결빙되거나 적설이 있는 도로와 같이 마찰이 작은 노면에서는 일반 노면과 비교했을 때 제동거리가 크게 증가하기 때문에 심각한 교통사고로 이어질 수 있다. 이에 블랙 아이스(Black ice)와 같은 노면 위험을 감지 하기 위한 노면 분류 기술에 대한 연구가 지금까지 지속적으로 이루어지고 있다. ESC(Electronic Stability Control) 시스템은 차량 자세 제어를 통해 마찰이 작은 노면에서 차량의 미끄러짐 및 전복을 방지하는 능동 안전시스템(Active safety system)이다. ESC 시스템의 성능을 위해서는 정확한 노면 마찰 계수(Road friction coefficient) 추정을 통한 노면 분류가 중요하다. 최근의 노면 분류 기술은 카메라, LiDAR 등의 이미 지 기반의 방법에 중점을 두고 연구가 진행되고 있다. 그러나 이러한 이미지 기반의 방법들은 정확도가 낮을 뿐만 아니라 높은 계산 복잡도의 문제를 가지고 있다. 이뿐만 아니라 높은 비용으로 인해 상용화 측면에서도 단점을 드러내고 있다. 본 연구에서는 그림1처럼 센서 융합 기술을 활용하여 이미지 기반 방법의 문제점을 해결하고자 한다. 차량 횡방향 동역학 모델(Vehicle lateral dynamic model)을 선형화하여 칼만 필터(Kalman filter)를 적용한 노면 마찰 계수 추정 알고리즘을 설계하고, 기계학습(Machine learning) 모델을 적용하여 블랙 아이스 검출 알고 리즘을 설계한다. 전기차 CAN 버스로부터 얻을 수 있는 차량 종방향 가속도(Vehicle longitudinal acceleration)를 제어 입력으로 하고, 요 레이트(Yaw rate)를 측정값으로 하여 칼만 필터에 적용하여 차량 종 방향 속도(Vehicle longitudinal velocity)와 차량 횡방향 속도(Vehicle lateral velocity), 요 레이트, 차량 횡방 향 힘(Vehicle lateral force)을 추정한다. 이때 전통적인 칼만 필터 대신 EKF-UI(Extended kalman filter with unknown input)를 적용하여 시스템 행렬의 크기를 줄여 계산 복잡도를 감소시키고 차량의 거동 변화 를 보다 정확하게 반영할 수 있도록 하였다. 추정된 차량 종방향 속도, 차량 횡방향 속도, 요 레이트를 통해 사이드 슬립 각(Side slip angle)을 구해 사이드 슬립 각과 차량 횡방향 힘의 관계를 이용해 특징들을 찾아 기계학습 모델(e.g. 앙상블 기법, SVM 등)을 적용하여 블랙 아이스를 검출할 수 있다. MATLAB/Simulink SW 및 CarSim을 사용하여 개발한 알고리즘의 성능을 검증하였으며, 본 연구의 결과는 ESC 시스템의 성능 을 개선시켜 차량의 미끄러짐으로 인한 교통사고의 예방에 도움이 될 것으로 예상한다. 여기에 스마트 타이 어(Smart tire)의 센서도 추가해 노면과 타이어 사이의 직접적인 데이터를 추가해 검출 성능을 높일 것이다.

Black ice, a thin and nearly invisible ice layer on roads and pavements, poses a significant danger to drivers and pedestrians during winter due to its transparency. We propose an efficient black ice detection system and technique utilizing Global Positioning System (GPS)-reflected signals. This system consists of a GPS antenna and receiver configured to measure the power of GPS L1 band signal strength. The GPS receiver system was designed to measure the signal power of the Right-Handed Circular Polarization (RHCP) and Left-Handed Circular Polarization (LHCP) from direct and reflected signals using two GPS antennas. Field experiments for GPS LHCP and RHCP reflection measurements were conducted at two distinct sites. We present a Normalized Polarized Reflection Index (NPRI) as a methodological approach for determining the presence of black ice on road surfaces. The field experiments at both sites successfully detected black ice on asphalt roads, indicated by NPRI values greater than 0.1 for elevation angles between 45o and 55o. Our findings demonstrate the potential of the proposed GPS-based system as a cost-effective and scalable solution for large-scale black ice detection, significantly enhancing road safety in cold climates. The scientific significance of this study lies in its novel application of GPS reflection signals for environmental monitoring, offering a new approach that can be integrated into existing GPS infrastructure to detect widespread black ice in real-time.

PURPOSES : Road surface conditions are vital to traffic safety, management, and operation. To ensure traffic operation and safety during periods of snow and ice during the winter, each local government allocates considerable resources for monitoring that rely on field-oriented manual work. Therefore, a smart monitoring and management system for autonomous snow removal that can rapidly respond to unexpected abrupt heavy snow and black ice in winter must be developed. This study addresses a smart technology for automatically monitoring and detecting road surface conditions in an experimental environment using convolutional neural networks based on a CCTV camera and infrared (IR) sensor data. METHODS : The proposed approach comprises three steps: obtaining CCTV videos and IR sensor data, processing the dataset acquired to apply deep learning based on convolutional neural networks, and training the learning model and validating it. The first step involves a large dataset comprising 12,626 images extracted from the acquired CCTV videos and the synchronized surface temperature data from the IR sensor. In the second step, image frames are extracted from the videos, and only foreground target images are extracted during preprocessing. Hence, only the area (each image measuring 500 × 500) of the asphalt road surface corresponding to the road surface is applied to construct an ideal dataset. In addition, the IR thermometer sensor data stored in the logger are used to calculate the road surface temperatures corresponding to the image acquisition time. The images are classified into three categories, i.e., normal, snow, and black-ice, to construct a training dataset. Under normal conditions, the images include dry and wet road conditions. In the final step, the learning process is conducted using the acquired dataset for deep learning and verification. The dataset contains 10,100 (80%) data points for deep learning and 2,526 (20%) points for verification. RESULTS : To evaluate the proposed approach, the loss, accuracy, and confusion matrix of the addressed model are calculated. The model loss refers to the loss caused by the estimated error of the model, where 0.0479 and 0.0401 are indicated in the learning and verification stages, respectively. Meanwhile, the accuracies are 97.82% and 98.00%, respectively. Based on various tests that involve adjusting the learning parameters, an optimized model is derived by generalizing the characteristics of the input image, and errors such as overfitting are resolved. This experiment shows that this approach can be used for snow and black-ice detections on roads. CONCLUSIONS : The approach introduced herein is feasible in road environments, such as actual tunnel entrances. It does not necessitate expensive imported equipment, as general CCTV cameras can be applied to general roads, and low-cost IR temperature sensors can be used to provide efficiency and high accuracy in road sections such as national roads and highways. It is envisaged that the developed system will be applied to in situ conditions on roads.