Subject:
|
RE: Line Following by Humans versus Bots
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 16 Nov 1999 19:20:28 GMT
|
Original-From:
|
Jim Thomas <[jim.thomas@trw.com]IHateSpam[]>
|
Viewed:
|
562 times
|
| |
|
|
An example of this effect is that spinning tunnel that Universal Studios has
(had?) that was also featured in an episode of the Six Million Dollar Man
(the Big Foot episodes, IIRC). The walls of the tunnel spin clockwise as
you move through it. It basically makes you *feel* like you are falling
over sideways -- very disorienting. As far as the fly goes, it would be
interesting (devious?) to get one of those "holding pattern" flies that
always flies straight then turns 90 degrees and put it in the sliding wall
hallway. Would it fly in arcs or just turn more or less than 90 degrees?
Inquiring minds want to know...
JT
> -----Original Message-----
> From: Luis Villa [mailto:liv@duke.edu]
> Sent: Monday, November 15, 1999 8:08 PM
> To: lego-robotics@crynwr.com
> Subject: Re: Line Following by Humans versus Bots
>
>
> I wish I had time to post more on this right now, but suffice
> it to say
> there has been a lot of research on how humans and animals do
> this type
> of thing. Staying within a hallway without bumping into the walls is
> actual done by taking a mental snapshot and comparing it to the next
> "frame"- in that way, we determine the relative motion of each of the
> areas we are looking at, and attempt to make those relative
> motions fit
> our ideas of what we should see (i.e., they should both be moving the
> same amount if we are going down the center of the hall, left
> wall should
> be moving more if we are turning left, etc.)
>
> The experiments that have done to "prove" this have been done
> mainly on
> flies. By putting them in a "hallway" where one wall is
> actually moving
> (scrolling, IIRC) they tend to turn towards or away from that
> wall, as
> appropriate.
>
> I can't find any URLs at the moment, but if you want to
> search for it,
> the technical term for what we produce is a "flow field"
> because we are
> determining how objects around us are "flowing." Look for
> that and see
> what you can find.
> -Luis
>
> P.S. Oh yeah- do I even need to mention that I know about this
> because it is the topic of extensive research, and is essentially
> impossible to do without video cameras and complex
> algorithms? Not much
> computation involved (a fly can do it, after all) but still complex.
>
> On Tue, 16 Nov 1999, C S Soh wrote:
> > I wonder if bots were made to follow parallel lines the way • we humans
> > do, would just 2 light sensors be necessary and sufficient for the
> > purpose? And would the bot move smoothly without jerking to • and fro like
> > a druken driver?
> >
> > --
> > C S Soh
> >
> > http://web.singnet.com.sg/~cssoh
> > ... where air is power
> >
> >
>
> ##############################################################
> #########
>
> Profanity is the one language that all programmers understand.
> -Anonymous
>
> ##############################################################
> #########
>
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
