From aaai97-owner@cc.gatech.edu Thu May 29 18:42:39 1997 Return-Path: aaai97-owner@cc.gatech.edu Date: Thu, 29 May 1997 18:17:42 -0400 (EDT) > From: Barry Werger <barry@cs.brandeis.edu> > Received: (from barry@localhost) by emerald.cs.brandeis.edu (8.8.4/8.8.3) id DAA04415; Thu, 29 May 1997 03:08:30 -0400 (EDT) > Message-Id: <199705290708.DAA04415@emerald.cs.brandeis.edu> > Subject: [tucker@cc.gatech.edu: Find Life: pens and colors] (fwd) > To: aaai97@cc.gatech.edu > Date: Thu, 29 May 1997 03:08:30 -0400 (EDT) > Cc: barry@berry.cs.brandeis.edu (Barry Werger) > X-Mailer: ELM [version 2.4 PL24 PGP2] > MIME-Version: 1.0 > Content-Type: text/plain; charset=US-ASCII > Content-Transfer-Encoding: 7bit > > > Dear Reid and other fellers: > > We are also using a three-color system, and would like to > emphasize that the actual selection of the colors is quite important: > they should be fluorescent if possible, as distinct from each other as > possible, and as distinct from the rest of the environment as > possible. > However, we'd like to have more meaningful markings than just > three colors and so suggest the following: > > Three colors: Black for rocks and danger zones, etc > Neon Lime Green for live Martians > Hot Pink for Dead Martians > > And: Pen Doors marked by large cylinders colored with two large > stripes of Neon Lime Green or Hot Pink separated by Black. > > These colors may need to change, depending on lighting and such. > > Simple patterns of colored stripes I imagine should be good markers > for both the color-based and other types of vision systems; though > objects that will change orientation (such as squiggle balls and > tennis balls) should be solid. We'd be happy to work with any such > simple patterns of three nice colors. > > > Also: we are considering putting some robots inside the pen to > do sorting of the martian stuff. Is it possible that the rules could > be adjusted to penalize stuff in the wrong pen at the end of the run > rather than penalizing stuff being put in the wrong door? Can the > interior pen separation be low (a few cm)? > > Thanks, > > Barry > > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > Barry Brian Werger Interaction Lab > Volen Center for Complex Systems > barry@cs.brandeis.edu Department of Computer Science > http://www.cs.brandeis.edu/~barry Brandeis University > Waltham, MA 02254 > > > --DAB25516.864889759/burdell.cc.gatech.edu-- > >