RBFNN-based nonsingular fast terminal sliding mode control for piezoelectric stack actuator

To enhance the performance of finite-time trajectory tracking control for piezoelectric stack actuator (PSA), this paper proposes an innovative control methodology by integrating Radial Basis Function Neural Networks (RBFNN) with Nonsingular Fast Terminal Sliding Mode Control (NFTSMC). In consideration of the inherent uncertainties within PSA systems, an NFTSMC strategy is formulated to ensure finite-time convergence of system states to the desired trajectory. To further refine control precision, the RBFNN is employed to compensate for lumped disturbances in real-time, with its weights adaptively updated via an adaptive law. The proposed controller incorporates both a robust control term and the RBFNN, effectively mitigating various external disturbances. Stability analysis, based on Lyapunov theory, confirms that the proposed RBFNN-NFTSMC scheme guarantees finite-time convergence of the trajectory tracking error, along with global stability. Simulation results substantiate the superior performance of the proposed control strategy compared to conventional methods.