Bachelor and Master Theses

Title: Investigating the effects of various control and data flows on the timing behavior of distributed software control systems
Subject: Computer Science
Level: Advanced
Description: Background:
Development strategies for real-time embedded systems in the automotive and other vehicular applications domain are to an extent based on model- and component-based development approach. This approach uses models to describe functions, structures and other design artifacts; and supports the development of large software systems by integration of software components. It raises the level of abstraction for software development and aims to reuse software components and their architectures. There are several technologies to support such development, for example, Rubus [1] that is a collection of methods and tools for model- and component-based development of dependable embedded real-time systems. Rubus is developed by Arcticus Systems [1] in close collaboration with Mälardalen University. Several international companies mainly use Rubus for development of control functionality in vehicles. The Rubus concept is based around the Rubus Component Model (RCM) [2] and its development environment Rubus-ICE (Integrated Component development Environment) [3] which include modeling tools, code generators, analysis tools and run-time infrastructure. The overall goal of Rubus is to be aggressively resource efficient and to provide means for developing predictable, timing analyzable and synthesizable control functions in resource-constrained embedded systems.

Goal:
The goal of this thesis is to investigate the timing behavior of distributed software control systems with respect to various trigger and data flows in their software architectures. The investigation should be conducted using the Rubus-ICE tool suite. Within this context, a prototype (control systems application) should be developed. The prototype should run on Rubus real-time operating system. End-to-end timing analysis (e.g. [4]) of the prototype should be performed with respect to various trigger and data flows. Further, the operation of the prototype should be demonstrated with respect to various trigger and data flows on hardware/simulated platform. A detailed comparative evaluation of the prototype with respect to various trigger and data flows should also be carried out.

References
[1] Rubus models and methods, http://www.arcticus-systems.com/index.php?pageId=6.
[2] K. Hänninen et.al., “The Rubus Component Model for Resource Constrained Real-Time Systems,” in 3rd IEEE International Symposium on Industrial Embedded Systems, June 2008, http://www.mrtc.mdh.se/index.php?choice=publications&id=1464.
[3] Rubus-ICE, http://www.arcticus-systems.com/index.php?pageId=18.
[4] S. Mubeen, J. Mäki-Turja, and M. Sjödin, Support for end-to-end response time and delay analysis in the industrial tool suite: Issues, experiences and a case study, in Computer Science and Information Systems, vol. 10, no. 1, pp 453-482, Jan. 2013, http://www.mrtc.mdh.se/index.php?choice=publications&id=3163
Company: Arcticus Systems AB, kontaktperson: Kurt-Lennart Lundbäck
Start date: 2016-01-18
Prel. end date: 2016-06-05
Presentation date: 2016-06-02
Student: Nandu Kuttikkat Sreevalsan nkn14001@student.mdh.se
Prerequisites: Good programming skills in C/C++ are mandatory. Basic knowledge of real-time embedded systems is a must.
IDT supervisor: Saad Mubeen
saad.mubeen@mdh.se, +4621103191
Examinator: Jukka Mäki-Turja
Jukka Mäki-Turja
jukka.maki-turja@mdh.se, +46-21101466
Misc: Basic knowledge of control theory can be a plus.

Rapport och bilagor

Size

Senaste uppdatering


  • Mälardalen University |
  • Box 883 |
  • 721 23 Västerås/Eskilstuna |
  • 021-101300, 016-153600 |
  • webmaster |
  • Latest update: 2017.08.23