On-line Parameter Estimation and Adaptation in Nonlinear Model-Based Control

The objective of this research is to demonstrate that improved control of nonlinear systems can be achieved in highly nonlinear operability regions when a periodically adapted nonlinear model is used in a nonlinear model predictive control (NMPC) strategy. In this paper, we develop a robust and efficient on-line parameter estimation scheme capable of adapting the nonlinear model for implementation in the nonlinear controller. We discuss the ramifications of nonlinear control in the absence of estimation, and we consider how the choice of parameters to be estimated impacts controller performance.