*3.2.* H<sup>∞</sup> *Robust Control*

In the second approach, estimation and feedback are jointly computed solving a H<sup>∞</sup> robust control regulator. Even if the classical separation of H<sup>2</sup> does not apply in H<sup>∞</sup> controls [14,15], a weakly coupled state observer and estimated state feedback can still be recognized. Then, maintaining the feedback coefficients of the estimated states of interest ( *c*ˆ, ˆ *c*˙, *p*ˆ, ˆ *δ*) and setting to zero the remaining ones relative to unobservable or uncontrollable modes (it can be verified that this has very little influence on the closed-loop poles), the strategy of Equation (5) can be implemented. It is interesting to note the similarity of the performances obtained between the two approaches, as shown in the next section.
