From Simulation Data to Test Cases for Fully Automated Driving and ADAS
Within this paper we present a new concept on deriving test cases from simulation data and outline challenging tasks when testing and validating fully automated driving functions and Advanced Driver Assistance Systems (ADAS). Open questions on topics like
- PDF / 1,081,770 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 51 Downloads / 224 Views
Audi Electronics Venture GmbH, Sachsstr. 20, 85080 Gaimersheim, Germany {christoph.sippl,harald.altinger}@audi.de 2 Department of Computer Science 7, Friedrich-Alexander-University, 91058 Erlangen, Germany {florian.inifau.bock,reinhard.german}@fau.de 3 Chair of Automotive Technology, Technical University of Munich, Boltzmannstr. 15, 85748 Garching, Germany [email protected] Abstract. Within this paper we present a new concept on deriving test cases from simulation data and outline challenging tasks when testing and validating fully automated driving functions and Advanced Driver Assistance Systems (ADAS). Open questions on topics like virtual simulation and identification of relevant situations for consistent testing of fully automated vehicles are given. Well known criticality metrics are assessed and discussed with regard to their potential to test fully automated vehicles and ADAS. Upon our knowledge most of them are not applicable to identify relevant traffic situations which are of importance for fully automated driving and ADAS. To overcome this limitation, we present a concept including filtering and rating of potentially relevant situations. Identified situations are described in a formal, abstract and human readable way. Finally, a situation catalogue is built up and linked to system requirements to derive test cases using a Domain Specific Language (DSL). Keywords: Virtual validation · ADAS Simulation · Test case generation · DSL
1
·
Fully automated vehicles
·
Introduction
Today’s driver assistance functions and emergency systems help to avoid accidents and support the driver in critical situations. As the system boundaries are clearly defined, test cases can easily be specified. On the contrary, Advanced Driver Assistance Systems (ADAS) and fully automated vehicles ensure safety and comfort while driving in a normal mode. Here, defining all relevant test cases poses problems for developers due to the large amount of dynamic objects, including pedestrians and cyclists in urban traffic as well as the variety of priority rules and traffic guidance. So far, comprehensive test concepts and structured c IFIP International Federation for Information Processing 2016 Published by Springer International Publishing AG 2016. All Rights Reserved F. Wotawa et al. (Eds.): ICTSS 2016, LNCS 9976, pp. 191–206, 2016. DOI: 10.1007/978-3-319-47443-4 12
192
C. Sippl et al.
test case generation, such as Equivalence Class Partitioning (ECP), Boundary Value Analysis (BVA) and Predicate Testing improve the efficiency of software testing as stated by Eo et al. [1], but are not well-suited to identify all possible and relevant situations for fully automated vehicles and ADAS. Therefore, simulation-based development and new concepts for virtual validation are needed, instead of testing new driving functions with the help of many thousands of test kilometres. There are already numerous tools for simulationbased function development and testing, thus driving simulators with realistic, environment-sensitive behaviour of road users (e
Data Loading...
