A computationally tractable online system-identification scheme based on subspace-system identification (SSI) was developed. The applicability of SSI algorithm was extended to identify a nonlinear process model and subsequently designed an output feedback robust controller based on the identified model. The applicability of the proposed identification scheme was demonstrated on a kinetic Monte Carlo realization of a simplified surface reaction scheme for the dynamics of NO oxidation by H2 on Pt and Rh surfaces. The system is characterized by two "coarse" stable stationary states. An output feedback controller to successfully switch the process from one stationary state to the other was designed. This is an abstract of a paper presented at the AIChE Annual Meeting (Salt Lake City, UT 11/4-9/2007).