From modeling to test case generation in the industrial embedded system domain

Abstract

Model-based testing (MBT) is the process of generating test cases from specification models representing system requirements and the desired functionality. The generated test cases are then executed on the system under test in an attempt to obtain a pass or fail verdict. While different MBT techniques have been developed, only a few target the real-world industrial embedded system domain and show evidence on its applicability. As a consequence, there is a serious need to investigate the use of MBT and the evidence on how modeling and test generation can improve the current way of manually creating test cases based on natural language requirements. In this paper, we describe an on-going investigation being carried out to improve the current testing processes by using the MBT approach within an industrial context. Our results suggest that activity and structure diagrams, developed under MBT, are useful for describing the test specification of an accelerator pedal control function. The use of MBT results in less number of test cases compared to manual testing performed by industrial engineers.