Docstoc

Embedded Microcomputer Systems

Document Sample
Embedded Microcomputer Systems Powered By Docstoc
					                               1




Embedded
Microcomputer Systems
Andrew Karpenko




Prepared for



      Technical Presentation
      February 25th, 2011
                                                                             2




A.R.Drone by Parrot
Augmented Reality Drone

• A helicopter with four rotors
• Internally stabilized
• Controlled by any device with WiFi
• Originally designed as a game
  accessory for the Apple iPhone and
  iPod touch platforms

                                       • On-board computer vision
                                       • Front and down facing cameras for
                                           video streaming and object
                                           detection
                                       •   Down facing ultrasonic distance
                                           telemeter
                                       •   6 Degree Of Freedom inertial
                                           measurement unit
                                       •   4 ARM7 Microprocessors
                                       •   1 ARM9 Microprocessor
                                        3




Make Platform
• Make Controller Kit
  • ARM7 Microprocessor
  • Ethernet
  • USB
  • 8 Analog Inputs
  • 8 Digital Inputs/Outputs
  • 4 LEDs



                • ARM7 Microprocessor
                    • 48MHz
                    • 256KB Memory
                                                                  4




Development Board
Custom development platform designed by Department of
Electrical Engineering at UW.



             LCD                                    Keypad


  Bread Board                                           Speaker


 Make
Controller                                                Motor
   Kit
                                  5




Projects
Completed over a 10 week period
in groups of two and four:


• Lab 1
   • Introduction
• Lab 2
   • Scheduling, Input/Output
• Lab 3
   • FreeRTOS, PWM and Drone
     Control
• Lab 4
   • Manual and Autonomous
     Drone Control
                                                                       6




Lab 1 - Introduction
              • Learn the basics of the Make Controller and the
 Controller     Development Board



              • Review and get familiar with the C programming
     C          language




   LEDs       • Flash Light Emitting Diodes at different frequencies




   LCD        • Display a message on the Liquid Crystal Display
                                                                     7




Lab 2 – Scheduling, Input/Output
         • Create a simple Operating System for handling multiple
  OS       tasks
         • Build a scheduler that runs tasks in a given time frame




  I/O    • Read values from an analog sensor
         • Write to and Read from digital Input/Output pins



         • Drive a motor using pulses
 Motor   • Vary the motor speed using values gathered from the
           analog sensors




  LCD    • Output motor speed and sensor values on the display
                                                                     8




Lab 3 – FreeRTOS, PWM and Drone Control

           • Transition to the FreeRTOS embedded Real Time
  RTOS       Operating System that guarantees executed code will
             meet deadlines




  PWM      • Implement Pulse Width Modulation to drive the motor
             based on sensor input




   ISR     • Implement an Interrupt Service Routine to measure the
             revolutions of the motor using a speed encoder



           • Read values from the keypad
 Control   • Generate drone commands based on keypad input
           • Fly the AR.Drone by sending commands to it
                                                              9




Lab 4 – Manual and Autonomous Control

  Manual      • Manually control the drone to navigate a
  Control       physical obstacle course


              • Use a Wii Classic Controller as the control
 Controller     interface for the drone


 Autonomous   • Create an algorithm that will autonomously
   Control      navigate the drone in a predefined path
                                                                               10


Quadrotor Operation
Manual Control


            left                                 left


                                 front                                 front

    back                                 back

                         right                                 right

        Spin Directions                                 Roll
                                                (Left and Right)



            left                                 left


                                 front                                 front

    back                                 back

                         right                                 right

                 Pitch                              Yaw
        (Forward and Back)                        (Rotation)
                                                      11


Communication
Manual Control

• Drone acts as a Router
• 3 Port UDP Communication
    •   5556: Commands to the drone
    •   5555: Video feeds from drone
    •   5554: Navigation data from the drone
• Different devices can connect to




                                               WiFi
  different ports




                               Ethernet
                                12


Commands
Manual Control

• Send directional commands
  •   Forward/Back
  •   Left/Right
  •   Rotate
  •   Up/Down


• Send all parameters in one
  command
  •   Back while rotating and
      ascending


• Control motor speeds
  directly (Dangerous)
                                                                         13


Controller
Manual Control
• Requires implementing the I2C (Inter-Integrated Circuit) bus




                                                                 Takeoff
Change
Camera




                                                  Forward
                            Up




                                                                  Land
                          Rotate               Left/Right
                            Down




                                                  Back



                           Wii Classic Controller Pro
                                                                 14


Computer Vision
Autonomous Control

• Drone can detect 2D tags
  and other drones
  •   Up to 4 tags
  •   Can tell them apart                A. Color pattern on
                                                outdoor hull
  •   Estimates distance to tags


• Detection works based on
  color pattern of outdoor hull,
  or stickers for the indoor hull

• Used for Augmented Reality
                                    B. Stickers on indoor hull
  games on the iPhone and
  iPod touch platforms
                                                           15


Path
Autonomous Control

Points are awarded based on
how close the drone is to the
starting point at the end of the
flight
                                                B
   1 m height                               Target


     H          Start
                                                       2m
                                                       height
  End
                          A

                        Target

                                                     4m x 4m
                                   1 m height        envelope
                          16


Our Approach
Autonomous Control



•   Computer Vision
•   Position Estimation
•   Closed Loop Control
•   …?
                                                    17




Summary
• A.R. Drone by Parrot
• Make Platform
• Development Board
• Projects 1 - 3
   • Lab 1 – Introduction
   • Lab 2 – Scheduling, Input/Output
   • Lab 3 – FreeRTOS, PWM and Drone Control

• Project 4 – Manual and Autonomous Drone Control
   • Manual Control
     • Quadrotor Operation
     • Communication
     • Commands
     • Controller
  • Autonomous Control
     • Computer Vision
     • Path
     • Our Approach
                                                                                                              18




Embedded
Microcomputer
Systems
Andrew Karpenko
University of Washington
Seattle
                               All images are property of their respected owners.
                               The following images were obtained online:

                               •   http://makezine.com/images/06/intefacecontroller1.jpg
                               •   http://images.bit-tech.net/content_images/2010/07/parrot-ar-drone-review/AR.Drone-06-
                                   b.jpg
                               •   http://www.uncrate.com/men/images/2010/01/wii-classic-controller-pro-xl.jpg
                               •   http://www.prepaid-wireless-guide.com/images/wifi-router.jpg
Prepared for                   •   http://media.digikey.com/photos/NXP%20Semi%20Photos/568-64-LQFP,SOT314-2.jpg
                               •   http://www.makingthings.com/store/media/catalog/product/cache/1/image/5e06319eda
                                   06f020e43594a9c230972d/k/i/kit20_askew.jpg
                               •   http://cheesycam.com/wp-content/uploads/2010/10/ar-drone-parrot-quadricopter.JPG
                               •   http://lasarobotics.org/downloads/Association/Fundraising/sponsor-
      ELP Onsite Visit             supporter_logos/NI%20Logo_large2.JPG
      Technical Presentation
                               Any other images found herein have been personally taken by Andrew Karpenko.
      February 25th, 2011

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:4/19/2013
language:English
pages:18