Model Predictive Control with generalized terminal state constraint

The properties of a particular terminal state constraint for the design of Model Predictive Control (MPC) laws are reviewed, and the technique is applied to an inverted pendulum example. The new approach, named “generalized” terminal state constraint, induces the recursive feasibility of the underlying optimization problem and recursive satisfaction of state and input constraints; moreover, it may provide, in general, a larger feasibility set with respect to existing approaches, given the same computational complexity.