Research on Adaptive Control Method Based on Intelligent Unloading Device

In response to the challenges posed by external disturbances and uncertainties in complex operational environments for intelligent unloading devices, this paper proposes an adaptive control method based on a radial basis neural network observer and a global fractional-order terminal sliding-mode surface. First, an N-joint unloading device tracking error dynamics model is established to accurately depict the evolution of system errors in each joint during unloading operations. Next, a radial basis neural network observer using cubic B-spline functions as the activation function is adopted to estimate unknown model terms and external disturbances in real time, providing reliable compensation information for subsequent controller design. Then, leveraging a global fractional-order terminal sliding-mode surface, an adaptive isochronous convergence rate algorithm is designed to ensure fast and robust convergence of joint errors. Finally, an adaptive global fractional-order terminal sliding-mode controller is developed, and simulation results confirm the effectiveness of the proposed method in terms of tracking accuracy, robustness, and convergence speed. This provides a viable control strategy for intelligent unloading devices in practical industrial applications.