Reliability of artificial intelligence for the automotive sector: Politecnico research receives an award

We are pleased to announce that the paper “Estimating the Impact of Soft Errors on AI-based Perception in Automotive”, developed within the Interdepartmental Center CARS and authored by researchers from the DIMEAS, DET, DAUIN, and DIATI Departments of Politecnico di Torino, has received the Best Paper Award at IEEE LATS 2025, one of the leading international conferences on electronic system reliability.

In assisted and autonomous driving systems, artificial intelligence is used to perceive the driving environment and support decision-making processes. The complexity of these systems introduces reliability issues, including soft errors—temporary hardware malfunctions that can affect algorithm processing and vehicle behavior.

To address this challenge, the Politecnico di Torino research team developed TIARA (Two-step IntegrAted Reliability Assessment), a methodology that enables efficient reliability estimation of AI-based systems from the early design stages.

For further information: Affidabilità dell’intelligenza artificiale per l’automotive: premiata la ricerca del Politecnico | Politecnico di Torino