VEHITS 2015 Abstracts


Area 1 - Connected Vehicles

Full Papers
Paper Nr: 5
Title:

VNetIntSim - An Integrated Simulation Platform to Model Transportation and Communication Networks

Authors:

Ahmed Elbery, Hesham Rakha, Mustafa Y. ElNainay and Mohammad A. Hoque

Abstract: The paper introduces a Vehicular Network Integrated Simulator (VNetIntSim) that integrates transportation modelling with Vehicular Ad Hoc Network (VANET) modelling. Specifically, VNetIntSim integrates the OPNET software, a communication network simulator, and the INTEGRATION software, a microscopic traffic simulation software. The INTEGRATION software simulates the movement of travellers and vehicles, while the OPNET software models the data exchange through the communication system. Information is exchanged between the two simulators as needed. As a proof of concept, the VNetIntSim is used to quantify the impact of mobility parameters (traffic stream speed and density) on the communication system performance, and more specifically on the data routing (packet drops and route discovery time).

Paper Nr: 11
Title:

Privacy Issues and Pitfalls in VANET Standards

Authors:

Sebastian Bittl and Arturo A. Gonzalez

Abstract: Wireless vehicular networks are about to enter the deployment stage in the next years with important progress being made in Europe and the USA. Thereby, one of the core concerns is privacy of vehicles and their drivers, especially in Europe. Prior work has regarded only a small sub-set of the information exposed by current standards to an attacker for vehicle tracking. Thus, we take a close look on the data contained on different protocol layers of an ETSI ITS system. We find that much data is very distinctive and can be used to identify static vehicle parameters such as manufacturer or even model. This greatly reduces the usability of formerly proposed cooperative pseudonym switching strategies. Many more constraints have to be applied for selecting cooperation partners significantly reducing their availability. Therefore, current techniques cannot provide the level of privacy defined in VANET standards. Suggestions for improving the security entity and facility layer of ETSI ITS are given to limit the impact of the found issues.

Paper Nr: 13
Title:

A Method for Managing Transportation Requests and Subdivision Costs in Shared Mobility Systems

Authors:

Gianmichele Siano, Mariano Gallo and Luigi Glielmo

Abstract: This paper presents an algorithm for managing the demand and supply in a shared transportation system. In particular we present a method, independent from the Geographic Information System (GIS), which processes drivers and passengers requests and ranks them in order to encourage matching and to propose the solution profitable for all. The basic idea is to give priority to the requests of passengers with more common route and avoid those with greater excess path. In the end, we propose a solution for the distribution of costs among the participants of shared travel based on the application of the Shapley value.

Area 2 - Intelligent Transport Systems and Infrastructure

Short Papers
Paper Nr: 7
Title:

Algorithms for the Hybrid Fleet Vehicle Routing Problem

Authors:

Fei Peng, Amy M. Cohn, Oleg Gusikhin and David Perner

Abstract: In the classical Vehicle Routing Problem (VRP) literature, as well as in most VRP commercial software packages, it is commonly assumed that all vehicles are identical in their characteristics. In real-world problems however, this is often not true. In many cases, fleets are made up of different vehicle types, which may vary by size, engine/fuel type, and other performance-impacting factors. Even in a homogeneous fleet, vehicles often differ by age and condition, which can greatly impact performance. Our research was specifically motivated by cases where the fleet contains vehicles that not only vary in performance, but this variation is a function of the arc type, such that a given vehicle might have lower cost on some arcs but higher cost on others. We refer to this as the Hybrid Fleet Vehicle Routing Problem (HFVRP). We propose two heuristic methods that take into account the vehicle-specific cost structures. We provide computational results to demonstrate the quality of our solutions, as well as a comparison with a Genetic Algorithm (GA) based method seen in the literature.

Area 3 - Intelligent Vehicle Technologies

Short Papers
Paper Nr: 8
Title:

Design and Implementation of Transportation Management System

Authors:

I. Ashour, M. Zorkany and M. Shiple

Abstract: This paper proposes an effective method of transportation management system. This proposed system is designed to interconnect public transport vehicles and bus stations to “Central Room” to monitor the vehicles & traffic status. Based on the collected data and via analyzing road condition, estimated arrival times are computed and transmitted to all relevant stations. The main structure of proposed system consists of Bus unit, station unit and main control centre with servers. Bus and station unit can be hardware unit or mobile android unit. Monitoring Busses based on GPS and GPRS applications. The data transferred between Bus units, station units and the main servers are managed via GPRS/UMTS link. At the server (Central Room) and based on the collected data from buses and via analyzing road condition, accurate arrival times will be computed (Via Neural network (NN) / Kalman Filter (KF)) and transmitted to all relevant stations. In this paper, we proposed a modified technique to predict bus arrival time depending on the two algorithms (NN & KF) simultaneously to take advantage of historical data (NN) with current data (KF). Achieving these main features will cause major improvements in public transport convenience and safety. Field tests were performed under real traffic situations in order to test the system.

Area 4 - Electric Vehicles

Short Papers
Paper Nr: 10
Title:

A Renewable Source Aware Model for the Charging of Plug-in Electrical Vehicles

Authors:

Jânio Monteiro and Mário S. Nunes

Abstract: The number of Electric Vehicles is estimated to continuously rise over the next years. While this trend is expected to lead to a reduction in CO2 emission, existing electrical grids have not been planned to support a large number of electrical vehicle’s batteries charging simultaneously. The integration of distributed production using renewable energy sources is seen as a solution to meet the requirements of battery charging. Renewable sources are however affected by variation and lack of predictability, due to the environmental factors they depend on, which are the cause of inefficiencies and mismatches in the required demand-response equilibrium. In these conditions, the model and the associated scheduling algorithms to use in medium to large charging parks play an important role, due to the implications it has in their operational costs and in the maximization of the return of investments made in renewable sources. In this paper we propose and evaluate a charging model that engages users to participate in demand response measures, by giving them the ability of selecting two energy components for the charging of their electrical vehicles, one of which varies according with the variable nature of renewable sources. Based in this model we propose one scheduling algorithm and compare it with several other solutions, demonstrating that the proposed solution is able of achieving a significant cost reduction with significant low computational complexity and processing times, while achieving a high ratio of renewable energy usage.

Area 5 - Intelligent Vehicle Technologies

Short Papers
Paper Nr: 14
Title:

Automatic Obstacle Classification using Laser and Camera Fusion

Authors:

Aurelio Ponz, C. H. Rodríguez-Garavito, Fernando García, Philip Lenz, Christoph Stiller and J. M. Armingol

Abstract: State of the art Driving Assistance Systems and Autonomous Driving applications are employing sensor fusion in order to achieve trustable obstacle detection and classification under any meteorological and illumination condition. Fusion between laser and camera is widely used in ADAS applications in order to overcome the difficulties and limitations inherent to each of the sensors. In the system presented, some novel techniques for automatic and unattended data alignment are used and laser point clouds are exploited using Artificial Intelligence techniques to improve the reliability of the obstacle classification. New approaches to the problem of clustering sparse point clouds have been adopted, maximizing the information obtained from low resolution lasers. After improving cluster detection, AI techniques have been used to classify the obstacle not only with vision, but also with laser information. The fusion of the information acquired from both sensors, adding the classification capabilities of the laser, improves the reliability of the system.

Area 6 - Intelligent Transport Systems and Infrastructure

Short Papers
Paper Nr: 16
Title:

An Intelligent Framework and Prototype for Autonomous Maintenance Planning in the Rail Industry

Authors:

C. J. Turner, A. Tiwari, A. Starr, I. Durazo-Cardenas and K. Blacktop

Abstract: This paper details the development of the AUTONOM project, a project that aims to provide an enterprise system tailored to the planning needs of the rail industry. AUTONOM extends research in novel sensing, scheduling, and decision-making strategies customised for the automated planning of maintenance activities within the rail industry. This paper sets out a framework and software prototype and details the current progress of the project. In the continuation of the AUTONOM project it is anticipated that the combination of techniques brought together in this work will be capable of addressing a wider range of problem types, offered by Network rail and organisations in different industries.

Area 7 - Intelligent Vehicle Technologies

Short Papers
Paper Nr: 17
Title:

Localizing Changes in Driver Behavior via Frequency-pattern-analysis

Authors:

M. Schneider, M. Sieber and B. Färber

Abstract: Explorative analysis of driver behavior, a key variable in the context of automotive research and development, can be tedious. The authors present a quick and easy method to identify changes in recorded driver behavior data. The method consists of a data processing algorithm that uses Fourier-series and statistical t-tests to identify points in time where changes in the frequency of the recorded signal occur. An exemplary use-case for the method is presented for driver steering torque data obtained in an experiment with an automatic obstacle-avoidance maneuver. The results allow for the assumption that changes in frequency of driver steering torque may mark meaningful, implicit changes in driver behavior even when driver behavior does not explicitly change, thereby making obvious the potential of the proposed analysis method.

Paper Nr: 18
Title:

Safety First? - V2X – Percived Benefits, Barriers and Trade-offs of Automated Driving

Authors:

Teresa Schmidt, Ralf Philipsen and Martina Ziefle

Abstract: Today, we are on the edge of increasing population and urbanization with an increasing portion of older people. These far-reaching societal developments necessitate novel mobility infrastructure concepts, in which a diverse population and a higher population density are considered. Safety in traffic situations is one of the most important and needs to be taken into account. A highly potent approach is to combine in-vehicle systems and vehicle sensors. Whereby the public perception and user acceptance of V2X-technology in general is insufficiently explored. Using a two-tier approach, in which both qualitative and quantitative data are combined, this research gains insights into human perceptions of V2X-technology, plausible trade-offs and basic fears. Results show safety as an important factor which should be included in further future research.

Area 8 - Intelligent Transport Systems and Infrastructure

Short Papers
Paper Nr: 19
Title:

The University as EV Ecosystem Hub - Education and Outreach to Accelerate EV Adoption

Authors:

Dunbar P. Birnie III

Abstract: The author’s university location is being developed as an alternative-fuelled-vehicle “ecosystem” that is serving both educational and research missions. In addition, it is assisting with gradual transition from gasoline-powered private vehicles to PHEV and EV thereby providing real positive regional environmental impacts. By highlighting the early phases of this transformation locally and including students in the discussion we hope to assist in accelerating this transformation for the future. This paper surveys our present status and provides data on the usage patterns as well as on the costs and practical difficulties encountered when considering hardware installation and making site selection.

Paper Nr: 20
Title:

Shaping the Current Waveform of an Active Filter for Optimized System Level Harmonic Conditioning

Authors:

Espen Skjong, Marta Molinas, Tor Arne Johansen and Rune Volden

Abstract: Harmonic voltages and currents in electrical systems, when present to a certain degree, represent not only a power quality problem but they are also strongly associated with the electrical system overall losses and they are arguably a source of instability and a safety concern. Mitigating harmonics distortion across the entire system by actively reducing harmonic currents propagation is an effective way of coping with these issues and can be dealt with the injection of a compensating current waveform with an active filter installed at a given bus. This paper shows how, by shaping the compensating current waveform in an optimal way, the overall electrical system harmonic distortion can be optimally reduced in a cost effective manner with a minimum size of the compensating device. The process of shaping this optimal compensating current is shown by how its components are defined by the optimization algorithm using the phase and amplitude of each harmonic as degrees of freedom in the process of finding the optimal waveform. A marine vessel distribution grid is used as representative example to prove the concept.

Paper Nr: 21
Title:

Aggregating and Managing Big Realtime Data in the Cloud - Application to Intelligent Transport for Smart Cities

Authors:

Gavin Kemp, Genoveva Vargas-Solar, Catarina Ferreira Da Silva, Parisa Ghodous and Christine Collet

Abstract: The increasing power of computer hardware and the sophistication of computer software have brought many new possibilities to information world. On one side the possibility to analyse massive data sets has brought new insight, knowledge and information. On the other, it has enabled to massively distribute computing and has opened to a new programming paradigm called Service Oriented Computing particularly well adapted to cloud computing. Applying these new technologies to the transport industry can bring new understanding to town transport infrastructures. The objective of our work is to manage and aggregate cloud services for managing big data and assist decision making for transport systems. Thus this paper presents our approach for developing data storage, data cleaning and data integration services to make an efficient decision support system. Our services will implement algorithms and strategies that consume storage and computing resources of the cloud. For this reason, appropriate consumption models will guide their use. Proposing big data management strategies for data produced by transport infrastructures, whilst maintaining cost effective systems deployed on the cloud, is a promising approach.

Paper Nr: 23
Title:

Socializing Public Transportation - Using Situational Context in Public Transportation to Get in Touch with People Around You

Authors:

Roman Roor, Olga Birth, Michael Karg and Markus Strassberger

Abstract: Unexpected delays or long traveling times lead often to people who get bored while using public transport. Whereas some might use their travel time for work or even enjoy the silence, there are still many people that would welcome an opportunity to use the spatio-temporal proximity to get to know others or meet with friends that are in the same train. This paper introduces a novel, smartphone based concept to bring people together while using or waiting for public transport. Based on location and personal preferences, suggestions can be made for getting in touch with nearby persons. We propose a recommendation system, which identifies the concrete public transport vehicle and compares the preferences with other users to create recommendations about people nearby who are also traveling in or waiting for a public transport vehicle.

Area 9 - Electric Vehicles

Short Papers
Paper Nr: 24
Title:

Advanced Driver Aid System for Energy Efficient Electric Bus Operation

Authors:

Teemu Halmeaho, Marko Antila, Jari Kataja, Paula Silvonen and Mikko Pihlatie

Abstract: Electric bus energy consumption is mainly due to the vehicle traction. Additionally, auxiliary systems such as cabin heating-cooling, air compressor, and power steering consume energy. One way to optimize the consumption is a Driver’s Aid System (DAS). Based on the route information, DAS provides the driver the optimal driving suggestions, and simultaneously may optimise the energy use of auxiliary systems. These approaches are discussed in the paper. When the optimal air compressor operation was introduced, vehicle energy consumption was decreased 1.6 %. In addition to guiding the auxiliary devices and the driver, prospects of using DAS as a communication hub for managing buses, their charging and to share information for a bus operator are discussed.

Area 10 - Intelligent Vehicle Technologies

Posters
Paper Nr: 1
Title:

The Effect of Acceleration and Deceleration Information of Preceding Vehicle Group on Fuel Economy of the Following Vehicle

Authors:

Shuichi Matsumoto

Abstract: Eco-driving and other environmentally-friendly behaviors have been gaining widespread acceptance. In order to encourage eco-driving efficiently, this study looked at the effect of preceding and pre-preceding vehicle’s acceleration-deceleration information on the following vehicle's gasoline mileage. As a result, the following was found: 1. By providing information to a following vehicle, the fuel consumption rate of the following vehicle can be reduced. 2. Subjects that improved their gasoline mileage tended to value pre-preceding vehicle information more than those that worsened it. 3. With the provision of information on the pre-preceding vehicle, the following vehicles started moving earlier. 4. The pre-preceding vehicle's acceleration information caused the following vehicle to increase accelerate gradually when starting to move. Therefore, it was suggested that sharing the information on preceding and pre-preceding vehicles was effective.

Area 11 - Intelligent Transport Systems and Infrastructure

Posters
Paper Nr: 9
Title:

Multilevel Modelling of Urban Transport Infrastructure

Authors:

Oleg Saprykin and Olga Saprykina

Abstract: This article covers the transportation processes modeling in the Intelligent Transportation System environment. The combined microscopic and mesoscopic simulation is included. This article is dedicated to solving the problem of data preservation during the transition from a microscopic to a mesoscopic model. The solution suggests modifying the multi-agent transportation system, and using artificial neural networks, considering implementation of the unique architecture of an intelligent agent which collects additional information to be forwarded to the next simulation level. The article describes the microsimulation process implementation in the multi-agent system MATSim. A Ward neural network (trained using the data transferred from the microscopic level) is used for the processing for the mesoscopic level.

Area 12 - Intelligent Vehicle Technologies

Posters
Paper Nr: 15
Title:

Fault Detection Architecture for Proprioceptive Sensors based on a Multi Model Approach and Fuzzy Logic Decisions

Authors:

Nicolas Pous, Dominique Gruyer and Denis Gingras

Abstract: In this paper a new fault detection architecture will be presented. Inspired by multi-model data fusion algorithms and fuzzy logic decisions, it consists in the comparison between the estimation of a dynamic mode using each sensor independently. This method is used to deal with important non-linearity and strong interaction with the environment usually encountered in the domain of the intelligent vehicles localization. The concept of analytic redundancy is also used to ignore model uncertainties.

Area 13 - Connected Vehicles

Posters
Paper Nr: 22
Title:

Methods of Interaction Between Multiprotocol Unit and Telematics Map Cloud Service

Authors:

Chuvatov Mikhail, Glazunov Vadim, Kurochkin Michail and Popov Serge

Abstract: Continuous access to the service from the moving vehicle improves safety and provides ecological compatibility of transport infrastructure functioning in big agglomerations conditions. The lack of the guaranteed signal level of global and local networks requires new approaches to form the strategy for connection continuity provision. We offer the technology enabling to form the transport facility network by appealing multiprotocol unit and telematics map. The method suggest using external data concerning global and local wireless networks in each vehicle. The approach involves collecting networks data by means of the multiprotocol units, transmitting of these data into cloud service of telematics map, the data generalisation and to meat the query about available networks in the vehicles vicinity. The completely automatic technology of data control is designed in such a way that it can provides external data for multiprotocol routing in integrated vehicle networks. In order to check the suggested approach we performed experiments and dives the information system prototype that demonstrated its efficiency. Technical feasibility of the information system was confirmed during the experiments.

Area 14 - Intelligent Transport Systems and Infrastructure

Posters
Paper Nr: 25
Title:

Traffic Flow Prediction from Loop Counter Sensor Data using Machine Learning Methods

Authors:

Blaž Kažic, Dunja Mladenić and Aljaž Košmerlj

Abstract: Due to increasing demand and growing cities, traffic prediction has been a topic of interest for many researchers for the past few decades. The availability of large amounts of traffic-related data and the emerging field of machine learning algorithms has led to a significant leap towards data-driven methods. In this paper, loop counter data are used to develop models that can predict traffic flow for several different prediction intervals into the future. In depth exploratory data analysis and statistical testing is performed to obtain good quality informative features. Several feature sets were compared by using different machine learning methods: Ridge Regression, SVR and Random Forests. The results show that in order to obtain good prediction results thorough feature extraction is just as or even more important than learning method selection. Good features enables us to use less complex methods, which run faster, are more reliable and easier to maintain. In conclusion, we address ideas regarding how predictions could be improved even further.