Development of self-diagnosis tests system using a DSL for creating new test suites for integration in a cyber-physical system

Abstract

Testing Cyber-physical systems (CPS) requires highly qualified engineers to design the tests since its computational part is programmed in low-level languages. The origin of this work arises from the need to find a solution that optimizes this problem and allows abstracting the current methods so that the tests can be created and executed more efficiently. We intend to do this by creating a self-diagnosis tests system that allows us to automate some of the current processes in the creation and execution of test suites. The work presented here addresses the problem by creating a new self-diagnosis tests system that will guarantee the reliability and integrity of the CPS. In detail, this paper begins by exposing a study on the current state of the art of test automation, Keyword-driven Testing (KDT) methodology and Domain-specific Languages (DSL). A new modular and extensible architecture is proposed for self-diagnosis tests systems based on two main concepts: the creation of a DSL combined with the use of the KDT methodology, as well as a methodology to extend it and integrate it into a CPS. A new self-diagnosis tests system has been proposed that applies the proposed architecture proving that it is possible to carry out the self-diagnosis in real-time of the CPS and allowing the integration of any type of test. To validate the implementation of the system, 28 test cases were carried out to cover all its functionalities. The results show that all test cases passed and, therefore, the system meets all the proposed objectives.