Shop OBEX P1 Docs P2 Docs Learn Events
11-year old's question about Scribbler line following abilities — Parallax Forums

11-year old's question about Scribbler line following abilities

dpetra5dpetra5 Posts: 2
edited 2009-11-19 23:25 in Robotics
I'm using my Dad's log in.· My 5th grade science fair project is to see whether my Scribbler robot can follow different colored paths.· I tested black, dark blue and red using electrical tape.· Why could my Scribbler only follow the black line?· My Dad says it has something to do with the light the path sensor used but that doesn't make sense to me.

Comments

  • SRLMSRLM Posts: 5,045
    edited 2009-11-19 05:52
    From the hackers hints (on the scribbler download page):
    Hacker Hints said...
    The line sensors will give their best guesses no matter what color surface is beneath the Scribbler.
    Light colors will be seen as white, and dark colors as black, but this isn’t as reliable as using true
    black and white.

    In short, the scribbler can only give you one of two answers: light or dark. Normally, line following robots use white as "light" and black as "dark", but you could also use something like sky blue as "light" and maroon as "dark", albeit with a possible higher chance of misreads.
  • dpetra5dpetra5 Posts: 2
    edited 2009-11-19 09:30
    This is his Dad, the blue was dark and on a white background. The red was also fairly dark and on a white background. Scribbler could not detect them at all. Just kept searching. I assumed it had something to do with infrared light sensors
  • BeanBean Posts: 8,129
    edited 2009-11-19 12:19
    I think you are right. It depends on how the colored tape reflect the infrared light. There probably is no way to know unless you try them. Some colors that at dark in visable light, might be quite refective to IR light.

    Bean.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    Does that byte of memory hold "A", 65, $41 or %01000001 ?
    Yes it does...


    ·
  • WhitWhit Posts: 4,191
    edited 2009-11-19 15:01
    dpetra5 (both),

    The reflectivity of the line does make a lot of difference. Usually black electrical tape is too reflective for most Scribblers to read. The best thing I have found is a black printed line (not copied). Even then, some inks are too reflective. A very flat black line is best.

    There is also·some variation in Scribblers. Some just "see" better than others in my experience.

    Great question!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Whit+


    "We keep moving forward, opening new doors, and doing new things, because we're curious and curiosity keeps leading us down new paths." - Walt Disney
  • BeanBean Posts: 8,129
    edited 2009-11-19 15:21
    I wonder if "friction" tape would work better. It is a cloth-like material that is a very flat black color.

    Bean.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    Does that byte of memory hold "A", 65, $41 or %01000001 ?
    Yes it does...


    ·
  • ercoerco Posts: 20,256
    edited 2009-11-19 16:50
    These guys are 100% right; it's all about high contrast and low reflectivity for the black lines to follow. Even black electrical tape is glossy & reflective, Scribbler can follow some types, not all. And IR light behaves quite differently than the visible light we see.

    For my Scribbler dock, the robot followed "tracks" I made from 1" strips cut from black posterboard. And I used the less-glossy side of the black posterboard. Background was white posterboard, glossy side up, and I had excellent results. I lightly taped the black strips down. This way, it is easy to pull up the strips and make a new configuration. For a more interesting path, you could also cut curves from the black·posterboard.

    http://www.botmag.com/articles/scribbler.shtml

    For your science fair project, you could use this method and try strips of different colored posterboard, or even construction paper, as long as the dark line has a matte (dull) finish. I would still use a bright shiny background, such as posterboard. Good luck!

    BTW, the line sensors' sensitivity is fixed by hardware (I'll bet the upcoming Scribbler 2 will have software-adjustable sensitivity!).·I've considered manually (and permanently) decreasing the· sensitivity by opening it up and adding a series resistor to each IR LED underneath, but haven't done it yet. Some experimentation would be necessary to determine the resistor's value, and that would depend on individual circumstances.

    This guy had it easy, following dull grout lines between shiny white tiles! http://www.youtube.com/watch?v=AmjvZfAzGq4

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    ·"If you build it, they will come."
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2009-11-19 17:36
    The best tape I've found for line following is a flat black "crepe" paper masking tape from McMaster-Carr. Once there, do a search on 76265A41 (1/2"), or 76265A42 (3/4"). You will have to have cookies enabled to do the search.

    -Phil

    Post Edited (Phil Pilgrim (PhiPi)) : 11/20/2009 2:03:28 AM GMT
  • WBA ConsultingWBA Consulting Posts: 2,934
    edited 2009-11-19 22:38
    So, theoretically, you could have a complete black surface that absorbs IR and use black tape that reflects IR and the scribbler would appear to follow an imaginary line since all a human would see is all black?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Andrew Williams
    WBA Consulting
    WBA-TH1M Sensirion SHT11 Module
    Special Olympics Polar Bear Plunge, Mar 20, 2010
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2009-11-19 22:46
    Yes, that's correct.

    -Phil
  • Steph LindsaySteph Lindsay Posts: 767
    edited 2009-11-19 23:05
    Good advice here!

    One more thing to keep in mind is the thickness of the material you are using for creating the line. The sensors under the Scribbler are designed to work at a fixed distance above the floor. The infrared light is shined down at a certain angle to bounce off the floor and into the angled receiver tubes to catch the reflected light just right.· If you make your line out of thick material, such as heavy cardboard, the distance from the line-surface to the sensors would decrease, and Scribbler might not be able to see the line. Likewise, if you were to put fatter tires on the wheels, the distance between the floor and the sensors would increase, and again the Scribbler might not be able to see the line.

    I found that some inkjet printers soak the paper so much that it becomes really warped and wrinkly, and the Scribbler can have a hard time seeing that because of the changing distance between the line and the sensors.· Regular toner seems to work well. The ink used in home printers that print photographs doesn't work at all. That black ink is actually very shiny and reflects infrared light just as well as the white paper. That was a big surprise!

    We discovered not everything that·looks black to the human eye is really absorbing infrared light. We can't see infrared light, so we don't know if a black object is absorbing it or reflecting it. Our robots with infrared detectors are better at this than we are!

    Once for an event, I had to build a short wall to go around a SumoBot robot-wrestling ring. These robots are using infrared to "see" each other in the ring, similar to the Scribbler's object-detector sensors in the front, so they can attack or evade each other. So, the wall had to be invisible to them, that is, to absorb infrared light. We went to a craft store with the little robots, and tested all kinds of poster board, felt, fabric and papers. The· To make matters more complicated, some the overhead fluorescent ceiling lights in the store were giving off infrared light the robots could see, messing up our tests. So, we would have to take a bolt of fabric across the store to test it. The sales staff sure was looking at us funny! We were surprised how many things would not work - even the black velvet fabric reflected infrared light!

    Finally, we found a heavy construction paper that was pretty good. We used the paper to make kind of a half-pipe curve up the walls, so any infrared light shined on it that was not absorbed would be reflected upwards and away from the robots.

    Have fun experimenting!
  • Beau SchwabeBeau Schwabe Posts: 6,566
    edited 2009-11-19 23:25
    Yes, good advice here, but this got me thinking....

    "A new, invisible to the naked eye and infrared (IR) readable, ink formulation based on silicon(IV) 2,3-naphthalocyanine bis(trihexylsilyloxide) as colorant is described. The ink formulation is based on absorption and reflection of light in the IR, in contrast to the currently employed mechanism of excitation and fluorescence detection. The ink absorbs strongly at 790 nm and has highly transmitting characteristics in the visible. A good contrast level was obtained with an illumination source emitting at 790 nm. The ink is stable, lightfast and applicable to offset printing, and can also be extended to other printing techniques."

    Invisible IR Ink


    www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6TFY-3YF4P4N-4&_user=10&_rdoc=1&_fmt=&_orig=search&_sort=d&_docanchor=&view=c&_searchStrId=1101383480&_rerunOrigin=google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=9ac208b4ae8263a0e789ec5525b343f9

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
Sign In or Register to comment.