Empty vehicles management may improve the average waiting time for vehicle delivery in the Personal Rapid Transit (PRT) network. In this study, original heuristic algorithm of empty vehicle management is presented. The algorithm is tested in several benchmark PRT structures under Feniks simulation environment. The results show that significant improvements of average waiting time may be achieved just because of the multi-parameter analysis of the present network state alone rather than by the predictive use of forecasted demand. The algorithm does not use any central database of demand and location of free vehicles. Instead, it assumes the local exchange of data between stations on their state and expected vehicles. Therefore it seems well tailored to a distributed implementation.

Data collection, and especially data annotation, are surprisingly time consuming and costly tasks for vehicle classification. Annotation is used to label examples of vehicles, manually outlining their shapes and assigning their correct classification, for use in classifier training and performance evaluation. This study presents a semi-automatic approach for the annotation of the vehicle samples recorded from roadside CCTV video cameras. Vehicles are detected by using automatic image analysis and classified into four main categories: car, van, bus and motorcycle/bicycle by using a vehicle observation vector constructed from the size, the shape and the appearance features. Unsupervised K-means clustering is used to automatically compute an initial class label for each detected vehicle. Then, in an iterative process, the output scores of a linear support vector machines classifier are used to identify the low confidence samples, for which the annotations are considered for manual correction. Experimental results are presented for both synthetic and real datasets to demonstrate the effectiveness and the efficiency of the authors approach, which significantly reduces the time required to generate an annotated dataset. The method is general enough that it can be used in other classification problems and domains that use a manually-created ground-truth.

Conventional methods of intergreen time design are based on deterministic traffic flow theory and are difficult to fully reflect traffic randomness especially driver's decision differences. This study is to develop a probabilistic model to extensively investigate the effects of traffic flow randomness and driver's decision errors on the determination of intergreen times. An analytical framework was proposed based on a safety reliability model earlier developed by the authors and validated based on field data at a typical high-speed intersection in Shanghai. Then, comprehensive sensitivity analysis was performed to look into the fluctuation of safety reliability towards critical variables as well as their variations and correlations. Two logistic regression models were developed to quantify the effects of each random component on the probabilities of risky behaviour and clearance failure, that is, safety reliability indicators for intergreen intervals. To facilitate easy applications of the proposed approach, a set of scenario analysis were done to calculate the required intergreen intervals under different traffic conditions and safety reliability levels. Application tables were finally provided for signal design, which are able to recommend more sophisticated intergreen times and thus aid the practitioners to select proper intergreen times according to local conditions without complicated calculation.

This study presents a data segmentation method, which was intended to improve the performance of the k-nearest neighbours algorithm for making short-term traffic volume predictions. According to the introduced method, selected segments of vehicle detector data are searched for records similar to the current traffic conditions, instead of the entire database. The data segments are determined on the basis of a segmentation procedure, which aims to select input data that are useful for the prediction algorithm. Advantages of the proposed method were demonstrated in experiments on real-world traffic data. Experimental results show that the proposed method not only improves the accuracy of the traffic volume prediction, but also significantly reduces its computational cost.

In this study a framework for the real-time trading of budgeted emission rights between a fleet of participating vehicles is presented. The trading problem is formulated as a utility maximisation or as a utility fairness problem, which can be solved in real time either in a centralised or in a distributed manner. In both cases, the authors illustrate the basic issues that arise when such a framework is realised in practice, and they show the efficacy of the approaches by providing several simulation examples and a realistic case study.

The automatic German toll system is both a software-intensive system – relying on the on-board units to collect the toll charges due – and a socio-technical system – the user can influence the system behaviour. The overall dynamic system behaviour emerges both from the technical implementation of the system and the users’ interaction with the system limiting the effectiveness of common software engineering practices. Simulations are an additional tool to reduce the operational risks and to improve the performance of system development projects. We have developed a realistic microscopic discrete event simulation model of the automatic tolling processes to simulate the overall system behaviour at a scale of 1:1. Taking this model our research investigates the effect of the HGV fleet behaviour on the dynamic system behaviour. To that end the system simulation model is accompanied by a separate model of the HGV driving patterns.

In this study, the authors introduce a highly dynamic and manageable average speed monitoring system. They define a privacy preserving security protocol in which vehicles hold a pseudonymous non-traceable private digital identity. Vehicle's identity is kept undisclosed for as long as the average speed over a designated road section does not exceed the specified limit. During the protocol execution, non-refutable evidences are built for vehicles and authorities, for future disputes. Finally, the security properties of the protocol are formally proven using BAN logic and, a metric is defined to measure the reliability and feasibility of the author's proposal.

In addition to the fuel-efficient design of vehicles, eco-driving technologies can further reduce the fuel consumption and carbon emissions of road transportation. One of the major issues in eco-driving technologies is how to determine the fuel savings or fuel-optimised operating strategies of the power train. We examine the periodic operation of autonomous vehicles in free-driving scenarios with the purpose of maximising fuel economy. The design of such strategies can be considered to be an optimal control problem, which is proved by π-test to have singular arcs because of the S-shaped engine fueling rate. The optimal operations are solved numerically by using the Legendre pseudospectral method, and many are found to demonstrate periodic behaviours, that is, pulse and gliding. In periodic operation, the engine switches between the minimum break specific fuel consumption point and the idling point, while the vehicle speed oscillates between its upper and lower bounds. The formation of periodic operation, as well as some key properties, is analysed and presented.

Every urban environment contains a rich set of dominant surfaces which can provide a solid foundation for visual odometry estimation. In this work visual odometry is robustly estimated by computing the motion of camera mounted on a vehicle. The proposed method first identifies a planar region and dynamically estimates the plane parameters. The candidate region and estimated plane parameters are then tracked in the subsequent images and an incremental update of the visual odometry is obtained. The proposed method is evaluated on a navigation dataset of stereo images taken by a car mounted camera that is driven in a large urban environment. The consistency and resilience of the method has also been evaluated on an indoor robot dataset. The results suggest that the proposed visual odometry estimation can robustly recover the motion by tracking a dominant planar surface in the Manhattan environment. In addition to motion estimation solution a set of strategies are discussed for mitigating the problematic factors arising from the unpredictable nature of the environment. The analyses of the results as well as dynamic environmental strategies indicate a strong potential of the method for being part of an autonomous or semi-autonomous system.

Numerous vehicles exist worldwide such as cars, motorcycles and bicycles. Although parking for such vehicles is available in many places, parking problems still always exist, such as full lots or a lack of lots. Commuters seeking a parking space expend time when spaces are occupied, and resources are wasted when parking spaces are empty. On the other hand, biking is a green vehicle in a fuel-shortage situation and also a good exercise for people. The Ubike system is a popular short-distance transit vehicle system in Taipei City that also has the parking problem. Therefore, this study uses two common regression schemes – linear regression and support vector regression (SVR) to predict the number of bicycles in Ubike stations to determine the number of available parking spaces. It also uses the proportional selection method to increase accuracy and reduce training time for SVR. Some evaluations are conducted to validate the feasibility of the two regression-based service availability prediction schemes for the Ubike system.

Road vehicle emissions have become a main aspect related to the quality of life and sustainability. The aim of this study is to develop a method to estimate pollutant emissions that takes into account the daily variations of traffic flow conditions and parameters relative to the different state of vehicles and that can easily be applied to evaluate the results of both on-line and off-line traffic management strategies in wide congested urban networks. The method has been tested in the real case of the city of Brindisi (Italy), demonstrating its reliability and efficiency.

Excess speeding on roads with a low-speed limit is a key error in drivers of all ages although the reasons for speeding can be significantly different. Drivers aged over 60 are generally more cautious, and take fewer risks than younger aged drivers. This study investigates whether Intelligent Speed Adaptation (ISA) technology can assist older drivers in maintaining vehicle speed. The technology can be employed in three ways: advisory (AISA), differential (DISA) and mandatory (MISA). Twenty-six drivers aged over 60 years old participated along with a comparison group of 16 experienced younger drivers aged under 60. All drivers completed four driving tasks in a driving simulator with and without ISA. Results show improvements in speed and lane-keeping performance vary depending on the type of ISA and driver age and training in effective use of ISA is needed for drivers of all ages. The study is one component of a wider research programme exploring how ITS could potentially help older people overcome some of the difficulties they experience with driving as they age and hopefully help them remain safe drivers for longer, a benefit both to the individual and society.