Dynamic Neural Networks for Motion Control of Redundant Manipulators

This book discusses the development and application of dynamic neural networks (DNNs) for solving complex motion control problems in redundant manipulators. Specifically, it presents a series of advanced DNNs, including noise-rejection DNNs, fuzzy-parameter DNNs, and so on, which are designed to optimize performance while ensuring robustness and computational efficiency. Based on the presented DNNs, this book further constructs a series of motion control schemes for redundant manipulators to address some key challenges such as cyclic motion, position and orientation tracking, and model-unknown scenarios. Each method is rigorously demonstrated for the convergence, and its effectiveness is validated through simulations and physical experiments. By integrating computational intelligence with control theory, this book provides a comprehensive framework for solving time-varying and noise-perturbed problems in robotics, making it a valuable resource for researchers and practitioners in the field.