Dynamic programming for 2D discrete nonlinear systems

Abstract

In this paper the dynamic programming approach is generalized to 2D discrete nonlinear systems. In particular, a generalized Bellman function is introduced and then to develop necessary and sufficient optimality conditions for a nonlinear optimization problem associated with the systems considered. Finally, with a view to application to examples, an alternative to this function is developed which gives more computationally tractable necessary conditions for optimality.