Model-based safety validation of the automated driving functio highway pilot

Abstract

In the development cycle of an Advanced Driver Assistance System (ADAS) emphasis is always placed on passenger safety, which directly correlates to adequate and extensive testing and validation procedures. However, testing and validation of ADAS systems is not a simple task and cannot be performed in a similar manner as conventional testing approaches. The main challenges is the vast amount of scenarios and environment parameter variations that might occur during operation. Current validation and testing procedures mostly rely on real world tests conducted on roads; however, because of the cost and complexity, these tests are not exhaustive and compromises on the scenario types and the number of considered parameters are made. In this paper we propose a model-based validation approach performed in a SiL (Soft-ware in the Loop) environment to validate an ADAS system under various conditions and the proposed methodology is presented on a Highway Pilot case study.