ARCHIVES

Recent years have seen significant advancements in the cooperative control of nonlinear multi-agent systems, driven by the demand for greater resilience, adaptability, and intelligence in modern cyber–physical environments. Conventional consensus control strategies often fall short when confronted with system nonlinearities, uncertain dynamics, communication delays, and emerging cyber threats. In response, contemporary research has shifted toward the integration of adaptive, fuzzy, and learning-enhanced control mechanisms, frequently combined with event-triggered and prescribed-time coordination strategies. Adaptive and fuzzy control schemes have been increasingly utilized to achieve secure and fault-tolerant consensus tracking in complex settings that involve actuator or sensor failures, switching dynamics, and stochastic disturbances. Parallel to this, reinforcement learning and adaptive dynamic programming techniques have enabled experience-driven optimization of cooperative behavior through value-based and Hamiltonian-guided learning frameworks. Event-triggered and fixed-time consensus protocols have also gained prominence for their ability to reduce communication burden while ensuring convergence within finite or pre-assigned time intervals. Additionally, recent studies have emphasized improving robustness against cyber-attacks, time delays, and external uncertainties through predictive, fuzzy, and hybrid control strategies. Overall, the emerging trend reflects a transition from classical model-based paradigms to hybrid intelligent approaches that leverage fuzzy logic, neural networks, and learning-based adaptation to achieve secure, efficient, and highly responsive coordination. This review consolidates these developments, identifies critical research gaps, and outlines future directions for advancing intelligent, fault-tolerant, and resource-efficient consensus control of nonlinear multi-agent systems.