Method and system for preventing wrong-way driving

A system that collects and analyzes real-time vehicle data to provide visual and audio warnings addresses the issue of wrong-way driving, preventing accidents by enhancing driver awareness and encouraging proactive actions.

KR103046353B1Active Publication Date: 2026-11-10SONGWOO INFORMATION & TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
KR1020250114340
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-08-18
Publication Date
2026-11-10
Estimated Expiration
2045-08-18

AI Technical Summary

Technical Problem

Existing systems fail to effectively prevent vehicles from driving in the wrong direction, leading to frequent accidents, particularly on highways and one-way streets, due to misinterpretation of navigation instructions or road structures.

Method used

A system that collects real-time vehicle information, analyzes the probability of wrong-way driving using a classification model, and provides real-time visual and audio warnings to drivers to prevent entry into wrong-way roads.

Benefits of technology

Effectively prevents vehicles from entering wrong-way roads by providing intuitive visual and audio warnings, enhancing driver awareness and enabling proactive measures to avoid collisions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112025093910902-PAT00001_ABST
    Figure 112025093910902-PAT00001_ABST
Patent Text Reader

Abstract

A method for preventing vehicles driving in the wrong direction is disclosed. According to one aspect of the present invention, a method for preventing vehicles driving in the wrong direction is provided, comprising: a collecting unit collecting driving information of a vehicle in real time; an analysis unit analyzing a probability regarding the possibility of a vehicle entering a wrong-way road based on the driving information; and a providing unit determining that, when the probability is greater than or equal to a first threshold value, the vehicle is at least entering a wrong-way road, and providing a real-time video of the vehicle driving in the wrong direction based on the driving information so that the vehicle voluntarily turns around before completing entry into the wrong-way road.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a method and system for preventing vehicles from driving in the wrong direction. Background Technology

[0003] With the recent increase in elderly drivers and the widespread reliance on navigation systems, accidents involving driving against the flow of traffic are becoming frequent, as drivers misinterpret road structures or misinterpret navigation instructions. Particularly on sections such as highways, one-way streets, and on-ramps, drivers often misidentify their entry direction and drive in the wrong direction; this frequently leads to head-on collisions, which carry a high risk of major casualties. Nevertheless, the reality is that currently, only passive warning measures, such as road signs and markings, are provided to prevent driving against the flow of traffic. Prior art literature

[0005] Republic of Korea Published Utility Model Publication No. 20-2015-0002081 (Published June 3, 2015) The problem to be solved

[0006] The present invention provides a method and system for preventing vehicles traveling in the wrong direction that can effectively prevent vehicles traveling in the wrong direction. means of solving the problem

[0008] According to one aspect of the present invention, a method for preventing a vehicle from driving in the wrong direction is provided, comprising: a collecting unit collecting driving information of a vehicle in real time; an analysis unit analyzing the probability of a vehicle entering a wrong-way road based on the driving information; and a providing unit determining that, when the probability is greater than or equal to a first threshold, the vehicle is entering at least a wrong-way road, and providing a real-time video of the vehicle driving in the wrong direction based on the driving information so that the vehicle voluntarily turns around before completing entry into the wrong-way road.

[0009] The driving information includes a vehicle video captured in real time by a video device installed around the road for driving in the wrong direction, and the step of analyzing the probability analyzes the probability based on the driving direction of the vehicle detected in the vehicle video, and the real-time video of driving in the wrong direction is a vehicle video from the time of provision and can be provided through a display device installed around the road for driving in the wrong direction.

[0010] The step of analyzing the probability may include a step of generating a multidimensional feature vector by comparing a driving direction history vector, which is a time-series vectorization of the position and direction of the vehicle within the vehicle image, with a reference driving direction vector of the wrong-way road, and a step of calculating the probability of the vehicle entering the wrong-way road through a classification model using the feature vector as input.

[0011] The step of generating a feature vector can generate a feature vector composed of multiple statistical values ​​including the average value, minimum value, change amount, and continuous interval characteristic value less than a set value of the cosine similarity by calculating the cosine similarity between the driving direction history vector and the reference driving direction vector.

[0012] A classification model can calculate probabilities by applying weights and biases to each dimension value of a feature vector to perform a linear combination, and then applying a non-linear function to the result of the linear combination.

[0013] The providing unit may further include a step of providing a broadcast inducing the vehicle to turn around through an audio device installed near the road for driving in the wrong direction.

[0014] Driving information includes the vehicle's location, direction, and speed collected in real time through communication with the vehicle, and the step of analyzing the probability may include a step of generating a multidimensional feature vector by comparing a driving direction history vector, which is a time-series vector of the vehicle's location and direction, with a reference driving direction vector of the wrong-way road, and a step of calculating the probability of the vehicle entering the wrong-way road through a classification model using the feature vector as input.

[0015] Real-time reverse driving video may be generated by visualizing driving information.

[0016] Real-time reverse driving video can be generated by arranging driving information in a time series to generate the driving trajectory of a vehicle, mapping the driving trajectory with road map information, and then converting it into a video through data processing for visual reproduction.

[0017] According to one aspect of the present invention, a vehicle driving in the wrong direction prevention system is provided, comprising: a collection unit that collects driving information of a vehicle in real time; an analysis unit that analyzes the probability of a vehicle entering a wrong-way road based on the driving information; and a providing unit that, when the probability is greater than or equal to a first threshold, determines that the vehicle is entering at least a wrong-way road and provides a real-time video of the vehicle driving in the wrong direction based on the driving information so that the vehicle voluntarily turns around before completing entry into the wrong-way road. Effects of the invention

[0019] According to the present invention, vehicles traveling in the wrong direction can be effectively prevented. Brief explanation of the drawing

[0021] FIG. 1 is a flowchart illustrating a method for preventing vehicles from driving in the wrong direction according to an embodiment of the present invention. FIG. 2 is a configuration diagram showing a reverse-driving vehicle prevention system according to one embodiment of the present invention. Specific details for implementing the invention

[0022] The present invention is capable of various modifications and may have various embodiments; specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the present invention. In describing the present invention, detailed descriptions of related prior art are omitted if it is determined that such detailed descriptions may obscure the essence of the present invention.

[0023] Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another.

[0024] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0025] Hereinafter, embodiments of the method for preventing vehicles driving in reverse and the system (100) according to the present invention will be described in detail with reference to the attached drawings. In describing with reference to the attached drawings, identical or corresponding components are given the same reference numerals, and redundant descriptions thereof will be omitted.

[0027] First, a method for preventing vehicles traveling in the wrong direction according to one embodiment of the present invention will be described.

[0028] According to the present embodiment, as illustrated in FIGS. 1 and 2, a method for preventing a vehicle from driving in the wrong direction is provided, comprising the steps of: a collection unit (110) collecting driving information of a vehicle in real time (S110); an analysis unit (120) analyzing the probability of a vehicle entering a wrong-way road based on the driving information (S120); and a providing unit (140) determining that if the probability is greater than or equal to a first threshold value, the vehicle is entering at least a wrong-way road, and providing a real-time video of the vehicle based on the driving information to the vehicle so that the vehicle voluntarily turns around before completing entry into the wrong-way road (S140).

[0029] According to this embodiment, vehicles traveling in the wrong direction can be effectively prevented.

[0030] Hereinafter, each step of the method for preventing vehicles driving in the wrong direction according to the present embodiment will be described with reference to FIGS. 1 and FIGS. 2.

[0032] The method for preventing vehicles from driving in the wrong direction according to the present embodiment can be performed by a computing device.

[0033] Step 110 allows the collection unit (110) to collect driving information of the vehicle in real time.

[0034] Driving information may include vehicle footage captured in real time through a video device, such as a camera installed around the road where driving in the wrong direction occurs.

[0035] Meanwhile, driving information may include the vehicle's location, direction, and speed collected in real time through communication with the vehicle.

[0036] At this time, communication with the vehicle can be carried out in a V2X manner, and this can be implemented through various wireless communication technologies such as DSRC, C-V2X, LTE, and 5G.

[0037] Among the driving information, the vehicle's location information can be collected as base coordinate values ​​such as GPS and GNSS, the vehicle's direction information can be obtained through a gyroscope sensor, steering angle sensor, etc., and the vehicle's speed information can be collected in real time through an OBD system, CAN network, etc.

[0038] Step 120 allows the analysis unit (120) to analyze the probability of a vehicle entering a reverse-direction road based on driving information.

[0039] Here, a road for driving against the flow of traffic can be understood as a road where vehicles drive against the flow of traffic when entering or driving.

[0040] The step of analyzing the probability (S120) can analyze the probability based on the driving direction of the vehicle detected in the vehicle image captured in real time through the video device.

[0041] The step of analyzing the probability (S120) may include the step of generating a feature vector (S121) and the step of calculating the probability (S122).

[0042] Step 121 can generate a multidimensional feature vector by comparing the position and direction of the vehicle in a vehicle image captured in real time through a video device (when analyzing the probability based on the driving direction of the vehicle detected in the vehicle image), and / or the driving direction history vector, which is vectorized in time series through communication with the vehicle, with the reference driving direction vector of the reverse driving road.

[0043] The analysis unit (120) can sort the vehicle's location and direction information into time series data and then convert it into a vector form to represent the vehicle's driving trajectory over time as a driving direction history vector.

[0044] Subsequently, by comparing the driving direction history vector with the reference driving direction vector of a predefined reverse-driving road, a multidimensional feature vector can be generated that quantifies the degree of direction alignment and deviation between the two vectors.

[0045] The step of generating a feature vector (S121) can generate a feature vector composed of a plurality of statistical values ​​including an average value, a minimum value, a change amount, and a continuous interval characteristic value less than a set value of the cosine similarity by calculating the cosine similarity between the driving direction history vector and the reference driving direction vector.

[0046] In the process of generating feature vectors, cosine similarity can be utilized to quantify the degree of similarity between the driving direction history vector and the reference driving direction vector.

[0047] Cosine similarity is a value calculated by dividing the inner product of two vectors by the product of their magnitudes. For vectors A (driving direction history vector) and B (reference driving direction vector), cosine similarity is defined as the inner product of A and B divided by the product of their magnitudes. This value can range from -1 to 1. A value closer to 1 indicates that the directions of the two vectors are similar, a value closer to 0 indicates that they are perpendicular to each other, and a value closer to -1 indicates that they are in opposite directions.

[0048] The average value of cosine similarity can serve as an indicator of how closely the vehicle's driving direction matches the standard driving direction of the reverse-traffic road across the entire driving section.

[0049] The minimum value of cosine similarity can reflect whether the vehicle's driving direction has abruptly deviated at a specific point in time, and the amount of change can indicate whether the similarity of the vehicle's driving direction has changed abruptly over time.

[0050] In addition, the characteristic value of a continuous interval below the set value refers to the length or frequency of intervals in which the cosine similarity is continuously maintained at or below the set value, and this can be used as a risk persistence indicator to determine whether the vehicle's tendency to drive in the wrong direction persists for a certain period of time or longer.

[0051] Multiple statistical values ​​are composed of multidimensional feature vectors and subsequently provided as input values ​​to a classification model, thereby improving the precision and reliability of reverse driving judgment.

[0052] Step 122 can calculate the probability of a vehicle entering a wrong-way road using a classification model with a feature vector as input.

[0053] Since the feature vector is the result of statistically analyzing the time-series difference between the vehicle's driving direction history and the reference driving direction of the wrong-way road, it can be utilized as a high-dimensional input that reflects the vehicle's overall driving behavior.

[0054] The analysis unit (120) inputs a feature vector into a classification model to derive the probability in real time that the vehicle is currently entering a road for driving in the wrong direction or has entered and is driving in the wrong direction.

[0055] This probability-based calculation method can enable quantitative judgment through statistical analysis and learning-based prediction, even in boundary situations where the vehicle's driving direction does not clearly differ from the standard driving direction of a wrong-way road or where judgment is ambiguous.

[0056] This can increase the accuracy of determining whether a vehicle is driving in the wrong direction and also improve the speed of real-time response.

[0057] A classification model can calculate probabilities by applying weights and biases to each dimension value of a feature vector to perform a linear combination, and then applying a non-linear function to the result of the linear combination.

[0058] Classification models can be machine learning-based and can be implemented as logistic regression, multilayer perceptron (MLP), neural network-based models, etc., and in some cases, can be composed of a combination of these.

[0059] Each dimension value (element) of the feature vector is a quantitative statistical value of the driving direction history, and the classification model can perform a linear combination of these dimension values ​​by applying weights and biases derived through learning.

[0060] The linear combination result calculated at this time can be interpreted as a score indicating the probability that the vehicle is driving in the wrong direction, and by applying non-linear functions such as sigmoid and softmax to it, the score can be converted into a probability value between 0 and 1.

[0061] Through this, the classification model can generalize and interpret various forms of vehicle driving data, and in particular, it is possible to implement a classification structure that responds with high sensitivity to driving patterns that include abnormal driving trajectories or directional deviations.

[0062] Step 140 can provide a real-time reverse driving video based on driving information to the vehicle, so that when the probability is greater than or equal to a first threshold, the providing unit (140) determines that the vehicle is entering at least a reverse driving road and the vehicle voluntarily turns around before completing entry into the reverse driving road.

[0063] By providing real-time video of driving in the wrong direction, the driver of the vehicle can receive a clear visual warning about the possibility of driving in the wrong direction, thereby enabling them to quickly take preventive measures such as voluntarily turning around, decelerating, or stopping.

[0064] Specifically, the real-time reverse driving video may be a vehicle video captured in real-time through a video device from the time it is provided.

[0065] The providing unit (140) determines that a calculated probability is greater than or equal to a first threshold value (e.g., 0.6 to 0.7, more specifically 0.65) as a warning sign of driving in the wrong direction, and in response, can provide the driver of the vehicle with a real-time video of the vehicle being filmed in real-time through a video device at that time as a real-time video of driving in the wrong direction.

[0066] Meanwhile, the real-time reverse driving video may be generated by visualizing driving information collected in real time through communication with the vehicle.

[0067] The providing unit (140) can determine that a calculated probability is greater than or equal to a first threshold value (e.g., 0.6 to 0.7, more specifically 0.65) as a warning sign of driving in the wrong direction, and in response, can generate a real-time driving in the wrong direction video by visually configuring driving information collected through recent communication with the vehicle and provide it to the driver of the vehicle.

[0068] Real-time reverse driving video displays changes in the vehicle's position, direction, speed, etc., collected through communication with the vehicle in a time-series manner, enabling the driver to intuitively recognize abnormalities in the current vehicle driving path.

[0069] Real-time reverse driving video can be generated by arranging driving information collected through communication with the vehicle in a time series to generate the vehicle's driving trajectory, mapping the driving trajectory with road map information, and then converting it into a video through data processing for visual reproduction.

[0070] More specifically, the vehicle's position, direction, and speed data can be arranged in chronological order to form a time-series driving trajectory, and the driving trajectory can be configured to allow visual comparison of the difference between the vehicle's direction of travel on the actual road and the reference driving direction by mapping the reference driving direction to pre-set road map information.

[0071] At this time, the driving trajectory can be represented by solid lines, dotted lines, directional arrows, etc., and the discrepancy with the standard driving direction of the road can be emphasized through visual elements such as color changes, flashing effects, and warning icons.

[0072] These visual elements can be rendered in real-time frame by frame to generate a continuous video.

[0073] Meanwhile, the video of driving in the wrong direction may be provided through a display device, such as a high-resolution electronic display installed around the road where driving in the wrong direction occurs, and / or through an in-vehicle display, such as a display within the vehicle's HUD or instrument panel.

[0074] The real-time reverse driving video provided in this way allows the driver to intuitively recognize that the vehicle's current driving direction is inconsistent with the road's standard driving direction, and can induce a stronger visual perception effect than a simple warning message.

[0075] In particular, by enabling the driver to recognize the risk of driving in the wrong direction early through the video before fully entering the wrong-way road, the driver can be encouraged to voluntarily take measures such as turning around, slowing down, or stopping.

[0076] The providing unit (140) may further include a step (S145) of providing a broadcast (S145) that induces the vehicle to turn around, such as "You are driving in the wrong direction. Please turn around," through an audio device such as a speaker installed around the wrong-way road.

[0077] Meanwhile, the providing unit (140) can provide a broadcast that induces the vehicle to turn around through communication with the vehicle.

[0078] At this time, the turnaround guidance broadcast provided to the target vehicle can be audibly output to the driver through the target vehicle's internal audio, speakers, etc.

[0079] The execution unit (150) may further include a step (S150) in which, if the probability is greater than or equal to a second threshold that is greater than a first threshold, the vehicle determines that it has entered a road for driving in the wrong direction and performs different measures to prevent driving in the wrong direction depending on the speed of the vehicle.

[0080] The execution unit (150) can determine that the vehicle has actually entered a reverse-traffic road if the calculated probability is greater than the second threshold (e.g., 0.8 to 0.9, more specifically 0.85) which is greater than the first threshold (e.g., 0.6 to 0.7, more specifically 0.65).

[0081] This judgment indicates that the situation requires active intervention beyond a simple warning level, and accordingly, the execution unit (150) can perform the reverse driving prevention measure differently in terms of intensity and method based on the vehicle's current speed.

[0082] In this case, the vehicle's speed serves as a key indicator for assessing the risk of driving in the wrong direction. By differentiating response levels according to each speed range—low, medium, and high—it is possible to maximize safety, such as preventing collisions caused by driving in the wrong direction, while reducing unnecessary control.

[0083] The step (S150) of performing measures to prevent driving in reverse can, when the speed of the vehicle is less than the first speed value, transmit a warning to the vehicle and surrounding vehicles through communication with the vehicle and surrounding vehicles, and transmit a request to the traffic control center to prepare for dispatch of a patrol vehicle through communication with the traffic control center.

[0084] This can be understood as being intended to respond to situations where the vehicle's speed is less than a first speed value (e.g., 20 to 40 km / h, specifically 30 km / h), that is, when the vehicle is judged to have started driving in the wrong direction while at rest or moving slowly.

[0085] In this case, since the vehicle's speed is not very high, the immediate risk of an accident is relatively low; however, by recognizing and warning of this early at the initial stage of entering the wrong-way road, further entry or acceleration of the vehicle can be prevented in advance.

[0086] The execution unit (150) can transmit a warning not only to the vehicle in question but also to surrounding vehicles, that is, adjacent vehicles within a set radius, through V2X-based communication, thereby inducing surrounding vehicles to drive evasively or decelerate.

[0087] These warnings can be implemented as instruction signals to induce the driver to decelerate, and can be immediately delivered to the driver through various user interfaces (HMI), such as instrument panel displays, HUDs, and in-vehicle audio notifications.

[0088] In addition, the execution unit (150) can request the traffic control center to prepare for the dispatch of patrol vehicles via V2X-based communication, thereby enabling on-site response to be carried out without delay when necessary.

[0089] The step (S150) of performing a reverse driving prevention measure can transmit a rapid deceleration command to the vehicle when the vehicle's speed is greater than or equal to a first speed value or less than a second speed value.

[0090] This can be seen as an initial response procedure performed in a section where the vehicle speed is above a first speed value (e.g., 20 to 40 km / h, specifically 30 km / h) and below a second speed value (e.g., 50 to 70 km / h, specifically 60 km / h), that is, when a vehicle traveling in the wrong direction is moving beyond a low speed state and is traveling at a certain speed or higher, but has not yet reached a medium speed.

[0091] In this case, the possibility of driving in reverse has increased, but since the vehicle speed is lower to some extent and there is a possibility that the driver will respond to the warning and voluntarily decelerate or stop, the execution unit (150) can transmit a rapid deceleration command to the vehicle via V2X-based communication when the vehicle speed is greater than or equal to the first speed value and less than the second speed value.

[0092] Such rapid deceleration commands can be implemented as instruction signals to induce deceleration in the driver and can be immediately delivered to the driver through various user interfaces (HMI), such as instrument panel displays, HUDs, and in-vehicle audio notifications.

[0093] The step (S150) of performing measures to prevent driving in reverse can transmit a braking command to the vehicle and surrounding vehicles and transmit a request for dispatch of patrol vehicles to the traffic control center when the vehicle's speed is greater than or equal to a second speed value or less than a third speed value.

[0094] This can be seen as a step to respond to a situation where the vehicle's speed is greater than or equal to a second speed value (e.g., 50 to 70 km / h, specifically 60 km / h) and less than a third speed value (e.g., 70 to 90 km / h, specifically 80 km / h), that is, when a vehicle is traveling in the wrong direction at a certain speed and the possibility of an actual collision accident is significantly increased, and accordingly, the execution unit (150) can transmit a braking command to the vehicle and surrounding vehicles.

[0095] A braking command is an instruction signal that more strongly induces the driver to decelerate or stop through V2X-based communication with the vehicle, and can be implemented as a higher-level warning than a sudden deceleration command.

[0096] The braking command can be conveyed as a visually and / or audibly enhanced warning through HMIs such as instrument panel displays, HUDs, and audio systems, and in some cases, may be linked with ADAS functions to induce semi-automatic braking control.

[0097] In addition, the same braking command is transmitted to surrounding vehicles via V2X-based communication, which can prevent collisions between vehicles and minimize traffic disruption.

[0098] In addition, the execution unit (150) can transmit actual dispatch requests for patrol vehicles through V2X-based communication with the traffic control center, thereby supporting rapid on-site response in cases where driving against traffic has developed into a real danger.

[0099] The step of performing a reverse driving prevention measure (S150) can, when the speed of the vehicle is greater than or equal to a third speed value, remotely control the vehicle and surrounding vehicles to brake, and transmit an emergency signal control request to the traffic control center to immediately change the signal located in the direction of travel of the vehicle to red.

[0100] This can be understood as a final and mandatory intervention measure to respond to a dangerous situation when the vehicle speed is above the third speed value (e.g., 70 to 90 km / h, specifically 80 km / h), that is, when a vehicle traveling in the wrong direction is traveling at high speed.

[0101] The execution unit (150) can directly access the control system of the vehicle and surrounding vehicles through V2X-based communication and force physical braking by remotely operating a braking device such as a brake control module.

[0102] This remote control is a measure that forcibly stops the vehicle regardless of the driver's will, thereby enabling an immediate response to abnormal high-speed driving in the wrong direction.

[0103] In addition, through real-time communication with the traffic control center, an emergency signal control request can be transmitted to immediately change the traffic signals at intersections or major roads located in the direction of travel of the vehicle traveling in the wrong direction to red.

[0104] Through this, it becomes possible to prevent major traffic accidents and ensure the safety of surrounding vehicles by blocking the further entry of vehicles traveling in the wrong direction and restricting the progress of other vehicles on the relevant road section.

[0106] Next, a reverse driving vehicle prevention system (100) according to one embodiment of the present invention will be described.

[0107] According to the present embodiment, as illustrated in FIGS. 1 and 2, a vehicle driving in the wrong direction prevention system is presented, comprising a collection unit (110) that collects driving information of a vehicle in real time, an analysis unit (120) that analyzes the probability of a vehicle entering a wrong-way road based on the driving information, and a providing unit (140) that, when the probability is greater than or equal to a first threshold, determines that the vehicle is entering at least a wrong-way road and provides a real-time video of the vehicle driving in the wrong direction based on the driving information so that the vehicle voluntarily turns around before completing entry into the wrong-way road.

[0108] According to this embodiment, vehicles traveling in the wrong direction can be effectively prevented.

[0109] Each component of the reverse-driving vehicle prevention system (100) according to the present embodiment will be described below with reference to FIGS. 1 and FIGS. 2, and specific details regarding the components may follow the details of the reverse-driving vehicle prevention method described above.

[0111] The collection unit (110) can collect driving information of the vehicle in real time.

[0112] Driving information may include vehicle footage captured in real time through a video device, such as a camera installed around the road where driving in the wrong direction occurs.

[0113] Meanwhile, driving information may include the vehicle's location, direction, and speed collected in real time through communication with the vehicle.

[0114] The analysis unit (120) can analyze the probability of a vehicle entering a reverse-direction road based on driving information.

[0115] The analysis unit (120) can analyze the probability based on the driving direction of the vehicle detected in the vehicle image captured in real time through the video device.

[0116] Meanwhile, the analysis unit (120) can generate a feature vector and calculate a probability.

[0117] The analysis unit (120) can generate a multidimensional feature vector by comparing the position and direction of the vehicle in the vehicle image captured in real time through the image device (when analyzing the probability based on the driving direction of the vehicle detected in the vehicle image), and / or the driving direction history vector, which is vectorized in real time through communication with the vehicle, with the reference driving direction vector of the reverse driving road.

[0118] The analysis unit (120) can calculate the cosine similarity between the driving direction history vector and the reference driving direction vector and generate a feature vector composed of multiple statistical values ​​including the average value, minimum value, change amount, and continuous interval characteristic value less than the set value of the cosine similarity.

[0119] The analysis unit (120) can calculate the probability of a vehicle entering a wrong-way road through a classification model using a feature vector as input.

[0120] A classification model can calculate probabilities by applying weights and biases to each dimension value of a feature vector to perform a linear combination, and then applying a non-linear function to the result of the linear combination.

[0121] The providing unit (140) can determine that the vehicle is entering at least a reverse-driving road when the probability is greater than or equal to a first threshold value, and can provide the vehicle with a real-time reverse-driving video based on driving information so that the vehicle voluntarily turns around before completing entry into the reverse-driving road.

[0122] Specifically, the real-time reverse driving video may be a vehicle video captured in real-time through a video device from the time it is provided.

[0123] Meanwhile, the real-time reverse driving video may be generated by visualizing driving information collected in real time through communication with the vehicle.

[0124] Real-time reverse driving video can be generated by arranging driving information collected through communication with the vehicle in a time series to generate the vehicle's driving trajectory, mapping the driving trajectory with road map information, and then converting it into a video through data processing for visual reproduction.

[0125] Meanwhile, real-time video of driving in the wrong direction can be provided through display devices such as high-resolution electronic display boards installed around the road where driving in the wrong direction occurs, and / or through in-vehicle displays such as the vehicle's HUD or a display within the instrument panel.

[0126] The providing unit (140) can provide a broadcast that induces the vehicle to turn around, such as "You are driving in reverse. Please turn around," through a sound device installed around the road where the vehicle is driving in reverse.

[0127] Meanwhile, the providing unit (140) can provide a broadcast that induces the vehicle to turn around through communication with the vehicle.

[0128] If the probability is greater than or equal to a second threshold that is greater than a first threshold, it is determined that the vehicle has entered a reverse-driving road, and a unit (150) may further include different reverse-driving prevention measures depending on the speed of the vehicle.

[0129] The execution unit (150) can, when the speed of the vehicle is less than the first speed value, transmit a warning to the vehicle and surrounding vehicles through communication with the vehicle and surrounding vehicles, and transmit a request to the traffic control center to prepare for the dispatch of a patrol vehicle through communication with the traffic control center.

[0130] The execution unit (150) can transmit a rapid deceleration command to the vehicle when the vehicle's speed is greater than or equal to a first speed value or less than a second speed value.

[0131] The execution unit (150) can transmit a braking command to the vehicle and surrounding vehicles and transmit a request for the dispatch of a patrol vehicle to the traffic control center when the vehicle's speed is greater than or equal to a second speed value or less than a third speed value.

[0132] The execution unit (150) can remotely control the vehicle and surrounding vehicles to brake when the vehicle's speed is greater than or equal to the third speed value, and transmit an emergency signal control request to the traffic control center to immediately change the signal located in the vehicle's driving direction to red.

[0134] Meanwhile, the components of the aforementioned embodiments can be easily identified from a process perspective. That is, each component can be identified as a separate process. Furthermore, the processes of the aforementioned embodiments can be easily identified from the perspective of the device components.

[0135] In addition, the technical details described above 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, etc., either individually or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiments, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. Hardware devices may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

[0137] Although an embodiment of the present invention has been described above, those skilled in the art may modify and change the present invention in various ways by adding, changing, deleting, or adding components, etc., without departing from the spirit of the present invention as described in the claims, and such modifications and changes are also to be included within the scope of the rights of the present invention. Explanation of the symbols

[0139] 100: Anti-wrong-way driving system 110: Collection Department 120: Analysis Department 140: Provider 150: Execution Unit

Claims

Claim 1 A collection unit collects driving information of a vehicle in real time; an analysis unit analyzes the probability of the vehicle entering a reverse-driving road based on the driving information; The providing unit includes the step of providing a real-time reverse driving video prepared based on the driving information to the vehicle, such that when the probability is greater than or equal to a first threshold, the vehicle is determined to be entering at least the reverse driving road, and the vehicle voluntarily turns around before completing entry into the reverse driving road; the driving information includes a vehicle video captured in real time by an image device installed around the reverse driving road; the real-time reverse driving video is the vehicle video from the time when the provision of the real-time reverse driving video begins, and is provided to the driver of the vehicle through a display device installed around the reverse driving road; the step of analyzing the probability includes analyzing the probability based on the driving direction of the vehicle detected in the vehicle video; and the step of analyzing the probability includes generating a multidimensional feature vector by comparing a driving direction history vector, which vectorizes the position and direction of the vehicle within the vehicle video in a time series, with a reference driving direction vector of the reverse driving road.The method includes a step of calculating the probability of the vehicle entering the wrong-way road through a classification model using the above feature vector as input, wherein the step of generating the feature vector comprises calculating the cosine similarity between the driving direction history vector and the reference driving direction vector to generate the feature vector composed of a plurality of statistical values ​​including the average value, minimum value, change amount, and continuous interval characteristic value below a set value of the cosine similarity, wherein the continuous interval characteristic value below a set value is the length or frequency of an interval in which the cosine similarity is continuously maintained below a set value, and is a risk persistence indicator for determining whether the vehicle's wrong-way driving tendency persists for a certain period of time or longer, and the method further includes a step of determining that the vehicle has entered the wrong-way road when the probability is greater than or equal to a second threshold value which is greater than the first threshold value, and performing different wrong-way driving prevention measures according to the speed of the vehicle, wherein the step of performing the wrong-way driving prevention measures includes, when the speed of the vehicle is less than a first speed value, providing a warning to the vehicle and surrounding vehicles through communication with the vehicle and surrounding vehicles. A method for preventing vehicles driving in the wrong direction, comprising: transmitting, and through communication with a traffic control center, transmitting a request for the dispatch of a patrol vehicle to the traffic control center; if the speed of the vehicle is greater than or equal to the first speed value and less than the second speed value which is greater than the first speed value, transmitting a rapid deceleration command to the vehicle; if the speed of the vehicle is greater than or equal to the second speed value and less than the third speed value which is greater than the second speed value, transmitting a braking command to the vehicle and surrounding vehicles, transmitting a dispatch request for the patrol vehicle to the traffic control center; and if the speed of the vehicle is greater than or equal to the third speed value, remotely controlling the vehicle and surrounding vehicles to brake, and transmitting an emergency signal control request to the traffic control center to immediately change the signal located in the direction of travel of the vehicle to red. Claim 2 delete Claim 3 delete Claim 4 delete Claim 5 A method for preventing vehicles going in the wrong direction according to claim 1, wherein the classification model performs a linear combination by applying weights and biases to each dimension value of the feature vector, and then calculates the probability by applying a non-linear function to the result of the linear combination. Claim 6 A method for preventing vehicles driving in the wrong direction according to claim 1, further comprising the step of the providing unit providing a broadcast that induces the vehicle to turn around through a sound device installed around the road for driving in the wrong direction. Claim 7 delete Claim 8 delete Claim 9 delete Claim 10 A collection unit that collects driving information of a vehicle in real time; an analysis unit that analyzes the probability of the vehicle entering a reverse-direction road based on the driving information; The presenter includes a providing unit that, when the probability is greater than or equal to a first threshold, determines that the vehicle is entering at least the reverse-driving road and provides the vehicle with a real-time reverse-driving video prepared based on the driving information so that the vehicle voluntarily turns around before completing entry into the reverse-driving road; the driving information includes a vehicle video captured in real time by an image device installed around the reverse-driving road; the real-time reverse-driving video is the vehicle video from the point in time when the provision of the real-time reverse-driving video begins, and is provided to the driver of the vehicle through a display device installed around the reverse-driving road; the analysis unit analyzes the probability based on the driving direction of the vehicle detected in the vehicle video; the analysis unit generates a multidimensional feature vector by comparing a driving direction history vector, which vectorizes the position and direction of the vehicle within the vehicle video in a time series, with a reference driving direction vector of the reverse-driving road; calculates the probability regarding the possibility of the vehicle entering the reverse-driving road through a classification model using the feature vector as input; and the analysis unit, the driving direction history vector and the Calculate the cosine similarity between reference driving direction vectors to generate the feature vector composed of a plurality of statistical values ​​including the average value, minimum value, change amount, and continuous interval characteristic value below a set value of the cosine similarity, wherein the continuous interval characteristic value below a set value is the length or frequency of the interval in which the cosine similarity is continuously maintained below the set value, and is a risk persistence indicator for determining whether the vehicle's tendency to drive in the wrong direction persists for a certain period of time or longer, and if the probability is greater than or equal to a second threshold value which is greater than a first threshold value, determine that the vehicle has entered the wrong direction road.A reverse driving prevention system further comprising an execution unit that performs different reverse driving prevention measures depending on the speed of the vehicle, wherein the execution unit, when the speed of the vehicle is less than a first speed value, transmits a warning to the vehicle and surrounding vehicles through communication with the vehicle and surrounding vehicles, and transmits a request for dispatch preparation of a patrol vehicle to the traffic control center through communication with the traffic control center, when the speed of the vehicle is greater than or equal to the first speed value but less than a second speed value greater than the first speed value, transmits a rapid deceleration command to the vehicle, and when the speed of the vehicle is greater than or equal to the second speed value but less than a third speed value greater than the second speed value, transmits a braking command to the vehicle and surrounding vehicles and transmits a request for dispatch of the patrol vehicle to the traffic control center, and when the speed of the vehicle is greater than or equal to the third speed value, remotely controls the vehicle and surrounding vehicles to brake, and transmits an emergency signal control request to the traffic control center to immediately change the signal located in the direction of travel of the vehicle to red.

Citation Information

Patent Citations

  • System of peventing reverse running using virtulal detection line in image

    KR1020220154463A

  • Danger detecting device, danger detecting method, and danger detecting program

    WO2021038619A1

  • Information output device and control program

    JP7573595B2

  • Reverse running prevention system at rest area

    KR1020220068783A

  • Apparatus for detecting reverse driving vehicle based on object image analysis and method therefor

    KR1020250030272A