Next-generation precision motion systems are lightweight to meet stringent requirements regarding throughput and accuracy. Such lightweight systems typically exhibit lightly damped flexible dynamics in the controller cross-over region. State-of-the-art modeling and motion control design procedures do not deliver the required model complexity and fidelity to control the flexible dynamical behavior. The aim of this paper is to develop a combined system identification and robust control design procedure for high performance motion control and apply it to a wafer stage. Hereto, new connections between system identification and robust control are employed. The experimental results confirm that the proposed procedure significantly extends existing results and enables next-generation motion control design.

The performance of robust controllers hinges on the underlying model set. However, at present it is unclear which properties of the physical system should be accurately identified to enable high performance robust control. The aim of this paper is to clarify the intimate relation between quality of certain physical system properties and the resulting control performance. Hereto, an extended robust-control-relevant system identification methodology and a new visualisation approach is developed that is applicable to multivariable systems. The developed methodology is applied to an industrial wafer stage system. Experimental results indeed confirm that the developed techniques contribute to clarifying the complex relation between system identification and robust control. I. INTRODUCTIONThe quality of approximate models hinges on their purpose. This observation has led to the development of controlrelevant system identification techniques, including [1], [2], that aim at identifying nominal models that accurately represent those phenomena that are relevant for subsequent control. A typical characteristic of such schemes is their inherently iterative nature [3].Iterative identification and control design approaches that are solely based on nominal models have resulted in various outcomes [4]. In fact, discrepancies between the true system and the model may result in divergence of these iterative schemes. An approach to enhance convergence properties is to explicitly incorporate robustness to modeling errors through robust control.In the case that robustness is explicitly addressed during control design, then identification techniques should deliver a model set that enables the design of a high performance controller. Several approaches have been pursued to characterize and identify such model sets for robust control. In [5], the identification of model sets for robust control is investigated in a prediction error framework, where uncertainty is quantified in the parameter space, and control-relevance has been related to the size of the class of stabilizing controllers. An alternative definition of robust-control-relevance of a model set has been adopted in, e.g., [6], which is directly related to the ability of the model set to indeed deliver a high performance robust controller. In the latter case, extensive use is made of the dual-Youla-Kučera parameterization [7], [8]. However, due to an untransparent connection between the size of model uncertainty and the control criterion, the results in [6] seem to be mainly useful for SISO perturbation blocks and may lead to unnecessarily conservative results.Although several aspects of iterative identification and robust control design schemes have been thoroughly investigated, including the influence of weighting filters [9]

Abstract-Next-generation precision motion systems are lightweight to meet stringent requirements regarding throughput and accuracy. Such lightweight systems typically exhibit lightly damped flexible dynamics in the controller cross-over region. State-of-the-art modeling and motion control design procedures do not deliver the required model fidelity to control the flexible dynamical behavior. In this paper, identification and control challenges are investigated and a novel approach for next-generation motion control is presented. The procedure is applied to a multivariable wafer stage, confirming a significant performance improvement.