Model-free adaptive sliding mode control with adjustable funnel boundary for robot manipulators with uncertainties

Area of Science:

• Robotics
• Control Systems Engineering
• Applied Mathematics

Background:

• Nominal dynamics models in robotics are often inadequate for real-world applications.
• Robot manipulators frequently operate under uncertain conditions, necessitating robust control strategies.
• Existing control methods may struggle with precise trajectory tracking in the presence of system uncertainties.

Purpose of the Study:

• To propose a novel model-free adaptive sliding mode control (SMC) strategy for robot manipulators.
• To address uncertainties in robot dynamics and improve tracking performance.
• To introduce an adjustable funnel boundary for strict error confinement.

Main Methods:

• Utilizing time delay estimation (TDE) to accurately estimate unknown robot dynamics, enabling a model-free approach.
• Implementing a modified funnel function to constrain trajectory tracking errors within a defined, adjustable boundary.
• Developing a funnel nonsingular fast terminal sliding mode control (FSFT-SMC) scheme based on the transformed error variable.
• Designing an adaptive update mechanism with a single adaptive parameter tuned via the sliding surface to compensate for TDE errors.

Main Results:

• The proposed control scheme demonstrates superior transient and steady-state tracking performance.
• Simulation and experimental results validate the effectiveness and high-precision tracking capabilities of the approach.
• The model-free nature and adjustable funnel boundary contribute to robust control under uncertainties.

Conclusions:

• The developed model-free adaptive FSFT-SMC with an adjustable funnel boundary offers a highly effective solution for robot manipulator control.
• The approach successfully manages system uncertainties and achieves precise trajectory tracking.
• This method provides a significant advancement for robotic applications requiring high accuracy and adaptability.