Type-3 Fuzzy Data-Driven Control of Heterogeneous Multi-Agent Systems
Yaqing Liang, Man‐Wen Tian, Du Changdong, Khalid A. Alattas, Afef Fekih, Ardashir Mohammadzadeh · IEEE Access · 2025
This study investigates the data-based consensus problem of discrete-time heterogeneous multi-agent systems (MASs) through the development of distributed protocols. A type-3 (T3) fuzzy logic system (FLS) is presented to effectively model the inherent nonlinearities present in MASs, allowing for more accurate representation of agent dynamics. Utilizing a Lyapunov stability approach alongside a linear matrix inequalities (LMIs) framework, the sufficient conditions are derived for designing the data-driven distributed control protocols. The results ensure that the consensus error dynamics are asymptotically stable, thereby guaranteeing the achievement of consensus among agents. Furthermore, the optimal linear quadratic regulator (LQR) problem is addressed within the context of heterogeneous MASs, providing an additional layer of control performance optimization. Also, the design procedure is analyzed under conditions of noisy data, which is critical for real-world applications where sensor inaccuracies are prevalent. To validate the theoretical findings, extensive simulations are conduced that showcase the effectiveness and robustness of the proposed methods. The implementation is demonstrated on a network comprising four multiple autonomous underwater vehicles (AUVs), hyper-chaotic systems, and various configurations of discrete-time MASs. The results highlight the advantages of suggested data-driven control approach, emphasizing its practicality and efficiency in achieving consensus in complex multi-agent environments. This work paves the way for future research in robust control strategies for heterogeneous physical and financial systems operating under uncertainty. For example, by the use of the suggested data-driven approach to achieve consensus among heterogeneous agents, the financial institutions can enhance their decision-making processes, and optimize performance under uncertainty.