Remote Sensing and Actuation Using Unmanned Vehicles E-Book

Haiyang Chao

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Serie: IEEE Series on Systems Science and Engineering
• Sprache: Englisch

Seitenzahl: 254

Veröffentlichungsjahr: 2012

Beschreibung

Unmanned systems and robotics technologies have become very popular recently owing to their ability to replace human beings in dangerous, tedious, or repetitious jobs. This book fill the gap in the field between research and real-world applications, providing scientists and engineers with essential information on how to design and employ networked unmanned vehicles for remote sensing and distributed control purposes. Target scenarios include environmental or agricultural applications such as river/reservoir surveillance, wind profiling measurement, and monitoring/control of chemical leaks.

Leseprobe

Contents

Cover

Series Page

Title Page

Copyright

Dedication

List of Figures

List of Tables

Foreword

Preface

Acknowledgements

Acronyms

Chapter 1: Introduction

1.1 Monograph Roadmap

1.2 Research Motivations

1.3 Monograph Contributions

1.4 Monograph Organization

References

Chapter 2: AggieAir: A Low-Cost Unmanned Aircraft System for Remote Sensing

2.1 Introduction

2.2 Small UAS Overview

2.3 AggieAir UAS Platform

2.4 OSAM-Paparazzi Interface Design for IMU Integration

2.5 AggieAir UAS Test Protocol and Tuning

2.6 Typical Platforms and Flight Test Results

2.7 Chapter Summary

References

Chapter 3: Attitude Estimation Using Low-Cost IMUs for Small Unmanned Aerial Vehicles

3.1 State Estimation Problem Definition

3.2 Rigid Body Rotations Basics

3.3 Low-Cost Inertial Measurement Units: Hardware and Sensor Suites

3.4 Attitude Estimation Using Complementary Filters on SO(3)

3.5 Attitude Estimation Using Extended Kalman Filters

3.6 AggieEKF: GPS-Aided Extended Kalman Filter

3.7 Chapter Summary

References

Chapter 4: Lateral Channel Fractional Order Flight Controller Design for a Small UAV

4.1 Introduction

4.2 Preliminaries of UAV Flight Control

4.3 Roll-Channel System Identification and Control

4.4 Fractional Order Controller Design

4.5 Simulation Results

4.6 UAV Flight Testing Results

4.7 Chapter Summary

References

Chapter 5: Remote Sensing Using Single Unmanned Aerial Vehicle

5.1 Motivations for Remote Sensing

5.2 Remote Sensing Using Small UAVs

5.3 Sample Applications for AggieAir UAS

5.4 Chapter Summary

References

Chapter 6: Cooperative Remote Sensing Using Multiple Unmanned Vehicles

6.1 Consensus-Based Formation Control

6.2 Surface Wind Profile Measurement Using Multiple UAVs

6.3 Chapter Summary

References

Chapter 7: Diffusion Control Using Mobile Sensor and Actuator Networks

7.1 Motivation and Background

7.2 Mathematical Modeling and Problem Formulation

7.3 CVT-Based Dynamical Actuator Motion Scheduling Algorithm

7.4 Grouping Effect in CVT-based Diffusion Control

7.5 Information Consensus in CVT-Based Diffusion Control

7.6 Simulation Results

7.7 Chapter Summary

References

Chapter 8: Conclusions and Future Research Suggestions

8.1 Conclusions

8.2 Future Research Suggestions

References

Appendix

A.1 List of Documents for CSOIS Flight Test Protocol

A.2 IMU/GPS Serial Communication Protocols

A.3 Paparazzi Autopilot Software Architecture: A Modification Guide

A.4 DiffMAS2D Code Modification Guide

References

Topic Index

IEEE Press445 Hoes LanePiscataway, NJ 08854

IEEE Press Editorial BoardJohn B. Anderson, Editor in Chief

Kenneth Moore, Director of IEEE Book and Information Services (BIS)

Technical ReviewersDongbing Gu, University of Essex

IEEE Press Series on Systems Science and Engineering

Copyright © 2012 by The Institute of Electrical and Electronics Engineers, Inc.

Published by John Wiley & Sons, Inc., Hoboken, New Jersey. All rights reserved

Published simultaneously in Canada.

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, [phone], fax [phone], or on the web at www.copyright.com. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, [phone], fax [phone], or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact our Customer Care Department within the United States at [phone], outside the United States at [phone] or fax [phone].

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic formats. For more information about Wiley products, visit our web site at www.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Chao, Haiyang.

Remote sensing and actuation using networked unmanned vehicles/Haiyang Chao, Yangquan Chen.

pages cm

Includes bibliographical references and index.

ISBN 978-1-118-12276-1 (hardback)

1. Geomorphology–Remote sensing. 2. Environmental monitoring–Remote sensing. 3. Vehicles, Remotely piloted. I. Chen, Yangquan, 1966- II. Title.

GB400.42.R4C46 2012

621.36'78–dc23

2012006660

To my parents and my brotherHaiyang Chao

To my familyYangQuan Chen

List of Tables

2.1 Small UAS Categories 17

2.2 Comparison of Physical Specifications of Autopilots

2.3 Comparison of Autopilot Functions

2.4 AggieAir2 UAS Specifications

2.5 GhostFinger Digital Video Camera Specifications

2.6 GhostFinger Digital Camera Specifications

2.7 Serial Mapping of Gumstix Console-VX Board

2.8 Communication Protocol Between Gumstix and Paparazzi

2.9 Packet Definition for Data from GhostGX2

2.10 Gumstix Resource Monitoring for GhostGX2

2.11 UAVs Developed (or Partly Developed) by H. Chao

2.12 Robustness of AggieAir UAS

3.1 IMU Categories

3.2 IMU Specification Comparisons

4.1 Change UAS Specifications

6.1 AggieAir UAS Specifications

7.1 Computational Time for Simulation and Control Results

7.2 Run Time for Simulation and Control Results

7.3 Comparison of Performance for Different Group Size

7.4 Comparison of Control Performance

7.5 Comparison of Total Neutralizing Material 163 tbtbxix

Foreword

This monograph is very timely and in many ways prophetic. In the United States, lawmakers just passed a long awaited bill setting requirements to integrate unmanned aerial systems (UAS) in the national airspace (NAS). Such requirements would set deadlines for full-scale integration in fall 2015. News of UAS also fills headlines as both their mission roles and numbers increase. It is estimated that over one-third of military aircraft in the United States are unmanned. As these numbers continue to grow, coupled with the opening of the NAS, challenges in technology, policy, and even ethics will remain in the forefront for years to come. As such, this monograph provides a much needed resource to root the community; academics, industrial practitioners, policymakers, and business developers stand to benefit on the foundations presented in this book.

Aviation giants all have divisions that are aggressively pursuing UAS programs. The conviction is that aerospace leadership into the 21st century will be determined by those who successfully commercialize unmanned aircraft. Such mindsets are thus pushing the boundaries of UAS roles with creative applications to meet untapped needs such as agricultural crop handling, wildlife monitoring, road traffic reporting, meteorological pattern detection, and forest wildfire mitigation. Realization of such applications is still underpinned by open research challenges. This monograph provides insight into both the engineering fundamentals and best practices that have resulted in years of field-proven work. The authors Haiyang Chao and YangQuan Chen are pioneers in UAS research and development. They also led award-winning teams in international UAS competitions and have logged years of flight time. Readers are indeed fortunate to benefit from their experiences as vetted in this book.

This book is striking in its clarity and comprehensiveness. Moreover, the book is underscored by examination and examples revolving around a low-cost UAS. From aerodynamic modeling fundamentals to actual construction and flight-testing, readers can implement their own real-world UAS. Such a UAS provides a focal point and test bed to apply topics like flight control, wide-area coverage, multi-UAV formations and aerial image processing.

Here to now, UAS design, build and fly has been mostly ad hoc. This book stands to be the definitive guide to analytically design and systematically engineer UAS. As such, this book will be a must for every individual including UAS designers and users. This is a very timely book and will help shape an tbtbxxi exciting future as UAS become part of our everyday experience.

Paul Y. Oh, ASME Fellow

Founding Chair of IEEE RAS TC on ARUAV

Associate Professor & Department Head

Mechanical Engineering & Mechanics Department

Drexel University, Philadelphia, PA, USA

Preface

Unmanned vehicles, including unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs), have been increasingly used to free human beings from dangerous, dull, and dirty jobs. Unmanned vehicles can serve as remote sensors for surveillance applications, or actuators for control purposes, or both. Typical remote sensing tasks include the monitoring of air quality, forest fire, and nuclear leaks, which all require measurements over a large scale (typically tens of square miles or even bigger). Example actuation tasks include fire extinguishing, and neutralizing control of chemical spills. One of the key challenges in unmanned vehicle developments is the design of autonomous capability or remote control capability. More importantly, unmanned vehicles need to combine navigation, communication, and computation capabilities with mission-specific payload to provide useful information. For instance, each aerial image requires synchronized position and orientation information for image interpretation purposes. Another key challenge using unmanned vehicles is how to deploy single or groups of vehicles optimally for different sensing or actuation tasks. Optimal paths are needed to extend the sensing range of unmanned vehicles, or to minimize the time taken for control missions.

This monograph focuses on how to design and employ unmanned vehicles for remote sensing and distributed control purposes in the current information-rich world. The target scenarios are environmental or agricultural applications such as river/reservoir surveillance, wind profiling measurement, monitoring/control of chemical leaks, and so on. This monograph comprises two parts. The first part is on the design and deploy of UAVs for remote sensing applications with a focus on real flight experiments. AggieAir, a small and low-cost unmanned aircraft system, is designed based on the remote sensing requirements from environmental monitoring missions. A summary of low-cost IMU hardware and attitude estimation software is detailed for small UAVs. The advanced lateral flight controller design problem is further introduced. Then the single-UAV-based remote sensing problem is focused with flight test results. The latter part of this monograph is on ground test and simulations to use networked unmanned vehicles in future sensing and control missions such as formation control, wind profile monitoring, and neutralizing control. Given the measurements from unmanned vehicles, the actuation algorithms are needed for missions such as the diffusion control. A consensus-based central Voronoi tessellation (CVT) algorithm is proposed for better control of the diffusion process. Finally, the monograph conclusion and some new research suggestions are presented. A Web site has been set up with extra videos, figures, and source codes (http://sites.google.com/site/haiyangchao/book_rsauv).

This book can benefit engineers or developers who want to develop their own low-cost UAV platforms for remote sensing, environmental tbtbxxiii monitoring, aerial image processing, wireless communication, flight control, cooperative control, or general robotics researches. More importantly, this book provides the authors' approach on using unmanned vehicles for formation control and actuation missions. Such missions will be very important especially in scenarios such as nuclear leaks. The authors hope that this book can help increase the uses of unmanned vehicles in both remote sensing and actuation applications.

Haiyang ChaoYangQuan Chen

January, 2012

Acknowledgments

The authors would like to thank other contributors of the AggieAir UAS development including Austin M. Jensen, Yiding Han, Long Di, Calvin Coopmans, and Daniel Morgan. Part of the results in Chapter 5 was collected with AggieAir platform by Austin M. Jensen from Utah Water Research Laboratory. The fractional flight controller is codeveloped by the authors together with Dr. Ying Luo and Long Di. The authors also want to thank other past OSAM UAV team members: Marc Baumann, Dr. Yongcan Cao, Dr. Hu Sheng, Aaron Avery, Mitch Humphries, Chris Hall, Aaron Quitberg, Norman Wildmann, and other MAS-net team members: William K. Bourgeous, Nathan Sorensen, Dr. Zhen Song.

Haiyang Chao would like to thank Dr. Wei Ren for joint work in the experimental validations on the MAS-net platform; Dr. Donn Cripps, Dr. Todd Moon, Dr. Vladimir Kulyukin, Dr. Jiming Jin and Dr. Hui Chen for their help in his Ph.D. dissertation work. Thanks also go to other CSOIS members for their help including Dr. Hyo-Sung Ahn, Dr. Christophe Tricaud, Rongtao Sun, Yashordhan Tarte, Tripti Bhaskaran, Varsha Bhambhani, Jessi Lianghan N.G., Shayok Mukhopadhyay, Dr. Yan Li, Dr. Bin Wang, Dr. Wei Sun, Dr. Yongshun Jin, Dr. Hongguang Sun. The authors want to especially thank MaryLee Anderson, Brandon Stark, and Jinlu Han for proof reading, Dr. Dongbing Gu for providing very useful reviewer comments, Mary Hatcher and Taisuke Soda from John Wiley & Sons for helping in publishing this book.

Last but not the least, the authors want to thank developers from the open-source society including Curtis Olson, Pascal Briset, Gautier Hattenberger, Anton Kochevar, Antoine Drouin, Felix Ruess, William Premerlani, Paul Bizard, Dr. JungSoon Jang, and all the Paparazzi UAV developers for sharing their open-source projects. The authors benefited a lot through reading their source codes and discussing details with them.

The work presented in this book is partly supported by the Utah Water Research Laboratory (UWRL) MLF Seed Grant (2006–2011) and the Vice President for Research Fellowship (2005–2006) from Utah State University. The authors are thankful to Professor Mac McKee for his original research vision on UAV application in water science and engineering and Professor Raymond L. Cartee for providing the USU farm at Cache Junction as the flight test field, and Professor H. Scott Hinton, the Dean of College of Engineering, for travel support to Maryland for the AUVSI Student UAS competitions in summer 2008 and 2009. Haiyang Chao would like to thank the Graduate tbtbxxv Student Senate of Utah State University for the Enhancement Award, and the NSF Institute for Pure and Applied Mathematics (IPAM) at UCLA for the travel support to participate in the 1 week-long workshop on “Mathematical Challenges on Sensor Networks.”

Acronyms

AGLabove ground levelARXauto-regressive exogenousAUVSIAssociation for Unmanned Vehicle Systems InternationalBFCSbody-fixed coordinate systemCCDcharge-coupled deviceCGcenter of gravityCOTScommercial off-the-shelfCPScyber-physical systemsCSOISCenter for Self-Organizing and Intelligent SystemsCVTcentroidal Voronoi tessellationsDCdigital cameraDCMdirection cosine matrixDOFdegree of freedomDPSdistributed parameter systemDVdigital video cameraECEFEarth-centered Earth-fixedENACEcole Nationale de l'Aviation CivileEKFextended Kalman filterFOCfractional order controlFOPIfractional order proportional integralFOPTDfirst order plus time delayFOVfield of viewGAgenetic algorithmGCSground control stationGFGhost FingerGF-DCGhost Finger digital cameraGF-DVGhost Finger digital video cameraGISgeographic information systemGPSglobal position systemgRAIDgeospatial real-time aerial image displayIDidentificationIGSinertial guidance systemIMUinertial measurement unitINSinertial navigation systemIOPIinteger order proportional integralIRinfraredLQGlinear quadratic GaussianLIDARlight detection and rangingLLHlatitude longitude height tbtbxxviiLSleast squaresLTIlinear time-invariantLTVlinear time-varyingMASnetmobile actuator and sensor networkMEMSmicroelectromechanical systemsMIMOmultiple inputs multiple outputsMZNmodified Ziegler–NicholsNIRnear infraredNNneural networkODEordinary differential equationOSAM-UAVopen-source autonomous multiple unmanned aerial vehiclePDEpartial differential equationPIproportional-integralPIDproportional-integral-derivativePPRZPaparazzi open-source autopilot projectPRBSpseudo random binary signalPTPpicture transfer protocolRCremote controlledRGBred–green–blueRPVremote piloted vehicleSDsecure digitalSISOsingle input single outputSO3special orthogonal group 3TWOGTiny without GPSUARTuniversal asynchronous receiver/transmitterUASunmanned aircraft systemUAVunmanned air/aerial vehicleUGVunmanned ground vehicleUSBuniversal serial busUSUUtah State UniversityUTMuniversal transverse mercatorUUVunmanned underwater vehicleUWRLUtah Water Research LabVCvideo cameraWVUWest Virginia UniversityXMLextensible markup language

Chapter 1

Introduction

1.1 Monograph Roadmap

This monograph focuses on how to design and employ unmanned systems for remote sensing and distributed control purposes in the current information-rich world. The target scenarios include river/reservoir surveillance, wind profiling measurement, distributed control of chemical leaks, and the like, which are all closely related to the physical environment. Nowadays, threats of global warming and climate change demand accurate and low-cost techniques for a better modeling and control of the environmental physical processes. Unmanned systems could serve as mobile or stationary sensors and actuators. They could save human beings from dangerous, tedious, and repetitious outdoor work, whether it is deep in the ocean or high up in the sky. With the modern wireless communication technologies, unmanned vehicles could even work in groups for some challenging missions such as forest fire monitoring, ocean sampling, and so on. However, unmanned systems still require physics-coupled algorithms to accomplish such tasks mostly in the outdoor unstructured environments. Questions such as what to measure, when to measure, where to measure, and how to control all need to be properly addressed. This monograph presents our approach about how to build and employ unmanned vehicles (ground, air, or combined) to solve the problem of distributed sensing and distributed control of agricultural/environmental systems.

1.1.1 Sensing and Control in the Information-Rich World

Advances in electronics technologies such as embedded systems, microelectromechanical systems, and reliable wireless networks make it possible to deploy low-cost sensors and actuators in large amounts in a large-scale system. This poses a problem for control scientists and engineers on how to deploy and employ those vast amount of networked sensors/actuators optimally. The sensors and actuators can be static or mobile, single or multiple, isolated or networked, all depending on the application scenario. The options for sensor and actuator types are shown in Fig. 1.1. For example, both the temperature probe (point-wise sensing) and the thermal camera (zone sensing) could be used to measure the temperature of the crop canopy in a given field of interest. But which one to use? Proper sensing techniques are essential for the high-precision farming that can support the sensing of a large-scale system with an acceptable cost. Thermal aerial images are better for this mission. On the other hand, there are also coarse agricultural applications, which only need the temperature probe due to the cost limits. Another typical example is to use unmanned vehicles to monitor the forest fires. It is intuitive to use multiple unmanned aerial vehicles (UAVs), since they could provide more real-time information. However, there are questions regarding what information to share among UAVs and how often to share.

Figure 1.1 Sensors and Actuators in an Information-Rich World.

Unmanned vehicles can add the mobility to the sensors and actuators, which is especially beneficial for most outdoor environment monitoring applications. Different kinds of sensors and actuators could be installed on the unmanned vehicles based on specific application scenarios, as shown in Fig. 1.2. For instance, contact sensors can be installed on unmanned underwater vehicles (UUVs) to make accurate measurements of the temperature and humidity of the sea current. Cameras or radars can be mounted on UAVs for a more complete view of a farm or a reservoir. Chemical sprayers could be installed on unmanned ground vehicles (UGVs) for neutralizing gas leaks or extinguishing fires.

Figure 1.2 Unmanned Vehicles as Mobile Sensors/Actuators.

In this monograph, the unmanned system is defined as the unmanned vehicle together with onboard payload sensors or actuators. The fundamental functions of a typical unmanned systems include the mobility, computation, decision making, communication, and sensing/actuation, as shown in Fig. 1.3. Most unmanned systems have a powerful embedded processor to coordinate all the functions and make decisions based on information collected from its own or shared from other neighboring vehicles. With the communication subsystems, groups comprising of heterogeneous unmanned systems can now be designed to cooperate with each other to maximize their capabilities and the team's collective performance.

Figure 1.3 System Structures of Unmanned Vehicles.

1.1.2 Typical Civilian Application Scenarios

This monograph focuses mostly on the monitoring and control of environmental or agricultural systems or processes, which are of course closely related to human beings. Such systems could be categorized into two groups: fast-evolving ones such as chemical spill, gas leak, or forest fire and slow-evolving ones including heat transfer, moisture changing, wind profiling, and the like. The objective of monitoring these kinds of systems is to characterize how one or several physical entities evolve with both time and space. One typical example is an agricultural farm, as shown in Fig. 1.4. Water managers are interested in knowing how the soil moisture evolves with time in a farm to minimize the water consumption for irrigations. However, the evolution of soil moisture is affected by many other factors such as water flows, weather conditions (e.g., wind), and vegetation types, which all require measurements over a large scale (typically tens of square miles or even bigger). For such missions, ground probe stations are expensive to build and can only provide sensor data with very limited range. Satellite images can cover a large area, but have a low spatial resolution and a slow temporal update rate. Small UAVs cost less money but can provide more accurate information from low altitudes with less interference from clouds. In addition, small UAVs combined with ground and orbital sensors can form a multiscale remote sensing system, shown in Fig. 1.5.

Figure 1.4 Typical Agricultural Field (Cache Junction, UT).

Figure 1.5 Water Watch Concept.

Other typical civilian applications of unmanned systems include:

Forest Fire Monitoring and Containment Control: The monitoring, prediction, and containment control of forest fires could greatly reduce the potential property damages. Unmanned systems have obvious advantages over manned vehicles because human operators are not required onboard.Fog Evolution or Chemical Leaking Monitoring and Control: The evolution of hazardous fogs under emergency conditions can cost human lives without accurate and real-time measurements from unmanned systems. Example harmless fog evolutions are shown in Fig. 1.6.Wind Field Measurement: The wind direction and wind speed could have a significant impact on the diffusion of heat, water, or wind powers. However, the wind field is hard to measure because of its high variation, both temporally and spatially. Unmanned vehicles can be easily sent into the air for accurate 3D measurements.Canopy Moisture Measurement and Irrigation Control: The moisture on the vegetation canopy represents how much water could be absorbed by the plants. This information can be used for accurate irrigation control. The large scale of most agriculture fields requires cheap sensing techniques.

Figure 1.6 Fog Evolution (Taken in Yellowstone National Park).

1.1.3 Challenges in Sensing and Control Using Unmanned Vehicles

The problem of monitoring an environmental field can be defined as below. Let Ω ⊂ R3 be a polytope including the interior, which can be either convex or nonconvex. A series of density functions 1, 1, 3, . . . are defined as i(q, t) [0, ∞), ∀q Ω. For instance, i could be wind direction, surface temperature, soil moisture level, and the like. The goal of monitoring a spatial–temporal process is to find the distribution of the required density functions: