Neural Adaptive Fixed-Time Consensus Control for Nonlinear Multi-Agent Systems with Prescribed Performance

This paper proposes a neural adaptive fixed-time consensus control approach for the nonlinear multi-agent system (MAS) with prescribed performance. First, by employing an error transformation, a prescribed performance consensus control strategy is presented, which can steer the consensus errors to the predetermined range in a fixed time. Then, the unknown nonlinear functions are approximated by utilizing neural networks and the adaptive backstepping technique. Meanwhile, a first-order filter is adopted to avoid taking derivatives of the virtual controller. With the proposed strategy, all signals of the MAS remain bounded and the fixed-time leader-following consensus control performance is ensured. Finally, these results are illustrated by simulations.