KR20190057501A - System and Method for Predicting Riskness of Road According to Weather Conditions - Google Patents

Abstract

The present invention relates to a system for predicting a road risk in accordance with weather conditions, and to a method thereof. According to the present invention, the method for predicting a road risk in accordance with weather conditions comprises the steps of: using a road surface prediction model and predicting a road surface state in accordance with weather conditions of a prediction target point; receiving a pre-stored road geometry structure with respect to the prediction target point and the predicted road surface state to use a road risk prediction algorithm and predicting the road risk of the prediction target point; constructing learning data which mapped road surface state information obtained by using a road surface sensor mounted on a moving object with weather data; and using the learning data and learning a road surface state prediction model which predicts the road surface state in accordance with weather conditions.

Description

TECHNICAL FIELD The present invention relates to a system and a method for predicting a road risk according to weather conditions,

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

Conventional techniques include a method of inducing safe driving in consideration of a forward situation in a vehicle, a method of maintaining a relative speed and a safe speed by using a yaw rate sensor in a curved road, a road visibility warning System, a system for calculating the safety distance of the vehicle from two hours between the speed of the vehicle and the vehicle ahead and back, and the like. However, such a conventional technique has a limitation in that firstly, the road surface condition, road surface friction coefficient, and road traffic information are not considered. Second, there is a problem of calculating the safety distance without taking into consideration various variables such as the driver's reaction time, the time to the conflict, and the like. Third, it is difficult for the driver to intuitively recognize and actually utilize the safety distance by calculating the safety distance instead of the safety speed.

In order to solve such a problem, Korean Patent Registration No. 10-1331054 discloses a safety speed calculation method and apparatus for considering the road surface information and road traffic information. According to the method and apparatus disclosed in Korean Patent No. 10-1331054, it is possible to improve the traffic safety by recommending the safety speed according to the speed information, the position information, and the road surface information collected from the survey vehicle capable of wireless communication with the GNSS receiver And can also be used to induce drivers to lower the speed by positively responding to traffic clogging, accident occurrence, sudden weather change, etc., thereby contributing to the prevention of additional secondary accidents and mitigation of delays There are advantages to be able to.

However, in Korean Patent No. 10-1331054, since the road surface condition must be continuously collected by a survey vehicle in real time, there is a great deal of cost and effort in operation. In addition, when the weather condition is suddenly changed after passing the survey vehicle, for example, when the vehicle suddenly changes its road surface state due to rain or snow after passing through a corresponding point on the road, it can not be reflected in real time.

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

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a system and a method for predicting a road condition by estimating a road condition according to a weather condition.

According to another aspect of the present invention, there is provided a method for predicting a road surface risk according to a weather condition, the method comprising the steps of: predicting a road surface state according to a weather condition of a forecasted point using a road surface state prediction model, And predicting the road risk of the predicted location using the road risk prediction algorithm in response to the predicted road surface state.

The method comprises the steps of: constructing learning data in which road surface state information obtained by using a road surface sensor mounted on a mobile body is mapped with weather data; and predicting a road surface state according to a weather state using the learning data, And a step of learning the road surface state prediction model.

According to an aspect of the present invention, there is provided a road surface condition prediction system according to a weather condition, including a road surface state prediction unit for predicting a road surface state according to a weather condition of a prediction target point using a road surface state prediction model, And a road risk prediction unit for predicting the road risk of the prediction target point using the road risk prediction algorithm based on the road geometry pre-stored for the prediction target point and the predicted road surface condition.

The system includes a database for storing learning data obtained by mapping road surface state information obtained by using a road surface sensor mounted on a moving object with weather data, and a database for predicting a road surface state according to a weather condition using the learning data. And a learning unit for learning the road surface state prediction model.

The road surface state information may include at least one of a road surface temperature, a road surface water film thickness, a road surface friction coefficient, an outside air temperature, and a road surface state type.

The road surface state type may be information that distinguishes the road surface state according to a predetermined standard.

The road risk prediction algorithm can predict the road risk by reflecting road geometry, road surface condition, and road traffic information.

The road geometry may include at least one of a road end slope, a road cross slope, and a road curvature.

The road traffic information may include at least one of traffic volume and traffic speed.

According to the present invention, learning data obtained by mapping road surface state and weather data to construct a road surface state prediction model is advantageous in that the road risk can be predicted only by meteorological data and road traffic information provided in real time.

1 is a block diagram showing a configuration of a road-risk prediction system according to a weather condition according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating an operation of a road-risk prediction system according to a weather condition according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention.

1 is a block diagram showing a configuration of a road-risk prediction system according to a weather condition according to an embodiment of the present invention.

1, the road risk prediction system according to the present invention includes a road surface state collecting unit 110, a weather information obtaining unit 120, a road traffic information obtaining unit 130, a database 140, A road surface condition predicting unit 160, and a road risk predicting unit 170. The learning unit 150 may include a road surface condition predicting unit 160,

The road surface state collecting unit 110 collects road surface state information and constructs learning data. The road surface condition collecting unit 110 may include a road surface sensor 111 and a GNSS receiving unit 115. A road information collecting vehicle (not shown) equipped with a road surface condition collecting unit 110 travels on the road and calculates a GNSS signal (not shown) obtained from the GNSS receiving unit 115 on the road surface state information obtained through the road surface sensor 111 (Latitude and longitude) obtained from the data acquisition time and the data collection time.

The road surface sensor 111 can obtain the road surface surface temperature, the thickness of the road surface water film, the outside air temperature, the road surface state type, and the like in a noncontact manner. The method of obtaining the road surface surface temperature, the road surface water film thickness, the outdoor air temperature, the road surface state type, and the like in a noncontact manner by using the road surface sensor 111 is well known and will not be described in detail here. Meanwhile, according to the embodiment, the road surface sensor 111 may include a function of measuring the road surface friction coefficient.

In this case, the road surface type may be dry, wet, wet, ice or snow with water, ice or snow or hoar frost according to predetermined criteria, And the like. Of course, the criteria for distinguishing the road surface type may vary depending on the embodiment. For example, it is possible to simply distinguish between the above-mentioned criteria such as drying, wetting, wetting, ice, snow, and the like, and it is also possible to classify the road surface state type into more steps.

The coefficient of road friction depends on the road surface type, road pavement material, road pavement type, road pollution situation, vehicle tire shape, tire rubber condition and load. Therefore, when the road surface friction coefficient is measured by the road surface sensor 111, the road surface friction coefficient is measured according to a predetermined standard.

The road surface state collecting unit 110 stores the collected road surface state information, location information, and time information in a memory (not shown) or transmits the collected road surface state information, location information, and time information to a predetermined destination on a real- It is possible.

The road geometry information may include a road end slope, a road cross slope, and a road curvature. The road geometry information can be obtained by using information previously constructed or by collecting information related to the road geometry as shown in Table 1 below while using the inertia sensor or the like mounted on the road information collection vehicle described above.

Road geometry

time
Latitude
Hardness
Road name
Direction crossing
inclination End
inclination
Curved
Radius of curvature

The weather information obtaining unit 120 can receive weather data from a weather server (not shown) operated by a weather station or a private weather service provider. The weather data may include information on temperature, rainfall, snowfall, wind speed, humidity, solar radiation, sunshine, and the like.

The database 140 may store road surface state information, road geometry, weather data, and the like. In particular, the database 140 may store learning data constructed by mapping the road surface state information collected by the road surface state collecting unit 110 to weather data. According to the embodiment, the road surface state information, the weather data, and the road geometry information are also mapped together to generate the learning data. In this case, the learning data can be expanded by using the learning data of other points having the same or similar road geometry even though the road surface state information for the specific point of the road and the learning data to which the weather data are mapped are not provided.

The learning data may be updated in the database 140 together with the corresponding weather data whenever the road surface state information is collected in the road surface state collecting unit 110. [

The learning unit 150 can learn a road surface state prediction model that predicts a road surface state according to a weather condition using the learning data constructed in the database 140. [ The learning unit 150 can learn the road surface state prediction model that predicts the road surface state corresponding to the weather data corresponding to a specific point on the road through artificial intelligence learning such as deep running or machine learning. Of course, when the learning data is generated while operating the road-risk prediction system according to the present invention, the road-state prediction model may be updated to reflect the learning data.

The road surface state predicting unit 160 can predict the road surface state according to the weather state of the prediction target point using the road surface state prediction model. Specifically, the road surface state predicting unit 160 may predict the road surface state type and / or the road friction coefficient of the predicted object point by inputting the weather data provided through the weather information obtaining unit 120.

The road surface state predicted by the road surface state predicting unit 160 may vary depending on the information that can be collected by the road surface sensor 111. [ For example, when the road surface state sensor 111 can not measure the friction coefficient and measures the road surface state type, the road surface state predicting unit 160 may be configured to predict the road surface state type. When the road surface sensor 111 measures the coefficient of friction, the road surface condition predicting unit 160 can directly estimate the road surface friction coefficient.

The road traffic information obtaining unit 130 may receive road traffic information in real time from a traffic server (not shown) operated by a provider that provides information related to traffic management subjects or road traffic. Here, the road traffic information may include the current traffic volume at each point on the road and the average traffic speed of the vehicle.

The road risk prediction unit 170 performs a function of predicting a road risk of the prediction target point using a road risk prediction algorithm. The road risk prediction algorithm can predict the road risk by reflecting the predicted road surface friction coefficient, the stored road geometry, and the road traffic information obtained in real time.

On the other hand, when the road surface friction coefficient is not measured by the road surface sensor 111, the road surface friction coefficient is estimated by reflecting road pavement material, vehicle average traffic speed, and the like to the road surface state type predicted by the road surface condition predictor 160, The estimated friction coefficient may be applied to the road risk prediction algorithm.

Surface condition type Road friction coefficient (the higher the average speed, the smaller) Dry concrete road surface 0.85-0.95 Dry asphalt road surface 0.8 to 0.9 Wet concrete road surface 0.4 to 0.6 Wet asphalt road surface 0.3 to 0.5 Freezing road 0.1 to 0.2

In order to obtain road surface friction coefficient estimates according to road pavement material, road surface state type, vehicle average traveling speed, etc., tables as shown in Table 2 may be prepared in advance, or conversion equations may be prepared. Of course, in order to use Table 2, information about the road pavement material of the predicted road point may be stored in the database 140 in advance.

The road risk can be calculated from a score of 1 to 100, or it can be calculated by classifying it as 'very dangerous', 'dangerous', 'normal', 'safe' or 'very safe'.

For example, the road risk prediction algorithm can be designed to increase the risk as the road surface friction coefficient becomes smaller. And the road geometry can design the road risk prediction algorithm to increase the risk as the road end tilt, the road crossing slope and the road curvature become larger. On the other hand, the road risk prediction algorithm can be designed to increase the risk as the traffic speed increases. In addition, the higher the traffic volume, the higher the risk of accidents between vehicles, so the road risk prediction algorithm can be designed to increase the risk. It is also possible to design a road hazard prediction algorithm to increase the road risk as the traffic increases, or to design the road so that the risk becomes higher as the traffic volume increases and the risk becomes lower after the traffic volume reaches a certain level. Also, the correlation between road geometry and traffic volume can be analyzed and reflected in the roadway risk prediction algorithm. In addition, the correlation between traffic volume and risk can be used to reflect the road risk prediction algorithm.

Of course, it is also possible to set the influence of road friction coefficient, road geometry and road traffic information on the road risk calculation differently depending on the time zone such as the season, the week, and the night.

FIG. 2 is a flowchart illustrating an operation of a road-risk prediction system according to a weather condition according to an embodiment of the present invention.

Referring to FIGS. 1 and 2, the road surface state information may be collected and the learning data mapped with the weather data may be constructed in the database 140 (S210). The road information collection vehicle mounts the road surface state collecting unit 110 and acquires the road surface state information while traveling on the road.

Next, the learning unit 150 may learn a road surface state prediction model for predicting the road surface state according to the weather state using the learning data constructed in the database 140 (S220). In step S220, when the weather data corresponding to a specific point on the road is input through artificial intelligent learning such as deep running or machine learning, the learning unit 150 generates a road surface state prediction model for predicting the road surface state corresponding thereto You can learn.

The steps S210 to S220 may be continuously performed in a manner of updating the road surface state prediction model by reflecting the generated learning data when the road risk prediction system is operated.

Thereafter, the road surface state predicting unit 160 receives the weather data from the weather server in real time (S230), calculates the road surface state according to the weather state of the prediction object point using the road surface state prediction model learned in step S230 (S240). In step S240, the road surface state predicting unit 160 predicts the road surface state type of the predicted object point by inputting the weather data provided through the weather information obtaining unit 120. [ Of course, the road surface state predicting unit 160 may directly predict the road friction coefficient according to the embodiment.

Meanwhile, the road traffic information obtaining unit 130 can receive the road traffic information from the traffic server in real time (S250). Here, the road traffic information may include the current traffic volume at each point on the road and the average traffic speed of the vehicle.

Next, the roadway risk prediction unit 170 can estimate the roadway risk of the prediction object point using the roadway risk prediction algorithm (S260). In step S260, the road risk prediction unit 170 may predict a road risk that reflects the road surface friction coefficient of the predicted object point, the previously stored road geometry, and the road traffic information obtained in real time.

The road surface friction coefficient of the predicted object point in step S260 may be determined by applying the predicted road surface friction coefficient in step S260 to the road risk prediction algorithm or by comparing the predicted road surface state type with road pavement material, It is possible to apply the road surface friction coefficient estimated based on the traveling speed and the like.

The risk of each road point obtained by the road risk prediction system according to the present invention can be provided to a vehicle terminal or a user terminal through a communication network or through an electric signboard installed on a road. Of course, it is also possible to calculate and provide the safe running speed of the vehicle according to the degree of danger obtained according to the embodiment.

Embodiments of the present invention include a computer-readable medium having program instructions for performing various computer-implemented operations. The medium records a program for executing the above-described method. The medium may include program instructions, data files, data structures, etc., alone or in combination. Examples of such media include magnetic media such as hard disks, floppy disks and magnetic tape, optical recording media such as CD and DVD, programmed instructions such as floptical disk and magneto-optical media, ROM, RAM, And a hardware device configured to store and execute the program. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, Of the right.

110: road surface condition collecting unit
120: weather information acquiring unit
130: Road traffic information obtaining unit
140: Database
150:
160: road road surface state predicting unit
170: road risk prediction unit

Claims (10)

Estimating the road surface state according to the weather condition of the prediction target point using the road surface state prediction model, and
Estimating a road risk of the prediction target point using a road risk prediction algorithm based on the road geometry pre-stored for the prediction target point and the predicted road surface state,
And estimating road risk according to weather conditions. The method of claim 1,
Collecting road surface condition information and constructing learning data mapped with weather data, and
A step of learning the road surface state prediction model for predicting a road surface state according to a weather condition using the learning data
And estimating a road risk according to a weather condition. 3. The method of claim 2,
The road surface state information includes:
A road surface friction coefficient, and a road surface state type,
Wherein the road surface state type is information obtained by dividing a road surface state according to a predetermined standard. 3. The method of claim 2,
The road risk prediction algorithm includes:
A method for predicting road risk according to weather conditions that predicts road risk by reflecting road geometry, road surface condition and road traffic information. 5. The method of claim 4,
The road geometry
At least one of a road end slope, a road crossing slope, and a road curvature,
Wherein the road traffic information includes at least one of traffic volume and traffic speed. A road surface condition predicting unit for predicting a road surface condition according to a weather condition of a prediction target point using a road surface condition prediction model, and
A road risk prediction unit for predicting a road risk of the prediction target point by using a road risk prediction algorithm based on the road geometric structure previously stored for the prediction target point and the predicted road surface state,
A road hazard prediction system according to a weather condition. The method of claim 6,
A database for storing learning data obtained by mapping collected road surface state information to weather data, and
A learning section for learning the road surface state prediction model for predicting a road surface state according to a weather condition using the learning data;
And estimating a road risk according to a weather condition. 8. The method of claim 7,
The road surface state information includes:
A road surface friction coefficient, and a road surface state type,
Wherein the road surface state type is information obtained by dividing a road surface state according to a predetermined criterion. 8. The method of claim 7,
The road risk prediction algorithm includes:
A road hazard prediction system according to weather conditions that predicts road risk by reflecting road geometry, road surface condition and road traffic information. The method of claim 9,
The road geometry
At least one of a road end slope, a road crossing slope, and a road curvature,
Wherein the road traffic information includes at least one of traffic volume and traffic speed.

Images