Simplified Nonlinear Programs for NMPC Based on Active Set Construction

The nonlinear program arising in nonlinear model predictive control can be simplified by constructing candidate active sets for the successor state. Instead of modifying the optimization algorithm directly, we use these active sets to anticipate the relevant constraints for the next time step and to solve a simplified nonlinear program. Since active sets are in general valid for a set of initial states, an inherent robustness with respect to additive disturbances results.