Bo Zhao, Yuan‐Xin Li, Zhongsheng Hou
This article focuses on the problem of dual-channel event-triggered (ET) adaptive neural network (NN) fixed-time (FXT) optimal consensus for nonlinear multiagent systems (MASs) with unknown control directions. A dual-channel ET FXT tracking strategy is applied to address the FXT optimal consensus problem. First, to save communication resources, ET mechanisms are designed for both the sensor-to-controller (S–C) channel and the controller-to-actuator (C–A) channel. In addition, to deal with the nondifferentiability issue of the virtual controllers caused by the ET mechanism, a switching transformation function is designed to reconstruct the output signals. Based on the reconstructed signals, an FXT control algorithm is devised by utilizing the backstepping technique, where the Nussbaum functions are designed to compensate for the negative impact of unknown control directions. Then, the actor-critic architecture-based reinforcement learning (RL) technique is implemented at each step to generate optimal control commands, ensuring optimal performance while ensuring consensus tracking. With the Lyapunov stability theory, it is verified that the designed control algorithm can guarantee the achievement of consensus tracking in a fixed time. Finally, the effectiveness of the designed algorithm is demonstrated by a representative simulation.