SdMobi05 - TESSOF Standardized Test Procedure for Highly Configurable Software Defined Mobility Systems in Operation.

Problem definition

With individually configurable mobility systems, manufacturers are setting themselves apart from their competitors in a saturated market. Due to the growing software share in these systems and trends - such as high-frequency over-the-air updates and increasing vehicle hardware runtimes - individualization is increasingly taking place at the software level and while the vehicle is still in operation. The resulting wealth of variants in mobility systems poses a major challenge for the efficient validation of these systems, for which standardized test procedures are currently lacking.

Project goals

The aim of the project is to develop a new generation of test procedures for the efficient and systematic validation of highly configurable mobility systems. The standardizable test procedures should be able to cope with the variability of the software in space (many configuration possibilities) and time (updates during life cycle). This will create procedures that can be applied throughout the lifecycle to achieve high software quality.

Implementation

Major deficits of current test procedures are addressed, which are mainly caused by a missing comprehensive variability model, inefficient configuration selection and test case generation. Based on a unified variability model, novel methods of risk-based adaptive sampling will be used to extract a set of configurations that are relevant for the verification of a system change. For these configurations, test cases will be generated in the next step using fuzzing techniques, which will finally be executed in a test process.

ITIV Involvement

• Establishment of a test process infrastructure for OTA updates in the form of an extensible cloud edge architecture.
• Create a CI/CD pipeline for updates with extensible interface for standardized test procedures and explore a maturity metric for software updates
• Explore and integrate adaptive sampling techniques and associated metrics for evaluating the necessary coverage of a software update
• Evaluate the experimental setup