Phooey on all that complex programming.  I want my mowbot to be as stupid as
a cochroach.  Did you know, a cochroach is so stupid, it cannot even
individually address it's own legs?  They have about four modes - go slow
with six legs, go fast with six legs, go with the right legs only, go with
the left legs only.  Robot scientists puzzle over how they clamber over tall
obstacles with no finesse, they just scrabble until they get over it.  No
finesse at all.

Now, who would argue that cochroaches are not wildly successful?

My mower is going to stumble aimlessly across the yard until it bumps into
the barrier at the edge, back up, turn a random number of degrees, and
wander aimlessly until it bumps into something again.  You get good coverage
by randomness.  Stupid as a cochroach!

Also, I don't want any complex rocket science edge sensing.  A line of big
dumb bricks is what I'm considering.  I'd rather not find my Bot mowing the
nieghbor's flowers when the house-of-cards edge sensing scheme fails.  As
long as the edge is a barrier the bot cannot climb over, then bumpers are
all that I need.

Besides, I've got to finish this project by next summer.

--Lawrence Lile

> There's one mowbot behavior question I'd like to get on the table: How are
you
> planning the cutting path? I'm planning on using bumpers to keep the bot
in
> an enclosed area for now, but I'm wondering if edge sensing, random
pattern,
> planned path (and how to measure that the path is maintained), or some
other
> technique is viable. At this point power efficiency is secondary to
getting
> good cutting coverage.
>
> Any thoughts?
>
> BAJ
>
> --
> http://www.piclist.com hint: The list server can filter out subtopics
> (like ads or off topics) for you. See http://www.piclist.com/#topics
>
>

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics