VEHITS 2018 Abstracts

Area 1 - Connected Vehicles
Full Papers Short Papers
Area 2 - Intelligent Transport Systems and Infrastructure
Full Papers Short Papers
Area 3 - Intelligent Vehicle Technologies
Full Papers Short Papers
Area 4 - Data Analytics
Full Papers Short Papers
Area 5 - Sustainable Transport
Short Papers
Full Papers
Paper Nr: 112
Title:

Integrated Provision of Heterogeneous Mobility Services
Authors:

Markus C. Beutel, Sevket Gökay, Eva-Maria Jakobs, Karl-Heinz Krempels, Fabian Ohler, Christian Samsel, Christoph Terwelp and Sara Vogelsang
Abstract: Advanced travel information systems (TIS) support intermodal traveling by combining heterogeneous modes of transportation (e.g., public transportation and electric vehicle sharing systems). Because of a large diversity of characteristics of the different transportation modes, the integration of data is highly complex. Facing the information complexity (such as different ways of utilization, diverse tariff systems, or accounting of an intermodal trip) mayor barriers occureven at different integration layers, e.g., the business model. In this paper, we present a method to develop advanced travel information systems, incorporating heterogeneous modes of transportation. As a key component of the proposed method specify an analysis phase which helps to digitize service processes and fundamentally determines system architecture, components, interfaces as well as the business model. Our Solution desrcibes and discusses the method, applied in the German research project Mobility Broker.

Area 1 - Connected Vehicles

Full Papers
Paper Nr: 33
Title:

Adaptive Decision Making based on Temporal Information Dynamics
Authors:

Tobias Meuser, Martin Wende, Patrick Lieser, Björn Richerzhagen and Ralf Steinmetz
Abstract: To increase road safety and efficiency, connected vehicles rely on the exchange of information. On each vehicle, a decision-making algorithm processes the received information and determines the actions that are to be taken. State-of-the-art decision approaches focus on static information and ignore the temporal dynamics of the environment, which is characterized by high change rates in a vehicular scenario. Hence, they keep outdated information longer than necessary and miss optimization potential. To address this problem, we propose a quality of information (QoI) weight based on a Hidden Markov Model for each information type, e.g., a road congestion state. Using this weight in the decision process allows us to combine detection accuracy of the sensor and the information lifetime in the decision-making, and, consequently, adapt to environmental changes significantly faster. We evaluate our approach for the scenario of traffic jam detection and avoidance, showing that it reduces the costs of false decisions by up to 25% compared to existing approaches.
Download
Paper Nr: 37
Title:

Cellular Bandwidth Prediction for Highly Automated Driving - Evaluation of Machine Learning Approaches based on Real-World Data
Authors:

Florian Jomrich, Alexander Herzberger, Tobias Meuser, Björn Richerzhagen, Ralf Steinmetz and Cornelius Wille
Abstract: To enable highly automated driving and the associated comfort services for the driver, vehicles require a reliable and constant cellular data connection. However, due to their mobility vehicles experience significant fluctuations in their connection quality in terms of bandwidth and availability. To maintain constantly high quality of service, these fluctuations need to be anticipated and predicted before they occur. To this end, different techniques such as connectivity maps and online throughput estimations exist. In this paper, we investigate the possibilities of a large-scale future deployment of such techniques by relying solely on lowcost hardware for network measurements. Therefore, we conducted a measurement campaign over three weeks in which more than 74,000 throughput estimates with correlated network quality parameters were obtained. Based on this data set—which we make publicly available to the community—we provide insights in the challenging task of network quality prediction for vehicular scenarios. More specifically, we analyze the potential of machine learning approaches for bandwidth prediction and assess their underlying assumptions.
Download
Paper Nr: 108
Title:

SNB-PPB: Social-network-based-privacy-preserving Broadcast for Vehicular Communications
Authors:

Yanheng Liu, Haifeng Zhu and Jian Wang
Abstract: Internet Key Exchange (IKE) can be used in vehicular communications in vehicular ad hoc networks (VANETs), but it requires a stable communication procedure, which is usually difficult to achieve because of the highly dynamic context of a VANET. In this paper, we propose a new method named social-network-based-privacy-preserving broadcast (SNB-PPB) for secure vehicular communications, which can omit the key exchange procedure. It transfers the information of online social networks to offline vehicular communications, dividing messages into different privacy levels and vehicles into different trust levels based on trust relations on social networks. The sender uses attributes on social networks and the message’s privacy level to encrypt messages; only those receivers who satisfy the corresponding trust level can decrypt the broadcast packet.To the best of our knowledge, this is the first time that social networks have been applied in VANETs to achieve secure vehicular communications and privacy preservation. Our simulation results show that our method enables the sender to decide the amount and percentage of vehicles that can decrypt the received broadcast.
Download
Short Papers
Paper Nr: 14
Title:

Open Problems when Mapping Automotive Security Levels to System Requirements
Authors:

Thomas Rosenstatter and Tomas Olovsson
Abstract: Securing the vehicle has become an important matter in the automotive industry. The communication of vehicles increases tremendously, they communicate with each other and to the infrastructure, they will be remotely diagnosed and provide the users with third-party applications. Given these areas of application, it is evident that a security standard for the automotive domain that considers security from the beginning of the development phase to the operational and maintenance phases is needed. Proposed security models in the automotive domain describe how to derive different security levels that indicate the demand on security, but do not further provide methods that map these levels to predefined system requirements nor security mechanisms. We continue at this point and describe open problems that need to be addressed in a prospective security framework for the automotive domain. Based on a study of several safety and security standards from other areas as well as suggested automotive security models, we propose an appropriate representation of security levels which is similar to, and will work in parallel with traditional safety, and a method to perform the mapping to a set of predefined system requirements, design rules and security mechanisms.
Download
Paper Nr: 44
Title:

Blind Estimation of OFDM Sampling Frequency Offset and Application to Power Line Communication in Aircrafts
Authors:

Navish Lallbeeharry, Rose Mazari, Virginie Degardin, Martine Lienard and Christophe Trebosc
Abstract: Orthogonal Frequency Division Multiplexing (OFDM) is a widely used technique but its practical implementation sometimes leads to issues related to the synchronization between the transmitter and the receiver. The sampling frequency offset playing a major role in the degradation of the link performance, various solutions have been proposed to cope with this effect. The objective of this paper is to present a simple blind estimation of the offset, calculated on each OFDM symbol and thus only based on the received data. A correction is then applied on the phase of the received signal. Synchronization of the receiver with a phased-locked loop is not treated in this paper. After illustrating this approach for an additional white Gaussian noise channel, a power line communication in an aircraft is envisaged. The architecture of the network is described and a parametric study is carried out to assess the performance of the proposed offset estimator and the phase correction technique.
Download
Paper Nr: 97
Title:

Don’t Talk to Strangers - On the Challenges of Intelligent Vehicle Authentication
Authors:

Alishah Chator and Matthew Green
Abstract: Vehicle-to-vehicle (V2V) communications offer an unprecedented opportunity to increase driver safety. At the same time, the use of computer networking technologies raises new concerns around information security and privacy. Specifically, V2V communications systems provide the opportunity for malicious individuals to transmit false data, with unknown effects on future vehicle systems. A number of proposals have been advanced in order to add authenticity guarantees to V2V systems using cryptographic techniques. Unfortunately, many of these proposals have a number of side effects related to efficiency and driver privacy. In this work we discuss these tradeoffs and explain why it is challenging to achieve all desired properties in a single system. We then suggest alternative approaches that may be more realistic than current proposals.
Download
Paper Nr: 99
Title:

Towards an Automated Flying Drones Platform
Authors:

Bruno Areias, Nuno Humberto, Lucas Guardalben, José Maria Fernandes and Susana Sargento
Abstract: Nowadays, some drone Flight Controllers (FCs) support basic automation (e.g., GPS waypoint, return-to-home, path flight, take-off/landing), although it requires direct drone connectivity (e.g., radio, base station/Ground Control Station (GCS)) and an extensive knowledge over technical details (e.g., assembly, configuration, battery maintenance, flight, etc.). This paper proposes a novel platform that offers an abstraction layer between the end-user and the drone itself, automating most of the drone flight requirements. The platform allows to perform high-level drones control (e.g., up, down, left, right, GPS go-to and stream follow, return-to-base, etc.) through end-user communications, contributing with a highly modular event-driven control platform, enabling development of more complex integrations between drones and other technologies. The obtained results show that the proposed automated flying drones’ platform is able to properly abstract and decouple the direct control, handling up to 32 drones with no significant impact on the performance. The platform is also capable of displaying and correlating sensor metrics obtained in real-time during flight.
Download
Paper Nr: 111
Title:

Introducing Cellular Network Layer into SUMO for Simulating Vehicular Mobile Devices’ Interactions in Urban Environment
Authors:

Siim-Toomas Marran, Artjom Lind and Amnir Hadachi
Abstract: During the last decade researchers have been demonstrating the importance of mobile data or CDR data in depicting the human mobility patterns. However, this type of data is not easy to get access to from mobile operators. Besides, in order to make this type of data available and enable their usage for the scientific communities the process can face many constraints that can constitute obstacle. From this perspective, this paper introduces a way to produce realistic real-life mobility logs through the traffic simulation tool SUMO, which has been enhanced with a cellular network layer to mimic cellular networking behavior.
Download
Paper Nr: 49
Title:

A Security Model for Dependable Vehicle Middleware and Mobile Applications Connection
Authors:

Shengzhi Zhang, Omar Makke, Oleg Gusikhin, Ayush Shah and Athanasios Vasilakos
Abstract: Nowadays automotive industry has been working on the connectivity between automobile and smartphones, e.g., Ford’s SmartDeviceLink, MirrorLink, etc. However, as the interoperability between the smartphone and automotive system increase, the security concern of the increased attack surface bothers the automotive industry as well as the security community. In this paper, we thoroughly study the attack vectors against the novel connection framework between automobile and smartphones, and propose a generic security model to implement a dependable connection to eliminate the summarized attack vectors. Finally, we present how our proposed model can be integrated into existing automotive framework, and discuss the security benefits of our model.
Download

Area 2 - Intelligent Transport Systems and Infrastructure

Full Papers
Paper Nr: 29
Title:

Weather-Tuned Network Perimeter Control - A Network Fundamental Diagram Feedback Controller Approach
Authors:

Maha Elouni and Hesham Rakha
Abstract: Inclement weather has been shown to increase congestion, justifying the need for weather-responsive traffic control. From one side, all existing weather-responsive controllers currently operate on freeways or limited road segments. From the other side, existing controllers operating on networks do not take into consideration the weather effect on the network fundamental diagram (NFD). This paper describes the development of a macroscopic weather-tuned perimeter controller. First, an NFD-based proportional-integral perimeter controller (PC) is implemented in INTEGRATION, tuned using clear weather data and then tested for clear and inclement weather conditions. In order to respond to weather changes, new sets of control parameters were tuned for each weather and given to the controller. This weather-tuned perimeter controller (WTPC) is compared to the regular PC. Simulation results show that the WTPC reduces congestion inside the protected sub-network better than PC. Also, it improves the performance of the full network (inside and outside the protected sub-network) in terms of average speed and total delay. Compared to the non-perimeter control case, WTPC increases the average speed of the entire network by 28.61% for rain and 42.64% for snow conditions. Total delay is decreased by 33.26% and 42.02% for rain and snow, respectively.
Download
Paper Nr: 35
Title:

Large-scale Agent-based Multi-modal Modeling of Transportation Networks - System Model and Preliminary Results
Authors:

Ahmed Elbery, Filip Dvorak, Jianhe Du, Hesham A. Rakha and Matthew Klenk
Abstract: The performance of urban transportation systems can be improved if travelers make better-informed decisions using advanced modeling techniques. However, modeling city-level transportation systems is challenging not only because of the network scale but also because they encompass multiple transportation modes. This paper introduces a novel simulation framework that efficiently supports large-scale agent-based multi-modal transportation system modeling. The proposed framework utilizes both microscopic and mesoscopic modeling techniques to take advantage of the strengths of each modeling approach. In order to increase the model scalability, decrease the complexity and achieve a reasonable simulation speed, the proposed framework utilizes parallel simulation through two partitioning techniques: spatial partitioning by separating the network geographically and vertical partitioning by separating the network by transportation mode for modes that interact minimally. The proposed framework creates multi-modal plans for each trip and tracks the travelers trips on a second-by-second basis across the different modes. We instantiate this framework in a system model of Los Angeles (LA) supporting our study of the impact on transportation decisions over a 5 hour period of the morning commute (7am-12pm). The results show that by modifying travel choices of only 10% of the trips a significant reduction in traffic congestion is achievable that results in better traffic flow and lower travel times.
Download
Short Papers
Paper Nr: 34
Title:

Heuristics for Improving Trip-Vehicle Fitness in On-demand Ride-Sharing Systems
Authors:

Sevket Gökay, Andreas Heuvels and Karl-Heinz Krempels
Abstract: On-demand ride-sharing services are emerging alternatives to classical transport modes. Combined with self-driving vehicles, this movement has potential to shape the future of our mobility. To make full use of the potential, such services need to be scalable with growing demand. Assigning real-time trip requests to vehicles such that the driving costs are minimized is computationally expensive, but has to be done fast. This work proposes an approach to reduce the processing time it takes to assign a trip request to a vehicle. The solution is a trip-vehicle fitness estimation framework that is flexible enough to utilize any fitness measure and is self-adjusting through feedback loops. We analyze the placement of a trip request within a vehicle schedule, present and implement three fitness measures. The resulting system is evaluated based on performance, customer satisfaction and vehicle costs criteria by running simulations. The evaluation results indicate significant performance improvement and noticeable improvements in terms of customer satisfaction and vehicle costs.
Download
Paper Nr: 48
Title:

Pictures of You, Pictures of Me - User Acceptance of Camera-technology in Intelligent Transport Systems
Authors:

Teresa Brell, Ralf Philipsen and Martina Ziefle
Abstract: The integration of connected and smart technology is a key factor of our future traffic system development. By integrating traffic participants into the technology development circle, possible trade-offs, obstacles and advances can be identified and further, an understanding of technology acceptance can be evolved. This paper will show, how camera-based technology in intelligent transport systems is evaluated from a user-centred perspective. The focus of this work lies on the identification and evaluation of perceived benefits and barriers, but also conditional and functional aspects are investigated as well as an overall acceptance picture. Results show, that the need for technology is not denied, but privacy concerns and a feeling of surveillance still restrain users.
Download
Paper Nr: 104
Title:

Flexible Multicriterial Agenda Planning in Public Transit Systems - An Intelligent Agent for Mobility-oriented Agenda Planning
Authors:

Felix Schwinger, Fabian Ohler and Karl-Heinz Krempels
Abstract: Planning a mobility-oriented agenda is a time-consuming and tedious task for many travelers. A person is required to collect information from different sources such as a map service, a business register, a calendar and a journey planner. However, she is mostly not interested in either planning the agenda or the journeys between different locations of the agenda; but is more interested in completing the tasks of the agenda. Therefore, we propose an intelligent agenda planning agent that aims to support people with this task. We integrate public transit schedules with additional spatial information from OpenStreetMap to create an information database for the agent. The agent can then plan tasks and appointments and the mobility between those items. First brief evaluations with a survey have shown, that the algorithm finds shorter agendas than most manually found agendas. However, participants of the survey criticized the temporal placement of tasks in the agenda.
Download
Paper Nr: 23
Title:

“Flight Map” Modelling Intellectual Geoinformation System for Urban Areas Cargo Delivery by Unmanned Aerial Vehicle
Authors:

Oleg Golovnin, Nikita Ostroglazov and Tatyana Mikheeva
Abstract: This article is dedicated to solving the problem of safe and secure cargo delivery in urban territory using unmanned aerial vehicles (multicopters) by modelling “flight map”, i.e. a system based on intellectual geoinformation system that develops an optimal and secure route for each UAV in the system and tracks all of them on electronic map online. The paper describes the results of algorithms analysis and assumes the algorithm showing higher test results.
Download
Paper Nr: 109
Title:

Intelligent Containers Network Concept
Authors:

Sergej Jakovlev, Audrius Senulis, Mindaugas Kurmis, Darius Drungilas and Zydrunas Lukosius
Abstract: In this paper, a novel approach is presented to increase the security of shipping containers transportation and storage in container yards. This approach includes wireless sensors networks with programmable modules to increase the effectiveness of the decision support functionality for operators’ onsite. This approach is closely related to the Container Security Initiative and is intended to deepen knowledge in the intelligent transportation research area. This paper examines an urgent challenge - secure of cargo transportation in containers, i.e., how quickly it is possible to detect dangerous goods in shipping containers without changing their tightness and hence rationally implements international security regulations all around the world. This paper contributes to the development of new approaches of shipping containers handling and monitoring in terms of smart cities and smart ports (for the development of the Smart Port initiative) for ports that have higher levels of security violations. This contribution is addressed as an informative measure to the general public working in the Information and Communications Technologies (ICT) research area.
Download

Area 3 - Intelligent Vehicle Technologies

Full Papers
Paper Nr: 1
Title:

Crashzam: Sound-based Car Crash Detection
Authors:

Matteo Sammarco and Marcin Detyniecki
Abstract: Connected vehicles, combined with embedded smart computation capabilities, will certainly lead to a new generation of services and opportunities for drivers, car manufacturers, insurance and service companies. One of the main challenges remaining in this field is how to detect key triggering events. One of these crucial moments is a car accident, for which not only smart connected vehicles can improve drivers’ safety as car accidents are still one of the main causes of fatalities worldwide, but also help them during minor, but very stressful moments. In this paper, we present Crashzam which is an innovative way to detect any type car accidents based on sound produced by car impact, while, so far, crash detection is only a prerogative of accelerometer sensor time series analysis, or its proxy: activation of the airbag. We describe the system design, the sound detection model, and the results based on a dataset with in-car cabin sounds of real crashes. We have beforehand built such dataset with real car accident sounds. Classification is built upon features extracted from the time and frequency domain of the audio signal and from its spectrogram image. Results show that the proposed model is able to easily identify crash sounds from other sounds reproduced in-car cabins. Moreover, considering that Crashzam can run on smartphones, it is a low cost and energy solution, contributing to the spreading of such a car safety feature and reducing delays in providing assistance when an accident occurs.
Download
Paper Nr: 10
Title:

A Multi-stage Centralized Approach to Formation Flight Routing and Assignment of Long-haul Airline Operations
Authors:

Malik Doole and Hendrikus G. Visser
Abstract: This paper describes the development of an optimization-based multi-stage centralized planning system for the efficient routing and assignment of extended flight formations in commercial airline operations. In an extended formation, where aircraft are longitudinally separated by 5-40 wingspans, a trailing aircraft can attain a reduction in induced drag at fixed lift, and consequently in fuel burn, by flying in the upwash of the leading aircraft’s wake. To organize the assembly of flight formations on a network-wide scale essentially two distinct approaches can be taken, viz., a centralized approach and a decentralized approach. Both approaches have distinct advantages and disadvantages. In this study a novel multi-stage method for flight formation assignment is proposed that combines the advantages of the decentralized approach (fast computation and reduced vulnerability to flight delays) with the main benefit of the centralized approach (a near-global optimum in terms of fuel savings). The multi-stage centralized approach that we propose is validated and subsequently demonstrated in a case study involving a wave of 267 eastbound transatlantic flights. In the case study fuel savings of 6.8% are recorded (relative to flying “solo”), while flying in formations comprising up to 16 aircraft.
Download
Paper Nr: 15
Title:

Fully Virtual Rapid ADAS Prototyping via a Joined Multi-domain Co-simulation Ecosystem
Authors:

Róbert Lajos Bücs, Pramod Lakshman, Jan Henrik Weinstock, Florian Walbroel, Rainer Leupers and Gerd Ascheid
Abstract: Advanced Driver Assistance Systems (ADAS) have evolved into comprehensive hardware/software applications with exploding complexity. Various simulation-driven techniques emerged to facilitate their development, e.g., model-based design and driving simulators. However, these approaches are mostly restricted to functional prototyping only. Virtual platform technology is a promising solution to overcome this limitation by accurately simulating the entire ADAS hardware/software stack, including functional and non-functional properties. However, all these tools and techniques are limited to their individual simulation environments. To reap their combined benefits, this paper proposes a joined frameworking to facilitate ADAS prototyping via full virtualization and whole-system simulation. For this purpose, an advanced automotive-flavor virtual platform was also designed, ensuring detailed, near real-time simulation. The benefits of the approach and the joined frameworks are shown by prototyping two ADAS applications in various system configurations.
Download
Paper Nr: 16
Title:

High Resolution Radar-based Occupancy Grid Mapping and Free Space Detection
Authors:

Mingkang Li, Zhaofei Feng, Martin Stolz, Martin Kunert, Roman Henze and Ferit Küçükay
Abstract: The high-resolution radar sensors have the ability to detect thousands of reflection points per cycle, which promotes the perception capability on a pixel level similar to video systems. In this paper, an occupancy grid map is created to model the static environment. The reflection amplitudes of all detection points are compensated, normalized, and then converted to the detection probability based on a radar sensor model. According to the movement of the ego vehicle, the a posteriori occupancy probability is computed to build the occupancy grid map. Thereafter the occupancy grid map is converted to the binary grid map, where the grids in the obstacle areas are defined as occupied. In order to eliminate the outliers, the connected occupied grids are clustered using the Connected-Component Labelling algorithm. Through the Moore-Neighbour Tracing algorithm the boundaries of the clustered occupied grids are recognized. Based on the boundaries, the interval-based free space detection is performed using the Bresenham's line algorithm. As mentioned, the occupancy grid map and the free space detection results obtained from radar road measurements match with the real scenarios.
Download
Paper Nr: 45
Title:

Overtaking Vehicle Detection Techniques based on Optical Flow and Convolutional Neural Network
Authors:

Lu-Ting Wu and Huei-Yung Lin
Abstract: As the rise of the intelligent vehicle applications in recent years, the development of onboard vision systems for advanced driving assistance has become a popular research topic. This paper presents a real-time system using a monocular camera mounted on the rear of a vehicle to perform overtaking detection for safe lane change operations. In this work, the possible overtaking vehicle is first located based on motion cues. The candidate is then identified using Convolutional Neural Network (CNN) and tracked for behavior analysis in a short period of time. We also propose an algorithm to solve the issue of repetitive patterns which is commonly appeared in the highway driving. A series of experiments are carried out with real scene video sequences recorded by a dashcam. The performance evaluation has demonstrated the effectiveness of the proposed technique.
Download
Paper Nr: 51
Title:

Time-controlled Neighborhood-driven Policy-based Network Selection Algorithm for Message Dissemination in Hybrid Vehicular Networks
Authors:

Oleg Oleinichenko, Yagmur Sevilmis, Karsten Roscher and Josef Jiru
Abstract: In vehicular ad hoc networks (VANETs), successful delivery of GeoUnicast and GeoBroadcast packets depends on scenario-specific aspects like vehicle density, distribution of vehicles on the road and type of the environment (e.g., urban, rural). These aspects can significantly influence the reliability of the connection between communication parties making traditional ITS-G5 based ad hoc networks unreliable. The absence of communication partners in range, long transmission distances, non-line-of-sight (NLOS) conditions are just a few examples that could hinder ITS-G5 transmissions. In this paper, we propose a Hybrid Policy-based Network Selection Algorithm that uses LTE to strengthen and complement ITS-G5 under critical conditions in which successful transmission over the ad hoc network is highly unlikely. The main objective is to use as less LTE transmissions as possible whilst maintaining high Packet Delivery Ratio (PDR) within defined delay constraints. The results, which are derived from extensive simulation campaigns, show a clear advantage of using the hybrid scheme over solely ITS-G5 or LTE.
Download
Paper Nr: 53
Title:

Towards Multi-Object Detection and Tracking in Urban Scenario under Uncertainties
Authors:

Achim Kampker, Mohsen Sefati, Arya S. Abdul Rachman, Kai Kreisköther and Pascual Campoy
Abstract: Urban-oriented autonomous vehicles require a reliable perception technology to tackle the high amount of uncertainties. The recently introduced compact 3D LIDAR sensor offers a surround spatial information that can be exploited to enhance the vehicle perception. We present a real-time integrated framework of multi-target object detection and tracking using 3D LIDAR geared toward urban use. Our approach combines sensor occlusion-aware detection method with computationally efficient heuristics rule-based filtering and adaptive probabilistic tracking to handle uncertainties arising from sensing limitation of 3D LIDAR and complexity of the target object movement. The evaluation results using real-world pre-recorded 3D LIDAR data and comparison with state-of-the-art works shows that our framework is capable of achieving promising tracking performance in the urban situation.
Download
Paper Nr: 67
Title:

Non-intrusive Distracted Driving Detection based on Driving Sensing Data
Authors:

Sasan Jafarnejad, German Castignani and Thomas Engel
Abstract: Nowadays Internet-enabled phones have become ubiquitous, and we all witness the flood of information that often arrives with a notification. Most of us immediately divert our attention to our phones even when we are behind the wheel. Statistics show that drivers use their phone on 88% of their trips, in 2015 in the United Kingdom 25% of the fatal accidents were caused by distraction or impairment. Therefore there is need to tackle this issue. However, most of the distraction detection methods either use expensive dedicated hardware and/or they make use of intrusive or uncomfortable sensors. We propose a distracted driving detection mechanism using non-intrusive vehicle sensor data. In the proposed method 8 driving signals are used. The data is collected, then two sets of statistical and cepstral features are extracted using a sliding window process, further a classifier makes a prediction for each window frame, lastly, a decision function takes the last l predictions and makes the final prediction. We evaluate the subject independent performance of the proposed mechanism using a driving dataset consisting of 13 drivers. We show that performance increases as the decision window gets larger. We achieve the best results using a Gradient Boosting classifier with a decision window of total duration 285 seconds which yields ROC AUC of 98.7%.
Download
Paper Nr: 69
Title:

Iterative Calibration of a Vehicle Camera using Traffic Signs Detected by a Convolutional Neural Network
Authors:

Alexander Hanel and Uwe Stilla
Abstract: Intrinsic camera parameters are estimated during calibration typically using special reference patterns. Mechanical and thermal effects might cause the parameters to change over time, requiring iterative calibration. For vehicle cameras, reference information needed therefore has to be extracted from the scenario, as reference patterns are not available on public streets. In this contribution, a method for iterative camera calibration using scale references extracted from traffic signs is proposed. Traffic signs are detected in images recorded during driving using a convolutional neural network. Multiple detections are reduced by mean shift clustering, before the shape of each sign is fitted robustly with RANSAC. Unique image points along the shape contour together with the metric size of the traffic sign are included iteratively in the bundle adjustment performed for camera calibration. The neural network is trained and validated with over 50,000 images of traffic signs. The iterative calibration is tested with an image sequence of an urban scenario showing traffic signs. The results show that the estimated parameters vary in the first iterations, until they converge to stable values after several iterations. The standard deviations are comparable to the initial calibration with a reference pattern.
Download
Short Papers
Paper Nr: 5
Title:

Adaptive Cruise Control for Electric Bus based on Model Predictive Control with Road Grade Prediction
Authors:

Jindong Bian, Bin Qiu, Yahui Liu and Haotian Su
Abstract: Adaptive Cruise Control (ACC) makes the driving experience safer and more pleasurable. To comprehensively deal with tracking capability and energy consumption issue of ACC-activated vehicle on rugged roads, this paper presents a MPC based vehicular following control algorithm with road grade prediction. A simulation model of ACC for electric bus based on MPC is built for analysing the performance of the algorithm. The simulation results show that road grade prediction can improve improves both energy consumption and tracking capability.
Download
Paper Nr: 7
Title:

Using the Projection-based Vehicle in the Loop for the Investigation of in-Vehicle Information Systems: First Insights
Authors:

Matthias Graichen, Lisa Graichen, Thomas Rottmann and Verena Nitsch
Abstract: Most driving simulators cannot replicate real driving dynamics and thus fail to convey a realistic driving experience. To overcome this issue, the Vehicle in the loop (VIL) had been developed, which combines a virtual visual environment with the realistic kinaesthetic feedback of a vehicle while driving on a closed test track. Previous VIL setups used a head-mounted display (HMD) for displaying the virtual environment. This limits the driver’s visual input to the virtual environment and makes it difficult to investigate potential research questions concerning driver interactions with in-vehicle information systems (IVIS). To address this issue, a new version of the VIL has been developed, which uses a projector for displaying the driving simulation on an inset in the windshield and two monitors mounted at the vehicle’s sides. This work presents the first application of the Pro-VIL for investigating IVIS and their impact on driving performance in safety critical situations. For this purpose, we built a setup for comparing the user experience when using either a gesture- or touch-based interaction system, and the observation of driver attention. Results support the overall practicability of the setup, but also revealed new challenges for experimental research design and execution.
Download
Paper Nr: 12
Title:

Autonomous Vehicle Simulation Model to Assess Potential Collisions to Reduce Severity of Impacts
Authors:

Alex Gilbert, Dobrila Petrovic, Kevin Warwick and Vasilis Serghi
Abstract: Autonomous vehicle safety has received much attention in recent years. Autonomous vehicles will improve road safety by eliminating human errors. However, not all automotive collisions can be avoided. A strategy needs to be developed in the event when an autonomous vehicle encounters an unavoidable collision. Furthermore, the vehicle will need to take responsibility for the safety of its occupants, as well as any other individuals, who may be affected by the vehicle’s behaviour. This paper proposes a control system to assist an autonomous vehicle to make a decision to reduce the risks to occupants potentially involved in highway motorway collisions. Before any decision can be made, the potential collisions need to be assessed for their effects. A quick and numerical method for evaluation of impact of potential collisions was developed. Assessing the Kinetic Energy of the vehicles before and after collisions is proposed as a method to assess the severity of collisions. A simulation model developed calculates the kinetic energy values and recommends an autonomous vehicle the motorway lane to drive into to cause the least severe collision impact. Different scenarios are defined and used to test the simulation model. The results obtained are promising and in line with the decision made by the subject expert.
Download
Paper Nr: 20
Title:

Optimal Control for Energy Management of Connected Hybrid Electrical Vehicles - Predictive Connectivity Compared to an Adaptive Algorithm
Authors:

Hamza Idrissi Hassani Azami, Stéphane Caux, Frederic Messine and Mariano Sans
Abstract: For fuel consumption andCO2 emissions reduction, an optimal predictive control strategy for connected hybrid electrical vehicles is proposed, and evaluated through a comparison to an adaptive strategy. The predictive strategy relies on the future driving conditions that can be predicted by intelligent navigation systems with realtime connectivity. The theory proposed for such real-time optimal predictive algorithm is based on Pontryagin minimum principle, a mathematical principle that provides general solutions for dynamic systems optimization with integral criteria, under given constraints. In this work, the energy management problem is mathematically modeled as an optimal control one, and optimal solutions are synthesized. The predictive optimal real-time algorithm is confronted to the adaptive method. Both control strategies are simulated on different driving cycles. The simulation results show the interest of predictive approaches for hybrid electrical vehicles energy management.
Download
Paper Nr: 21
Title:

Anticipating Driver Actions via Deep Neural Networks and New Driver Personalization Technique through Transfer Learning
Authors:

Sahim Kourkouss, Hideto Motomura, Koichi Emura and Eriko Ohdachi
Abstract: Anticipating driving behaviours is a promising technology for novel advanced driver assistance systems. In recent years, predicting a driver’s future action became an important element to preventive safety technologies and has been advancing greatly contributing to a reduction in road accidents. In this paper, we propose a deep learning network that anticipates driving actions based on information of subject vehicle as well as surrounding vehicles and environment. By re-using a network trained on a great number of various drivers’ data with different driving behaviours and linking it to a particular driver with particular taste we propose a method that enables the anticipation of driving behaviours that can be tailored to each driver individually, leading to improved user experiences. We experimentally test our method for acceleration, deceleration and brake profile anticipation task using actual driving data. Our results demonstrate the effectiveness of our approach, achieving a great improvement when anticipating for individuals.
Download
Paper Nr: 26
Title:

Base Algorithms of Environment Maps and Efficient Occupancy Grid Mapping on Embedded GPUs
Authors:

Jörg Fickenscher, Frank Hannig, Jürgen Teich and Mohamed Essayed Bouzouraa
Abstract: An accurate model of the environment is essential for future Advanced Driver Assistance Systems (ADASs). To generate such a model, an enormous amount of data has to be fused and processed. Todays Electronic Control Units (ECUs) struggle to provide enough computing power for those future tasks. To overcome these shortcomings, new architectures, like embedded Graphics Processing Units (GPUs), have to be introduced. For future ADASs, also sensors with a higher accuracy have to be used. In this paper, we analyze common base algorithms of environment maps based on the example of the occupancy grid map. We show from which sensor resolution it is rational to use an (embedded) GPU and which speedup can be achieved compared to a Central Processing Unit (CPU) implementation. A second contribution is a novel method to parallelize an occupancy grid map on a GPU, which is computed from the sensor values of a lidar scanner with several layers. We evaluate our introduced algorithm with real driving data collected on the autobahn.
Download
Paper Nr: 27
Title:

Intention-based Prediction for Pedestrians and Vehicles in Unstructured Environments
Authors:

Stefan Kerscher, Norbert Balbierer, Sebastian Kraust, Andreas Hartmannsgruber, Nikolaus Müller and Bernd Ludwig
Abstract: Motion prediction for holonomic objects in unstructured environments is an ambitious task due to their high freedom of movement compared with non-holonomic objects. In this paper, we present a method for inferring the future goal of holonomic objects by a heuristic generation of target points (tp) and following discriminating decision making. The target points are generated, in a manner that covers the most common motion hypotheses like following or staying, safety relevant motion hypotheses like crossing future ego trajectories or the movement to special points of interest, e.g. gained from a map. Subsequently, for each considered object a trajectory to the inferred target point will be planned. Finally, the uncertainty of the trajectory is estimated by applying a Kalman Filter with a dynamically adjusted process noise matrix. An additional benefit of this concept is its ability to cope with a different quality of context knowledge, so it can produce sound results even at poor structured environments.
Download
Paper Nr: 30
Title:

Crowdsourced System to Report Traffic Violations - RoadCop: Bi-Modular System
Authors:

Maryam Jameela, Hammad Afzal, Khawar Khurshid and Asad Waqar Malik
Abstract: With increasing demand of transportation, implementation of the traffic regulations has become a major challenge for the developing countries. Most of the traffic accidents occur due to violation of traffic rules, thus, resulting in loss of human lives and property. The developed countries have addressed the situation by deploying surveillance systems at intersection, but the solution can be expensive; therefore, due to the cost factor the solution is out of reach for many underdeveloped countries. In order to overcome the situation, a framework is proposed that is based on crowdsourced model to report violations supplemented by the video evidence. The user reports are evaluated in multiple phases. In the first stage, spam is eliminated through evaluation, and associated user profiles are blocked. In the second stage, traffic law experts evaluate the report and on every valid report the users are rewarded with incentive points while ensuring the anonymity. The system is evaluated for usability, advantages to authorities, citizen involvement, skills and resources required and transparency. The results of functional testing indicate that the participants appreciated the purpose of the application and found it quite easy to use. With a large-scale deployment and an effective mechanism to identify offender, this system can lead to much improved implementation of traffic regulations.
Download
Paper Nr: 42
Title:

Improving Range Prediction of Battery Electric Vehicles by Periodical Calculation of Driver Parameters based on Real Driving Data
Authors:

Kurt Kruppok, Tobias Walter, Reiner Kriesten and Eric Sax
Abstract: Due to the battery's limited storage capacity, it is important to reduce energy consumption of electric vehicles. Depending on the average speed, an aggressive driving behaviour can result in an up to 40% higher energy consumption compared to an economic one. In this work, we propose a methodology, which calculates driver parameters based on measured real drive speed and acceleration profiles as well as signposted speed limits. The presented approach compares the energy consumption and driver parameters between our past estimation and the real drive speed profile in order to continuously improve the energy demand estimation for the remaining distance. Thus, this paper provides a procedure to increase the accuracy of energy demand estimation for battery electric vehicles which helps to reduce the range anxiety. In future work, it will be used within a navigation assistance system that supports the driver in reaching his destination with a low battery charge.
Download
Paper Nr: 75
Title:

Isolation Forest for Anomaly Detection in Raw Vehicle Sensor Data
Authors:

Julia Hofmockel and Eric Sax
Abstract: A vehicle generates data describing its condition and the driver’s behavior. Sending data from many vehicles to a backend costs money and therefore needs to be reduced. The limitation to relevant data is inescapable. When using data collected from a vehicle fleet, the normality can be learned and deviations from it identified as abnormal and thus relevant. The idea of learning the normality with the Replicator Neural Network and the Isolation Forest is applied to the identification of anomalies and the reduction of data transfer. It is compared how good the methods are in detecting anomalies and what it means for the traffic between vehicle and backend. It can be shown that the Isolation Forest beats the Replicator Neural Network. When reducing the transfered amount of data to 7%, in average more than 80.63% of the given anomalies are included.
Download
Paper Nr: 77
Title:

On Social Interactions and the Emergence of Autonomous Vehicles
Authors:

Carolina Centeio Jorge and Rosaldo J. F. Rossetti
Abstract: Nowadays and in the contemporary age, the reality of an all-autonomous traffic seems closer and closer. However, this transition period casts a lot of cards onto the table. Although technology can be replacing people at the driver seat, it has not as yet gained our full trust in what concerns communication in real time and safety. Humans interact on a daily basis in their various activities, and traffic is no exception. Most actions performed on the road rely on our perception of others’ awareness and potential reactions. For instance, pedestrians seek for an eye contact before crossing the road, drivers seek for a gesture before starting a manoeuvre, and so forth. Thus, the question remaining is what happens when someone is seeking such a communication interaction and the car has no driver, nor has it someone who even knows what the car is doing. Moreover, people seating in the car might be performing any other activities but driving. Other questions also arise such whether people will accept the idea of trusting self-driving vehicles, or whether will they feel safe when walking amongst such machines. In this paper we pursue a rather social perspective and will raise questions, covering the literature so as to understand what practitioners, researchers and the industry have been doing to overcome the lack of confidence in self-driving cars and improve their trustworthiness towards more efficient and smarter mobility, as well as to identify trends and approaches to answer these emerging questions.
Download
Paper Nr: 78
Title:

When Should the Chicken Cross the Road? - Game Theory for Autonomous Vehicle - Human Interactions
Authors:

C. W. Fox, F. Camara, G. Markkula, R. A. Romano, R. Madigan and N. Merat
Abstract: Autonomous vehicle localization, mapping and planning in un-reactive environments are well-understood, but the human factors of complex interactions with other road users are not yet developed. This study presents an initial model for negotiation between an autonomous vehicle and another vehicle at an unsigned intersections or (equivalently) with a pedestrian at an unsigned road-crossing (jaywalking), using discrete sequential game theory. The model is intended as a basic framework for more realistic and data-driven future extensions. The model shows that when only vehicle position is used to signal intent, the optimal behaviors for both agents must include a non-zero probability of allowing a collision to occur. This suggests extensions to reduce this probability in future, such as other forms of signaling and control. Unlike most Game Theory applications in Economics, active vehicle control requires real-time selection from multiple equilibria with no history, and we present and argue for a novel solution concept, meta-strategy convergence, suited to this task.
Download
Paper Nr: 82
Title:

HACIT: A Privacy Preserving and Low Cost Solution for Dynamic Navigation and Forensics in VANET
Authors:

Kevin Decoster and David Billard
Abstract: The current architecture for VANET related services relies on a Client-Server approach and leads to numerous drawbacks, such as network congestion due to the bottleneck problem or, more importantly, data privacy concerns. Indeed, because of the network topology, traffic must go through nodes which limit the bandwidth and thus bound the overall network capacity. Finally, user data is collected and stored in servers, used by third-party services. However, these providers are known to treat lightly user privacy by selling or using the data for their own purposes (Beresford and Stajano, 2004). By use of a decentralized and distributed communication protocol (\textit{i.e.} D2D), one can overcome these problems by spreading the communication burden to all nodes in the mesh. By means of cryptographic techniques, we can ensure that the shared data is secured and controlled at the end-user side. This paper presents a study and proposes a proof of concept of a decentralized and distributed information system by means of a dynamic navigation system for VANET, using a low-cost solution such as Wifi or LTE-direct new 3GPPP protocol. This system preserves user privacy and is augmented with forensics capabilities.
Download
Paper Nr: 84
Title:

An Extended Hybrid Anomaly Detection System for Automotive Electronic Control Units Communicating via Ethernet - Efficient and Effective Analysis using a Specification- and Machine Learning-based Approach
Authors:

Daniel Grimm, Marc Weber and Eric Sax
Abstract: Due to the increasing number of functions fulfilled by ECUs in a vehicle, there is a need for new networking technologies offering more bandwidth than e.g. Controller Area Network. Automotive Ethernet is one of the most promising candidates and already used in modern cars. However, currently there is the open issue of detecting and preventing cyber attacks via this well known networking technology. In this paper we present the extension of our hybrid anomaly detection system for ECUs to improve the security and safety of vehicles using Automotive Ethernet. The system combines specification- and machine learning-based anomaly detection methods. The features, necessary for the machine learning part, are selected to enable the detection of anomalies in real-time and with respect to the automotive specific communication scheme. Finally, the detection performance and the applicability of different machine learning algorithms is evaluated in a simulation environment based on synthetic and well defined anomalies.
Download
Paper Nr: 91
Title:

AUTOCITS Pilot in Lisbon - Perspectives, Challenges and Approaches
Authors:

Cristiano Premebida, Pedro Serra, Alireza Asvadi, Alberto Valejo, Ricardo Fonseca, Rui Costa, Lara Moura and Conceição Magalhães
Abstract: In this paper we describe the Cooperative Intelligent Transport System (C-ITS) framework, the case-studies and the connected autonomous vehicles to be deployed in the AUTOCITS pilot in Portugal. AUTOCITS project - which stands for Regulation Study for Interoperability in the Adoption of Autonomous Driving in European Urban Nodes, has the main goal of contributing to the deployment of C-ITS in Europe, namely in Spain, France and Portugal, by carrying out pilots using connected and autonomous vehicles and by contributing to the regulations/legal framework on autonomous driving. The Lisbon Pilot will be conducted in a motorway (A-9) and in an urban node (in Lisbon city), where autonomous vehicles (AVs) and instrumented vehicles, all equipped with C-ITS instruments, will be evaluated according to three scenarios: dedicated lanes, shared lane and road without restrictions. In the first scenario, traffic control vehicles will also be present during the tests. In this paper we focus on the Lisbon Pilot: the main technical challenges for infrastructure, test cases and scenarios, and the perspectives on the Portuguese legislation.
Download
Paper Nr: 92
Title:

Signaling Game-based Approach to Improve Security in Vehicular Networks
Authors:

Abdelfettah Mabrouk, Abdellatif Kobbane and Mohammed EL Koutbi
Abstract: Secure communication between vehicle nodes is significant in Vehicular Ad Hoc Networks (VANETs). To guarantee public safety on the roads, vehicular networks need an appropriate security mechanism to protect them from various malicious attacks. In this paper we present an intrusion detection system available to detect internal malicious nodes. When an accident appear on the road, the vehicles must have information about this, but the existence of malicious nodes, the information will be deleted from the network. Because of this, we have adopted a mathematical model based on coalition and signaling game theory to design an Intrusion Detection Game (IDG) modeling the interaction between malicious nodes and the Coalition Head that equipped with Intrusion Detection System (CH-IDS) agent and seek its Bayesian Nash Equilibrium (BNE) for the optimal detection strategy.
Download
Paper Nr: 93
Title:

Development of a Comprehensive Walking Path System in Hong Kong
Authors:

Lilian S. C. Pun-Cheng
Abstract: Walkability has been defined as the extent to which the urban environment is pedestrian friendly. This article presents a case in Hong Kong of how to develop a walking path system to enable users to choose a pedestrian-friendly route. It is found that different details of land configuration can result in varying paths. Such differences can be significant in contributing not only to an accurate system, but also in convincing and stimulating people to walk more according to their own preference.
Download
Paper Nr: 94
Title:

Autonomous Vehicle Architecture Inspired by the Neurocognition of Human Driving
Authors:

Mauro Da Lio, Alice Plebe, Daniele Bortoluzzi, Gastone Pietro Rosati Papini and Riccardo Donà
Abstract: The realization of Autonomous vehicles is recognized as a relevant objective for the modern society and constitutes a challenge which in the last decade is concentrating a growing interest, involving both manufacturers and research institutes. The standard approach to the realization of automated driving agents is based on a well-known paradigm, consisting of the sense-think-act scheme. Even though this implements an understandable and agreeable logic, a driving agent based on such an approach needs to be tested and qualified at a level of reliability which requires a huge experimental campaign. In this position paper the scope of the problem of automated driving is widened into the cognitive sciences, where the inspiration is taken to reformulate the underlying paradigm of the automated agent architecture. In the framework of the EU Horizon 2020 Dreams4Cars Research and Innovation Action project the challenge is to design and train an automated driving agent which mimics the known human cognitive architecture and as such is able to learn from significant situations encountered (either simulated or experienced), rather than simply applying a set of fixed rules.
Download
Paper Nr: 96
Title:

Creation of Transport Network Thematic Layer in GIS via Remotely Sensed Data
Authors:

Tatyana Mikheeva and Alexander Fedoseev
Abstract: The problem of transport net GIS thematic layer creation based on remotely sensed data with limit spectral feature space is a subject of discussion. The proposed solution is based on a hybrid approach which involves shared use of both cluster and contour approaches to earth remote sensing data processing. It aims to unlock the synergies as a result of combining the spectral immanent properties of multispectral image with the spatial information of panchromatic image. The results are also used to create a dynamic transport objects thematic layer.
Paper Nr: 24
Title:

Controlled Emission Zone Pollution Resource Management in 5G C-ITS
Authors:

Tomasz Mach and Wei Guo
Abstract: An innovative pollution resource management scheme is proposed to tackle air pollution. The scheme introduces a novel concept of a pollution grant, a centralised pollution grant scheduler and accompanying pollution grant signalling between the scheduler and controlled polluting vehicle or a stationary source in a co-operative ITS environment. The scheme is analysed and discussed qualitatively as it can be effectively applied to controlled emission zones in cities and as a result, can improve the pollution control fairness, effectiveness and efficiency. The scheme can be implemented as a new pollution resource management function in 5G wireless base station within MEC architecture to leverage its low latency capabilities in parallel to its traditional radio resource management role.
Download
Paper Nr: 38
Title:

Scale-Invarinat Kernelized Correlation Filter using Convolutional Feature for Object Tracking
Authors:

Mingjie Liu, Cheng-Bin Jin, Bin Yang, Xuenan Cui and Hakil Kim
Abstract: Considering the recent achievements of CNN, in this study, we present a CNN-based kernelized correlation filter (KCF) online visual object tracking algorithm. Specifically, first, we incorporate the convolutional layers of CNN into the KCF to integrate features from different convolutional layers into the multiple channel. Then the KCF is used to predict the location of the object based on these features from CNN. Additionally, it is worthying noting that the linear motion model is applied when do object location to reject the fast motion of object. Subsequently, the scale adaptive method is carried out to overcome the problem of the fixed template size of traditional KCF tracker. Finally, a new tracking update model is investigated to alleviate the influence of object occlusion. The extensive evaluation of the proposed method has been conducted over OTB-100 datasets, and the results demonstrate that the proposed method achieves a highly satisfactory performance
Download
Paper Nr: 39
Title:

Motion Prediction Influence on the Pedestrian Intention Estimation Near a Zebra Crossing
Authors:

Júlia Škovierová, Antonín Vobecký, Miroslav Uller, Radoslav Škoviera and Václav Hlaváč
Abstract: The reported work contributes to the self-driving car efforts, more specifically to scenario understanding from the ego-car point of view. We focus on estimating the intentions of pedestrians near a zebra crossing. First, we predict the future motion of detected pedestrians in a three seconds time horizon. Second, we estimate the intention of each pedestrian to cross the street using a Bayesian network. Results indicate, that the dependence between the error rate of motion prediction and the intention estimation is sub-linear. Thus, despite the lower performance of motion prediction for the time scope larger than one second, the intention estimation remains relatively stable.
Download
Paper Nr: 47
Title:

Automated Driving System based on Roadway and Traffic Conditions Monitoring
Authors:

Pavel Beresnev, Anton Tumasov, Dmitry Tyugin, Denis Zeziulin, Valery Filatov and Dmitry Porubov
Abstract: In article development of a concept of advanced driver-assistance systems is considered. It consists of several subsystems such as a warning about leaving the line, warning the driver about the possibility of collision with an obstacle in the lane, detection of traffic signs. The concept of the system receives visual information about the road scene and decides on the need for to correct the course and speed parameters of the traffic. The component structure of system is shown. A number of methods are proposed (Hough Transform, bird's eye view, etc.) to solve the task. Tests of a concept of advanced driver-assistance systems are carried out. Based on the results of the tests, a technical task will be formulated for conducting development work.
Download
Paper Nr: 70
Title:

Bus Schedule Rationalisation - An Analysis of Trip Completion Times
Authors:

Shankar Venkatagiri, Gaurav Kumar and Munish Kaushik
Abstract: Public transit systems offer a smart option to reduce congestion in Indian cities. Due to the poor service quality of public bus transit operators, more commutes are now being completed using private transport, exacerbating traffic problems. In this paper, we examine AVL data generated by public buses in Bengaluru and identify a problem of schedule compliance for buses plying a popular route. We then undertake a time series analysis of the trip run times. We finalise on an ARIMA model and derive a forecast of completion times. We conclude with recommendations for trip scheduling.
Download
Paper Nr: 79
Title:

Precise Vehicle Positioning for Indoor Navigation via OpenXC
Authors:

Yusuf Turk, Baturay Ozcan and Sezer Gören
Abstract: We propose a method for vehicle positioning for indoor locations like parking garages. Our method does not require other external positioning systems such as GPS. Instead, we determine the vehicle position from the vehicle data obtained from an OpenXC dongle attached to the OBD-II interface of the vehicle. An accompanying smartphone application which connects with the dongle via Bluetooth is developed. This application calculates the position of the car and applies the algorithms proposed in this paper to the vehicle data received from the interface. The route of the vehicle is then constructed and displayed on the smartphone screen. As a future work, an assistant application will be developed that guides the driver to the spot where the car was parked before.
Download
Paper Nr: 80
Title:

Usage Profile Rating of Suitability to E-Vehicles Utilizing a Physical Consumption Model
Authors:

Florian Hertrampf, Sebastian Apel and Steffen Späthe
Abstract: The project “Wohnungswirtschaftlich integrierte netzneutrale Elektromobilitat in Quartier und Region” (WINNER) aims to integrate shared electric vehicles, smart local grids and renewable energy in tenant households. This paper focuses on how to find the model of an electric vehicle (consumption, recharging, usage) which perfectly matches the requirements of particular carsharing stations. This approach utilizes usage profiles of conventional combustion vehicles. Each profile describes booking time and distance. Applying that information to a rating model which simulates the driving task and charges the vehicle between usages should be able to tell how much bookings might be handled by an electric vehicle. Within this paper, we give an introduction to our simulation system. This covers the data model, transforming bookings into driving tasks, and the consumption and charging model itself. Further, we validate the model by using high detailed data captured on regular routes as well as booking sets with electric vehicles. This validation shows an average relative error of 10% for high detailed data from and an average relative error for booking information with known consumptions of 5 %. Finally, we present the application of our simulation system to make a decision based on historical booking information. This application example shows that 90% usages at some station might be handled with electric vehicles, while others should not be replaced.
Download
Paper Nr: 101
Title:

Analysis of the GNSS Error Distribution for the Generation of a Cooperative Environment Model for Advanced Driver Assistance Systems
Authors:

Florian Alexander Schiegg, Tobias Frye and Florian Wildschütte
Abstract: In the context of rising traffic automation, the generation of a reliable environmental model plays a key role. By sharing their information, vehicles and infrastructure are able to set up cooperative environmental models of considerably increased accuracy. The GNSS-based localization receives special attention in this regard, since it allows switching from vehicle relative coordinates to absolute and vice versa. While the focus of most related work lies on improving the mean of the GNSS fix, the work at hand analyses its error distribution. Field tests were performed on various scenarios and compared with simulations. Finally, a utility function is proposed, revealing the amount of information carried by every description parameter of the respective distribution.
Download
Paper Nr: 103
Title:

Fault Tolerance in the Traffic Management System of a Last-mile Transportation Service
Authors:

Koji Hasebe, Shohei Sasaki and Kazuhiko Kato
Abstract: A last-mile transportation system has previously been developed using semi-autonomous driving technologies. For traffic management of this system, travel requests are gathered at a central server and, on this basis, the operation schedule is periodically updated and distributed to each vehicle via intermediate servers. However, the entire traffic system may stop if the central server malfunctions owing to an unforeseen event. To address this problem, we propose a fault-tolerant mechanism for the traffic management system of a last-mile transportation service. We use a modified primary-backup (or so-called passive) replication technique. More precisely, one of the intermediate servers is chosen as the central server and the other intermediate servers periodically receive state-update messages from the central server, allowing them to update their state to match that of the central server. If the central server fails, one of the node servers is selected to take over as the new central server. The present paper also demonstrates the availability through failure patterns in experiments with a prototype implementation.
Download

Area 4 - Data Analytics

Full Papers
Paper Nr: 4
Title:

Data Mining Applied to Transportation Mode Classification Problem
Authors:

Andrea Vassilev
Abstract: The recent increase in processing power and in the number of sensors present in today’s mobile devices leads to a renewed interest in context-aware applications. This paper focuses on a particular type of context, the transportation mode used by a person or freight, and adequate methods for automatically classifying transportation mode from smartphone embedded sensors. This classification problem is generally solved by a searching process which, given a set of design choices relative to sensors, feature selection, classifier family and hyper parameters, etc., find an optimal classifier. This process can be very time consuming, due to the number of design choices, the number of training phases needed for a cross validation step and the time necessary for one training phase. In this paper, we propose to simplify this problem by applying three data mining tools - Principal Component Analysis, Mahalanobis distance and Linear Discriminant Analysis - in order to clean the data, simplify the problem and finally speed up the searching process. We illustrate the different tools on the transportation mode classification problem.
Download
Paper Nr: 36
Title:

Travel Time Modeling using Spatiotemporal Speed Variation and a Mixture of Linear Regressions
Authors:

Mohammed Elhenawy, Abdallah A. Hassan and Hesham A. Rakha
Abstract: Real-time, accurate travel time prediction algorithms are needed for individual travelers, business sectors, and government agencies. They help commuters make better travel decisions, avert traffic congestion, help the environment by reducing carbon emissions, and improve traffic efficiency. Travel time prediction has begun to attract more attention with the rapid development of intelligent transportation systems (ITSs), and is considered one of the more important elements required for successful ITS subsystems deployment. However, the stochastic nature of travel time makes accurate prediction a difficult task. This paper proposes travel time modeling using a mixture of linear regressions. The proposed model consists of two normal components. The first component models the congested regime while the other models the free-flow regime. The means of the two components are modeled by two linear regression equations. The predictors used in the linear regression equation are selected out of the spatiotemporal speed matrix using a random forest machine-learning algorithm. The proposed model is tested using archived data from a 74.4-mile freeway stretch of I-66 eastbound connecting I-81 and Washington, D.C. The experimental results show the ability of the model to capture the stochastic nature of travel time and to predict travel time accurately.
Download
Paper Nr: 66
Title:

Survey of Public Transport Routes using Wi-Fi
Authors:

João Ribeiro, André Zúquete and Susana Sargento
Abstract: An important aspect in improving public transport efficiency is collecting information regarding travelers’ routes, usually represented as an Origin Destination (OD) matrix. Most public transportation systems implement fare collection systems that can provide the accurate origins of travelers’ routes but not accurate destinations. In this paper we look at Wi-Fi, more specifically 802.11 data-link layer, as a candidate to provide OD matrix estimations. We present a system and an algorithm capable of collecting information, complemented with positioning and time, regarding Wi-Fi capable devices inside a bus. A system is also presented to implement this concept using minimal requirements. An implementation of this system was deployed in a public bus to collect data for several months. This resulted on over 71929 traveler routes collected in 127 different days. This data was contextualized and mapped to an OD system in order to demonstrate how it can be used to generate OD matrix estimations.
Download
Short Papers
Paper Nr: 6
Title:

Integrated System for Collecting and Reporting Crash and Citation Data
Authors:

Alexander Paz, Cristian Arteaga and Carlos Gaviria
Abstract: Currently, the collection of crash and citation data is performed by law enforcement agents without taking full advantage of existing state-of-the-art technologies. Availability of communication networks and recent developments in software technology provide opportunities to collect data in an easier, faster, and more accurate manner. Key challenges in collecting of this type of data include standardizing and capturing the right location where crashes occur as well as minimizing the exposure time of law enforcement agents at the scene. This paper describes the development of a state-of-the-art crash and citation data collection system and geospatial database, hosted by a remote server, a mobile application, and a web portal. The proposed system takes full advantage of Geographic Information and Positioning Systems to capture location data and provide tools to create scene diagrams. The proposed system was designed and implemented in cooperation with law enforcement agencies and data users to meet the needs of various stakeholders.
Download
Paper Nr: 65
Title:

An Information System for Bus Travelling and Performance Evaluation
Authors:

Leandro Ricardo, Susana Sargento and Ilídio C. Oliveira
Abstract: A wide vehicular network has a huge potential to collect city-data, specially with respect to city mobility, one of the top concerns of the municipalities. In this work, we propose the use of the mobility data generated by the movement of the connected buses to deliver a new set of tools to support both the bus passengers and bus fleet operator use cases. Considering the bus passengers, it is possible to build smart schedules, which deliver an estimated time of arrival based on the city dynamics along time, and that can be accessed directly in the smartphone. Considering the bus fleet operator, it is possible to characterize the behaviour of buses and bus lines. Using the GPS trace of buses and map-matching algorithm, we are able to discover the line each bus is assigned to. Estimated times of arrival and predictions are implemented recurring to time estimations and predictions, using both data mining and machine learning approaches. Proof-of-concept applications were implemented to demonstrate the real-life applicability, including a mobile app for the citizens, and a web dashboard for the fleet operator.
Download
Paper Nr: 88
Title:

Stability Analysis for Adaptive Behavior - (Position Paper)
Authors:

Emil Vassev and Mike Hinchey
Abstract: One of the biggest challenges related to the research and development of autonomous systems is to prove the correctness of their autonomy. Nowadays, autonomous and adaptive systems are the roadmap to AI and the verification of such systems needs to set boundaries that will provide the highest possible guarantees that AI will be safe and sound, so trust can be established in its innocuous operation. In this paper, the authors draw upon their work on integrating stabilization science as part of a mechanism for verification of adaptive behavior. Stability analysis is studied to find an approach that helps to determine stable states of the behavior of an autonomous system. These states are further analyzed to determine behavior trajectories and equilibrium orbits. KnowLang, a formal method for knowledge representation and reasoning of adaptive systems, is used as a platform for stability analysis of autonomous systems.
Download
Paper Nr: 100
Title:

Automatic Fault Detection using Cause and Effect Rules for In-vehicle Networks
Authors:

Alexander Kordes, Sebastian Wurm, Hawzhin Hozhabrpour and Roland Wismüller
Abstract: In-vehicle networks (IVNs) connect Electronic Control Units (ECUs) for automotive applications. Most of the communication on the IVNs directly affect the comfort or even the safety of the driver. Therefore, it is necessary to monitor these systems in order to find the cause and effect of a fault. Current developments use plausibility checks in automotive ECUs to enhance safety and security. Within the LEICAR project in cooperation with INVERS GmbH we focus on all sensors signals recorded directly from CAN bus IVNs for this positional paper. Even without the knowledge of the sensors semantics it is possible to extract cause and effect rules for all recorded sensor signal relationships of the vehicle, map them in a graph and extract certain situations. The proposed solution detects direct and slowly evolving changes even if they propagate across several involved sensor values. For the automatic fault containment we extract features from the cause and effect rules to train a machine learning model in order to make predictions on new data. Besides that it is possible to implement optimized error checking procedures for the involved ECUs.
Download
Paper Nr: 64
Title:

Decision Support Dashboard for Traffic and Environment Analysis of a Smart City
Authors:

Jorge Pereira, Susana Sargento and José Maria Fernandes
Abstract: Porto city has an in-place infrastructure of fixed and moving sensors in more than 400 buses and roadside units, with both GPS and mobility sensors in moving elements, and with environmental sensors in fixed units. This infrastructure can provide valuable data that can extract information to better understand the city and, eventually, support actions to improve the city mobility, urban planning, and environment. Our system provides a full stack integration of the information into a city dashboard that displays and correlates the data generated from buses and fixed sensors, allowing different visualizations over the traffic and environment in the city, and decisions over the current status of the city. A good example relates bus speed variation with possible anomalies on the road or traffic jams. Visualizing such information and getting alarms on anomalies can be a valuable tool to a city manager when taking urban planning decisions to improve the city mobility.
Download

Area 5 - Sustainable Transport

Short Papers
Paper Nr: 25
Title:

Performance Evaluation of Square Coupled Coils at Different Misalignments for Electric Vehicle Battery Charging

Authors:

P. Srinivasa Rao Nayak, Kishan Dharavath, Radhakrushna Dey, K. Sundareswaran and Sishaj P. Simon
Abstract: Wireless Power Transfer (WPT) for electric vehicle battery charging is an advancing battery charging technology. The crucial part in the WPT system is the coupling coil structure and it plays a major role in effective power transfer. This paper describes the mutual inductance and flux distribution characteristics of the square coupled coils with different misalignments also to make more realistic for Electric Vehicle (EV) battery charging applications the coupled coils are designed with and without core and chassis. The evaluation contains the mutual inductance gets affected by distance between the coils, lateral and angular misalignment effects. The results of the analysis are used in the implementation of the wireless EV battery charging system.
Download
Paper Nr: 76
Title:

Vehicle Fleet Prediction for V2G System - Based on Left to Right Markov Model
Authors:

Osamu Shimizu, Akihiko Kawashima, Shinkichi Inagaki and Tatsuya Suzuki
Abstract: The regulations for internal combustion vehicles, CO2 or NOx emission or noise and so on, are strengthened. Therefore EV (electric vehicle)'s market is expanding. The amount of EV get more, the amount of electric get more and the impact for grid that are voltage fluctuation and frequency fluctuation is concerned. V2G (Vehicle to Grid) can solve this problem, but it has a constraint that EV’s battery can be used during it parked. So as the basic technology, the prediction the vehicles’ state that is driving or parked is important. In this research, machine learning algorithm for predicting vehicle fleet's states is developed. The data for study and test is obtained by person-trip survey. The algorithm is based on left to right Markov-model. The states are stay or drive from an area to an area. Future state probability is predicted using the latest observed state and state transition probability. As the result, the prediction error of stay is less than the prediction error of drive. Therefore study data and test data are separated into sunny day and rainy day, the prediction error becomes less.
Download
Paper Nr: 87
Title:

Shifting Speed and Belt Behavior of Model CVT (Continuously Variable Transmission) with Push and Pull Type V-belt Driven on Semi-Transparent Pulleys - Influence of Stiffness of V-belt in Clamping Direction
Authors:

Shinnosuke Nomura, Kazuya Okubo and Toru Fujii
Abstract: The purpose of this study is to investigate influence of the stiffness of V-belt in clamping direction on shifting speed of V-belt type CVT (Continuously Variable Transmission). Model CVT with push and pull type V-belt was prepared with semi-transparent pulleys made of epoxy resin in order to observe the belt behaviour in the pulley groove. The stiffness of the belt in clamping direction was changed to investigate the influence on the shifting speed in which the cross sections of belts were reduced as the alternative types of belts. At the case where the belt pitch radius was increased, the behaviour of elements of the push type V-belt in the pulley groove indicated that the remarkable radial slip between the element and pulley was not occurred. It was suggested that the pitch radius of the belt entering into pulley groove was depended on the deformation of the belt in compression in clamping direction in pulley groove. It was shown that the shifting speed was increased by reducing the stiffness of belts in clamping direction regardless of the belt type.
Download
Paper Nr: 110
Title:

Structuring of Methods to Estimate Benefits of Partial Networking
Authors:

Alexandr Vasenev
Abstract: Partial Networking, as a mechanism for moving-to-sleep and waking-up embedded systems, is beneficial for saving energy within a vehicle (or within other complex distributed systems). Even though a number of models exist which identify benefits of partial networking, they often address rather specific cases. Moreover, these fragmented efforts do not necessarily make explicit which methodological steps were taken. Explicating and analysing methodologies of existing research is beneficial to construct an overarching structure how to estimate potential energy savings for partial networking implementations. This structure can be used to select which steps to take to investigate the savings, and how to construct an argument for presenting the findings. This paper describes initial results of such a research. It reviews several models, illuminates their (sometimes not explicitly documented) methods, and outlines a generalized sequence for estimating partial networking benefits. Besides, it provides a list of questions to consider when introducing partial networking. The outlined methods and the analysis can be of interest to other domains interested in energy savings, such as smart grids, smart cities, and internet of things.
Download
Paper Nr: 9
Title:

The Feasibility of using V2G to Face the Peak Demand in Warm Countries
Authors:

Ibrahem A. Almansour, Enrico H. Gerding and Gary Wills
Abstract: As a result of the very difficult weather in Saudi Arabia during the summer, there is too high power peak demand in the grid and this is expected to increase in the next decade. To fix this problem, power consumers should participate in the power production. Vehicle-to-grid (V2G), one of the efficient sustainable technologies, can offer this opportunity. It is defined as a concept where electric vehicle (EV) provides electric to the grid when parked. This investigation looks at the feasibility of using V2G to mitigate the problem of highest electricity peak demand in the summer period in one of the warmest countries of the world (Saudi Arabia). We conduct a survey in order to serve this issue and we use information from Saudi Arabia electricity authority. We found that, V2G is a promising solution to the peak demand challenge in the summer in Saudi Arabia since there is about 80% of the sample interested in using V2G technology. Moreover, 90% of the participants used their vehicles less than 4 hours daily. Furthermore, in the summer period, most of the participants park their vehicles for the longest time between 13:00 to 18:00, which is the peak demand period.
Download