Uniformly optimal control of linear time-invariant plants: Nonlinear time-varying controllers (Q1077354)

This paper continues the study of the advantages of using nonlinear compensators for linear systems. It is shown that, for the problem of minimizing the operator norm of the map that sends exogenous inputs into regulated variables, nonlinear controllers do not better than linear ones (assuming that the original plant is linear and time-invariant). If robust control is desired, the result is no longer true.