Event-Triggered Adaptive Preassigned Finite-Time Consensus Control for Multiagent Systems With Nonlinear Faults

• Published in: IEEE transactions on cybernetics Vol. 54; no. 12; pp. 7392 - 7403
• Main Authors: Salmanpour, Yasaman, Arefi, Mohammad Mehdi, Cao, Jinde
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.12.2024
• Subjects: Artificial neural networks; Asymptotic tracking control; Consensus control; Convergence; Estimation; finite-time consensus control; nonaffine nonlinear faults; Nonlinear dynamical systems; nonlinear multiagent systems (MASs); Stability analysis; Switches; switching event-triggered control (ETC)
• ISSN: 2168-2267; 2168-2275; 2168-2275
• DOI: 10.1109/TCYB.2024.3443352

This article investigates a novel neuro-adaptive barrier Lyapunov function (BLF)-based event-triggered preassigned finite-time consensus control with asymptotic tracking for the nonlinear multiagent systems. The proposed approach is designed to broaden the scope of application by considering the high-order nonstrict-feedback dynamics of each agent with dynamic uncertainties subject to external disturbances and nonaffine nonlinear faults. A neural network (NN) is employed to approximate the unknown nonlinear terms. By fusing the NNs and Butterworth low-pass filter technique, the issues arising from the nonaffine nonlinear fault are addressed. To save the communication resources, a novel dynamic event-triggered mechanism based on an enhanced switching threshold is suggested. Additionally, a novel concept called the preassigned finite-time performance function (PFTPF) is defined to improve the transient and steady-state performances as well as providing faster response. The key feature of the proposed adaptive BLF-based control based on the bound estimation method is the introduction of a smooth function with decreasing variable which not only ensures that all the signals remain bounded and the synchronization errors are restricted within the PFTPF but also guarantees that the tracking errors asymptotically converge to zero. Finally, an illustrative example is provided to verify the feasibility of the proposed control approach.