New results and new developments in Model Predictive Control

The aim of the work reported in this thesis is to investigate some fundamental issues, such as stability and robustness, and some new trends in the research on Model Predictive Control (MPC, also referred to as RHC, i.e. Receding Horizon Control), and then to conduct some implementations of MPC or RHC. In the theoretical work on stability and robustness of MPC, four new MPC algorithms are developed. The first algorithm employs non-linear terminal control law; the second one calculates terminal region and terminal weighting matrix separately and covers some popular existing MPC algorithms; the third one discards traditional terminal penalty but still guarantees stability and robustness; and the fourth one is developed for uncertain constrained systems with time-varying delays. Besides, a case study is conducted where MPC is applied to the flight control system of a nonlinear missile with non-axisymmetric airframes. The concept of RHC is then introduced into management and operational research. A novel methodology of integrating the RHC strategy with genetic algorithm is proposed first. Some case studies are reported focusing on introducing the concept of RHC into ATM (air traffic management) system, where three classical ATM problems, namely, arrival sequencing and scheduling at airports, online optimizing free-flight paths for commercial aircraft, and airport capacity management, are investigated by using RHC based methodology. Both theoretical algorithms and case studies are proved successful by simulation results.