WIXLIB

Table of ContentsIntel RealSenseCamera software structureDefining termsTesting the 3D sensor in standalone modeRunning ROS nodes for visualizationVisual data using Image ViewerVisual data using rvizNavigating with TurtleBotMapping a room with TurtleBot 2160161161162163163165169170Autonomous navigation with TurtleBot 2175Navigating to a designated locationNavigating to waypoints with a Python script using a map183185Defining termsBuilding a mapHow does TurtleBot accomplish this mapping task?Defining termsDriving without steering TurtleBot 2rviz controlHow does TurtleBot accomplish this navigation task?Defining TurtleBot's position on a mapDefining waypoints on a mapUsing Python code to move TurtleBotTurtleBot at final goal pointSLAM for TurtleBot 3Autonomous navigation with TurtleBot 3rqt reconfigureExploring ROS navigation furtherSummaryChapter 5: Creating Your First Robot Arm (in Simulation)Features of XacroBuilding an articulated robot arm URDF using XacroSpecifying a namespaceUsing the Xacro property tagExpanding XacroUsing roslaunch for rrbotUsing the Xacro include and macro tagsAdding mesh to the robot armControlling an articulated robot arm in GazeboAdding Gazebo-specific elementsFixing the robot arm to the worldViewing the robot arm in GazeboAdding controls to XacroDefining transmission elements for 25

Table of ContentsAdding a Gazebo ROS control pluginCreating a YAML configuration fileCreating a control launch fileControlling your robot arm with the ROS command lineControlling your robot arm with rqtTrying more things in rqtSummaryChapter 6: Wobbling Robot Arms Using Joint ControlIntroducing BaxterBaxter, the research robotBaxter SimulatorBaxter's armsBaxter's bend jointsBaxter's twist jointsBaxter's coordinate frameControl modes for Baxter's armsBaxter's grippersBaxter's arm sensorsLoading Baxter softwareInstalling Baxter SDK softwareInstalling Baxter SimulatorConfiguring the Baxter shellInstalling MoveIt!Launching Baxter Simulator in GazeboBringing Baxter Simulator to lifeWarm-up exercisesFlexing Baxter's armsUntucking Baxter's armsWobbling armsControlling arms and grippers with a keyboardControlling arms and grippers with a joystickControlling arms with a Python scriptRecording and replaying arm 67270272Baxter's arms and forward kinematicsJoints and joint state publisherUnderstanding tf273273278Commanding the joint angles directlyrviz tf framesViewing a tf tree of robot elementsIntroducing MoveIt!Planning a move of Baxter's arms with MoveIt!281282283284286A program to move Baxter's arms to a zero angle position[ vi ]279

Table of ContentsAdding objects to a scene288Position of objectsPlanning a move to avoid obstacles with MoveIt!Configuring a real Baxter setupControlling a real BaxterCommanding joint position waypointsCommanding joint torque springsDemonstrating joint velocityAdditional examplesVisual servoing and graspingInverse kinematicsMoving Baxter's arms with IKUsing a state machine to perform YMCASummaryChapter 7: Making a Robot FlyIntroducing quadrotorsWhy are quadrotors so popular?Defining roll, pitch, and yawHow do quadrotors fly?Components of a quadrotorAdding sensorsQuadrotor communicationsUnderstanding quadrotor sensorsInertial measurement unitQuadrotor condition sensorsPreparing to fly your quadrotorTesting your quadrotorPre-flight checklistPrecautions when flying your quadrotorFollowing the rules and regulationsUsing ROS with UAVsIntroducing Hector QuadrotorLoading Hector QuadrotorLaunching Hector Quadrotor in GazeboFlying Hector outdoorsFlying Hector indoorsIntroducing Crazyflie 2.0Controlling Crazyflie without ROSCommunicating using Crazyradio PALoading Crazyflie ROS softwareSetting up udev rules for Crazyradio[ vii 325330334336337338340

Table of ContentsPre-flight checkFlying Crazyflie with teleop341342Details of teleop xbox360.launchFlying with a motion capture systemFlying multiple CrazyfliesIntroducing BebopLoading bebop autonomy softwareTesting Bebop communicationsFlying Bebop using commands344346346347349351352Take offLanding352353SummaryChapter 8: Controlling Your Robots with External DevicesCreating a custom ROS game controller interfaceTesting a game controllerAlternative test of a game controllerUsing the ROS joy packageControlling Turtlesim with a custom game controller interfaceCreating a custom ROS Android device interfaceInstalling Android Studio and toolsInstalling a ROS–Android development environmentDefining termsIntroducing ROS–Android developmentCreating ROS nodes on Arduino or Raspberry PiUsing ArduinoInstalling Arduino IDE softwareInstalling ROS–Arduino softwareUltrasonic sensor control using ROS and ArduinoOther applications using ROS and ArduinoUsing Raspberry PiInstalling ROS on the Raspberry 378386387387SummaryChapter 9: Flying a Mission with CrazyflieMission componentsKinect for Windows v2Crazyflie operationMission software structureOpenCV and ROSLoading software for the missionInstalling libfreenect2Installing iai kinect2[ viii ]388389391391392392394395396399

Table of ContentsUsing the iai kinect2 metapackage402kinect2 bridge and kinect2 viewerkinect2 calibration402403Setting up the missionDetecting Crazyflie and a target408409Using Kinect and OpenCVHow to track CrazyflieHow to control CrazyflieCrazyflie control states414417420420Flying CrazyflieHovering in place426426Identifying markers in a color imageDetecting and viewing markers with OpenCVUsing ROS services to control takeoff and landUsing PID control for hover and flightUsing an observer nodeWhat makes hover work?409412421423425427Flying to a stationary target428Learned lessons431What makes target detection work?430Logging messages with rosout and rospy431Summary432Chapter 10: Controlling Baxter with MATLAB 433Index445Installing the MATLAB Robotics System ToolboxCheck the MATLAB and Robotics System Toolbox versionsView the Robotics System Toolbox commands for ROSUsing MATLAB Robotics System Toolbox and Baxter SimulatorInstalling Baxter messages in MATLABRunning Baxter Simulator and MATLABMaking Baxter moveSummary[ ix ]434434434435435437440444

PrefaceBeing excited about learning ROS and working with ROS robots such as Baxter andTurtleBot is the beginning of a big adventure. The features and benefits of ROS aresubstantial, but the learning curve is steep. Through trial and error, we have forageda path through many of the ROS applications trying everything. In this book, wehope to present to you the best of our knowledge of ROS and provide you withdetailed step-by-step instructions for your journey. Our approach centers on usingthe ROS robots that are featured, namely TurtleBot, Baxter, Crazyflie, and Bebop, aswell as simulated robots—Turtlesim and Hector.This book provides introductory information as well as advanced applicationsfeaturing these ROS robots. The chapters begin with the basics of setting upyour computer and loading ROS and the packages for ROS robots and tools.Straightforward instructions are provided with troubleshooting steps for when thedesired results are not achieved. The building blocks of ROS are described first inthe simulation Turtlesim, then on each of the featured robots. Starting with basicROS commands, the ROS packages, nodes, topics, and messages are explored to gainan overall knowledge of these ROS robotic systems. Technical information on theseexample robots is provided to describe the robot's full capabilities.ROS encompasses a full spectrum of software concepts, implementation, andtools that attempt to provide a homogeneous view of the complex systems andsoftware integration required in robotics. Extensive libraries of sensor and actuatordrivers and interfaces are already in place, as well as the latest and most efficientalgorithms. What ROS doesn't provide directly is imported from other prevailingopen source projects such as OpenCV. ROS also possesses a spectrum of time-savingtools to control, monitor, and debug robot applications: rqt, rviz, Gazebo, dynamicreconfigure, and MoveIt, to name a few.[ xi ]

PrefaceIn the pages that follow, each of these areas will be incrementally introduced to thereader as part of the robot examples. With TurtleBot, the subjects of navigation andmapping are explored. Using Baxter, joint control and path planning are describedfor your understanding. Simple Python scripts are included to provide examples ofimplementing ROS elements for many of these robots. These robots are all availablein simulation to accomplish the exercises in this book. Furthermore, instructions areprovided for you to build and control your own robot models in simulation.The power of ROS, the variety of robots using ROS, and the diversity and support ofthe widespread ROS community make this adventure worthwhile. Extensive onlinetutorials, wiki instructions, forums, and tips and tricks are available for ROS. So diveinto the pages of this book to begin your adventure with ROS robotics!What this book coversChapter 1, Getting Started with ROS, explains to you the advantages of learning ROSand highlights the spectrum of robots currently using ROS. Instructions for installingand launching ROS on a computer running an Ubuntu operating system are provided.An overview of the ROS architecture is given and its components are described. TheTurtlesim simulation is introduced, and used to provide a deeper understanding ofhow the components of ROS work and a familiarity with ROS commands.Chapter 2, Creating Your First Two-Wheeled ROS Robot (in Simulation), introduces youto the ROS simulation environment of Gazebo. We will lead you through the steps tocreate your first robot simulation (a two-wheeled differential-drive base) and teachthe structure of the Universal Robotic Description Format. The use of the ROS toolrviz and Gazebo are detailed to enable you to display your robot and interact with it.Chapter 3, Driving Around with TurtleBot, introduces you to real ROS robots,TurtleBot2 and the recently available TurtleBot 3. These mobile base robots canbe used in the simulation environment of Gazebo if you do not own one. ROScommands and Python scripts are used to control TurtleBot through a variety ofmethods. The ROS tool rqt is introduced, and subsets of its plugins are used tocontrol TurtleBot and monitor its sensor data.Chapter 4, Navigating the World with TurtleBot, explores visual sensors and the abilityfor a robot to map its environment. The 3D sensor options for TurtleBot's visionsystem are described and their setup and operation using ROS enables TurtleBotto navigate autonomously. The knowledge of the Simultaneous Localization andMapping techniques is applied in combination with TurtleBot's navigation stack tomove about in the mapped environment.[ xii ]

PrefaceChapter 5, Creating Your First Robot Arm (in Simulation), provides a gentle introductioninto the complexity of robotic arms. A simulated robot arm is designed and builtusing the macro language of Xacro. Controllers for the arm are created to operate thearm in Gazebo. Through developing the controllers for this arm, an insight into themechanics and physics of a simple robot arm is offered.Chapter 6, Wobbling Robot Arms Using Joint Control, takes a deeper look at theintricacies of controlling robotic arms. Baxter has two 7 degree-of-freedom arms anda number of other sensors. Baxter Simulator is available as open source software touse for the instructions in this chapter. Examples are provided for control of Baxter'sarms using position, velocity, and torque modes with control for both forwardand inverse kinematics. The ROS tool MoveIt is introduced for motion planning insimulation and execution on either a real or simulated Baxter.Chapter 7, Making a Robot Fly, describes a growing area of ROS robotics—unmannedair vehicles. This chapter focuses on quadrotors, and an understanding of quadrotorhardware and flight control is provided. Instructions for downloading andcontrolling the simulated quadrotor Hector are supplied. With skills from flying asimulated quadrotor, you can move on to control a real Bitcraze Crazyflie or ParrotBebop. Quadrotor control is via teleoperation or ROS topic/message commands.Chapter 8, Controlling Your Robots with External Devices, presents a number ofperipheral devices you can use for controlling a ROS robot. Joystick controllers,controller boards (Arduino and Raspberry Pi), and mobile devices have ROSinterfaces that can be integrated with your robot to provide external control.Chapter 9, Flying a Mission with Crazyflie, incorporates many of the ROS componentsand concepts presented in this book into a challenging mission of autonomous flight.The mission involves the Crazyflie quadrotor flying to a "remote" target all mappedthrough a Kinect 3D sensor. This mission uses ROS message communication andco-ordinate transforms to employ the Kinect's view of the quadrotor and targetto orchestrate the flight. Flight control software for Crazyflie using PID control isdescribed and provided as part of the mission software.Chapter 10, Controlling Baxter with MATLAB , delves into a new realm ofcommunicating with and controlling ROS robots through MATLAB and its RoboticsSystem Toolbox. Baxter, the two-armed robot introduced in Chapter 6, Wobbling RobotArms Using Joint Control, will be used to show how to set up a ROS robot in MATLABby adding custom messages into the Robotics System Toolbox. Communication andcontrol of Baxter and his arms will be accomplished using MATLAB scripts and ROScommands.[ xiii ]

PrefaceWhat you need for this bookThe format of this book is intended for the reader to follow along and performthe instructions as the information is provided. The reader will need a computerideally with Ubuntu 16.04 (Xenial Xerus) installed. Other Ubuntu versions andLinux distributions may work, as well as macOS, Android, and Windows, butdocumentation for those versions will need to reference the ROS wiki(http://wiki.ros.org/kinetic/Installation).The version of ROS that this book was written around is Kinetic Kame, which is thecurrent release recommended for stability. Its end of life is targeted for April 2021.All software used in this book is open source and freely available for download anduse. Instructions for downloading the software are found in the chapter where thesoftware is introduced. In Chapter 1, Getting Started with ROS, instructions are givenfor downloading and setting up the ROS software environment.Our preferred method to download software is the use of Debian packages. Whereno Debian packages exist, we refer to downloading the software from repositoriessuch as GitHub.Gazebo simulation performs intensive graphics processing, and the use of adedicated graphics card is advised but not required.Peripheral devices, such as 3D sensors, Xbox or PS3 controllers, Arduino orRaspberry Pi controller boards, and Android mobile devices are optional equipment.Who this book is forIf you are a robotics developer, whether a hobbyist, researcher, or professional, andare interested in learning about ROS through a hands-on approach, then this book isfor you. You are encouraged to have a working knowledge of GNU/Linux systemsand Python.ConventionsIn this book, you will find a number of text styles that distinguish between differentkinds of information. Here are some examples of these styles and an explanation oftheir meaning.[ xiv ]

PrefaceCode words in text, directory names, filenames, file extensions, and pathnames areshown as follows: "The Terminal commands rostopic and rosnode have a numberof options "A block of code is set as follows: ?xml version '1.0'? robot name "dd robot" !-- Base Link -- link name "base link" visual origin xyz "0 0 0" rpy "0 0 0" / geometry box size "0.5 0.5 0.25"/ /geometry /visual /link /robot To avoid repeating previous code blocks, but provide with placement of new codeblocks, previous code left for reference is abbreviated and grayed-out as follows: ?xml version '1.0'? robot name "dd robot" !-- Base Link -- link name "base link" /link !-- Right Wheel -- link name "right wheel" Any command-line input is written as follows: rosrun rqt reconfigure rqt reconfigureOutput from command is written as:[ INFO] [1427212356.117628994]: Starting turtlesim with node name /turtlesimNew terms and key words are shown in bold.Words that you see on the screen, for example, in menus or dialog boxes, appear inthe text like this: "By clicking on the Add button on the Displays panel."[ xv ]

PrefaceURL references are shown as: http://www.ros.org/about-ros/Warnings or important notes appear in a box like this.Tips and tricks appear like this.Reader feedbackFeedback from our readers is always welcome. Let us know what you think aboutthis book—what you liked or disliked. Reader feedback is important for us as it helpsus develop titles that you will really find useful and enjoyable.To send us general feedback, simply email feedback@packtpub.com, and mentionthe book's title in the subject of your message.If there is a topic that you have expertise in and you are interested in either writingor contributing to a book, see our author guide at www.packtpub.com/authors.Customer supportNow that you are the proud owner of a Packt book, we have a number of things tohelp you to get the most from your purchase.Downloading the example codeYou can download the example code files for this book from your account athttp://www.packtpub.com. If you purchased this book elsewhere, you can visithttp://www.packtpub.com/support and register to have the files emailed directlyto you.You can download the code files by following these steps:1. Log in or register to our website using your e-mail address and password.2. Hover the mouse pointer on the SUPPORT tab at the top.3. Click on Code Downloads & Errata.4. Enter the name of the book in the Search box.[ xvi ]

Preface5. Select the book

the Python programming language. His interest in robotics, image processing, and Python started with that project. After his graduation, Lentin worked for 3 years at a start-up company focusing on robotics and image processing. In the meantime, he learned to work with famous robotics software platforms, such as Robot Operating System (ROS), V-REP,