|
[DPRG] Allen Robots (Will's emotions)
Subject: [DPRG] Allen Robots (Will's emotions)
From: Randy M. Dumse
rmd at newmicros.com
Date: Fri Mar 9 17:07:21 CST 2007
> Anybody else wrting robot code that is basically the
> behavior-based or "subsumption"
> architecture, while adding a supervisor or monitor layer
> above that to control weights and ordering of the lower level
> behaviors?
I've shown you (dpa) my code in my Mini-Sumo. (Anyone can see it
on our web site under downloads pages, if the code I think is
there is up). Basically, the robot has two tricks, Line
Following and Mini Sumo fighting. There is a supervisory machine
that enables the lower machines. When the Supervisor is idle, so
is everything else. When the supervisor is in the line following
"state of mind" (could we say this was the line following mood?
as if it were an emotion?) a line following machine is allowed
to leave the idle state. When the supervisor is in the Sumo
"state of mind" two machines are allowed to leave the idle
state, one that attacks forward, and another that backs away
when it sees an edge. The Back Away subsumes the Attack.
So I feel I have 1) Behavior Based 2) Subsumption running under
a 3) Supervision layer, which has states somewhat like emotions,
because changing mode completely alters who owns the inputs
(sensors) and (outputs). Yet some inputs cause the Supervisor to
change state, like entering and leaving emotions.
Randy
www.newmicros.com
More information about the DPRG mailing list
|
