Distributed Data-Driven Adaptive Iterative Learning Control for Multiagent Systems with DoS Attacks

doi:10.1109/jiot.2026.3741550

Distributed Data-Driven Adaptive Iterative Learning Control for Multiagent Systems with DoS Attacks

2026-10-08

0

PRE

AI

DOI：10.1109/jiot.2026.3741550

Original

Original request for help

Save

Abstract

En 中文

This paper investigates a distributed data-driven adaptive iterative learning control (AILC) approach for the consensus tracking problem of uncertain nonlinear multi-agent systems (MASs) subjected to two types of denial-of-service (DoS) attacks. A novel attack characterization framework is first established along the iteration axis based on a switching topology, wherein models for connectivity-maintained and connectivity-broken DoS attacks are formulated. In order to achieve the consensus tracking, a data-driven model is established for the consensus error, based on which the proposed data-driven AILC scheme is then developed. Subsequently, by employing the contraction mapping principle, the convergence of the consensus error is rigorously proven, ensuring that the consensus tracking objective is achieved under connectivity-maintained DoS attacks. The analysis is then extended to hybrid DoS attack scenarios, in which the attacks alternate between connectivity-maintained and connectivity-broken phases. Within this extended framework, it is further demonstrated that the MASs can still achieve consensus tracking control despite the presence of hybrid DoS attacks, provided certain conditions are satisfied. Finally, numerical simulations are conducted to verify the effectiveness and robustness of the proposed control method.

Keywords:

Data-driven adaptive iterative learning control

multiagent systems

denial-of-service attacks

Distributed Data-Driven Adaptive Iterative Learning Control for Multiagent Systems with DoS Attacks

Abstract

Journal

Organization

Cited Papers