The Autonomous Urban Transport Optimisation project, a ground-breaking collaboration between A2A and the AIDA – Artificial Intelligence Driving Autonomous team of the Politecnico di Milano, led by Prof. Sergio Matteo Savaresi from the Department of Electronics, Information and Bioengineering, has officially started.

Part of the research programme of MOST (National Centre for Sustainable Mobility), the project launched in Brescia, where an electric Fiat 500 completed its first kilometre in fully autonomous driving mode, marking the beginning of an innovative experiment. The objective is to redefine the car-sharing experience: users will no longer need to locate vehicles themselves, and service providers will no longer require large fleets to ensure accessibility.

Through this initiative, the Politecnico di Milano, A2A, and MOST aim to develop a car-sharing model that is unparalleled in Europe. By combining the potential of electric vehicles with autonomous driving, the project introduces a new paradigm for urban mobility. It seeks to reduce the number of cars circulating in cities while making car sharing more practical and user-friendly.

The project is a testament to Italian technological excellence, powered by the advanced autonomous driving system developed by the AIDA team. The solution integrates a cutting-edge hardware setup—featuring state-of-the-art sensors, actuators, networking services, and computing units—with artificial intelligence algorithms designed to replicate human driving behaviour. This system ensures exceptional precision and safety standards. The autonomous vehicles, capable of operating at low speeds (up to 30 km/h), can independently park, transfer themselves to the next customer, or head to a charging station, significantly simplifying service management and reducing operational risks.

The trial, conducted on public roads, has received authorisation from the Ministry of Infrastructure and Transport and the Municipality of Brescia, adhering to the guidelines set by the Smart Road ministerial decree. Scheduled for completion by November 2025, the project involves one to two monthly tests.

During each trial, a supervisor will be present on board the vehicle to intervene if necessary, supported by a control room that will provide real-time monitoring of operations. This approach introduces an innovation in autonomous driving: remote operators or onboard supervisors will only step in during complex situations, enhancing the service's overall safety and reliability. Additionally, a safety car will accompany the autonomous vehicles during tests to inform road users of the ongoing experiment on public roads.