Robot Learning Human Skills and Intelligent Control Design

Learning from demonstrations (LfD) is a method for robots to acquire skills from human demonstrators, enabling them to perform new tasks quickly. This book discusses advanced LfD-based learning and control approaches to improve robot dexterous manipulation, including transferring human arm variable stiffness to the robot, modeling motion and impedance profiles, learning correlation between signals, and using neural network enhanced control strategies. Examples of simulation and experiments are included to facilitate better understanding.

Format: Hardback
Length: 174 pages
Publication date: 17 June 2021
Publisher: Taylor & Francis Ltd

In recent years, the field of robotics has witnessed a remarkable surge in the intelligence and capabilities of robots, with a growing expectation for them to handle a diverse array of tasks. One particularly significant area of focus is the acquisition of manipulation skills by robots, drawing inspiration from human counterparts. To achieve this goal, a wide range of learning algorithms and techniques have been developed and successfully implemented across various robotic domains. Among these methods, learning from demonstrations (LfD) stands out as a powerful approach, enabling robots to rapidly acquire skills by observing and mimicking human demonstrators.

This book serves as a comprehensive resource for exploring advanced LfD-based learning and control approaches aimed at enhancing the dexterous manipulation capabilities of robots. It begins by providing an introduction to the simulation tools and robot platforms employed in the authors' research. The focus then shifts to the crucial aspect of enabling robots to learn human-like adaptive skills, which involves transferring human users' arm variable stiffness to the robot based on online estimation from muscle electromyography (EMG). The book further explores the modeling of motion and impedance profiles using dynamical movement primitives, enabling the planning and generalization of these profiles for new tasks. Moreover, it delves into the learning of correlations between signals collected from demonstration, such as motion trajectory, stiffness profile estimated from EMG, and interaction force, utilizing statistical models like hidden semi-Markov models and Gaussian Mixture Regression.

Several widely used human-robot interaction interfaces, including motion capture-based teleoperation, are presented to facilitate human-robot interaction and the transfer of movements between humans and robots in both simulation and real-world environments. Furthermore, the book discusses the use of neural networks to improve the performance of robot manipulation, highlighting their potential for enhancing the robots' ability to learn and adapt to complex tasks.

In conclusion, this book offers a valuable contribution to the field of robotics by presenting cutting-edge research on advanced LfD-based learning and control approaches for improving the dexterous manipulation capabilities of robots. It provides a comprehensive overview of the simulation tools, robot platforms, and learning algorithms employed in the research, as well as the application of these techniques to various robotic tasks. By leveraging the power of LfD and other advanced learning methods, robots can be programmed to perform complex tasks with greater efficiency and accuracy, paving the way for a future where robots and humans collaborate seamlessly in a wide range of applications.

Weight: 422g
Dimension: 160 x 242 x 19 (mm)
ISBN-13: 9780367634360