"Explore" could be as simple as wandering aimlessly about without
bumping into things, or making a map to follow at a future point.
Things change. Objects move, or appear and disappear (for a robot.)
This might make a map obsolete really fast.
Since it (would) already know how to map once, why not simply run the
mapping algorithm any time the bot moves? Using an 'old' map and then
adapting on-the-fly to any changes... sounds like twice the trouble.
Some localization device (Pololu beacons for instance) installed on
some immovable object, then you're just a vector (or two) away from
exact positioning, with 'standard' object-avoidance code.
While detailed 'memory of a location' sounds desireable, I wonder
about the utility of it. The same results might be obtained in an
easier manner.
I'm posting a file about Map Creation. (Not mine, but very
informative.)
Good luck to you in this ambitious adventure!
--- In oopic@yahoogroups.com, "jasuk1970" <mail@...> wrote:
>
> In my case I'm pretty new to all this too. So my plans may change
> radically :).
>
> My main idea is to put robot at point A (A known point) Then make
it
> explore the area.
>
> My ideas currently behind exploration is to create a routine that
> will turn a servo a known amount. Then use this amount as a unit of
> the map. (For example one wheel rotation). The robot will on move
> forward and rotate on the spot (By 90 Degrees). It will find
> obsticals using IR sensors and black of parts of the map as un-
> passable.
>
> Once the map has been finished. I will put the robot at point A,
> select point B on the map by grid coordinate and then the PC will
> work out the quickest way to the target (Using 45 degree rotations)
>
> If a non plotted obstacle gets detected, the robot will wait a
> certain number of seconds (E.g. for the cat to get out the way :) )
>
> Before relaying this as a new blockage to the PC and asking for a
new
> route.
>
> The whole system hinges on the ability to accuratly measure
movement.
>
> Jas
>
> --- In oopic@yahoogroups.com, "design14me2" <design14me2@> wrote:
> >
> > Ahhh my first post. Yes I'm a nubbie.
> >
> > You guys are on a very interesting topic for me. This of course
is
> > the same question for an extension of my own project. I got my
> first
> > OOPIC about 10 days ago and still brain storming on how I'm going
> to
> > get to my final result.
> >
> > My question is:
> >
> > How do you plan on generating the map? I came up with storing
the
> > map or terrain in the computer.
> >
> > My reasoning is:
> >
> > 1. The map file could get rather large and become too much for
> > the onboard memory.
> >
> > 2. The computer could do more calculations and would be faster
> > than the (OOPIC). And all I would need to do is tell the
vehicle's
> > brain (OOPIC) to turn right, left, climb, and stop or come home.
> >
> > 3. How would the vehicle know where it was in relation to the
> > map? If the vehicle started at point "A" on its way to point "B"
> It
> > would need GPS. Or would it?
> >
> > 4. To could the OOPIC handle all that data while attempting to
> > fly, drive or crawl its way around?
> >
> > OOPIC-S
> > Visual Basic
> > Various CAD programs
> >
> > Lamont
> >
>
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/oopic/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/oopic/join
(Yahoo! ID required)
<*> To change settings via email:
mailto:oopic-digest@yahoogroups.com
mailto:oopic-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to:
oopic-unsubscribe@yahoogroups.com
<*> Your use of Yahoo! Groups is subject to:
No comments:
Post a Comment