Decision-making tools for smarter, safer mobility

Connected and automated vehicles have the potential to revolutionise transportation, but they struggle to manage real-world complexities. The EU-funded EVENTS project addressed this challenge, enabling automated vehicles to navigate unpredictable scenarios such as vulnerable road users, unstructured roads and adverse weather.

Transport and Mobility

The vision of the EVENTS project is to create safer, smarter and more reliable automated vehicles. By equipping vehicles with advanced perception and decision-making systems, it hopes to reduce accidents and enable connected and automated vehicles (CAVs) to handle adverse and unstructured environments. “EVENTS set out to ensure that automated vehicles can navigate not just the ideal conditions they were designed for but also the unexpected and challenging scenarios of the real world,” says project coordinator Angelos Amditis, R&D director from the Institute of Communication and Computer Systems at the National Technical University of Athens in Greece. The project focuses on developing next-generation perception algorithms and decision-making systems that enable vehicles to sense and respond to their environments with greater precision. A key aspect is expanding the operational design domain (ODD) of autonomous systems, enabling them to perform safely in challenging driving conditions. EVENTS also introduces self-assessment systems to monitor and detect sensor malfunctions or errors, enhancing operational reliability. These advancements are supported by smarter decision-making tools capable of dynamically adapting to complex driving environments.

From research to real-world solutions

The EVENTS approach combines rigorous research with practical applications. This includes targeted experiments to test the performance of autonomous systems’ suboptimal conditions, such as unstructured roads, adverse weather conditions such as fog or low winter sun, and the presence of vulnerable road users such as pedestrians and cyclists. Central to the project are machine learning algorithms developed for perception, motion prediction and decision-making. Prototype vehicles equipped with sensor suites were used to collect data and validate these solutions through a combination of simulations and real-world testing, ensuring both scalability and reliability. A fusion of sensors delivered high performance without relying on expensive hardware, while adaptive algorithms enabled vehicles to handle even the most complex and unpredictable scenarios. By sharing its findings with international bodies to guide the development of rigorous industry standards, the project is helping to nurture global standardisation. The work will build trust in autonomous technology and help ensure its widespread adoption. “Our goals are more than technical milestones – they are stepping stones to safer roads and a more sustainable future of mobility,” Amditis adds. The consortium faced challenges such as the scarcity of real-world data for rare edge cases, difficulties in replicating adverse weather conditions, and regulatory barriers. Despite these obstacles, EVENTS continued to thrive. “We were tested in many ways,” Amditis remarks. “But every challenge was an opportunity to adapt and improve. This resilience is what makes EVENTS stand out.”

The road ahead

EVENTS has already achieved many of its key milestones, including simulation-based data generation, innovative datasets and advanced algorithms. And as the project enters its final phase, its innovative methodologies are set to shape the future of automated driving. “Our work is not just about advancing technology – it’s about making transportation safer, more accessible and more sustainable for everyone,” concludes Amditis.