BLF-based neural dynamic surface control for stochastic nonlinear systems with time delays and full-state constraints

Abstract

This paper investigates the adaptive neural tracking control problem for a class of stochastic nonlinear systems with time delays and full-state constraints in a unified framework for the first time. The time-delay terms of the controlled systems are compensated by novel Lyapunov–Krasovskii functionals. The asymmetric barrier Lyapunov function (BLF) is adopted to guarantee that the full states are always restricted within prescribed constraints. RBF neural networks are utilised to approximate the lumped unknown functions in the design process. Furthermore, the dynamic surface control (DSC) technique is employed to simplify the process of control design significantly. Stability analysis shows all closed-loop signals are SGUUB, and full-state constraints are not violated. Finally, simulation results confirm the effectiveness of the proposed control scheme.

KEYWORDS:

• Stochastic nonlinear systems
• full-state constraint
• Lyapunov–Krasovskii functionals
• neural networks
• barrier Lyapunov function
• DSC

Disclosure statement

No potential conflict of interest was reported by the author(s).