A model-based approach to test automation for context-aware mobile applications

Current testing tools for mobile applications do not provide sufficient support for context-aware application testing. In addition to regular input vectors (e.g. touch events, text entry) context parameters must be considered (e.g. accelerometer data interpreted as shake gestures, GPS location data, etc.). A multitude of possible application faults resulting from these additional context parameters requires an appropriately selected set of test cases. In this paper, we propose a model-based approach to improve the testing of context-aware mobile applications by deducing test cases from design-time system models. Using a custom-built version of the calabash-android testing framework enhanced by an arbitrary context parameter facility, our approach to test case generation and automated execution is validated on a context-aware mobile application.