Robot Learning by Visual Observation - Tapa dura

Acerca del autor

ALEKSANDAR VAKANSKI is a Clinical Assistant Professor in Industrial Technology at the University of Idaho, Idaho Falls, USA. He received a Ph.D. degree from the Department of Mechanical and Industrial Engineering at Ryerson University, Toronto, Canada, in 2013. The scope of his research interests encompasses the fields of robotics and mechatronics, artificial intelligence, computer vision, and control systems.

FARROKH JANABI-SHARIFI is a Professor of Mechanical and Industrial Engineering and the Director of Robotics, Mechatronics and Automation Laboratory (RMAL) at Ryerson University, Toronto, Canada. He is currently a Technical Editor of IEEE/ASME Transactions on Mechatronics, an Associate Editor of The International Journal of Optomechatronics, and an Editorial Member of The Journal of Robotics and The Open Cybernetics and Systematics Journal. His research interests include optomechatronic systems with the focus on image-guided control and planning.

De la contraportada

This book presents an overview of the methodology for robot learning from visual observations of human demonstrated tasks, with a focus on learning at a trajectory level of task abstraction

The content of Robot Learning by Visual Observation is divided into chapters that address methods for tackling the individual steps in robotic observational learning. The book describes methods for mathematical modeling of a set of human-demonstrated trajectories, such as hidden Markov models, conditional random fields, Gaussian mixture models, and dynamic motion primitives. The authors further present methods for generation of a trajectory for task reproduction by a robot based on generalization of the set of demonstrated trajectories. In addition, the book

• Discusses methods for optimization of robotic observational learning from demonstrations, such as formulation of the task planning step as a constrained optimization problem
• Focuses on regression approaches for task reproduction, such as spline regression and locally weighted regression
• Concentrates on the use of vision sensors for capturing motions and actions demonstrated by a human task expert, as well as addresses the use of vision sensors for robust task execution by a robot learner

In times of a growing worldwide demand for automation and robotics applications, as well as an aging population and a shrinking work force, the development of robots with capacity to learn by observation and abilities for visual perception of the environment with vision sensors emerges as an important means to mitigate the aforementioned problems. The book is a valuable reference for university professors, graduate students, robotics enthusiasts, and companies that seek to develop robots with such abilities.