Multi-resolution-Modeling for Testing and Evaluation of VANET Applications

Abstract

The evaluation and testing of future driving assistance systems based on Vehicular Ad-hoc Networks (VANETs) in real testbeds is difficult due to the need for repeatable scenarios and large-scale experiments. Therefore, a novel framework based on multi-resolution modeling is presented to test automotive software both accurately and efficiently in large-scale scenarios using virtual test drives in a simulated environment. The approach enables the precise and large-scale evaluation of real-world implementations. This is done through the synchronized execution of simulation models of multiple resolutions representing the vehicle and network domain, as well as the applications encapsulated in virtual Electronic Control Units. This paper provides a detailed and formal description of the applied multi-resolution methodology and explains the developed generic testing platform and the components it is comprised of.