To LUGNET HomepageTo LUGNET News HomepageTo LUGNET Guide Homepage
 Help on Searching
 
Post new message to lugnet.roboticsOpen lugnet.robotics in your NNTP NewsreaderTo LUGNET News Traffic PageSign In (Members)
 Robotics / 8280
8279  |  8281
Subject: 
Re: Line Following by Humans versus Bots
Newsgroups: 
lugnet.robotics
Date: 
Wed, 17 Nov 1999 16:55:16 GMT
Original-From: 
Paul Speed <pspeed@augustschell!spamcake!.com>
Viewed: 
938 times
  
C S Soh wrote:

Luis Villa wrote:

On Tue, 16 Nov 1999, Ralph M. Deal wrote:
On Tue, 16 Nov 1999, Stefano Franchi wrote:

Although it's an "old" book now by scientific standards, Valentino
Brateiberg's Vehicles (MIT press I believe, still in print) contains a
clear and accessible discussion of this topic, plus reference to the
standard literature. I suppose you may then proceed from there. And it's
required reading for any RCX'er anyway...

The reference is Vehicles by Valentino Braitenberg, MIT PRess 1984.
(Took awhile to find that!)

Wish I had a copy.    Ralph    Deal@kzoo.edu

Actually, it is pretty cheap on amazon (14 + shipping)- check it out here:
http://www.amazon.com/exec/obidos/ASIN/0262521121/o/qid=942778058/sr=8-1/102-8483281-8060851


Thanks, I'll have to get myself a copy.

Judging from the lively discussions in an accompanying thread on
line followers, it seems that following a single black line is far
from being a trivial matter.

You are right that with one stationary light sensor, line
following is not a trivial matter.  Change the constraint and it
starts to become a little more trivial... we're just splitting the
problem by focusing on both the mechanical and programming domains.


It seems to me more natural to be keeping within a hallway (or the
walls of a canyon in the Star Wars games). And it's alright for bots
to bump into walls but not humans, of course.

Sure, my 'bot can bump into walls but it's impossible for it
to know where it is or what direction it is facing without some other
tricks.  Just imagine approaching a T junction with an angle just
slightly off of the traveled hallway.  If the 'bot clips the left
corner of the T any number of things can go wrong.  The 'bot may
never notice and yet be turned enough that now it thinks it's
traveling down the left branch.  Or it may detect the slight hit
and assume that a wall is there and miss the left branch altogether.

This also does not sound trivial.  Not to mention that we
are still a long way from being able to work up a decent maze
traversal algorithm.  So as hallway following solutions become more
complicated, line following seems more attractive given it's
positional benefits.


I would really like to know who or how was it decided that bots
should follow a single line in the first place. Is it just to save
on black paint or the number of light sensors?

Any clues?

The primary motivation is because a hallway is wide and a
line is narrow.  This means that when the 'bot is on a line it has a
reasonably good idea where it is.  Also, in maze following, when it
reaches an intersection it can see all of its various choices without
randomly bumping around for five minutes.  And even after five minutes
of randomly bumping around, there might still be one option the 'bot
missed.

It will certainly be interesting to continue to see the
various solutions in both areas.
-Paul (pspeed@progeeks.com, http://www.progeeks.com/)



Message has 2 Replies:
  Re: Line Following by Humans versus Bots
 
Paul Speed wrote: <snip> <snip> (...) Thanks for the thoughts. I'm not sure if it's been done before, but I'm now making a playing field where the bot will have to keep within two black lines using 2 RCX light sensors to see how it compares to (...) (25 years ago, 18-Nov-99, to lugnet.robotics)
  Re: Line Following by Humans versus Bots
 
(...) I am really agreeing with you, but since yesterday I begun thinking a little in other pathways... The specifications for a general line following robot have to include also a specification for the kind of line/pathway the robot can follow. So (...) (25 years ago, 18-Nov-99, to lugnet.robotics)

Message is in Reply To:
  Re: Line Following by Humans versus Bots
 
(...) Thanks, I'll have to get myself a copy. Judging from the lively discussions in an accompanying thread on line followers, it seems that following a single black line is far from being a trivial matter. It seems to me more natural to be keeping (...) (25 years ago, 17-Nov-99, to lugnet.robotics)

22 Messages in This Thread:










Entire Thread on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact

This Message and its Replies on One Page:
Nested:  All | Brief | Compact | Dots
Linear:  All | Brief | Compact
    
Active threads in Robotics

 
Verified and Trusted Team of Hackers
1 hour ago
Custom Search

©2005 LUGNET. All rights reserved. - hosted by steinbruch.info GbR