Enhanced Control Strategies for Residual-Vibration Suppression and Adaptive Tracking in Flexible Manipulator Under Unknown Bounded Disturbances

Abstract

This paper proposes a hybrid control strategy for flexible-link manipulators that combines offline trajectory planning with adaptive tracking to suppress residual elastic vibration (REV) under unknown bounded disturbances. A rigid–flexible coupling dynamics model is first established using the floating-frame method and Lagrange's equations, forming a differential–algebraic system as the foundation for controller design and optimization. A parameterized cosine-based joint trajectory is optimized via particle swarm optimization to reduce endpoint deformation. The resulting trajectory is tracked using an adaptive sliding-mode backstepping controller enhanced with a nonlinear disturbance observer. Simulations show that the optimized trajectory reduces REV amplitude and shortens motion time compared with the reference trajectories considered in this study. Under the considered numerical conditions, the proposed controller yields smaller residual deformation than the selected proportional–derivative reference controller in both the disturbance-free and disturbed cases. These results indicate that, within the adopted modeling assumptions, disturbance description, observer applicability range, and actuator-amplitude-constrained setting, the proposed strategy can improve tracking performance and vibration suppression for the flexible manipulator.