Robust output tracking for strict-feedback systems using neural-net based approximators for nonlinearities

We consider a class of SISO nonlinear systems in strict-feedback form with additional stable zero dynamics and unknown nonlinearities. The only assumption we make on these nonlinearities is that when they are approximated in terms of radial basis functions, the corresponding optimal parameters lie in a known compact set. We address the question of designing a robust controller under which the system output tracks a given signal arbitrarily well, and all signals in the closed-loop system remain bounded.