Polycopters

From Vague Hope Wiki
Revision as of 16:40, 30 January 2011 by Haku (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Background reading

Hardware choices

CPU

  • Android phone - free GPS/gyro/cam/Wifi, but long boot time - even with a separate watchdog, crashed phone = crashed copter. Still need to break out an interface to motor control board.
  • Arduino 2009 / UNO - nice hardware as has A+D I/O built in + neat shields for motor control. GPS/gyro/cam/control method will still need to be worked out. Memory space too limited for full autonomy?
  • Arduino Mega - as above, more expensive but more memory, enough for full auto?

Altitude sensor

  • Barometer - least sensitive, probably best for high-flying vehicles. Needs to be kept away from the wash from the propellers.
  • Ultrasound sonar - very sensitive for low-level flight, but tops out at a few metres.
  • IR laser - Pete suggested this on my blog, I'd have thought performance would degrade outdoors, waiting to hear back. He's probably the expert here.

Motors & Propellers

Number-crunching needed here to figure out what kind of lift is generated by affordable components, and what that limits our chassis/hardware weight to.

Hardware required for behaviours

Behaviours Wanted Hardware Needed
Autopilot: takeoff, land, hover, up, down, pitch, roll, move wrt vehicle orientation 3-axis gyro & accelerometer
Autopilot: move wrt north

Autonomy: local time-based behaviours (think LOGO)

2/3 axis magnetometer
Autopilot: set altitude Barometer / ground-aimed sonar / IR laser rangefinder
Autonomy: outdoor location-based behaviours, accuracy ~10m GPS
Remove magnetometer requirement while in GPS coverage (not cost-effective) GPS Heading Unit
Outdoor position accuracy ~1m RTK GPS
Outdoor position accuracy ~1cm RTCM GPS
Autonomy: indoor location-based behaviours, accuracy ~1m Multi-band RF receiver capable of triangulation, plus transmitters
Indoor landing accuracy ~10cm Camera with image recognition (too much code for Arduino Mega)
Autonomy: dynamic avoidance Outward-aimed sonar or radar
Autonomy: dynamic docking Radar

Geekspeak

  • AP/Autopilot - low-level control processing, e.g. correcting for tilt due to wind, keeping level for landing. Think "Pilot".
  • Autonomy - high-level control processing, e.g. navigating to certain points, loitering. Think "Navigator".
  • RF - radio frequency
  • RTK / RTCM - Real Time Kinematic / something, read this: http://linux.die.net/man/5/rtcm-104 Very little chance we'll use either.

Objectives

Object 1: Platform that hovers stationary in the air.

Parts

(handy reference: http://aeroquad.com/wiki/index.php/Parts_List)

TODO: Find UK-bases suppliers for these components.

Objective 1

  • Frame - balsa and ducktape? - +1 from Ian, this has got to be the easiest way to start
  • Motors
  • Motor controllers - Arduino shield?
  • Props
  • Battery
  • Logic circuitry - Arduino? I already have an Arduino Mega. -はく
  • sensors - at least 3-axis gyro and 3-axis accelerometer?

HobbyKing