1: \begin{abstract}

2: 		Operator splitting techniques have recently gained popularity in convex optimization problems arising in various control fields.

3: 		Being fixed-point iterations of nonexpansive operators, such methods suffer many well known downsides, which include high sensitivity to ill conditioning and parameter selection, and consequent low accuracy and robustness.

4: 		As universal solution we propose \emph{SuperMann}, a Newton-type algorithm for finding fixed points of nonexpansive operators.

5: 		It generalizes the classical \KM{} scheme, enjoys its favorable global convergence properties and requires exactly the same oracle.

6: 		It is based on a novel separating hyperplane projection tailored for nonexpansive mappings which makes it possible to include steps along any direction.

7: 		In particular, when the directions satisfy a Dennis-Moré condition we show that \emph{SuperMann} converges superlinearly under mild assumptions, which, surprisingly, do not entail nonsingularity of the Jacobian at the solution but merely metric subregularity.

8: 		As a result, \emph{SuperMann} enhances and robustifies all operator splitting schemes for structured convex optimization, overcoming their well known sensitivity to ill conditioning.

9: 	\end{abstract}