Kinematic Control of Redundant Robot Arms Using Neural Networks

Redundant robot arms offer more joint motion than a task strictly requires—but turning that flexibility into controlled, useful movement calls for careful mathematics. This research-focused book develops neural-network approaches to the kinematic control problem, connecting algorithm design with theoretical analysis and simulation.

Across three parts, Shuai Li, Long Jin, and Mohammed Aquil Mirza address serial manipulators, parallel robots, and cooperative robot-arm motion. The result is a focused technical treatment for readers who want to examine how neural dynamics can support real-time control and redundancy resolution.

From serial arms to cooperative robots

The opening section explores neural-network methods for serial robot-arm control, including zeroing, adaptive dynamic programming, projection, and robust controller designs. It also considers neural learning and control co-design, as well as using neural networks to address robot singularities.

The book then turns to Stewart platforms, covering neural-network-based control and learning-and-control co-design. Its final section addresses zeroing neural networks for motion generation in cooperative robot arms.

Algorithms, analysis, and simulation

Chapters develop control problems and neural-network schemes alongside theoretical work on properties such as stability, convergence, and optimality. Simulations examine examples including position regulation, reference-path tracking, redundancy resolution, and control in the presence of noise. MATLAB-related modeling and simulation work is also noted by the publisher.

This balance of mathematical development and computational examples gives readers a way to follow both how the proposed methods are formulated and how their behavior is investigated. The chapter-by-chapter organization makes it possible to focus on a particular robot type or control approach.

For robotics and control researchers

Written for graduate students and academic or industrial researchers, this book is suited to readers with an interest in robot kinematics, neural dynamics, mechatronics, or mechanical engineering. It assumes an appetite for technical detail and is especially relevant to those studying neural-network-assisted control of redundant manipulators and related systems.

For a research library or an advanced robotics reading list, this volume brings several neural-network control approaches together in one structured study.