Loop detectors distributed on freeways are very vulnerable and could be damaged or malfunctioned due to improper sealing, pavement deterioration. This may lead to poor travel time estimation as most of existing methodologies require detailed data collected from numerous detectors along a specified freeway route. To address this problem, this study proposes an effective and reliable methodology for real-time freeway travel time estimation with data from sparse detectors. In contrast to the existing methods, the proposed methodology requires significantly less number of detectors but maintains fairy good performance on travel time estimation. The proposed methodology utilises a self-organised mapping algorithm to cluster the detectors with similar traffic patterns. The data collected from the representative detectors within each cluster is then employed to estimate the travel time based on a support vector regression model. The case studies conducted for three selected freeway routes in Northern California over 3 weeks demonstrate that the proposed methodology accurately captures the fluctuation of travel time induced by the variations of traffic states. The estimated results are exceptionally accurate with smaller mean errors and root-mean-squared errors compared with the benchmark values obtained from the well-known performance measurement system in California.

Novel ubiquitous traffic sensors such as floating car data (FCD) are getting extended due to the use of 24 h connected smartphones and global positioning systems. Road conditions such as travel speeds in each road link and mobility demand can be monitored by measurements coming from moving vehicles consisting of geolocation and speed information with timestamps. Map-matching is the process needed to identify the corresponding road link on a digital map and define the position of the geolocated vehicle on this link, overcoming positioning errors. Matching processes in urban environments are more prone to error due to the topology and features of city road networks. In this study, the accuracy of the map-matching is discussed depending on the road configuration for FCD in urban and interurban scenarios, under sampling frequencies ranging from 5 to 60 s. Concretely, in this analysis, three matching techniques have been evaluated against road density, nominative speed limit, edge length and edge count values in order to quantify the impact of these variables on the matching accuracy.

A Controller Interface Device (CID) is a hardware that connects a signal controller to simulation software. Running traffic simulation with an actual controller is called hardware-in-the-loop simulation (HILS) and is an important tool for simulating the operations of traffic signals. In practice, it has been found that the existing traditional CIDs have some limitations, particularly when simulating multiple signals. In this regard, a Virtual CID (VCID) is proposed in this study. VCID can connect traffic signal controllers to a micro-simulation software through the National Transportation Communications for ITS Protocol (NTCIP). A case study is conducted that uses the same traffic networks to compare the performance of VCID to CID. Queue length, travel time and simulation stability are considered in the evaluation. Results show that VCID could provide accurate and reliable simulation results. In comparison to traditional CIDs, VCID has more advantages as it does not depend on any hardware device and is easy to operate. In addition, since the communications are transformed through Ethernet, thus providing a convenient learning and testing environment for those who do not own a physical signal control laboratory. Therefore, VCID will have a broader prospect for development than CID.

High-speed electric multiple unit (HSEMU) is a complex non-linear system which runs under three typical operating modes (OMs) including traction, braking, and coasting. With the increasing traffic density of the high-speed railway, the conventionally manual control strategies may be inapplicable for the HSEMU to maintain a good running performance. To enhance the running performances, in this study, a novel multiple OM (MOM) running model is designed to accurately describe the non-linear relationship between running speed and controlling force. By utilising the advantages of adaptive neuro-fuzzy inference system (ANFIS) on non-linear modelling, this study proposes an MOM-ANFIS model of HSEMU. On the basis of the established MOM-ANFIS model, a new speed controller incorporated with OM selection mechanism is designed to achieve speed control of HSEMU followed by a stability analysis of the closed-loop system. Comparative experimental results using practical running data show that the proposed MOM-ANFIS model displays better modelling accuracy and the corresponding control strategy achieves improved speed control performances for the HSEMU.

Short-term traffic forecasting is becoming more important in intelligent transportation systems. The k-nearest neighbour (kNN) method is widely used for short-term traffic forecasting. However, the self-adjustment of kNN parameters has been a problem due to dynamic traffic characteristics. This study proposes a fully automatic dynamic procedure kNN (DP-kNN) that makes the kNN parameters self-adjustable and robust without predefined models or training for the parameters. A real-world dataset with more than one year traffic records is used to conduct experiments. The results show that the DP-kNN can perform better than the manually adjusted kNN and other benchmarking methods in terms of accuracy on average. This study also discusses the difference between holiday and workday traffic prediction as well as the usage of neighbour distance measurement.

The car-following model is an important micro-traffic model for simulating car-following behaviour in traffic engineering and research studies. Conventional car-following models are always presented using mathematical equations reflecting ideal traffic conditions. In the big data era, data-driven models become a popular trend. In this study, a data-driven car-following model based on the rough set theory is proposed to consider information hidden in a field data set. On the basis of field data obtained from measurement devices such as the next generation simulation (NGSIM) trajectory data set, and using the methods of the rough set theory, an optimal decision rule set is established. Redundant attributes and redundant attribute values are removed for simplifying the car-following behaviour decision problem. Attribute significance and weights are computed for selecting matching rules. A car-following behaviour decision algorithm is designed to choose appropriate rules to determine the follower's velocity according to current observations. Simulations illustrate that the proposed data-driven car-following model can simulate the micro-traffic behaviour of followers well.

This study introduces an agent-based dynamic feedback-control toll pricing strategy that accounts for the trip purpose, travel time reliability, departure time choice and level of income such that the toll revenue is maximised while maintaining a minimum desired level of service on the managed lanes. An agent-based modelling was applied to simulate drivers’ learning process based on their previous commuting experience. The study also analysed how drivers’ heterogeneity in value of time, and value of reliability for each trip purpose will influence route decisions and thus affect the optimal toll rates. Comparative evaluation between the newly developed strategy, the strategy currently deployed on Interstate 95 express lanes, and another strategy previously developed by the authors shows that the agent-based strategy produced a steadier increase in toll rate during the peak hours and a significantly higher toll revenue at speeds higher than 45 mph.

The study presents the methodology and system architecture of an integrated urban road network travel time prediction framework based on low-frequency probe vehicle data. Intended applications include real-time network traffic management, vehicle routing and information provision. The framework integrates methods for receiving a stream of probe vehicle data, map matching and path inference, link travel time estimation, calibration of prediction model parameters and network travel time prediction in real time. The system design satisfies three crucial aspects: computational efficiency of prediction, internal consistency between components and robustness against noisy and missing data. Prediction is based on a multivariate hybrid method of probabilistic principal component analysis, which captures global correlation patterns between links and time intervals, and local smoothing, which considers local correlations among neighbouring links. Computational experiments for the road network of Stockholm, Sweden and probe data from taxis show that the system provides high accuracy for both peak and off-peak traffic conditions. The computational efficiency of the framework makes it capable of real-time prediction for large-scale networks. For links with large speed variations between days, prediction significantly outperforms the historical mean. Furthermore, prediction is reliable also for links with high proportions of missing data.

Real-time vehicle counting can efficiently improve traffic control and management. Aiming to efficiently collect the real-time traffic information, the authors propose an effective vehicle counting system for detecting and tracking vehicles in complex traffic scenes. The proposed algorithm detects moving vehicles based on background subtraction method with ‘low-rank + sparse’ decomposition. For accurately counting vehicles, an online Kalman filter algorithm is used to track the multiple moving objects and avoid counting one vehicle repeatedly. The proposed method is evaluated on three publicly available datasets, which include seven video sequences with various challenging scenes for detection performance evaluation, and another two video sequences for vehicle counting evaluation. The experimental results demonstrate a good performance of the proposed method in terms of both qualitative and quantitative evaluations.