DEGO-ILC: an iterative learning variable gain controller for enhancing trajectory tracking precision in industrial robots

Purpose This paper aims to propose a data-driven iterative learning control (ILC) scheme based on swarm intelligence optimization to address the issues of excessive gain parameters, inaccurate models and slow convergence in iterative learning algorithms and improve the accuracy and robustness of industrial robot trajectory tracking. Design/methodology/approach The ILC gain matrix is formulated using a two-dimensional Gaussian distribution (GD). A differential evolution gain optimization (DEGO) method is then developed to determine the optimal GD coefficients, improving tracking accuracy. Finally, an iterative learning strategy is implemented, and physical experiments with various trajectories validate the results. Findings The proposed ILC method based on DEGO converges faster than conventional approaches, achieving higher trajectory tracking accuracy with the optimized gain matrix compared to conventional ILC methods. Originality/value This work introduces a self-adaptive ILC framework that prioritizes error data from previous iterations over treating all past errors equally. By incorporating a Gaussian-distribution-based gain adjustment strategy and heuristic optimization, the proposed method enhances tracking accuracy and adaptability, significantly improving trajectory tracking performance in industrial robots under nonlinear dynamics and model uncertainties.