Feedback linearization of discrete-time systems (Q1095854)

This paper is concerned with the feedback linearization problem for nonlinear discrete-time systems \(x(t+1)=f(x(t),u(t))\). That is, when does there exist a state feedback \(u=k(x,v)\) and a state space transformation \(x=\Phi (z)\), such that the resulting system is of the form \(z(t+1)=Az(t)+Bv(t)\). Necessary and sufficient conditions for the local solvability of this problem are given. These conditions ae expressed in tems of controllability distributions associated to the system, and are equivalent to those stated by \textit{J. W. Grizzle}, Analysis and optimization of systems, Proc. 7th Int. Conf., Antibes/France 1986, Lect. Notes Control Inf. Sci. 83, 273-281 (1986).