Search

KR-20260066865-A - System for Detecting Road Traffic Condition by Using V2X Data

KR20260066865AKR 20260066865 AKR20260066865 AKR 20260066865AKR-20260066865-A

Abstract

A road traffic condition detection system using V2X data is disclosed. The present embodiment provides a road traffic situation detection system using V2X data that divides areas on a road into a matrix form, defines the areas as recovery cells, collects road traffic status information using V2X (Vehicle to Everything) data obtained through communication with a vehicle for each cell, and processes the road traffic status to detect sudden situations for not only autonomous vehicles but also general vehicles.

Inventors

  • 박경국

Assignees

  • 주식회사 오스코

Dates

Publication Date
20260512
Application Date
20241105

Claims (9)

  1. A collection unit that collects V2X data from vehicles on the road from the RSU (Road Side Unit); A quality factor calculation unit that quantifies the quality factor in cell units for each area included in a virtual matrix corresponding to the road based on the above V2X data; A color code generation unit that generates a color code corresponding to the above quality factor; An image generation unit that maps the above color code to the above cell unit and generates a virtual matrix mapped to the above color code as an image; A labeling unit that generates labeling data labeled as general communication situations and unexpected situations based on a color code pattern on the above image; A modeling unit that generates a sudden event detection engine by inputting the above image and the above labeling data as training data and performing training; An incident detection unit that detects an incident on a new image including a virtual matrix mapped with a new color code using the above-mentioned incident detection engine; A road traffic situation detection device using V2X data characterized by including
  2. In paragraph 1, The above collection unit A road traffic condition detection device using V2X data, characterized by collecting V2X data including the speed, location, and direction of the vehicle using V2I (Vehicle-to-Infrastructure) messages and PVD (Prove Vehicle Data) messages through communication with the vehicle via the RSU.
  3. In paragraph 1, A virtual matrix generation unit that extracts latitude and longitude data from the above V2X data, generates the above virtual matrix set as virtual coordinates on a road to process the above latitude and longitude data in a planar coordinate system by coordinate transformation, distinguishes lanes by rows within the virtual matrix, distinguishes lanes by columns within the virtual matrix, and divides a plurality of areas of the above virtual matrix into cells; A road traffic situation detection device using V2X data characterized by additionally including
  4. In paragraph 1, The above quality factor calculation unit A road traffic situation detection device using V2X data, characterized by extracting coordinate data and speed data of the vehicle included in the above V2X data, generating coordinate values by converting the coordinate data, generating speed values by converting the units of the speed data, and converting the coordinate values into an XY coordinate system with a reference point on the latitude and longitude data as the origin to calculate one of the quality factors.
  5. In paragraph 4, The above quality factor calculation unit A road traffic situation detection device using V2X data, characterized by selecting a cell corresponding to the coordinate value among the virtual matrix above, reflecting the speed value in the corresponding cell, calculating the speed value as an average speed value per calculation cycle, and applying it as one of the quality factors.
  6. In paragraph 5, The above quality factor calculation unit A road traffic situation detection device using V2X data, characterized by calculating the average speed value for each of the above cells by reflecting weights at regular intervals to a value between 0 and 10 or between 0 and 100, and then applying it as one of the quality factors.
  7. In paragraph 1, The above quality factor calculation unit A traffic status verification unit that calculates the traffic status and travel time of the vehicle on the road based on factors stored for each column within the virtual matrix. A road traffic situation detection device using V2X data characterized by additionally including
  8. In paragraph 1, The above color code generating unit If the above quality factor is less than a preset first threshold, the corresponding cell in the virtual matrix is mapped in red, and If the above quality factor is greater than or equal to a preset second threshold and less than or equal to a third threshold, the corresponding cell within the virtual matrix is mapped to yellow, and A road traffic situation detection device using V2X data, characterized by generating a color code that maps the corresponding cell in the virtual matrix to green when the quality factor exceeds a third threshold.
  9. In paragraph 1, The above labeling part is, A road traffic situation detection device using V2X data, characterized by distinguishing and labeling general traffic situation patterns and sudden situation patterns based on the pattern to which the color code is applied and mapped to each cell within the virtual matrix.

Description

System for Detecting Road Traffic Condition by Using V2X Data One embodiment of the present invention relates to a road traffic situation detection system using V2X data. The following description merely provides background information related to the present embodiment and does not constitute prior art. The most fundamental and core technology of Intelligent Transportation Systems (ITS) is the collection of road traffic information. Various collection methods, such as point detection, section detection, and individual vehicle detection, have been developed to gather data on vehicle operation status on the roads. A representative technology of the point detection method utilizes loop detectors. By embedding loop detectors in the road and detecting vehicles through changes in the magnetic field as they pass, this technology collects traffic speed and volume through the installation of double loops and has been used as the most common and accurate data collection method. Subsequently, to replace physical loop detectors, image-based detection methods were developed that recognize vehicles on the road using cameras or radar to detect traffic volume and speed at specific points, largely replacing the loop method. The section detection method is a method that collects traffic conditions in a section by installing license plate recognition devices at two points within a specific section and utilizing the license plate number recognized at the starting point and the installed license plate number recognized at the ending point, based on the distance between the sections and the time at the time of recognition. In addition to the video method using a license plate recognition system, the section traffic information collection method has also been developed to collect the traffic status of the corresponding section by utilizing the communication of Hi-Pass terminals to collect the relevant IDs. As mentioned above, although technology for optimal communication information collection has been continuously developed using the latest sensor technology of the time, side effects such as many errors and high maintenance costs have occurred due to fundamental sensor limitations and environmental influences. Conventional methods for collecting road traffic information are based on Vehicle Detection Systems (VDS), which are traffic volume survey devices that collect road traffic information (traffic volume, speed, occupancy rate, etc.) and transmit it to a center. However, since this method collects representative values for each section, it has limitations as detailed and precise data. Therefore, it is necessary to establish a foundation for providing more effective and precise data by collecting detailed and precise road traffic information through the establishment of an infrastructure collection system that can further subdivide roads along the transverse and longitudinal axes, classify them by lane and micro-cell, and measure traffic conditions in the corresponding cells. With the recent advancement of Cooperative Intelligent Transport Systems (C-ITS) and autonomous vehicle technologies, technologies are being developed to precisely exchange and collect vehicle location and operation information through communication between vehicles and between vehicles and infrastructure. Autonomous vehicles and cooperative autonomous vehicles perceive and assess their surroundings using their own sensors to operate the vehicle autonomously or provide the information necessary for operation. The data collected by the vehicle utilizes high-performance GPS and vehicle sensors to ensure high accuracy and stability. However, in reality, there are limitations to sensor perception, and various problems arise that cannot be resolved solely through the collection of location data. To overcome these limitations, technologies are being developed to support more effective and safer operation by providing and collecting necessary information through mutual cooperation between vehicles and between vehicles and infrastructure. V2X communication is necessary for the exchange of information between vehicles and between vehicles and infrastructure, and various communication methods such as WAVE and C-V2X have been developed. In order to ensure scalability and compatibility of various services along with physical communication methods, standardization of service messages is also underway, and a message set defining specifications for mutually necessary datasets has been developed, and in Korea, work on establishing KS is also in progress. In Korea, the KS message set is being developed based on the U.S. industrial standard message known as SAE J2735. Representative messages in this message set include BSM (Basic Safety Message), PVD (Prove Vehicle Data), SPaT (Signal Phase And Timing Message), MAP (Map Data), and TIM (Traveler Information Message). Conventional incident detection methods operate by detecting events such as the falling, stopping, or drivin