An Iterative‐Learning‐Based Finite‐Time Prescribed Performance Control for Periodic Trajectory Tracking

An iterative learning‐based prescribed performance control (IL‐PPC) scheme for high‐precision tracking of periodic trajectories in nonlinear systems is proposed. Although conventional prescribed performance control (PPC) ensures transient and steady‐state bounds, it may suffer from lingering errors in periodic tasks. To address this, the IL‐PPC scheme is developed by integrating an iterative learning (IL) law into a PPC architecture. In this scheme, a novel prescribed performance function is first developed to guarantee tracking error convergence within a predefined time while inherently suppressing transient overshoot. The PPC architecture is then constructed within a backstepping framework, which incorporates an extended state observer to estimate and reject lumped uncertainties in real time. To eliminate repetitive error components, an IL law is further embedded into the PPC architecture, which utilizes historical trial data to iteratively improve performance. Comparative simulations on a robotic manipulator demonstrate that the proposed IL‐PPC scheme achieves superior tracking accuracy and robustness, outperforming conventional performance‐based methods.