Abstract:This paper proposes a distributed model predictive control algorithm for the consensus of nonlinear second-order multi-agent systems. At each update time, all the agents are permitted to optimize. A positively invariant terminal region and a corresponding auxiliary controller are developed for each agent. Furthermore, time-varying compatibility constraint is presented to denote a degree of consistency between the assumed trajectories and the actual trajectories of each agent. Given the designed terminal ingredients (terminal region, auxiliary controller, and terminal cost) and compatibility constraints, the recursive feasibility and closed-loop stability of the whole system are guaranteed. The simulation results are given to illustrate the effectiveness of the proposed approach. Copyright (c) 2016 John Wiley & Sons, Ltd.

Distributed Model Predictive Control for Consensus of Nonlinear Second-Order Multi-Agent Systems