Edit

Note Viewer

Electrical 2016

Attendance

Note Info

Students
Mentors
Date & Time
Location & Author

  • Sarah H.

  • Adam P.

  • 3/5/2016

  • 3 hours

  • Shapland Shop

  • Elizabeth S.

Work Completed

  • Got Xbox controller triggers working like buttons, supporting whileHeld commands.  Left trigger momentarily switches to mecanums while held
  • Mech controls while using Xbox still needs work
  • Fixed various motor ports being reversed, etc. on practice bot.  
  • Helped Neil find place to mount second camera.  No feed to dashboard yet, probably need to use different build of mjpg_streamer, or run second instance on different port
  • Wired shooter and feeder
  • Jay did a bunch of scouting stuff.  Crimped several ethernet cables.
  • Discussed alternate options for shooter encoders.

Challenges

Adam's Spotify keeps playing Huey Lewis and the News

Work for Next Meeting

  • Test grip with new exposure control.  Find value we like and get RPi startup script to set that prior to firing up streamer/grip
  • Finish rewiring & testing practice bot. Get sensors mounted & wired.  Everything we can do while waiting on feeder & shooter
  • find a way to rotate camera
  • Find best way to see/modify camera settings via command line on linux.  Play with brightness/exposure/etc. to get the best isolation of targets in imagebefore GRIP gets ahold of the stream.  Then adjust GRIP file to make the best use of that.
    • On some websites, they connect the video to the HSV filter twice so that there are two intervals for each. Might want to check this out.
  • Work with mechanical to find position for new forward-facing camera
  • Drive correction needs changes to handle crossing the 0-to-360 border without making the robot spaz out
  • Start autonomous code
  • Work on vision processing
    • Faster this is done the more testing we can do with the shooting
  • Need to SSH in and shutdown Pi nicely when disabled or match over.
  • Find good settings for mjpg-streamer (Look into the fork)
    • keep optimizing (It seems that brightness and contrast being near zero works extreemly well
  • Check on spike supply and maybe order more
  • Work on wiring of RGB LEDs
  • Look at Nolan's new and improved controls list
  • Started code to make indexer smarter by using (yet unmounted) limit switch
  • Sarah and Kyle did some test scouting for Jay

On Schedule?

nope nope nope