Comments on "Asymptotic state tracking in a class of nonlinear systems via learning-based Inversion"

In the above paper, a learning approach to asymptotic state tracking for a class of nonlinear systems was proposed. The central idea of the approach is to first establish an equilibrium which is in essence an oscillatory steady-state, then incorporate a learning mechanism to eliminate the discrepancy between the desired and actual control. The paper presented a new analysis tool for the learning control problems, which is quite different from most existing tools developed insofar for learning control analysis. In this note, we would like to address some existing technical problems in the above paper, as well as the potential limitations in implementation.