Authors: Stefan Müller, Eberhard Karls Universität Tübin...; Dennis Hospach, Eberhard Karls Universität Tübi...; ; Oliver Bringmann, Eberhard Karls Universität Tü...;
Abstract:
To achieve a high test coverage of Advanced Driver Assistance Systems, many different environmental conditions have to be tested. It is impossible to build test sets of all environmental combinations by recording real video data. Our approach eases the generation of test sets by using real on-road captures taken at normal conditions and applying computer-generated environmental variations to it. This paper presents an easily integrable framework that connects virtual prototypes with varying sensor perceptions. With this framework we propose a method to reduce the required amount of on-road captures used in the design and validation of vision-based Advanced Driver Assistance Systems and autonomous driving. This is done by modifying real video data through different filter chains. With this approach it is possible to simulate the behavior of the tested system under extreme conditions that rarely occur in reality. In this paper we present the current state of our virtual prototyping framework and the implemented plug-ins.
Publication Date: 2015/03/04
Location of Publication: Methoden und Beschreibungssprachen zur Modellierung und Verifikation von Schaltungen und Systemen (MBMV2015), Chemnitz, DE
Keyword: Specification/Virtual Prototyping
