Intelligent System Solutions for Auto Mobility and Beyond

Infrastructure supported autonomous driving is getting in the focus of current research. In this work, we investigated the usage of a traffic monitoring infrastructure combined with the environmental model of an autonomous driving vehicle for localization. The forwarded environmental model of the infrastructure contains tracked road users, but the used tracking algorithm is unknown. The result is based on a two-stage transformation process with optimized fusion and tracking by a Kalman filter. Experiments show, that the algorithm provides a consistent localization at the first time step.

A low cost communication alternative could allow better V2I communications. In this way, the alternative would complement the 5G technology, driving the reliability of the system up. For that purpose, a selection of low cost technologies, which cover a short/middle range, was carried out. Then, theirs capabilities were tested and compared by means of a test-bed, achieving knowledge about theirs strengths and weaknesses. The final goal is to apply the technology in a real proof of concept.

New technologies have been progressively integrated into vehicles during the last thirty years. Commercial vehicles today can be seen as 2 tonne IoT devices on wheels that continuously collect high-quality information not only from the internal performance and behaviour of the vehicle but also from the external environment. The adoption of cutting-edge technologies like 5G, Edge Computing and Artificial Intelligence (AI) will be essential pillars for the actual implementation of new Intelligent Transportation Systems (ITS) leveraging Vehicle-to-everything (V2X) paradigm. This paper is focused on the design and implementation of a connected vehicle-based system to enable new services and applications to be developed, by exploiting high-quality data collected from onboard sensors and ECUs and leveraging state-of-the-art machine learning technologies.

Automated driving is supposed to alleviate driver’s workloads and improve traffic efficiency and road safety. In order to achieve these goals, it is crucial to test the automated systems thoroughly. Since pure testing on test tracks and public roads is highly time consuming and still cannot cover all critical scenarios, usage of simulation and driving simulators can speed up the testing process. The TrustVehicle project aims at improving SAE Level 3 automated driving (L3AD) functionalities, especially in critical situations and under harsh weather conditions. Therefore, co-simulations, driving simulators and real-world tests are used to test the trustworthiness and availability of the developed functionalities.

An overview of the approach used within the project is presented in this paper. Based on one of the four use cases, the used toolchain and corresponding workflow is described.

The paper presents the concepts behind the developments of technologies, electronics components, communication concepts, embedded systems modules for designing fail-aware, fail-safe, fail-operational (HW and SW redundancy) electronic components and systems architecture that enable the introduction of autonomous/automated driving capabilities of level 4 and 5 future vehicles categories. Focus is given to the automotive standards such as ISO 26262, ISO/PAS 21448. The paper is based on findings of the AutoDrive ECSEL project [1] and gives examples of safety use case scenarios for heterogeneous V2X low latency communication between the autonomous vehicle and the environment.

Highly automated/autonomous vehicles using extended features like Vehicle to Vehicle (V2V) or Vehicle to Infrastructure (V2I), cognitive systems for decision taking, needing extensive perception features and sophisticated sensor functions, cause a considerable shift in safety and cybersecurity (trustworthiness) co-engineering and assurance. To achieve trust of the public/users, standards and certification/qualification are challenged, not comparable to conventional “singular vehicle only” issues. The paper highlights the necessary evolution in the automotive and related standardization landscape, including ethics guidelines and recent activities, and the consequences from upcoming UNECE (United Nations Economic Commission for Europe) regulations. An Overview on ongoing work in large European ECSEL projects, SECREDAS and AutoDrive, including standardization, is provided.

With growing connectivity in consumer and industrial applications, the need for security rises proportionally. Compromises in security design of e.g. autonomous driving systems endanger not only material goods but may threaten personal life of humans. Besides the demand for functional safety of complex, connected systems, cyber security is paramount. As the traditional verification flow is only dealing with functionality, reliability and safety aspects, a trusted design flow extends this by adding aspects of hardware vulnerabilities in verification and certification. Consequently, without full trust in the globally distributed development and production process, semiconductor manufacturers need to check that no malicious modifications are inserted. Physical layout verification is achieved via a comparison of the recovered chip layout extracted from an application against the design data. To enable trust, a comprehensive figure of merit has been defined to evaluate the correlation of the layout extraction to design data. Using this figure of merit (FoM), it is shown on a 40 nm sample that ultra-high scanning speeds still deliver adequate image quality.

Infrastructure planning of transport and mobility is a very complex and challenging issue for planners and decision-makers. Cooperative Intelligent Transport Systems (C-ITS) allow here not only improved information and planning in singular vehicle but optimization of the overall traffic and consideration of special needs. Understanding the design, structure and integration of the necessary infrastructure and other road users must become an integral part of developing modern vehicles and modern infrastructures. There are already existing concepts for special service permission for accessing road and traffic services which can improve public safety, decrease the environmental impact and improve the overall traffic flow. In this work, we analyse a CPS road network scenario, with both general and emergency processes, where vehicles and infrastructure are able to communicate with each other concerning dependability. We highlight challenges and propose building blocks towards a holistic system engineering method. Furthermore, we give a short overview of state-of-the-art methods proposed by standards and discuss their shortcomings concerning the raised dependability issues.

This study researched pros and cons of cooperative automated busses by means of a simulation model that renders the influence of bus concepts on key parameters, like: travel time, frequency, capacity, energy consumption and costs. Two existing bus routes are taken as base context. The bus concepts consisted of a completely autonomous bus (assumed SAE level 5), a partially automated bus (level 4) and a conventional bus. The results showed that an autonomous bus causes increased travel time for travelers. However, concepts that combine partially automated driving on dedicated lanes with cooperative driving show potential benefits from more frequent operation. Furthermore, the possibility to remain in the bus when waiting for a connection may be experienced as a comfort enhancement.

Just over one quarter of the EU-28 population lives in rural areas, possibly reflecting a trend towards leaving inner city areas in search of more (affordable, qualitative, open, peaceful) space, in suburbia, towns, or the countryside. Despite the positive connotations associated with rural dwelling, rural areas present conditions that require further support towards the adoption of smart integrated mobility solutions. The lower population density makes running public transport at high frequency inefficient, expensive, limited and not meeting the local demand. This, in turn, provokes an increase in the share of private cars ownership among rural inhabitants, which has negative environmental and socioeconomic impacts. Addressing these challenges, the EU-funded project “SMARTA 2-Sustainable Shared Mobility Interconnected with Public Transport in European Rural Areas” sets to deploy, implement and evaluate four demonstration sites in East Tyrol (AT), Trikala (GR), Águeda (PT) and Brasov (RO) involving sustainable, shared and integrated rural mobility solutions interconnected with public transport and supported by multimodal travel information services.

The rapid development of intelligent control techniques has brought changes to the automotive industry and led to the development of autonomous or self-driving vehicles. This paper presents self-driving autonomous vehicle (AV) shuttles in the smart city context and discusses different aspects necessary to consider when deploying AV shuttles. The smart city testbed is described where one of the key units is a self-driving last-mile shuttle bus named ISEAUTO. The objective of the ISEAUTO project is to establish a smart city testbed on the university campus where different types of experiments about future urban mobility can be studied. Vehicle and pedestrian safety is experienced as a key safety issue in the paper.

In the development of highly automated driving, strong focus is laid on the takeover and the improvement of takeover quality. Some research has shown that the complexity of a traffic situation has an influence on the takeover. However, different approaches towards complexity in driving exist and the topic has so far not been addressed sufficiently. In this study, a differentiation between subjective- and objective complexity is drawn. Their impact on eye movement patterns is evaluated and compared to the resulting takeover quality. Results of a driving simulator study show that objective and subjective complexity have an influence on several eye movement patterns. These eye movement patterns serve as an indicator of the resulting takeover quality. Furthermore, traces of the eye movement patterns are compared to predicted traces of the cognitive model for the takeover task. It can be shown that the cognitive model predicts visual traces in different traffic situations well. In order to support individual drivers during a takeover, it is thus important to consider complexity measurements in the development of cognitive assistance systems. Based on information about the environment and the cognitive model for the takeover task, a cognitive assistance system can be developed. In addition to that, eye tracking information further improves cognitive assistance systems.

The PRESLEEP project is aimed at the fine assessment and validation of the proposed proprietary methodology/technology, for the automatic detection and prediction of the transition between the behavioural states of a subject (e.g. wakefulness, drowsiness and sleeping) through a wearable Cyber Physical System (CPS). The Intellectual Property (IP) is based on a combined multi-factor and multi-domain analysis thus being able to extract a robust set of parameters despite of the, generally, low quality of the physiological signals measured through a wearable system applied to the wrist of the subject. An application experiment has been carried out at AVL, based on reduced wakefulness maintenance test procedure, to validate the algorithm’s detection and prediction capability once the subject is driving in the dynamic vehicle simulator.

One research area within the development of automated vehicles deals with the impact analysis on traffic flow by numerical simulation. This study investigates human drivers’ acceptance while interacting with different levels of automated vehicles on highways including on- and off-ramps. Reactions between conventional, human driven vehicles (CV) and automated vehicles (AV) were tested using a driving simulator. Gaps and headways between vehicles were recorded and analyzed. The analysis indicates similar behavior between CVs and aggressive AVs (short headways) while prudent AVs were perceived less favorable by the test drivers. Additionally, long headways showed more disturbance in traffic flow than shorter headway setups of the automatic distance control (ACC).

One of the main reasons for contested innovations to fail is the negligence of societal needs and public acceptance in due time in the development phase. In the specific case of the connected automated vehicle (CAV), there is an important degree of scepticism based on the awareness of the complexity and the risks of this technology. The SUaaVE project aims to make a change in the current situation of public acceptance of CAV by enhancing synergies amongst social science, human factors research and automotive market. The main ambition in SUaaVE is the formulation of ALFRED, defined as a human centred artificial intelligence to humanize the vehicle actions by understanding the emotions of the passengers of the CAV and managing corrective actions in vehicle for enhancing trip experience.

The EVC1000 project (Electric Vehicle Components for 1000 km daily trips) aims at developing brand-independent components and systems, and demonstrates them through an integrated wheel-centric propulsion architecture and EV (Electric Vehicle) management approach implemented on two different EVs.

The project relies on in-wheel motor and provides new chassis components and integrated controllers. Moreover a compact centralised drive for in-wheel motor axles, based on Silicon Carbide technology, targeting superior levels of functional integration and failsafe operation will be integrated in the EVs.

With the increasing adoption of electromobility, it has become crucial to balance the consumer demand and offer from electric vehicle charging points. To improve end-users’ electromobility experience and access to related services, the NeMO project developed a Hyper-Network of tools and services. As a service, this paper describes the artificial neural network (ANN) models that are built to predict the occupancy statuses of electric vehicle charging points in Barcelona. The models are designed to do both short- and long-term predictions of the next 2 h and next week, respectively, considering the different power levels. We propose to combine similar charging points by behavior with no geolocational information to reduce the computational cost by considering the high number of charging stations in the future.

Nowadays, the high level of atmosphere pollution strongly influenced by industry and transportation is one of the most relevant social problems. In order to reduce the amount of pollutant discharged into the atmosphere, some technological breakthroughs have been developed in vehicle electrification area. The main objective of the European project STEVE is to introduce Electrified L-category Vehicles in urban transport systems. However, the main drawback is their limited driving range thus, optimal energy management systems are crucial here. In this paper the newly developed Eco-navigation Android application, which generates the most optimal route based on speed and elevation profiles for use in e-bikes, will be presented. The Eco-routing app is going to be implemented in Calvià (Mallorca, Spain) during 2020.

The paper proposes a robust and ultra-fast short circuit detection method based on the voltage dip in the half-bridge due to the presence of stray inductance. For the application of the inverter in electromagnetic suspension, the short circuit is detected in less than 100 ns, which is a promising solution against the Fault Under Load due to a Single-Event Burnout failure type.

Automated driving will significantly influence our society. The development towards connected and automated mobility offers solutions to many of today’s societal challenges. The further development of automated and connected vehicles to market maturity requires extensive test procedures that meet the highest safety requirements. This paper will highlight key activities, which are necessary from a public authority’s point of view to test and implement them on domestic roads. It describes Austria’s strategy including its concrete measures and criteria to fulfill a holistic approach for enabling car manufacturer and scientific institutes to test and further develop technologies under real conditions, based on the active steering role from the public authority to ensure the safe and sustainable implementation.