Data-Driven Approaches for Resilient and Sustainable Urban Mobility

The transportation sector is responsible for 23% of energy-related CO2 emissions. Decarbonizing transportation is challenging, as it is still 92% dependent on non-renewable resources. However, current transport decarbonization-related policies are insufficient to decrease CO2 emissions to the expected level. Therefore, strategic approaches to reducing emissions from urban transport are critical to addressing the challenges of climate change. In this talk, we present our recent research activities on a framework to build the next level of innovative data-driven traffic light strategies as the most impactful action to reduce CO2 emissions within the context of urban mobility for Connected and Autonomous Cars. This Framework is committed to embracing the next generation of Edge-AI, benefiting from the ease of implementation and increased computation power toward more composable, distributed, and federated intelligence, as well as security by design frameworks. Powerful eye-bird-view multimodal data fusion approaches feed AI models for accurate CO2 and urban noise level predictions, that feed to dashboards for awareness purposes. Advanced reinforcement learning techniques make use of urban noise predictions to implement the best traffic light strategy in real time. We will also discuss the challenges to achieve resilience by proactively detecting misbehaving entities within Vehicle-to-Everything settings