Adaptive Iterative Learning Based Trajectory Tracking Control of Robotic Arm

This paper proposes an adaptive iterative learning control method to solve the robotic arm trajectory tracking problem. Given uncertain time-varying parameters and external perturbations in the practical robotic arm model, an alignment condition mechanism is introduced, enabling the current iteration’s initial state to rely on the previous iteration’s end state, thus overcoming the strict zero initial error requirement of traditional iterative learning control algorithms. The proposed control scheme integrates position and velocity error feedback, parameter estimation, and perturbation compensation for high-precision trajectory tracking. Theoretical analysis via a composite energy function shows that the system’s tracking error monotonically converges to zero as the number of iterations increases. Comparative verification confirms that the proposed controller’s trajectory tracking ability significantly outperforms other controller.