Environment-Aware and Human-Centric Software Testing Framework for Cyber-Physical Systems

Abstract

The functionalities, actuations and effects that are produced by an application of a cyber physical system (CPS) are usually consumed by users while they perform their daily activities. Therefore, it is critical to ensure that they do not interfere with human activities and do not harm the people who are involved in the CPS. In this paper, we propose a framework to test and verify the reliability and safety of CPS applications in the perspectives of CPS environments and users. The framework provides an environment-aware testing method by which the efficiency of testing CPS applications can be improved by prioritizing CPS environments, and by applying machine learning techniques. The framework also includes a metric and an algorithm by which we can test and choose the most effective services that can deliver effects from their associated physical devices to users. In addition, the framework provides a computational model to test whether a CPS application may cause a cognitive depletion or contention problems for users.