Consideration on the Time-Based Evaluation of the Traffic Information Prediction

최근 내비게이션에서는 실시간 교통정보와 과거의 교통정보를 가공하여 미래의 교통정보를 예측하는 패턴 교통정보를 같이 활용하여 빠른 길을 안내해주고 있다. 그러나 현재 사용되는 패턴 교통정보는 과거의 정보를 가공하여 교통정보를 예측하기 때문에 특별한 상황(유고, 날씨 등)에서는 예측이 정확하지 않는 문제점을 가지고 있다. 그래서 본 연구에서는 빠른 길을 찾기 위해 실시간으로 운전자들이 요청하는 경로탐색 데이터를 분석하여 가까운 미래 운전자들이 위치할 도로의 교통 혼잡도를 미리 파악하여 패턴 교통정보 보다 정확한 예측 교통정보를 제시하였다. 연구결과 첫째, 연구지역의 정체경로인 양재에서 마포간 차량속도 비교에서는 기존 상습정체 도로의 속도가중치 정확도가 3km/h에서 18km/h의 오차율이 발생하였지만, 본 연구의 Real 예측 교통 정보를 적용한 결과는 1km/h에서 5km/h의 오차율이 발생하였다. 둘째, 경로 품질에서 기존의 경로보다 최대 약 9분, 평균 약 3분 일찍 목적지에 도착하여 예측 교통정보 결과의 신뢰성을 입증할 수 있었다. 셋째, 기존의 경로탐색 결과 보다 혼잡도를 미리 예측하여 혼잡이 발생할 도로에 대해 회피되는 경로탐색 결과를 도출할 수 있었다. 따라서 본 연구결과의 경로탐색 비교를 통해 교통량에 대한 예측정보를 획득할 수 있었으며 이를 활용하여 실시간 빠른 길 탐색이 가능하고, 향후 교통 흐름을 분산 시키는데도 도움이 될 것으로 판단된다. Recent navigation systems provide quick guide services, based on processing real-time traffic information and past traffic information by applying predictable pattern for traffic information. However, the current pattern for traffic information predicts traffic information by processing past information that it presents an inaccuracy problem in particular circumstances(accidents and weather). So, this study presented a more precise predictive traffic information system than historical traffic data first by analyzing route search data which the drivers ask in real time for the quickest way then by grasping traffic congestion levels of the route in which future drivers are supposed to locate. First results of this study, the congested route from Yang Jae to Mapo, the analysis result shows that the accuracy of the weighted value of speed of existing commonly congested road registered an error rate of 3km/h to 18km/h, however, after applying the real predictive traffic information of this study the error rate registered only 1km/h to 5km/h. Second, in terms of quality of route as compared to the existing route which allowed for an earlier arrival to the destination up to a maximum of 9 minutes and an average of up to 3 minutes that the reliability of predictable results has been secured. Third, new method allows for the prediction of congested levels and deduces results of route searches that avoid possibly congested routes and to reflect accurate real-time data in comparison with existing route searches. Therefore, this study enabled not only the predictable gathering of information regarding traffic density through route searches, but it also made real-time quick route searches based on this mechanism that convinced that this new method will contribute to diffusing future traffic flow.

Accurate traffic prediction is crucial to alleviating traffic congestion in cities. Existing physical sensor-based traffic data acquisition methods have high transmission costs, serious traffic information redundancy, and large calculation volumes for spatiotemporal data processing, thus making it difficult to ensure accuracy and real-time traffic prediction. With the increasing resolution of UAV imagery, the use of unmanned aerial vehicles (UAV) imagery to obtain traffic information has become a hot spot. Still, analyzing and predicting traffic status after extracting traffic information is neglected. We develop a framework for traffic speed extraction and prediction based on UAV imagery processing, which consists of two parts: a traffic information extraction module based on UAV imagery recognition and a traffic speed prediction module based on deep learning. First, we use deep learning methods to automate the extraction of road information, implement vehicle recognition using convolutional neural networks and calculate the average speed of road sections based on panchromatic and multispectral image matching to construct a traffic prediction dataset. Then, we propose an attention-enhanced traffic speed prediction module that considers the spatiotemporal characteristics of traffic data and increases the weights of key roads by extracting important fine-grained spatiotemporal features twice to improve the prediction accuracy of the target roads. Finally, we validate the effectiveness of the proposed method on real data. Compared with the baseline algorithm, our algorithm achieves the best prediction performance regarding accuracy and stability.

Location based services (LBS) are capturing global spotlights increasingly, in which users tend to pick up navigation service as their favorites. This paper examines the dynamic routing problems based on real-time traffic information in LBS. At first, the overall structure of dynamic navigation system is illustrated, which is composed of four ingredients: traffic data collection, digital road map, static routing and dynamic re-routing. Traffic data collection is the task of traffic information center (TIC), where various sources of data are fused into link travel times that are used as criteria to choose candidates in path determining. Digital road map includes digital road network and historical traffic data, where the topology relationship of road network is elementarily explored and the method to represent statistical traffic data is referred to. Routing with historical traffic data and dynamic re-routing with live traffic data are two main functions of dynamic navigation service. The flow of dynamic re-routing algorithm is exhibited. Lastly, a first-in-first-out (FIFO) time-depend shortest path algorithm is discussed for dynamic routing, which adopts the adaptive A^* algorithm based on binary heap priority queue and RB-tree.

Common bus route navigation system provides distance and transferring information based on the user-determined start point and end point on the map. This paper presents the research and experiment of an advanced bus route navigation system in Beijing, which can additionally provide travel time information from start to end. The travel time is predicted based on statistical and real-time traffic information. In Beijing, about 10000 taxies send GPS data to a traffic information center to generate real-time traffic data, and the center accumulates the historical traffic data for generating statistical database. Field test of our system was carried out in January 2007, and the experiment result shows the real-time and statistical traffic information can be used not only for car navigation system, but also for public transit service. In our experiment, the average error rate of predicted travel time is about 10%, which is comparable to car navigation system.

This paper makes the first step towards mining citywide traffic congestion correlation by utilizing traffic related information from social media. Traffic congestion correlation mining, namely studying which road segments close to each other are highly likely to occur congestion simultaneously, is especially important to help many real applications, such as traffic prediction, traffic control, and urban transportation planning. Traditional traffic data collected from various sensors and other equipments are costly to obtain and hard to scale up to cover a entire city. With the rising popularity of social media, it is common for the public transportation systems and governments to share real time traffic information with the public through social media. It provides us great opportunities to study the traffic conditions of a city with the rich and easily available online data. However, it is also very difficult to use social media data to mine the citywide traffic congestion correlation due to the following major challenges: (1) Social media data like tweets in Twitter are usually noisy and hard to process, especially those tweets posted by individuals. (2) There lacks a method to study the citywide traffic congestion correlation. In this paper, instead of crawling all the traffic related tweets of a city, we only focus on utilizing the tweets posted by some particular organizations or governments. Tweets posted by them are more accurate and formal, thus it is much easier for traffic information extraction. We regard the traffic congestion correlation mining task as a spatio-temporal frequent pattern mining problem by considering each tweet reporting the traffic congestion of a particular road segment as a spatio-temporal item. A spatio-temporal frequent pattern mining algorithm TC_Apriori is also proposed to discover the road segment co-occurrence patterns in congestion. We use the tweets reporting the traffic information of Chicago to evaluate the proposed approach, and the results show that the proposed approach can effectively discover the road segment co-occurrence patterns in congestion.

Urban traffic situation information is the basis of effective traffic guidance and traffic control. In order to promote communication and practical applications of traffic information, then can provide full, accurate, real time and sufficient traffic information service to public, government and correlative enterprises, in this paper, the authors proposed the dynamic traffic information collection and processing methods of intelligent transportation systems (ITS) common information platform (CIP) based on floating car technology. According to the characteristics of the ITS CIP data collection, they were required large scale, accurate and dynamic data collecting. A complete set of dynamic traffic information processing methods based on floating car technology were studied, including road data storage by two queues, error data elimination by taximeter, map matching by region dividing method and boundary rectangle, and traffic condition calculation by circle queues and A* algorithm. The application of the methods in the ITS CIP of Guangzhou City proved that the methods were quite effective and valuable to be popularized.

For providing drivers with robust traffic information and Optimizing the energy management of Hybrid Electric Vehicles (HEVs), it is important to predict traffic information accurately with past traffic information. As acquisition of the traffic information have been easier by the development of Intelligent Transportation System (ITS), active study on traffic prediction is currently underway. Multi-Layer Perceptron (MLP) model have been widely utilized for predicting traffic information since it is appropriate to represent the non-linear characteristics inherent in traffic prediction. However, the MLP model doesn't reflect local dependencies of traffic data and is prone to noise in traffic data. Convolutional Neural Networks (CNN) based model, on the other hand, can capture the local dependencies of traffic data and is less prone to disturbance in data. In this paper, we use temporal data and speed data collected on main roads in Seoul, South Korea to construct traffic prediction models. The speed data which are collected by every 5 minutes are provided by Ministry of Land, Infrastructure and Transport in South Korea. We construct the CNN based model and two MLP models which predict traffic speed and compare performance of the prediction models in this paper. The comparison results show that the CNN based model's prediction performance is higher than the prediction performance of the other two MLP models.

Because of shortage of raw materials, the energy-saving issues become popular in Vehicular Ad-hoc Networks (VANETs). A VANET is formed by traveling vehicles with communicating capability and thus it brings various applications. Navigation system is one of important applications in VANETs. The traditional navigation system usually plans a shortest path for users according to geographic maps but the planned path may become a slower one due to the traffic congestion. A congestion path not only delays the traveling time but wastes fuel. So the real-time traffic information should be considered while constructing a navigation path. In this paper, we propose a navigation system to find a fuel-saving navigation path by considering real-time traffic information. In order to estimate the traffic load on the future path, we maintain a table of historical traffic information by Intelligent Transport Systems (ITS). Furthermore, our system can dynamically change the navigation path by updating traffic information. In the simulation results, our approach can save more than 20% fuel consumption when vehicles travel in a traffic congestion environment.

This paper proposes a unified spatio-temporal model on the basis of STARIMA (Space-Time AutoRegressive Integrated Moving Average) for short-term road traffic prediction. The contributions of this paper are as follows. First, we develop a physically intuitive approach to traffic prediction that captures the time-varying spatio-temporal correlation between traffic at different measurement points. The spatio-temporal correlation is affected by the road network topology, time-varying speed, and time-varying trip distribution. Distinctly different from previous black-box approaches to road traffic modeling and prediction, parameters of the proposed approach have physically intuitive meanings which make them readily amenable to suit changing road and traffic conditions. Second, unlike some existing techniques which capture the variation of spatio-temporal correlation by a complete redesign and calibration of the model, the proposed approach uses a unified model which incorporates the physical factors potentially affecting the variation of spatio-temporal correlation into a series of parameters. These parameters are relatively easy to control and adjust when road and traffic conditions change, thereby greatly reducing the computational complexity. Experiments using two set of real traffic traces demonstrate that the proposed approach has superior accuracy compared with the widely used ARIMA (AutoRegressive Integrated Moving Average) and is only marginally inferior to that obtained by constructing multiple STARIMA models for different time of the day, however with a much reduced computational and implementation complexity.

Effect of real-time traffic information on the relaxation process is studied. The provision of traffic information on a real-time basis to individual vehicles is expected as an effective way to distribute the traffic flow to a network of roads. However, it has been also pointed out that real-time information prevents the distribution of the traffic flow on a network of roads because of the concentration of the traffic flow to the recommended route. We can see that the real-time traffic information based on averaged velocity or travel time give rise to fluctuation with chaotic structure on the relaxation process of traffic flow formed by cellular automata. It is also proposed that the real-time traffic information composed by differentiation of averaged velocity of vehicle is effective to reduce the fluctuation of traffic flow.

Adaptive Decision Spatio-temporal neural ODE for traffic flow forecasting with Multi-Kernel Temporal Dynamic Dilation Convolution

As the existing traffic information system based on the fixed traffic information center needs high cost for construct, and the other traffic information system based on inter vehicle communication pays more attention to safe driving information but pays less attention to traffic congestion information, especially for the traffic congestion information in the road network. This paper studies and puts forward a pattern for automatic traffic information service which realizes the function of automatic collection and dynamic transmission of traffic information in the road network on the basis of VANET, and with no need for fixed traffic information center. According to this pattern, we build the models for automatic collecting, integrating and processing, transmitting of traffic information, and automatic generating and dynamic updating of traffic information which covers the whole road network. Finally, in this paper, we also carry out traffic simulation for these models, analyze and evaluate rationality of these models and transmission effect.

An IoV enables secure transmission and storage of real-time traffic accident information among IoV devices like Vehicles, Road Side Unit (RSU), and Edge Servers to reduce unintended loss. Regardless of the advantages, accurate traffic accident prediction and decentralized data storage remain challenges for a reliable IoV system. We propose IntelligentChain, a blockchain and machine learning-based edge server for IoV that enables low-latency IoV device registration and secure traffic accident information communication between them. IntelligentChain then implements novel decentralized trust, consensus procedures for blockchain-based traffic accident information storage and predicts localized area traffic accidents using LSTM recurrent neural networks. IntelligentChain has been implemented on the Raspberry Pi 4. The experiment results illustrate that IntelligentChain achieves 90% prediction accuracy on the Kaggle dataset while decreasing transaction approval delay.

Arrival time dependent shortest path finding is an important function in the field of traffic information systems or telematics. However large number of mobile objects on the road network results in a scalability problem for frequently updating and handling their real-time location. In this paper, we propose a query processing method in MANET(Mobile Ad-hoc Network) environment to find an arrival time dependent shortest path with a consideration of both traffic and location in real time. Since our method does not need a centralized server, time dependent shortest path query is processed by in-network way. In order to reduce the number of messages to forward and nodes to relay, we introduce an on-road routing, where messages are forwarded to neighboring nodes on the same or adjacent road segments. This routing method allows to collect traffic information in real time and to reduce the number of routing messages. Experiments show that the number of forwarded messages is reduced in an order of magnitude with our on-road routing method compared to LAR-like method. At best, our method reduces about 57 times less messages.

Information and communication technologies have improved the quality of Intelligent Transportation Systems (ITS). The real-time traffic information has traditionally been collected via stationary vehicle detectors or GPS-based probe cars. Compared to the traditional ways, estimating traffic information from Cellular Floating Vehicle Data (CFVD) is more cost-effective, and easier to acquire. In this paper, this study proposes a novel approach to evaluate the relation of call arrival, handover, traffic flow, and traffic density. Moreover, the traffic speed is estimated by the proposed approach according to CFVD. Through the analytical analysis, this study analyzes the effects of traffic information (e.g. traffic flow and vehicle speed) and communication behaviors (e.g. call arrival rate and call holding time) on handovers and call arrivals. In the simulation, this study compares the estimated traffic information with the real traffic information. The experiment results show that the accuracy of traffic speed estimation is 89.75%. Therefore, the proposed approach can be used to estimate traffic speed from CFVD for ITS.