Inverse-Free and Data-Driven Motion Tracking Control for Redundant Robot with Fuzzy Recurrent Neural Network
Min Yang, Siying Zhu, Hui Zhang
Abstract
Precise motion tracking control with unknown structural knowledge and noise disturbance for redundant robots remains a critical and unresolved challenge. This article proposes a novel data-driven fuzzy discrete recurrent neural network (D<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>-FDRNN) model to address two fundamental limitations of existing models: dependency on known kinematic knowledge and fixed sampling schemes. First, a Jacobian pseudo-inverse estimator is developed to reconstruct the manipulator’s necessary kinematic knowledge using input and output data, eliminating the need for explicit Jacobian inversion. Second, a fuzzy logic-based adaptive sampling strategy dynamically adjusts the step size to balance computational efficiency and tracking precision. In addition, a Kalman filter algorithm is applied to reduce the impact of noise. Rigorous proofs confirm the model’s exponential convergence and noise immunity. To validate the proposed D<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>-FDRNN model, simulations and physical experiments are carried out. The source code is available at https://github.com/YingluckZ/DD-FDRNN.git.
BibTeX
@inproceedings{iros2025_inversefreeandda,
title = {Inverse-Free and Data-Driven Motion Tracking Control for Redundant Robot with Fuzzy Recurrent Neural Network},
author = {Min Yang and Siying Zhu and Hui Zhang},
booktitle = {IROS 2025},
year = {2025}
}