction information, and the road shape information.

5. the information acquisition means acquires vehicle information indicating a drive system or a type of tires of a vehicle traveling on the road section; 2. The information processing apparatus according to claim 1, wherein the estimating means estimates the degree of driving difficulty based on the traffic volume prediction information, the snow depth prediction information, and the vehicle information.

6. an acquisition means for acquiring a data set in which snow depth information indicating information on past snow depth in an area including a sample road section, traffic volume information indicating past traffic volume on the sample road section, and evaluation information on the driving difficulty when a vehicle drives on the sample road section are associated with each other; a generation unit that generates an estimation model based on a plurality of the data sets, the estimation model estimating a degree of driving difficulty when a vehicle travels on a road section to be estimated at a first time point in the future by inputting information on an expected traffic volume on the road section to be estimated for a period determined based on the first time point and information on snow depth in an area including the road section to be estimated at the first time point; A generating device comprising:

7. The generating device according to claim 6, characterized in that the evaluation information is information obtained by judging driving difficulty based on images from an on-board camera or probe information when the vehicle travels along the sample road section.

8. 7. The generating device of claim 6, wherein the dataset further includes at least one of the following information: road characteristics related to snow melt or snow accumulation of the sample road section; a shape related to the ease of driving on the sample road section during snowfall; a drive system of a vehicle that traveled on the sample road section; and a type of tires of the vehicle.

Images