Adaptive Neural Network Finite-Time Fault-Tolerant Control for Nonlinear Multi-agent Systems

The paper proposes a novel adaptive neural network (NN) finite-time fault-tolerant control (FTC) scheme for nonlinear MASs subject to actuator faults and unknown symmetric output dead-zone. Unlike conventional asymptotic consensus strategies that exhibit slow convergence, the developed finite-time consensus framework ensures rapid convergence by introducing an arbitrary small positive constant and Nussbaum functions to simultaneously resolve the unknown control direction problem. Theoretical analysis demonstrates that all closed-loop signals remain bounded. Finally, simulation studies on MASs validate the superiority of the proposed method in terms of convergence speed, tracking accuracy and fault tolerance.