Event-triggered practical predefined-time control for Euler-Lagrange systems against deception attacks

An event-triggered practical predefined-time controller is constructed in this paper for a class of Euler-Lagrange systems under deception attacks. A novel speed function is introduced to reduce the computational complexity, while novel coordinate transformations are used to reduce the controller design complexity. Combining the novel speed function and barrier Lyapunov function, the closed-loop system converges to any desired neighborhood within the preset timeframe with excellent transient performance. Due to the presence of deception attacks, the Nussbaum gain method and fuzzy universal approximation are used to eliminate attack effects, while the event-triggered mechanism based on the relative threshold is introduced to reduce communication resources and the probability of being attacked. Finally, the efficiency of the proposed method is validated through a robotic arm simulation.