Neural-Network-Based Adaptive Funnel Control for Strict-Feedback Systems with Tracking Error and State Constraints

In order to deal with the error/state constraints of the strict-feedback systems with the unknown quantization pareme-ters and disturbances, an adaptive control strategy based on funnel variables and neural networks (NNs) is proposed. The original strict-feedback systems is transformed into a new systems without constrains by some funnel variables, and then a dynamic surface control (DSC) is used to stabilize the transformed systems, which guarantees the tracking error in a preset ferformance funnel and states in some bounded regions. The unknown disturbances are estimated and compensated by the NNs, and the quantization error of the control input is also considered by an adaptive way. Furthermore, the feasibility conditions of the virtual controllers in barrier Lyapunov function (BLF) are removed. The convergence of the transformed control systems is proved through the Lyapunov theory. Lastly, some simulations illustrate the effectiveness of the proposed control strategy.