This paper proposes a novel fractional variable-order super-twisting sliding mode controller (FVO-STSMC) for robust trajectory tracking in robotic manipulators affected by random disturbances and parameter uncertainties. This paper aims to address issues of chattering and adaptability in conventional sliding mode controllers, particularly in stochastic multi-input multi-output systems.
The FVO-STSMC uses a variable-order fractional derivative, where the order is dynamically adjusted using a hyperbolic tangent function of the tracking error combined with a time-dependent sinusoidal term. The controller is analyzed theoretically and validated through simulations on a three-degree-of-freedom (DOF) planar robotic manipulator under Gaussian noise and parameter uncertainties.
Theoretical analysis confirms mean-square boundedness of tracking errors in the presence of stochastic disturbances and parameter variations. Simulation results demonstrate that the FVO-STSMC achieves lower tracking errors than both fixed-order fractional and standard super-twisting controllers, highlighting its superior robustness and precision.
This paper presents the first use of a variable-order fractional super-twisting sliding mode controller with dynamic order adjustment for robotic manipulators. The proposed method significantly advances adaptability, precision and robustness, providing valuable contributions to the field of advanced control systems.
