An Energy-aware Mutation Testing Framework for EAST-ADL Architectural Models

Early design artifacts of embedded systems, such as architectural models, represent convenient abstractions for reasoning about a system’s structure and functionality. One such example is the Electronic Architecture and Software Tools-Architecture Description Language (EAST-ADL), a domain-specific architectural language that targets the automotive industry. EAST-ADL is used to represent both hardware and software elements, as well as related extra-functional information (e.g., timing properties, triggering information, resource consumption). Testing architectural models is an important activity in engineering large-scale industrial systems, which sparks a growing research interest. Modern embedded systems, such as autonomous vehicles and robots, have low-energy computing demands, making testing for energy usage increasingly important. Nevertheless, testing resource-aware properties of architectural models has received less attention than the functional testing of such models. In our previous work, we have outlined a method for testing energy consumption in embedded systems using manually created faults based on statistical model checking of a priced formal system model. In this paper, we extend our previous work by showing how mutation testing] can be used to generate and select test cases based on the concept of energy-aware mutants– small syntactic modifications in the architectural model, intended to mimic real energy faults. Test cases that can distinguish a certain behavior from its mutations are sensitive to changes in the model, and hence considered to be good at detecting faults. The main contributions of this paper are: (i) an approach for creating energy-related mutants for EAST-ADL architectural models, (ii) a method for overcoming the equivalent mutant problem (i.e., the problem of finding a test case which can distinguish the observable behavior of a mutant from the original one), (iii) a test generation approach based on UPPAAL Statistical Model Checker (SMC), and (iv) a test selection criteria based on mutation analysis using our MATS tool.