iDriving aims to deliver a TRL 6 prototype based on seven key pillars, all designed to work synergistically to create an interactive, accurate and efficient solution to improve infrastructure safety of urban and secondary rural roads. (i) A Safety Criteria Catalogue (SCC) for all road users for secondary and urban roads. This establishes a benchmark for safety standards and KPIs, forming the foundational blueprint that guides all aspects of the iDriving system. (ii) Common communication solutions, ensuring a seamless and real-time exchange of data across a diverse range of sensors from various vehicles and road infrastructures. (iii) Innovative sensor integration on compact packages. These can be easily deployed and maintained on both vehicles and road infrastructure, particularly on secondary and urban roads, thus acting as core data collection or independent processing hubs for onsite analysis. (iv) Efficient and accurate monitoring services. This includes visual inspection of the road network, vehicle behavior analysis, and the analysis of weather conditions, all crucial for proactive road safety management. (v) Intelligence to maintenance operations to reduce the costs and impact on other users through advanced tools like 3D visualization, defect detection at the local level, and optimal traffic management. (vi) AI-based warning mechanisms that can provide real-time alerts about hazardous conditions or accidents either on road infrastructure with signs or lightning or directly to vehicle users via enhanced interfaces. The final pillar ensures the seamless integration of all these elements, bringing together sophisticated AI-driven analytics, machine learning, sensor fusion, and cutting-edge simulations into a (vii) Digital Twin of the road infrastructure. This facilitates the anticipation, identification, and response to a wide range of road conditions and events in real-time.

MKLAB is responsible for the coordination of the iDriving project and also has the role of Technical Manager. It has also a critical role in the project by leading a WP on Intelligent Edge Computing and Monitoring for Safer and Well-maintained Roads, aiming to implement real-time edge computing for road monitoring, use AI to analyze road user behavior, identify maintenance needs, provide real-time environmental hazard alerts, and create 3D AI-driven representations for enhanced safety and maintenance. To this end, MKLAB will develop and implement advanced visual edge detection using deep learning for real-time object and anomaly identification on embedded devices, autonomous road infrastructure defect detection via computer vision, and 3D road environment modeling using Neural Radiance Fields (NeRF) for enhanced safety and maintenance insights within the iDriving ecosystem.