Work

Resilient Secure Consensus Tracking Control for Nonlinear Multi-Agent Systems: A Reinforcement Learning-Based Adaptive Approach

Year: 2026

Type: article

Abstract: This paper proposes an optimized consensus tracking control scheme based on reinforcement learning for a class of nonlinear multi-agent systems (MASs) subject to denial-of-service (DoS) attacks. Considering that communication channels may be corrupted by attackers, a distributed state observer with a switching mechanism is developed to estimate the state information between the leader and followers. To address the issue where the observed variable is limited to the first-order derivative due to DoS attacks, a low-pass filter chain is employed to obtain higher-order derivatives. Furthermore, with the aid of an auxiliary system, a disturbance observer is constructed to estimate external disturbances. To achieve optimized control, an identifier-actor-critic neural network is employed to approximate the solution of the Hamilton–Jacobi–Bellman (HJB) equation. Finally, this proposed optimized consensus control can smoothly steer the nonlinear MASs to realize the consensus of multiple system states, and the effectiveness of this control scheme is verified by simulations.

Source: IEEE Internet of Things Journal

DOI: 10.1109/jiot.2026.3729599

Authors: Zhen Gao, Guangdeng Zong, Huanqing Wang, Xudong Zhao, Guangjing Song

Institutions: Bohai University, Tiangong University, Weifang University

Language: English

Topic: Distributed Control Multi-Agent Systems

Subfield: Control and Systems Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $2,645

Funder: National Natural Science Foundation of China