On control problems with bounded state variables