Analýza rizik V2V komunikace a její usměrnění s použitím principů funkční bezpečnosti

Abstract

The dissertation deals with the analysis of risks associated with driving trucks in convoys. Among other things, attention is paid to V2V communication between vehicles and its impact on the safety of the entire system. The work is based on the principles and procedures defined by ISO 26262 ed.2 and directly applies them to the operation of vehicles with varying degrees of autonomy.

The aim of the thesis is to identify and evaluate the risks associated with the operation of trucks in convoy mode, to determine ASIL safety integrity levels, and to propose appropriate safety objectives and risk reduction mechanisms for selected risks. To this end, methods described in ISO 26262 ed.2, such as HARA, FTA, or FMEA, are used and subsequently applied to selected subsystems and their safety functions.

Part of the work is also devoted to the use of neural networks in safety-critical applications of autonomous vehicles. The work assesses the specific behavior of neural networks in terms of functional safety, in particular their nonlinear behavior, limited predictability, and the problem of their validation. The content of the work is a discussion of various approaches to their safe integration into individual autonomous vehicle systems.