Nonlinear stability-constrained model predictive control with input and state constraints

In stability-constrained model predictive control (SCMPC) a stability constraint is propagated from stage to stage to limit magnitude of the state vector in a controllable form. For the unconstrained case, a sufficient condition that can be easily evaluated guarantees the stability constraint is a feasible contraction mapping. This paper presents sufficient conditions for guaranteed asymptotic stability when SCMPC is applied to nonlinear systems with arbitrary constraints on the control and state.