A Multiobjective Output-Feedback Framework for Fully Distributed Resilient Consensus of Multiagents Under Stochastic DoS-Attacks

초록

This article deals with a multifaceted output feedback consensus control approach for multiagent systems (MASs) under stochastic denial-of-service (DoS) cyber-attacks. First, a fully distributed resilient cooperative state estimation approach to estimate the leader's state by followers over a sensor network has been devised. The proposed estimation approach ensures robust state estimation, exponential convergence, and stability of the adaptive mechanism against perturbations. Then, a new multiobjective fully distributed secure output-feedback consensus control approach under DoS attacks is proposed for achieving the leader-following consensus among MASs over a directed network. Several objectives, like fully distributed consensus, reliance on both leaders' and followers' outputs, robustness against disturbances, exponential convergence of estimation, and consensus errors, have been considered in the proposed resilient consensus methodology. In contrast to existing methods, the proposed secure consensus approach does not require the global information of network connectivity, is applicable to directed communication topologies, is robust against perturbations, and employs outputs of both leader and followers for feedback. An illustrative simulation study is presented for F-18 aircraft systems to demonstrate the effectiveness of the proposed methodologies for secure cooperative state estimation and resilient consensus control.

키워드