Master thesis 30 ECTS: Multi-processor scheduling through optimisation methods
Scheduling of safety-critical real-time systems
At Saab we continuously strive forward and challenge the limits of what is technically possible. We collaborate with colleagues around the world that all share our challenge – to make the world a safer place.
Saab serves the global market of governments, authorities, and corporations with products, services and solutions ranging from military defence to civil security. Aeronautics is one of the business areas within Saab and is, among other things, responsible for the development, production, marketing, sales, and support of the Gripen system.
In a modern aircraft, there are multiple computers, interconnected through a communication network, that interact with sensors, actuators, and mechanical components within the aircraft. This system, called an aircraft’s avionic system, is responsible for many critical functions such as steering, navigation, and communication. It is of utmost importance that such functions behave as expected. One part of ensuring this can be to schedule all computer activities beforehand. This leads to the topic of this thesis project: static scheduling of multi-processor systems.
The embedded real-time systems grow ever larger, increase in complexity, and become more interconnected. With this comes scheduling problems of increasing difficulty. To utilise system resources efficiently, and thereby avoid excessive hardware requirements, it is not enough to find any schedule, one must find a schedule of sufficient quality. At Saab, we therefore research methods to solve such large-scale scheduling problems in an efficient manner.
Description of the master thesis
This Master Thesis aims to further Saab’s competence within a specific area related to scheduling. The work will involve design of an optimisation method, implementation of the method, and evaluation against an existing collection of input data. The specific direction of the project will be decided together with your supervisor(s). It could, for example, be to investigate how graph-based machine learning can be used to guide the search of a metaheuristic, how a certain decomposition design can aid in faster solving a subproblem that arises in a scheduling tool, or if approximation algorithms can help with “warmstarting” other scheduling methods. Other directions for the project may also be possible.
This Master Thesis is suitable for 1-2 students with interest in optimisation and/or scheduling.
You are at the end of your studies within science/engineering and about to start your Master Thesis work for 30 ECTS. Knowledge in the area of optimisation at the level of a introductory undergraduate course is expected. Basic proficiency in programming is also required. Additional knowledge or experience within optimisation and/or algorithms is meriting.
What you will be a part of
Saab is a company full of opportunities, with more than 15,500 employees and operations on all continents. Gripen E is one of the most technologically advanced products currently being developed in Sweden and your project will be a part in ensuring the long-term sustainability and development capability of the Gripen E.
Last application day
Hannes Uppman, PhD, contact/supervisor
Emil Karlsson, PhD, contact/supervisor
Martin Gollvik, Manager
If you aspire to help create and innovate whilst developing yourself in a challenging team setting, Saab may well have the perfect conditions for you to grow. We pride ourselves on a nurturing environment, where everyone is different yet we share the same goal – to help protect people.