NEWS - 2024/10/07

Hi-Drive Summer School 2nd Edition: Navigating the Future

The Hi-Drive project, funded by the European Union, aims to address the diverse challenges facing the widespread deployment of automated driving (AD) systems across Europe. It focuses on improving the robustness and scalability of AD systems in a variety of complex environments. As part of this initiative, the second edition of the Hi-Drive Summer School took place in 2024, bringing together researchers, industry experts, and PhD candidates from across the field of automated mobility. Held over several days, the event served as a platform to share knowledge, discuss ongoing challenges, and present innovative solutions for the future of AD systems.

During the Summer School, numerous presentations covered topics essential to the advancement of automated driving technology. Leading academic figures and researchers provided insights into the fundamental challenges of AD systems. Professors from renowned universities across Europe focused on key technical issues such as sensor fusion, decision-making algorithms, and the integration of machine learning techniques to enhance the perception and reaction of autonomous vehicles. Additionally, industry professionals emphasized the importance of simulation and real-world testing environments to ensure the reliability of AD systems under varying traffic and weather conditions. These sessions offered both theoretical and practical frameworks for understanding how AD technologies could be adapted to real-world complexities.

Other presentations focused on the applications of AD technologies in diverse geographical and regulatory environments. PhD candidates and young researchers presented case studies from ongoing projects, sharing real-world data and experiences from automated driving pilots in both urban and rural environments. These discussions covered how AD systems handle unpredictable road behaviours, legal frameworks across different EU countries, and the integration of connected systems to improve traffic flow. Industry representatives also shared updates on collaborative projects between academic institutions and the private sector, illustrating how partnerships drive the development of scalable AD solutions.

One of the contributors to this edition of the Summer School was Gerard Franco-Panadés, a PhD candidate from the Universitat Politècnica de Catalunya. Gerard presented his poster titled “A Comparative Study of a Gradient Boosted Decision Trees Model and a Recurrent Neural Network Long-Short Term Memory Model for Predicting Interurban Roads Accident Risk.” His research focused on developing predictive models that use machine learning to analyse data from GPS and infrastructure sensors to improve the efficiency of traffic systems. His work also examined how predictive analytics could be applied to mitigate road accidents by predicting accident risk, contributing significantly to discussions around data-driven solutions in automated driving systems.

The Hi-Drive Summer School concluded with several key takeaways. One prominent conclusion was the need for enhanced collaboration between academia, industry, and regulatory bodies to address the complex technical, legal, and societal challenges of AD deployment. The event also underscored the importance of improving the safety and adaptability of AD systems, particularly in unpredictable environments. Additionally, participants emphasized the need for more comprehensive testing and validation frameworks to ensure the smooth integration of AD technologies across different regions and traffic scenarios.