[Synapse elist] art

[Leonel Moura]

(Wade and Paul)

Since 2001 I did build some art robots that generate original 
drawings and paintings based on emergent processes.
One of these robots, named RAP (Robotic Action Painter) is since 
February 2007 at the American Museum of Natural History in New York 
making drawings without any human assistance.
Once a week the personnel change the felt pens. That's all.
The robot is always alert but only performs when there is someone looking.
RAP also decides when the art work is ready. And when it does goes to 
the corner and signs.
<http://www.leonelmoura.com/rap.html>http://www.leonelmoura.com/rap.html
In two weeks I will be showing at my Gallery the first 12 drawings 
RAP did in the Museum
<http://www.leonelmoura.com/upcoming.html>http://www.leonelmoura.com/upcoming.html


"My ambition for the  DrawBots project is to make a group of robots 
that can manifest co-operative drawing behaviour" (Paul)
You should check ArtSBot (Artistic Swarm Robots), done in 2003.
<http://www.leonelmoura.com/artsbot.html>http://www.leonelmoura.com/artsbot.html
One of the things that may interest you in the context of the 
Drawbots project is the question of fitness.  I don't use fitness as 
it decreases the creative autonomy of the robot. Although I use 
thresholds which is not the same, but some critics may consider it to be.
However, it is possible to combine some kind of fitness with emergent 
processes without losing autonomy. If for example the fitness is not 
predetermined but generated by the robot itself as a kind of memory 
of favorable situations. Considering we are talking about painting 
robots, a favorable situation would be a place with a certain color 
concentration.
Also when Wade says that "it is always the creators rules within the 
system which governs the work" it is not taking in consideration many 
things. Like emergence, the capacity of the robot to gather 
information by its own means, stigmergy, environmental interaction, 
cooperative behavior and so on.
To dismiss autonomy because there is an algorithm "determining" the 
behavior of a machine is to forget that algorithms are everywhere. In 
trees, in ants, in us. And we can argue that they are also in rocks, 
in sand, in galaxies. And for sure in weather or in stock markets.
Hence the question is not so much to know if there is an algorithm, 
but if it works. In nature some algorithms work better than others 
and it is called evolution.

Anyway, from the beginning (2001) my project was to aim at the most 
possible autonomy of the machine. By that time the most advance and 
innovative idea was emergent behavior of a swarm of robots, stemming 
from the work of Marco Dorigo with the Ant Algorithms and the concept 
of Swarm Intelligence.
By using robots I took the ant algorithm to the real world and prove 
it works. By replacing pheromones with color I achieved a 
self-organizing system capable of generating original aesthetical compositions.
Thus my contribution, and to the best of my knowledge innovative 
step, was to apply such ideas to the artistic realm. I believe to 
have demonstrated that it is possible to make machines able to create 
a kind of art - or artificial creativity if you prefer- which is 
independent from the human artist that in fact started the process.