Main research findings
Frequent automated software testing is a crucial task for modern software development. It has the goal to evaluate a software's functionality and be confident about its quality after recent changes and before the integration of new features or its deployment into the actual production environment.
Further challenges are introduced when testing software for cyber-physical systems that integrate both software and dedicated hardware components, e.g. industrial robots or embedded devices.
This thesis explores how machine learning and constraint optimization can be leveraged to achieve the desired efficiency and to create an intelligent testing process. Specifically, we contribute new methodology for the test suite optimization process of test case prioritization, test case scheduling, and test case selection and assignment. All of these steps are relevant to decide which test cases are most relevant and when to execute them on which test hardware, e.g. an industrial robot. The results of this thesis have been published in international venues and are in production usage at our industrial partner, ABB Robotics Norway.
Before the defence,Helge Spieker presented his trial lecture"Modularization techniques in artificial neural networks".
The PhD defence and trial lecture willbe fully digital.
- Professor Lars Grunske,Department of Computer Science, Humboldt University of Berlin, Germany
- Professor Christel Vrain,Laboratoire d'Informatique Fondamentale d'Orléans (LIFO), University of Orléans, France
- Professor Xing Cai,Department of Informatics, University of Oslo, Norway
- Chief Research Scientist Arnaud Gotlieb,Simula Research Laboratory, Norway
- Professor Magne Jørgensen, Department of Informatics, UiO, Norway
- Dr. Morten Mossige, ABB Robotics, University of Stavanger, Norway
Chair of defence
- Associate Professor Petter Nielsen, Department of Informatics, UiO