This Project deals with the problem of avoiding performance degradation of a control system despite of the presence of model uncertainty. The presence of uncertainty is an inherent fact of every model, so it must be dealt with when thinking of a control system. On the other hand it is well known the trade off between uncertainty and performance: uncertainty always imposes a constraint (sometimes very hard) on achievable performance. This way, it is difficult to deal with both aspects in order to get a robust control system without being excessively conservative. The usual approach is that of adding robustness constraints in the controller design step in order to get a Robust controller. Those constraints will be depending on the uncertainty formulation and sometimes not well suited for performance specification. The final result uses to be an algorithm not always easy to apply. More concisely, this project attempts to a) tackle this problem from the point of view of the control configuration instead as the more classical approach of direct robust control design; b) to show the higher potential of two degree of freedom control configurations with respect to more classical configurations; c) develop a reformulation of well known control algorithms (optimal-predictive controllers) explicitly formulated on those control configurations and taking advantage of it, and d) show how it is possible to extend local controller performance to larger operating regions by using two degree of freedom control configurations.
|Effective start/end date||13/12/04 → 13/12/07|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.