Multiparametric nonlinear integer programming and explicit quantized optimal control

This paper deals with multiparametric nonlinear integer programming problems where the optimization variables belong to a finite set and where the cost function and the constraints depend in an arbitrary nonlinear fashion on the optimization variables and in a linear fashion on the parameters. We examine the main theoretical properties of the optimizer and of the optimum as a function of the parameters, and propose a solution algorithm. The methodology is employed to investigate properties of quantized optimal control laws and optimal performance, and to obtain their explicit representation as a function of the state vector.