Stationary average consensus for high‐order multi‐agent systems

Abstract

The stationary average consensus problem is studied for high-order multi-agent systems (MASs) under balanced directed networks. The objective is to bring the positions of agents to the average of their initial positions while allowing all the remaining states (e.g., velocity, acceleration, and higher-order states) to converge to zero. To this end, the authors propose two consensus protocols for high-order MASs in two cases: (i) state-feedback control, which assumes that each agent has access to its own states as well as the relative positions of its neighbours; and (ii) output-feedback control, where each agent measures only its own position and the relative positions of its neighbours. Two case studies are given to illustrate the advantages and effectiveness of the proposed protocols. In the first case study, a state-feedback controller is designed for consensus-based formation control of a team of vertical take-off and landing aircraft. In the second case study, an output-feedback consensus protocol is designed for a third-order MAS and compare it with a recent technique from the literature.