Nonparametric Bayesian Learning for Collaborative Robot Multimodal Introspection

ISBN-13: 9789811562624 / Angielski / Twarda / 2020 / 137 str.

Xuefeng Zhou; Hongmin Wu; Juan Rojas

Spis treści

Introduction to Robot Introspection.- Nonparametric Bayesian Modeling of Multimodal Time Series.- Incremental Learning Robot Complex Task Representation and Identification.- Nonparametric Bayesian Method for Robot Anomaly Monitoring.- Nonparametric Bayesian Method for Robot Anomaly Diagnose.- Learning Policy for Robot Anomaly Recovery based on Robot.

O autorze

Dr. Xuefeng Zhou is an Associate Professor and Leader of the Robotics Team at Guangdong Institute of Intelligent Manufacturing, Guangdong Academy of Science. He received his Ph.D. degree in Manufacturing and Automation from the South China University of Technology in 2011. His research mainly focuses on motion planning and control, force control and legged robots. He has published more than 40 journal articles and conference papers.

Dr. Hongmin Wu is a Researcher at Guangdong Institute of Intelligent Manufacturing, Guangdong Academy of Science. He received his Ph.D. degree in Mechanical Engineering from Guangdong University of Technology, Guangzhou, China, in 2019. His research mainly focuses on robot learning, autonomous manipulation, deep learning and human­–robot collaboration. He has published more than 20 journal articles and conference papers.

Dr. Juan Rojas is an “100 Young Talents” Associate Professor at the Guangdong University of Technology in Guangzhou, China, where he works at the Biomimetics and Intelligent Robotics Lab (BIRL). Dr. Rojas currently researches robot introspection, human intention prediction, high-level state estimation and skill acquisition for manipulation tasks. He has published more than 40 journal articles and conference papers. Dr. Rojas serves as an Associate Editor of Advanced Robotic Journal since 2019 and Associate Editor of IEEE International Conference on Intelligent Robots and Systems (IROS) since 2017.

Dr. Zhihao Xu is a Researcher at Guangdong Institute of Intelligent Manufacturing, Guangdong Academy of Science. He received his Ph.D. degree in Control Science and Engineering from Nanjing University of Science and Technology, China, in 2016. His research mainly focuses on intelligent control theory, motion planning and control and force control. He has published more than 30 journal articles and conference papers.

Prof. Shuai Li is a Ph.D. Supervisor and Associate Professor (Reader) at the College of Engineering, Swansea University, UK. He received his Ph.D. degree in Electrical and Computer Engineering from Stevens Institute of Technology, New Jersey, USA, in 2014. His research interests are robot manipulation, automation and instrumentation, artificial intelligence and industrial robots. He has published over 80 papers in journals such as IEEE TAC, TII, TCYB, TIE and TNNLS. He serves as Editor-in-Chief of the International Journal of Robotics and Control and was the General Co-Chair of the 2018 International Conference on Advanced Robotics and Intelligent Control.

Blurb

This open access book focuses on robot introspection, which has a direct impact on physical human–robot interaction and long-term autonomy, and which can benefit from autonomous anomaly monitoring and diagnosis, as well as anomaly recovery strategies. In robotics, the ability to reason, solve their own anomalies and proactively enrich owned knowledge is a direct way to improve autonomous behaviors. To this end, the authors start by considering the underlying pattern of multimodal observation during robot manipulation, which can effectively be modeled as a parametric hidden Markov model (HMM). They then adopt a nonparametric Bayesian approach in defining a prior using the hierarchical Dirichlet process (HDP) on the standard HMM parameters, known as the Hierarchical Dirichlet Process Hidden Markov Model (HDP-HMM). The HDP-HMM can examine an HMM with an unbounded number of possible states and allows flexibility in the complexity of the learned model and the development of reliable and scalable variational inference methods.

This book is a valuable reference resource for researchers and designers in the field of robot learning and multimodal perception, as well as for senior undergraduate and graduate university students.