IVORY - ‘AI for Vision Zero in Road Safety’ is an industrial doctorates network aiming to develop a new framework for optimal integration of Artificial Intelligence (AI) in road safety research, and train a new generation of leading researchers in the field. It addresses the UN Sustainable Development Goals target 3.6 and the EC Vision Zero strategy, of halving traffic fatalities by 2030 and eliminating them by 2050. IVORY addresses the lack of common understanding of the challenges and opportunities of AI for road safety by means of 4 research goals: it aims to develop (i) responsible, fair and impactful AI for road safety, (ii) new ways of road user support and human-vehicle-environment interaction, (iii) new scalable and equitable AI technologies for proactive infrastructure safety management, (iv) a sustainable knowledge sharing network on AI for road safety. IVORY outputs will not only provide more robust user support through AI in vehicle automation, but will also allow to responsibly and proactively manage the persistent problems of existing conventional, low-automation transport systems, so that new opportunities for global road safety impact can emerge. Moreover, IVORY takes a design-for-values approach for AI in road safety, operationalising the ethical principles of justice and explainability, and providing efficient AI solutions also for disadvantaged groups (e.g. vulnerable road users, low-to-middle-income countries). IVORY consists of 4 academic and 6 non-academic beneficiaries, and 12 associated partners, joining from engineering, data science and ethics of technology disciplines, from 11 countries. 13 young researchers will receive high-level doctoral education, industrial exposure, local training, and 8.5 ECTS of network-wide training on key advanced, core and transferable skills. IVORY will create an on-line learning & networking platform for AI in road safety, to be available after the end of the project for future researchers in this field.

Manufacturers of automated systems and the manufacturers of the components used in these systems have been allocating an enormous amount of time and effort in the past years developing and conducting research on automated systems. The effort spent has resulted in the availability of prototypes demonstrating new capabilities as well as the introduction of such systems to the market within different domains. Manufacturers of these systems need to make sure that the systems function in the intended way and according to specifications which is not a trivial task as system complexity rises dramatically the more integrated and interconnected these systems become with the addition of automated functionality and features to them. With rising complexity, unknown emerging properties of the system may come to the surface making it necessary to conduct thorough verification and validation (V&V) of these systems. VALU3S aims to design, implement and evaluate state-of-the-art V&V methods and tools in order to reduce the time and cost needed to verify and validate automated systems with respect to safety, cybersecurity and privacy (SCP) requirements. This will ensure that European manufacturers of automated systems remain competitive and that they remain world leaders. To this end, a multi-domain framework is designed and evaluated with the aim to create a clear structure around the components and elements needed to conduct V&V process through identification and classification of evaluation methods, tools, environments and concepts that are needed to verify and validate automated systems with respect to SCP requirements. The implemented V&V methods as well as improved process workflows and tools will also be evaluated in the project using a comprehensive set of demonstrators built from 13 use cases with specific SCP requirements from 6 domains of automotive, industrial robotics, agriculture, Aerospace, railway and health.

The objective of this project is to setup a framework for the definition, development, testing and validation of a context-aware ‘safety tolerance zone’ for on-road driving, within a smart Driver and Road Environment Assessment and Monitoring System (i-DREAMS). Taking into account, on the one hand, driver-related background factors (age, driving experience, safety attitudes and perceptions, etc.) and real-time risk-related physiological indicators (e.g. fatigue, distraction, stress, etc.), and on the other hand, driving task-related complexity indicators (e.g. time of day, speed, traffic intensity, presence of vulnerable road users, adverse weather, etc.) a continuous real-time assessment will be made to monitor and determine if a driver is within acceptable boundaries of safe operation (i.e. safety tolerance zone). Moreover, safety-oriented interventions will be developed to prevent drivers from getting too close to the boundaries of unsafe operation and to bring back the driver into the safety tolerance zone. These interventions will be composed of both real-time interventions, i.e. in-vehicle while travelling, and other interventions aimed at enhancing the knowledge, attitudes, perceptions and behavioural reaction of drivers with respect to safety-related technologies, situations and behaviours. Application areas will include: new road safety interventions, improved driver well-being and transfer of control between human and vehicle. Initial testing will take place in a driving simulator environment after which promising interventions will be tested and validated under real-world conditions in a testbed consisting of 600 drivers in total across 5 EU countries. Market roadmaps will be developed to support smooth transition of the investigated technologies to the market and experience from use cases in different European countries will be used to disseminate best practices.