Less costly manufacturing of cars thanks to software in time-critical computer systems
Jan 31, 2014 | Research/Cooperation
Mikael Åsberg’s doctoral thesis in Computer Science shows how the software in time-critical computer systems, used for instance in cars, can be made faster, safer and more effective. In the long term, this would lead to a less costly manufacturing of cars thanks to a more effective integration of software components in the final stage of the development process.
Mikael Åsberg presents his doctoral thesis Synthesis and Synchronization Support for Hierarchically Scheduled Real-Time Systems on 31 January at MDH in Västerås. The thesis is about structuring software in time-critical computer systems, that is software which must be able to complete tasks within restricted time limits, such as a few micro- or milliseconds. This type of software can be found in for instance cars, airplanes, TV sets and mobile phones.
– I think my thesis offers great usefulness and due to the fact that I present my results in terms of practical software solutions, I also think that they can be applied more quickly in companies. Just installing the new software is enough to be able to use them. For the automobile industry, for instance, this means lower cost and fewer software-related problems, says Mikael Åsberg.
Research results are available online
Modern time-critical computer systems are complex and difficult to analyse, certify and modify, and academic research is involved in the development of many new technologies for facilitating the analysis of time-critical systems. Mikael Åsberg’s thesis is focused on one of these new technologies. It divides the time period that is available to software in the processor into several large segments which, seen as discrete units, function as separate time intervals. These intervals can then be divided into smaller segments, and so on. As a result of this process, different software components must adhere to a strictly hierarchical time schedule, which eliminates for instance time-related software problems. The division can be made in time before the integration of different software components, which means that the integration stage at the end of the development process will have fewer unexpected negative consequences, which in turn prevents delays in projects and consequently also counteracts increased development expenses.
– I hope that this technology will make time-critical systems easier to analyse through dividing the system into smaller parts and then making it possible to analyse the smaller parts separately. In this manner, software systems can be developed faster and used more safely. The results in themselves have a practical application in software that we have developed. The plan was for the results to be directly applicable in real systems, and what makes this research project unique is that we have focused on making the results available to the public, in order for other researchers to be able to use the practical results for further developments, says Mikael Åsberg.
A position at Zenterio awaits Mikael
Mikael Åsberg was born and raised in Västerås, where he has also been a student in the Computer Science Programme at MDH 2004 – 2008. After Master’s degrees in both Computer Engineering and Computer Science, he continued to study Computer Science as a doctoral student at MDH. He has been living in Linköping for six months, and right after the public defence of his doctoral thesis, he will start working at Zenterio in Linköping, a company which develops complex software for digital TV boxes.
– Working with my thesis, I have learnt a great deal about what different time-critical systems look like in practice and the ways in which other researchers work with the challenges of this area, and this is knowledge that I hope to be able to apply to Zenterio’s systems. The most important thing for me at the moment is to start working in the industry and learn how computer systems are developed, but in the future I might also be interested in an academic career, says Mikael Åsberg.