Qr-model based testing

Abstract

As reactive and embedded systems continuously interact with their environment, it is important to test as many as possible interactions. The use of qualitative models of the environment and hardware has the potential to provide test cases that might not be considered with traditional testing methods. We present an approach that derives abstract test cases from such models using qualitative reasoning, which is a well known artificial intelligence technique to represent and reason about physical behavior. For this purpose we introduce the underlying concepts of qualitative reasoning, show the test case generation process, and provide the results of a case study. Copyright 2008 ACM.