Dealing with uncertainty in perception system for the characterization of driving situation

Situation characterization requires numerous data in order to recognize real driving situation. The quality of data is important to assume the quality of situation recognition and characterization. The aim of this work is to describe imperfections in sensor data processing and their implications in decision processing. Expected results are a granularity definition to model and propagate imperfection. In this paper we focus on uncertainty characterization deriving from the reliability and confidence in sensors and data processing.