KR20200068055A - System and Method for Predicting Road Surface State - Google Patents

Abstract

The present invention relates to a method for predicting a road surface state to correctly predict the road surface state in accordance with a weather situation and a system thereof. According to the present invention, the method comprises the following steps of: acquiring weather information about a prediction target point from which the road surface state is measured; and inputting weather data generated by processing the acquired weather information and a drying speed ratio previously calculated for the prediction target point to a road surface state prediction model to predict the road surface state of the prediction target point. The road surface state prediction model can be trained by using, as training data, the road surface state at an observation time of an observation point, a drying speed ratio of the observation point, an accumulated precipitation accumulated from a precipitation start time to a precipitation finish time of the observation point, a temporal distance from the precipitation finish time to the observation time, and an average temperature and a standard temperature deviation calculated for a plurality of time sections which are formed by dividing time from the precipitation start time to the precipitation end time in accordance with a preset method.

Description

System and method for predicting road surface state}

The present invention relates to a road surface condition prediction method and system, and more particularly to a road surface condition prediction method and system according to weather conditions.

Weather changes on the road affect the road surface condition and, consequently, the frictional force of the road. Decreasing the frictional force on the road increases the braking distance of the vehicle. The change in the braking distance of the vehicle has a great influence on the safety of the vehicle.

Korean Registered Patent No. 10-1331054 collects real-time information on road surface conditions such as wetting, drying, water film, snow, ice, and fog from a road surface sensor installed on the roadside, and calculates the safety speed based on this to calculate the safety speed. A configuration for providing crab information is disclosed.

However, Korean Registered Patent No. 10-1331054 requires a lot of cost and effort to operate because it is necessary to continuously collect the road surface condition by the survey vehicle in real time. In addition, when the weather condition suddenly changes after the investigation vehicle has passed, for example, when the road surface condition suddenly changes due to rain or snow after the vehicle passes the corresponding point on the road, there is a disadvantage that it cannot be reflected in real time.

Korean Registered Patent No. 10-1331054 (Registration Date: November 19, 2013)

Therefore, a technical problem to be solved by the present invention is to provide a road surface condition prediction method and system capable of accurately estimating a road surface condition according to weather conditions at a relatively low cost.

The road surface condition prediction method according to the present invention for solving the above technical problem comprises obtaining weather information on a road surface condition prediction target point, and weather data processing the obtained weather information and the prediction target point. And predicting the road surface condition of the prediction target point by inputting a pre-calculated dry speed ratio to the road surface condition prediction model.

The road surface condition prediction model includes a road surface condition at an observation point of an observation point, a drying rate ratio of the observation point, and cumulative precipitation that measures accumulated precipitation from the point of occurrence of precipitation to the point of precipitation at the point of observation, and the end of precipitation. The average distance and the standard temperature deviation calculated for a plurality of time periods divided by a predetermined method for a time from a time point to a time point between the observation time point and a time point from precipitation occurrence to the observation time point may be learned as learning data.

The drying speed ratio can be calculated using road geometry data.

The drying speed ratio at a predetermined point (P _sensor ) of the road may be defined by Equation 1 below.

[Equation 1]

Drying speed ratio = SA _{Vertical section} /TA _{Vertical section} ,

The predetermined point is located between the road sections connecting two neighboring _{local maximum points} (P _{local maximum peak 1} , P _{local maximum peak 2} ), and the TA _{Vertical section} is the local maximum point (P _{local minimum peak} ) located in the road section. wherein a two local maximum point area of a triangle formed of a (P _{local maximum peak1,} P _{local maximum peak2),} SA _{Vertical section} is the local maximum point (P _{local minimum peak),} wherein the predetermined point of (P _sensor), and the other side of the point (P _It is the width of the triangle made of _opposite ), the opposite point is located in the road section, and the line segment connecting the predetermined point (P _sensor ) and the opposite point may be parallel to the line segment connecting the two local maximum points. .

The weather data processed from the obtained weather information includes cumulative precipitation, which measures the accumulated precipitation from the point of occurrence of precipitation to the end of precipitation, and the time distance between the end of precipitation and the time of observation. It may include the average temperature and the standard deviation of the temperature, respectively calculated for a plurality of time periods divided by a predetermined method with respect to time until the observation point.

The plurality of time periods may include a first time period from a precipitation occurrence time to a prediction time period, a second time period from a precipitation occurrence time to a termination time, and a third time period from the precipitation end time to the observation time. have.

It may include a computer-readable recording medium containing a program for performing the road surface condition prediction method according to the present invention for solving the above technical problem.

The program includes a set of instructions for obtaining weather information for a point to be predicted for a road surface condition, and a road surface condition prediction model for a weather data processed from the acquired weather information and a pre-computed drying speed ratio for the predicted point. It may include a set of instructions for predicting the road surface condition of the predicted point by inputting to the.

The road surface prediction system according to the present invention for solving the above technical problem includes: a weather information acquisition unit for obtaining weather information for a road surface condition prediction target point, and weather data and the prediction processing the input weather information. A road surface condition prediction unit configured to predict a road surface condition of the predicted target point by inputting a pre-calculated dry speed ratio for the target point into a road surface condition prediction model.

According to the present invention, it is possible to provide a method and system for predicting a road surface condition that can accurately estimate a road surface condition according to a weather condition at a relatively low cost.

1 is a block diagram of a road surface condition prediction system according to an embodiment of the present invention.
2 is a view provided to explain a speed ratio of drying of a road according to an embodiment of the present invention.
3 is a flowchart illustrating an operation of a road surface condition prediction system according to an embodiment of the present invention.

Then, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art to which the present invention pertains may easily practice.

1 is a block diagram of a road surface condition prediction system according to an embodiment of the present invention.

Referring to FIG. 1, the road surface condition prediction system according to the weather situation according to the present invention includes a road information collection unit 110, a weather information acquisition unit 120, a database 130, a learning unit 140, and a road surface state Prediction unit 150 may be included.

The road information collection unit 110 performs a function of collecting road surface state information and constructing learning data. The road information collecting unit 110 may include a road surface sensor 111 and a GNSS receiving unit 115. As the road information collecting vehicle (not shown) equipped with the road information collecting unit 110 drives the road, from the GNSS signal obtained from the GNSS receiving unit 115 to the road surface state information obtained through the road surface sensor 111 It can be collected by matching the obtained location information (latitude, longitude) and data collection time. According to an embodiment, some of the road information collection units 110 may be installed on a road surface to include a sensor that collects road surface conditions.

The road surface sensor 111 may acquire the road surface type of the road in a non-contact manner. The road surface type can be divided into dry, moist, wet, etc., and more specifically, Slush, Ice or Snow with water, Ice and snow, or It is also possible to distinguish by adding more frost (Snow or Hoar Froast). The criteria for classifying the road surface type may vary depending on the embodiment.

According to an embodiment, the road surface sensor 111 may acquire a road surface temperature, a road surface film thickness, and an outside temperature as well as a road surface type in a non-contact manner. A method of obtaining the surface temperature of the road surface, the surface thickness of the road surface, the outside air temperature, and the type of the road surface by using the road surface sensor 111 in a non-contact manner is well known, so detailed descriptions thereof are omitted here.

The road information collection unit 110 may store the collected road surface state information, location information, and time information in a memory (not shown), or transmit it to a destination determined in real time or at predetermined intervals through a communication network (not shown). have.

Meanwhile, the road information collection unit 110 may collect road geometry data. For example, an inertial sensor or the like may be mounted on a road information collection vehicle equipped with the road information collection unit 110 to collect road geometry data while driving on the road.

The geometry data of the road includes basic information such as the longitudinal slope and coordinates and altitude of the road collected through the road information collection unit 110, and whether a curve section is generated through additional processing, curve radius, curve part length, pavement type And the like. Depending on the embodiment, the road geometry data may be pre-built information, or may be separately collected and used through the above-described road information collection vehicle or a separate mobile observation device.

The meteorological information acquisition unit 120 may receive meteorological information from a meteorological server (not shown) operated by a meteorological agency or a private meteorological operator. The weather information may include information about temperature, precipitation, rainfall, snowfall, wind speed, humidity, insolation, and sunshine. Precipitation is the amount including rainfall and snowfall. For example, data measured by the Meteorological Agency's Automated Surface Observing System (ASOS) observation equipment may be provided.

The database 130 may store road surface state information, road geometry data, and weather information. In particular, the database 130 maps road surface condition information collected from the observation points at the road surface condition collection unit 110, weather data processed weather information corresponding to the observation points, and maps the calculated drying speed ratio for the observation points. It is possible to store learning data constructed in the form. The weather data processing weather information corresponding to the observation point and the calculated drying rate ratio for the observation point are described in detail below.

The learning unit 140 may learn a road surface condition prediction model that predicts a road surface condition according to a weather condition using the training data built in the database 140. The learning unit 140 uses a machine learning model such as logistic regression, decision tree SVM (Support Vector Machine), k-NN (k-Nearest Neighbor), and deep learning. By doing so, the road surface condition prediction model can be trained.

The road surface condition prediction unit 150 uses the weather data obtained by processing the weather information obtained for the prediction target point and the pre-computed drying speed ratio for the prediction target point to the road surface condition prediction model learned by the learning unit 140. By inputting it, the road surface condition of the predicted point can be predicted.

The road surface condition prediction model according to the present invention includes the road surface condition at the observation point of the observation point, the drying rate ratio of the observation point, the cumulative precipitation measuring the accumulated precipitation from the point of occurrence of precipitation to the point of observation, and the end of precipitation. It can be learned by using the calculated average temperature and temperature standard deviation for a plurality of time periods divided by a predetermined method with respect to the time, the distance between the start point and the observation point, and the time from the occurrence of precipitation to the point of observation.

In general, there are various surrounding environments that affect road surface conditions such as buildings around the road, roadside trees, pavement types of roads, and road slopes. It is known that these road surroundings do not directly affect the road surface change, but rather the speed of the road surface change. Of course, in order to increase accuracy, it is good to consider all the surrounding environments that affect the road surface condition, such as the buildings around the road, the number of roads, the type of pavement, and the slope of the road. However, it was found that considering the longitudinal slope of the road is sufficient to explain the degree of moisture remaining on the road to predict the road surface change due to precipitation at a relatively low cost.

The longitudinal slope is a variable indicating whether the current road is downhill or uphill. When the longitudinal slope is negative, the road takes the form of a downhill slope, and when it is positive, it takes the form of an uphill slope. The downhill and uphill paths of the road affect the water thickness at the observation point. If it is the smallest part that changes from downhill to uphill, the water may accumulate and the thickness of the water film may be thick. Conversely, the part of the maximum point that changes from the uphill to the downhill may have a relatively thin film thickness. The water film thickness affects the drying rate ratio on the road surface.

2 is a view provided to explain a speed ratio of drying of a road according to an embodiment of the present invention.

Referring to FIG. 2, a drying speed ratio at a specific point (P _sensor ) of a road may be defined and calculated by Equation 1 below.

[Equation 1]

Drying speed ratio = SA _{Vertical section} /TA _{Vertical section} ,

Assuming that the position between the drying speed points to calculate the ratio (P _sensor) the adjacent two local maximum point (P _{local maximum peak1,} P _{local maximum peak2)} connecting road segment that, TA _{Vertical section} is located on the road segment local maximum point can be defined as the area of a triangle formed by _{(local minimum peak} P) and the two local maximum point _{(P local maximum peak1, P local} maximum peak2). And SA _{Vertical section} can be defined as the area of the triangle consisting of a local maximum (P _{local minimum peak} ), point (P _sensor ) and the opposite point (P _opposite ).

Here, the opposite point (P _opposite ) is a point located in the road section. Point (P _sensor) and the line segment connecting the other end point (P _opposite) can be determined in parallel with the line segment connecting the two local maximum point _{(P local maximum peak1, P local} maximum peak2).

Each point _{(P sensor, P opposite, P} local maximum peak1, P local maximum peak2, P local minimum peak) may be an intermediate point in the transverse direction of the road.

In Equation 1, by using the ratio of the width, not the length, it is possible to impart explanatory power to the water capacity of the corresponding slope section. This is because the height difference between the points can be used instead of the width of the longitudinal section, but if the longitudinal slope is different even at the same height, the thickness of the water film may also be different.

On the other hand, if you want to give more accurate explanatory power to the drying speed ratio of the section, you may need to consider the cross slope. This is because when precipitation occurs, most of the water is discharged to the shoulder by the cross slope in addition to being affected by the longitudinal slope. However, after most of the water is discharged to the shoulder by the cross slope, the moisture remaining on the road surface may vary depending on the width of the longitudinal section at the corresponding point. Therefore, the drying speed ratio may have sufficient explanatory power without considering the transverse slope. Of course, as described above, a more accurate model can be generated when the cross slope or other road surrounding environment is additionally considered, but in consideration of cost, only the drying speed ratio according to the longitudinal slope of the road is considered.

Weather information is the variable that has the most direct effect on the change of road surface conditions. Snow and rain cause changes in the road surface of the road, and temperature has a direct effect on changes in the road surface condition. However, accurate prediction may be difficult if weather information is not processed and used as an explanatory variable to predict road conditions. The most important part of the processing of meteorological variables is time. The road surface is also affected by the current precipitation, but it is also affected by the rain that fell an hour ago. In particular, snow is affected by the current road surface condition even if it falls 2 to 3 days before the point of view. Therefore, time must be considered when processing weather information into explanatory variables.

In this embodiment, the accumulated precipitation is measured from the point of occurrence of precipitation to the end of precipitation, the temporal distance between the point of observation from the end of precipitation, and the time from the point of occurrence of precipitation to the point of observation. The average temperature and the standard deviation of temperature, respectively calculated for the divided time periods, were considered as explanatory variables. Here, the observation point corresponds to the prediction target point in the process of predicting the road surface state after training the model, and the observation point corresponds to the prediction point.

Accumulated precipitation (acc_precipitation) is a measured value of accumulated precipitation from the time of precipitation to the end of precipitation. The cumulative precipitation value is an explanatory variable that directly or indirectly affects the current road surface condition. Precipitation refers to the amount of precipitation caused by rain and snow, so ASOS can use precipitation per minute and snowfall per hour to process cumulative precipitation. ASOS precipitation per minute provides cumulative precipitation information from 00:00 to the present. However, if the time of rainfall is before 00: 00, it is difficult to see the total amount of precipitation that occurred just before, so additional work is required. Snowfall can be used to supplement information about snow that was not accurately measured at precipitation per minute. Generally, snowfall and precipitation are expressed in a ratio of 10:1. That is, if 1 cm of snowfall is observed, it is recorded as 0.1 mm as precipitation. However, when viewed from the viewpoint of changes in the state of the road surface, the rate of change of the road state is relatively slower than rain because the snow evaporates while melting. In addition, considering that snow occurs in winter, the time required for the road surface to become dry increases because the snow freezes and repeats recording due to cold weather. For this reason, the expression of precipitation in snowfall can be replaced by simply converting units.

The temporal distance between the end of precipitation and the time of observation (time_dist_min_pet_st) is a variable that describes the effect of previously occurring precipitation on the current road surface condition. Precipitation may occur at the time of observation, and may end an hour or a day ago. If you predict the condition of the road during the rain in the summer, the road surface is wet. However, no matter how strong the rain is, if it rained three hours ago, the road may be dry if road management is good. The road surface is always wet or frozen during precipitation. Therefore, since the time when the road surface actually changes is the end of precipitation, the temporal distance between the two points can be defined as the time between the end of precipitation and the observation time, not the start of precipitation.

The temperature is defined as three sections from the time of precipitation to the point of observation, and the average and standard deviation can be calculated and used to express temperature information in each section. Average temperature has a direct effect on the rate at which moisture on the road surface evaporates. If the average temperature is high, the speed at which the road surface dries increases, and when it is low, it decreases. In addition, if the average temperature of the road surface is low, the road surface may freeze. However, the road surface is not always frozen when the average temperature is low. This is because the road surface in winter can be affected by the precipitation occurring 2 to 3 days ago, so the rain that rained 3 days ago may freeze due to the strong cold that occurred 2 days ago. In winter, the condition of the road surface is affected by vision. In the morning and evening, the temperature is below zero, but there are examples of maintaining the room temperature in the morning and during the day. The standard deviation of temperature can be used as a variable to describe the temperature that changes over time. Standard deviations are effective in predicting road surface conditions at similar average temperatures.

As for the temperature, the average temperature and the standard deviation of the temperature calculated for a plurality of time periods divided by a predetermined method may be used as explanatory variables. For example, as shown in Table 1, a plurality of time intervals is the first time interval from the time of precipitation to the observation time, the second time interval from the time of precipitation occurrence to the end of precipitation, and the third time interval from the time of precipitation termination to the time of observation. It can be divided by and average temperature and standard deviation for each section can be obtained.

Table 1 shows weather data processed as explanatory variables in relation to weather information in the present invention.

Explanatory variable Justice acc_precipitation Accumulated precipitation from the time of precipitation to the end of precipitation time_dist_min_pet_st The temporal distance between the end of precipitation and the point of observation avg_temperature_pet_cst Average temperature from the end of precipitation to the point of observation sd_temperature_pet_cst Standard deviation of temperature from the end of precipitation to the point of observation avg_temperature_pst_pet Average temperature from the time of precipitation to the end of precipitation sd_temperature_pst_pet Standard deviation of temperature from the time of precipitation to the end of precipitation avg_temperature_pst_st Average temperature from the time of precipitation to the point of observation sd_temperature_pst_st Standard deviation of temperature from the point of occurrence of precipitation to the point of observation

In this embodiment, as described above, only the drying speed ratio corresponding to the longitudinal slope of the road is used for the environment around the road. Other factors, such as surrounding buildings, the type of pavement on the road, and the cross slope of the road, can also affect the road surface condition. Therefore, it is preferable to learn and use the road surface condition prediction model according to the present invention for each road section where the same road pavement state, road environment, road management, etc. can be expected. In addition, even on the same road, the road surface condition prediction model may be learned and used by season.

3 is a flowchart illustrating an operation of a road surface condition prediction system according to an embodiment of the present invention.

Referring to FIGS. 1 to 3, first, road surface state information may be collected to build weather data processed by weather information and learning data mapped to a drying rate ratio of a road surface in the database 130 (S310 ).

Next, the learning unit 140 may learn a road surface condition prediction model for predicting a road surface condition according to a weather condition using the learning data built in the database 130 (S320). The learning unit 140 uses a machine learning model such as logistic regression, decision tree SVM (Support Vector Machine), k-NN (k-Nearest Neighbor), and deep learning. By doing so, the road surface condition prediction model can be trained.

Steps S310 to S320 may be continuously performed by updating the road surface condition prediction model by reflecting the learning data when additional learning data is generated.

Afterwards, the road surface condition prediction unit 150 predicts a road surface condition prediction model learned by the learning unit 140 for the weather data obtained by processing the weather information obtained for the prediction target point and the pre-computed drying speed ratio for the prediction target point. The road surface condition of the predicted target point may be predicted by inputting in (S330).

The road surface condition predicted in step S330 may be applied to a system that applies to a road risk prediction algorithm by applying a road friction coefficient estimated by reflecting a road pavement material, a vehicle average traffic speed, and the like to the road surface type. Of course, in addition to this, information on the road surface condition may be provided to a vehicle terminal, a user terminal, or may be displayed through a billboard installed on the road. In addition, it is possible to provide a road surface condition for an autonomous vehicle, etc. to assist in safe driving.

The embodiments described above may be implemented by hardware components, software components, and/or combinations of hardware components and software components. For example, the devices, methods, and components described in the embodiments include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors (micro signal processors), microcomputers, and field programmable gates (FPGAs). It can be implemented using one or more general purpose computers or special purpose computers, such as arrays, programmable logic units (PLUs), microprocessors, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of understanding, a processing device may be described as one being used, but a person having ordinary skill in the art, the processing device may include a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that may include. For example, the processing device may include a plurality of processors or a processor and a controller. In addition, other processing configurations, such as parallel processors, are possible.

The software may include a computer program, code, instruction, or a combination of one or more of these, and configure the processing device to operate as desired, or process independently or collectively You can command the device. Software and/or data may be interpreted by a processing device, or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodied in the transmitted signal wave. The software may be distributed over networked computer systems, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, or the like alone or in combination. The program instructions recorded in the medium may be specially designed and configured for the embodiments or may be known and usable by those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs, DVDs, and magnetic media such as floptical disks. Includes hardware devices specifically configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include high-level language code that can be executed by a computer using an interpreter, etc., as well as machine language codes produced by a compiler. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described by the limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, the described techniques are performed in a different order than the described method, and/or the components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or other components Alternatively, even if replaced or substituted by equivalents, appropriate results can be achieved.

110: road information collection unit
111: road sensor 115: GNSS receiver
120: weather information acquisition unit
130: database
140: learning department
150: road surface condition prediction unit

Claims (11)

Obtaining weather information for a point to predict the road surface condition, and
Predicting the road surface condition of the predicted target point by inputting the weather data processed from the obtained weather information and a pre-computed drying speed ratio for the predicted target point into a road surface condition prediction model.
Including,
The road surface condition prediction model,
The road surface condition at the observation point of the observation point, the drying rate ratio of the observation point, the cumulative precipitation measuring the accumulated precipitation from the point of occurrence of precipitation to the end of precipitation, and the time between the end of precipitation and the point of observation A method for predicting road surface conditions obtained by learning average temperature and temperature standard deviation calculated for a plurality of time periods divided by a predetermined method for time from a distance and a precipitation occurrence time to the observation time. In claim 1,
The road speed condition prediction method, wherein the drying speed ratio is calculated using road geometry data. In claim 2,
The drying speed ratio at a predetermined point (P _sensor ) of the road is defined by Equation 1 below,
[Equation 1]
Drying speed ratio = SA _{Vertical section} /TA _{Vertical section} ,
The predetermined point is located between road sections connecting two neighboring _{local maximum points} (P _{local maximum peak 1} , P _{local maximum} peak 2 ),
_{Vertical section} TA is the area of a triangle formed of a local maximum point _(peak P _{local minimum)} and the two local maximum point (P _{local maximum peak1,} P _{local maximum peak2)} located on the road segment,
SA _{Vertical section} is the area of the triangle consisting of the local maximum (P _{local minimum peak} ), the predetermined point (P _sensor ) and the opposite point (P _opposite ),
The opposite point is located in the road section, the predetermined point (P _sensor ) and the line segment connecting the opposite point is a road surface condition prediction method parallel to the line segment connecting the two local maximum points. In claim 3,
The weather data processing the obtained weather information,
The cumulative precipitation, which measures the accumulated precipitation from the point of occurrence of precipitation to the end of precipitation, the time distance between the end of precipitation and the point of observation, divided by a predetermined method for the time from the point of occurrence of precipitation to the point of observation A method for predicting road surface conditions, including the average temperature and standard deviation of temperature, respectively, calculated for a plurality of time periods. In claim 4,
The plurality of time intervals,
A road surface condition prediction method including a first time period from a precipitation occurrence time to a prediction time, a second time period from a precipitation occurrence time to a termination time, and a third time period from the precipitation end time to the observation time. A computer-readable recording medium containing a program for performing a road surface condition prediction method,
The above program,
A set of instructions for obtaining weather information for a point to predict road surface conditions, and
A set of instructions for predicting the road surface condition of the predicted target point by inputting the weather data processed from the obtained weather information and a pre-computed dry speed ratio for the predicted target point into a road surface condition prediction model
Including,
The road surface condition prediction model,
The road surface condition at the observation point of the observation point, the drying rate ratio of the observation point, the cumulative precipitation measuring the accumulated precipitation from the point of occurrence of precipitation to the end of precipitation, and the time between the end of precipitation and the point of observation A computer-readable recording medium trained with learning data of average temperature and temperature standard deviation calculated for a plurality of time periods divided by a predetermined method with respect to time from the time of occurrence of distance and precipitation to the time of observation. A weather information acquisition unit that acquires weather information for a point to be predicted for road surface conditions, and
A road surface condition prediction unit for predicting a road surface condition of the predicted target point by inputting the weather data processed from the input weather information and a pre-calculated dry speed ratio for the predicted target point into a road surface condition prediction model
Including,
The road surface condition prediction model,
The road surface condition at the observation point of the observation point, the drying rate ratio of the observation point, the cumulative precipitation measuring the accumulated precipitation from the point of occurrence of precipitation to the end of precipitation, and the time between the end of precipitation and the point of observation A road surface condition prediction system learned with learning data of average temperature and temperature standard deviation calculated for a plurality of time periods divided by a predetermined method with respect to time from the time of occurrence of distance and precipitation to the observation time. In claim 7,
The road speed condition prediction system, wherein the drying speed ratio is calculated using road geometry data. In claim 8,
The drying speed ratio at a predetermined point (P _sensor ) of the road is defined by Equation 1 below,
[Equation 1]
Drying speed ratio = SA _{Vertical section} /TA _{Vertical section} ,
The predetermined point is located between road sections connecting two neighboring _{local maximum points} (P _{local maximum peak 1} , P _{local maximum} peak 2 ),
TA _{Vertical section} is the area of the triangle _{consisting of the local maximum} point (P _{local minimum peak} ) and the two _{local maximum points} (P _{local maximum peak 1} , P _{local maximum peak 2} ) located in the road section,
SA _{Vertical section} is the area of the triangle consisting of the local maximum (P _{local minimum peak} ), the predetermined point (P _sensor ) and the opposite point (P _opposite ),
The opposite point is located in the road section, the predetermined point (P _sensor ) and the line segment connecting the opposite point is a road surface condition prediction method parallel to the line segment connecting the two local maximum points. In claim 9,
The weather data processing the obtained weather information,
Accumulated precipitation, which measures the accumulated precipitation from the point of occurrence of precipitation to the end of precipitation, the temporal distance between the end of precipitation and the point of observation, divided by a predetermined method for the time from the point of occurrence of precipitation to the point of observation A road surface condition prediction system comprising an average temperature and a standard temperature deviation, respectively, calculated for a plurality of time periods. In claim 10,
The plurality of time intervals,
A road surface condition prediction system including a first time period from the time of precipitation occurrence to the observation time, a second time period from the time of precipitation occurrence to the end of precipitation, and a third time period from the end of precipitation to the time of observation.

Images