The important aspect in achieving sustainable development in highly automated drive is safety analysis of its functions and critical scenarios. For L4 or L5 automated vehicle, it performs more complex driving tasks and has more interaction with other traffic participants, which brings considerable challenges for identifying and analyzing safety critical scenarios. In this paper, we propose a methodology which provides a solution on how to generate safety critical scenarios from real-world traffic data and how to use them to in the assessment of functional safety. Based on this, we form the theoretical model of the development and safety evaluation of the automatic driving functions. Backward collision warning and anti-collision lane changing are two effective measures to control the risk and reduce the damage.

CICTP