Decentralised finite‐time consensus for second‐order multi‐agent system under event‐triggered strategy. Issue 4 (8th January 2020)

• Record Type: Journal Article
• Authors: Zhang, Liang; Zhang, Zexu; Lawrance, Nicholas; Nieto, Juan; Siegwart, Roland
• Is Part Of: IET control theory & applications. Volume 14:Issue 4(2020)
• Page Start: 664
• Page End: 673
• Publication Date: 2020-01-08
• Subjects: graph theory -- stability -- multi‐robot systems -- multi‐agent systems -- control nonlinearities -- distributed control -- network theory (graphs) -- decentralised control; event‐triggered schemes -- event‐triggered strategy -- agent states -- update frequency -- actuators -- consensus accuracy -- control protocol -- second‐order multiagent system -- finite‐time consensus problem -- SOMAS -- event‐triggered control strategy -- ideal controller behaviour -- desired update rate -- finite‐time control theory -- decentralised event‐triggered nonlinear protocol -- undirected communication graph -- connected communication graph; Control theory -- Periodicals; Automatic control -- Periodicals; 629.8312
• DOI: 10.1049/iet-cta.2019.0865
• Languages: English
• ISSNs: 1751-8644

Abstract

Abstract : In this technical note, the authors address the finite‐time consensus problem of a second‐order multi‐agent system (SOMAS) under an event‐triggered control strategy. This problem arises from the case where actuators in a SOMAS fail to follow the ideal controller behaviour exactly, due to either the failure of an embedded processor to maintain the desired update rate or a large time constant in the actuators, while still trying to maintain some performance merits from the finite‐time control theory. A decentralised event‐triggered non‐linear protocol is proposed based on only local information for the undirected and connected communication graph. The rigorous proof is then presented and shows that the consensus can be achieved in finite time even under event‐triggered schemes. Owing to the event‐triggered strategy, they proposed a technique that bounds the steady errors of agent states into a small region around zero in finite time and reduces the update frequency of actuators while excluding the Zeno behaviour. Finally, simulation results are provided to validate the correctness and effectiveness of the proposed method and demonstrate that it can adjust a composed performance metrics with respect to the convergent rate, consensus accuracy, and update frequency of actuators by tuning the parameters in the control protocol.