Subject:
|
Re: Line Following by Humans versus Bots
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Tue, 16 Nov 1999 14:55:00 GMT
|
Viewed:
|
488 times
|
| |
|
|
Luis Villa wrote:
>
> 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.
Fascinating, I tried the search but unfortunately nothing related turned
up.
Coincindentally, there's another thread on line following based on the
conventional single black line. There is the suggestion that the line
has to be somewhat modified (? cheating) to make it easier for the bot
to follow.
Carrying this further, I'm suggesting that maybe if bots were given two
parallel lines within which they must stay (just like humans), they
might do a better tracking job than following a single line.
Well, I guess the next step is to make a test pad to see if the idea
will work.
--
C S Soh
cssoh@singnet.com.sg
http://web.singnet.com.sg/~cssoh
...where air is power
|
|
Message is in Reply To:
 | | 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 (...) (25 years ago, 16-Nov-99, to lugnet.robotics)
|
22 Messages in This Thread:
 
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
