[mowbot] mapping
robin nospam at acm.org
Thu, 26 Sep 1996 03:01:00 +0100
Dave Everett <Deverett nospam at vir.idx.com.au> wrote:
> If we construct the sensors properly and connect them efficiently to the
> behaviours, the robot will perform very well without any explicit
> programming. Just by using the grass sensors discussed before, sensors
> to determine concrete/grass, some bump sensors and a low battery sense,
> should be sufficient to achieve some very useful emergent behaviours.
I agree. But I wouldn't want to rule out having a map in there too just
on ideology.
> Like Connell, I am not keen on persistent states, preferring a mostly
> reactive system.
Emergent behaviour works better with more information, and recent history
of actions is certainly relevant---though it would be interesting to
see just how good it can be without such knowledge.
Robin.
--
Copyright (C) 1996 R.M.O'Leary <robin nospam at acm.org> All rights reserved.
For licence to copy, see http://dragon.swansea.linux.org.uk/mowbot/