Abstract
Mechatronic technologies are used in a wide range of industries, from aerospace to automotive, manufacturingand even to personal devices, such as cd/dvd players. Although their multidisciplinary natureprovides great functionalities, it is still one of the substantial challenges which frequently impede theirdesign process. Apart from this problem, an early system design evaluation while adhering to adaptabledesign requirements is still missing. In this paper we propose a SysML-based method for an IntelligentConceptual Design Evaluation of mechatronic systems, abbreviated as SysDICE. Particularly, wecontribute by, firstly, making use of SysML as a common modeling language for the engineering teaminvolved in the design process and secondly, by adopting a widely used, in artificial intelligence, patternrecognition tool, namely non-parametric regression, to support a multi-alternative design mechanism,with the aim of attaining the best combination of components' alternatives that suits a set of prioritizednumerical requirements. To evaluate our framework, we have conducted two design experiments: (1) atwo-wheel differential drive robot, and (2) a quad-rotor unmanned aerial vehicle. Results prove how ourframework can assist system engineers and support the design process.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 24th Benelux Conference on Artificial Intelligence (BNAIC 2012) |
| Editors | Jos W.H.M. Uiterwijk, Nico Roos, Mark H.M. Winands |
| Place of Publication | Maastricht |
| Publisher | Department of Knowledge Engineering, Maastricht University |
| Publication status | Published - 2012 |