Trends in Control and Decision-Making for Human Robot Collaboration Systems

Transcription

1 Trends in Control and Decision-Making for Human Robot Collaboration Systems

2 Yue Wang Fumin Zhang Editors Trends in Control and Decision-Making for Human Robot Collaboration Systems 123

3 Editors Yue Wang Department of Mechanical Engineering Clemson University Clemson, SC USA Fumin Zhang School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta, GA USA ISBN ISBN (ebook) DOI / Library of Congress Control Number: MATLAB is a registered trademark of The MathWorks, Inc., 3 Apple Hill Drive, Natick, MA , USA, Springer International Publishing Switzerland 2017 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. Printed on acid-free paper This Springer imprint is published by Springer Nature The registered company is Springer International Publishing AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

4 To our advisors, students, and family

5 Preface This book aims to present new developments in control and decision-making theory in the field of human robot interaction (HRI). Despite advances in robotics, sensing, and autonomy, human participation is still an essential component in various operations, especially under uncertain and complex environments. In particular, human robot collaboration (HRC) systems integrate the strengths of human in terms of high-level decision-making, flexibility, dexterity, vision processing, etc., with robot s capabilities of performing repetitive tasks in dangerous situations in order to realize the full potential of autonomous systems. Therefore, a considerable amount of effort has been made in this area, which however still lacks qualitative analysis, performance prediction, and guarantees. This offers little insight into the effective control and decision-making for the collaboration operations, especially in real time. To fill the gap, the chapters in this book discuss in detail the development of new control and decision-making algorithms for HRC systems for guaranteed joint human robot system performance. Chapter 1 provides an overview of the extant works in HRI and detailed introduction of main contributions of each chapter. Chapters 2 9 present methods for one human robot pair collaboration and Chaps develop the control and coordination algorithms for humans to collaborate with multiple robots and swarms. Both physical (e.g., force, vision) and psychological human factors (e.g., trust and regret) are embedded into control and decision-making approaches such as nonlinear control, shared control, switching control, optimal control, and sequential detection. The considered applications include transportation, healthcare, manufacturing, and defense. Robot experiments and simulations with human-in-the-loop are conducted to demonstrate the effectiveness of the proposed algorithms. Clemson, SC, USA Atlanta, GA, USA April 2016 Yue Wang Fumin Zhang vii

6 Acknowledgements The editors would like to thank all the chapter authors and reviewers that work together on this book. A special acknowledgment goes to all the graduate students in the Interdisciplinary and Intelligent Research (I 2 R) Laboratory in the Mechanical Engineering Department at Clemson University, whom assisted the editors to review chapters and provided useful feedbacks to improve the quality of the book. The first editor would like to thank the support from the National Science Foundation under Grant No. CMMI The editors would also like to thank Springer Verlag and its staff for the professional support. ix

7 Contents 1 Introduction... 1 Yue Wang and Fumin Zhang 1.1 Overview Collaboration Between One Human Robot Pair Collaboration Between Human and Multiple Robots/Swarms... 6 References Robust Shared-Control for Rear-Wheel Drive Cars Jingjing Jiang and Alessandro Astolfi 2.1 Introduction Problem Formulation, Definitions, and Assumptions Design of the Shared-Control Law with Measurements of Absolute Positions Design of the Feedback Controller Shared-Control Algorithm Disturbance Rejections Design of the Shared Control Without Measurements of Absolute Positions Design of the Feedback Controller Shared-Control Algorithm Case Studies Case I: Turning Without Absolute Positioning Case II: Driving on a Road with Parked Cars Case III: Emergency Breaking Conclusions References xi

8 xii Contents 3 Baxter-On-Wheels (BOW): An Assistive Mobile Manipulator for Mobility Impaired Individuals Lu Lu and John T. Wen 3.1 Introduction System Description Experimental Platform: BOW System Kinematics Control Algorithm Baseline Shared-Control Algorithm Free-Space Mode and Contact Mode Application to the BOW User Interface Object Pick-Up and Placement Task Board Cleaning Task Conclusion References Switchings Between Trajectory Tracking and Force Minimization in Human Robot Collaboration Yanan Li, Keng Peng Tee and Shuzhi Sam Ge 4.1 Introduction Dynamic Models Robot Model Human Arm Model Unified Model Trajectory Tracking Control Design Control Objective Selection of Cost Functions Optimal Control Simulations Simulation Settings Change of Weights Adaptation of Desired Trajectory Conclusions Appendix: Proof of Lemma References Estimating Human Intention During a Human Robot Cooperative Task Based on the Internal Force Model Ehsan Noohi and Miloš Žefran 5.1 Introduction Internal Force Model Problem Formulation... 86

9 Contents xiii Existing Models Proposed Model Discussion Method Apparatus Procedure Results Validation of the Model Statistical Analysis of the Internal Force Features Initial Grasp Force Magnitude Final Grasp Force Magnitude Internal Force Energy Difference Between Initial and Final Grasp Forces Internal Force Variation Negotiation Force Negotiation Force Versus Object Velocity Proposed Cooperation Policy Conclusion References A Learning Algorithm to Select Consistent Reactions to Human Movements Carol Young and Fumin Zhang 6.1 Introduction Background Expert-Based Learning Binary Learning Algorithms Analysis Performance Consistency Adaptiveness Tie Breaking Expanded Dual Expert Algorithm Performance Analysis Consistency and Adaptiveness Simulation Dual Expert Algorithm Expanded Dual Expert Algorithm Experiment Setup Results Conclusions References

10 xiv Contents 7 Assistive Optimal Control-on-Request with Application in Standing Balance Therapy and Reinforcement Anastasia Mavrommati, Alex Ansari and Todd D. Murphey 7.1 Introduction Assistive Control Synthesis Calculating a Schedule of Optimal Infinitesimal Actions Computing the Control Duration Human Robot Interaction in Assisted Balance Therapy Related Work: Assist-as-Needed Techniques Interactive Simulation Study Human Robot Communication in Posture Reinforcement: A Short Study Conclusion References Intelligent Human Robot Interaction Systems Using Reinforcement Learning and Neural Networks Hamidreza Modares, Isura Ranatunga, Bakur AlQaudi, Frank L. Lewis and Dan O. Popa 8.1 Introduction HRI Control: Motivation and Structure Overview of the Proposed Approach Inner Robot-Specific Loop Outer Task-Specific Loop Control Task-Specific Outer Loop Control Method: An LQR Approach Learning Optimal Parameters of the Prescribed Impedance Model Using Integral Reinforcement Learning Simulation Results Conclusion References Regret-Based Allocation of Autonomy in Shared Visual Detection for Human Robot Collaborative Assembly in Manufacturing S.M. Mizanoor Rahman, Zhanrui Liao, Longsheng Jiang and Yue Wang 9.1 Introduction The Hybrid Cell for Human Robot Collaborative Assembly Detection Problem Formulation with Focus on the Selected Assembly Task Description of the Problem

11 Contents xv Problem Formulation Bayesian Sequential Decision-Making Algorithm for Allocation of Autonomy Inclusion of Regret in Bayesian Decision-Making Algorithm for Allocation of Autonomy Illustration of the Decision-Making Approach Illustration of the Optimal Bayesian Decision-Making Approach Illustration of the Regret-Based Modified Decision-Making Approach Implementation Scheme of the Regret-Based Bayesian Decision-Making Approach for the Assembly Task The Overall Scheme in a Flowchart Measurement of Sensing Probability and Observation Cost Measurement Method for Regret Intensity Experimental Evaluation of the Regret-Based Bayesian Decision-Making Approach Objective Hypothesis The Evaluation Criteria The Experiment Design Subjects The Experimental Procedures Evaluation Results and Analyses Conclusions and Future Innovations References Considering Human Behavior Uncertainty and Disagreements in Human Robot Cooperative Manipulation José Ramón Medina, Tamara Lorenz and Sandra Hirche 10.1 Introduction Human Robot Cooperative Manipulation Cooperative Manipulation Control Challenges in Physical Human Robot Interaction Reactive Assistants Proactive Assistants Interaction Wrench Decomposition Nonuniform Wrench Decomposition Matrices Effective and Internal Wrenches Load Share and Disagreement Optimal Robot Assistance Considering Human Behavior Uncertainty and Disagreements

12 xvi Contents Anticipatory Assistance Based on Learned Models The Two-Dimensional Translational Case Experiments Conclusions Appendix References Designing the Robot Behavior for Safe Human Robot Interactions Changliu Liu and Masayoshi Tomizuka 11.1 Introduction The Safety Issues and Existing Solutions Safety Problems in HRI: Conflicts in Multiagent Systems Safe Control and Exploration Modeling the Human Robot Interactions The Agent Model The Closed-Loop System Information Structure The Safety-Oriented Behavior Design The Safety Principle The Safety Index The Safe Set Algorithm (SSA) The Control Algorithm Online Learning and Prediction of Humans' Dynamics Applications The Safe Exploration Algorithm (SEA) The Safe Set in the Belief Space Learning in the Belief Space A Comparative Study Between SSA and SEA Combining SSA and SEA in Time Varying MAS Topology The Control Algorithm The Learning Algorithm Performance Discussions The Energy Based Methods Limitations and Future Work Conclusion References

13 Contents xvii 12 When Human Visual Performance Is Imperfect How to Optimize the Collaboration Between One Human Operator and Multiple Field Robots Hong Cai and Yasamin Mostofi 12.1 Introduction Human and Robot Performance in Target Classification [4] Optimizing Human Robot Collaboration for Target Classification Predetermined Site Allocation Optimized Site Allocation Numerical Results Collaboration Between the Human Operator and One Robot [4] Predetermined Site Allocation Optimized Site Allocation Conclusions References Human-Collaborative Schemes in the Motion Control of Single and Multiple Mobile Robots Antonio Franchi 13.1 Introduction Modeling of the Robot and the Interactions Mobile Robot Communication Infrastructure Human Robot Interface A Taxonomy of Collaborative Human Robot Control Physical Domain of the Robots Degree of Autonomy from the Human Operator Force Interaction with the Operator Near-Operation Versus Teleoperation Physical Interaction with the Environment Use of Onboard Sensors Only A Taxonomy of Collaborative Human Multi-robot Control Level of Centralization Master Leader Followers Schemes Formation-Orthogonal Control Schemes Group-Property Preservation Schemes Physical Interaction with Contact Conclusions References

14 xviii Contents 14 A Passivity-Based Approach to Human Swarm Collaboration and Passivity Analysis of Human Operators T. Hatanaka, N. Chopra, J. Yamauchi and M. Fujita 14.1 Introduction Intended Scenario and Control Goals Control Architecture and Passivity Convergence Analysis Synchronization in Position Control Mode Synchronization in Velocity Control Mode Passivity of the Human Operator Decision Process Experimental Setup and Approach Analysis on Human Passivity in Position Control Mode Analysis on Human Passivity in Velocity Control Mode Analysis on Individual Variability Summary References Human Swarm Interactions via Coverage of Time-Varying Densities Yancy Diaz-Mercado, Sung G. Lee and Magnus Egerstedt 15.1 Introduction Human Swarm Interactions via Coverage The Coverage Problem Centralized Coverage of Time-Varying Densities Distributed Coverage of Time-Varying Densities Designing Density Functions Diffusion of Drawn Geometric Configurations Control of Gaussian Functions Robotic Experiments Conclusions References Co-design of Control and Scheduling for Human Swarm Collaboration Systems Based on Mutual Trust Xiaotian Wang and Yue Wang 16.1 Introduction Swarm Setup Dynamic Timing Model and Collaboration Delay Cooperative Control for Swarm Agents Collaboration Framework Trust Model

15 Contents xix Human Performance Model Swarm Performance Model Human Attention Preference Fitness Real-Time Scheduling Simulation Results Parameter Setup Results and Discussions Conclusions References Index