The focus of this research is on the estimation of traffic density from data obtained from Connected and Autonomous Probes (CAPs). CAPs pose an advantage over expensive and invasive infrastructure such as loop detectors. CAPs maneuver their driving trajectories, sensing the presence of adjacent vehicles and distances to them by means of several electronic sensors, whose data can be used for more sophisticated traffic density estimation techniques. Traffic density has a highly nonlinear nature during on-congestion and queue-clearing conditions. Closed-mathematical forms of the traditional density estimation techniques are incapable of dealing with complex nonlinearities, which opens the door for data-driven approaches such as machine learning techniques. Deep learning algorithms excel in data-rich contexts, which recognize nonlinear and highly situation-dependent patterns. Our research is based on an LSTM (Long short-term memory) neural network for the nonlinearity associated with time dynamics of traffic flow. The proposed method is designed to learn the input-output relation of Edie’s definition. At the same time, the method recognizes a temporally nonlinear pattern of traffic. We evaluate our algorithm by using a microscopic simulation program (PARAMICS) and demonstrate that our model accurately estimates traffic density in Free-flow, Transition, and Congested conditions.

Recent literature suggests a need for a more realistic representation of driver behavior. In an effort to integrate prospect theory as a potential descriptive method into traveler behavior, it is important to determine the validity of previous estimated parameters obtained from empirical studies and interviews of people who are not in the same fast-paced, dynamic, travel time disutility setting as real drivers. A genetic algorithm is used to simultaneously estimate the parameters of the cumulated prospect theory (CPT) value and weight functions as well as the coefficients of the random utility model; this procedure leads to estimates that have a higher likelihood value and statistical significance than an equivalent expected utility–based logit model or a CPT-based logit model using the empirical values developed earlier. The value function parameters generally conform to conclusions from previous literature. The weight function parameters, however, suggest that drivers in a fast-paced changing environment with multiple subjects for prospect evaluation may become overwhelmed by the certainty effect.

도로에서 발생하는 포트홀과 같은 돌발상황으로 인한 교통사고는 지속적으로 증가하고 있다. 하지만 도로관리청은 제한된 인력과 예산으로 인해 실시간으로 돌발상황을 관리하는 것은 매우 어렵다. 또한 도로이용자는 자신의 위치와 해당 도로의 관리청이 누구인지 알 수 없어 신고하는 방법이 까다로운 실정이다. 본 연구에서는 위치정보 기반 국민참여형 도로불편 신고 앱의 개발을 위한 UI를 제시하였다. 특히 도로에서 신고가 가능하도록 5가지 사용자 니즈를 도출하고 이를 해결할 수 있는 앱 기능을 제시하였다. 약 4년간의 운영을 통해 앱 이용자의 만족도를 모니터링 하였으며, 총 2,212명이 만족도 평가에 참여하였다. 범위는 0점에서 최대 5점까지이며, 전체 평균점수는 4.8점으로 나타났으며, 개발된 UI 및 신고 및 연계기능 등에서 양호한 평가를 받는 것으로 판단된다. 향후 앱을 통한 신고 및 처리결과 데이터가 지속적으로 축척되어 분석된다면, 도로관리청에서도 이를 활용한 효율적인 도로관리체계가 구축될 수 있을 것으로 기대된다.