Event-Triggered Model Predictive Control for Compartmental Systems with Application to Congestion Control of Air Traffic Networks

Air traffic management aims to mitigate congestion in air traffic networks mainly caused by capacity constraints of air centers. In this paper, an air traffic network is effectively modeled as a compartmental system, and a model predictive control (MPC) approach with a steady state-input constraint is proposed to mitigate traffic congestion and achieve the departure demand as best as possible. An event-triggered scheme is designed to trigger the solution of the MPC optimization problem when necessary, leading to reduced computational and communication burden. Recursive feasibility of the proposed approach and asymptotic evolution of the system to a steady point are analyzed. The effectiveness of the proposed approach is demonstrated by a five-inflow and three-outflow air traffic network with ten air centers.