Adaptive Neural Network Control for Nonlinear Multi‐Agent Systems With Actuator Faults and Asymmetric Time‐Varying State Constraints
Year: 2025
Type: article
Abstract: ABSTRACT In the paper, we design an adaptive neural network fault‐tolerant controller (FTC) with the distributed sliding‐mode estimator. This controller is suitable for the a type of nonlinear multi‐agent systems (MASs) containing the actuator faults and the asymmetric time‐varying state constraints (ATVSCs). In practical applications, the actuator faults and ATVSCs will affect the stability of most systems and lead to security problems. The proposed controller ensures that system states remain within asymmetric time‐varying constraints and maintain stable operation under actuator faults. Neural network (NN) is employed to approach unknown functions in design procedure. We introduce a compensation function by the radial basis function neural networks (RBF NNs) to prevent taking derivative of virtual controller repeatedly. In the meantime, the asymmetric time‐varying barrier Lyapunov functions (ATVBLFs) are designed to keep state in the constraint range of asymmetric time‐varying. In addition, a distributed sliding‐mode estimator is designed to estimate reference signal, which simplifies error and makes the system respond quickly to external disturbances. For every agent system, there is only a need for one parameter adaptive law, which reduces the computational complexity of the controller. Finally, Lyapunov stability is proven through theoretical analysis, and it is proved to be feasible by an example.
DOI: 10.1002/rnc.7974
Authors: Guowei Dong, Peitong Xu, Ke‐Wen Li
Institution: Liaoning University of Technology
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
Awards: G7007359228, G8304752702
OpenAlex ID: W4409238524