triggering, mapping, post mapping and fuzzy logic

Posted by Armando Menicacci on June 10, 2008 at 10:13am in Theoretical Investigations

Back to Theoretical Investigations Discussions

Hello,
I'd like to start an exchange about post mapping strategies like intelligent agents and especially fuzzy logic. In fact I believe that fuzzy logic can help to "capture" certains states of the interpreter that can be interpreted as controllers. For exemple I want the nervousness of an interpreter to be the controller of some event (sound, image or whatever). A little bit as the Mnm object for max. First thing could be to test different codes, different formulas in Max, Pd, Isadora, etc.

Tags: code, fuzzy, logic, mapping, post

▶ Reply to This

Replies to This Discussion

Permalink Reply by Marlon Barrios Solano/producer on June 11, 2008 at 1:08am: Hello Armando,
Cool,
I like thie question and the investigation in some kind of generative mapping to give the systems some kind of generative agency beyond one to one mapping and randomness.
I found this paper on the Mnm object
I think that also is very important to define how will we interface nervousness in the behavior level (sensing strategy) and also how the system will interpret the selected indicators and differentiate it from "hesitation" or "excitement". The key concepts here are interpretation as pattern recognition+inference and mapping based in generative compositional rules.
just some starting thoughts!
marlon; ▶ Reply to This

Permalink Reply by Armando Menicacci on June 13, 2008 at 8:33pm: Hi Marlon,
in fact I don't think it is important here to discuss how we interface something with something. But what we interface: in fact a fex sensor can be very interesting on a dancer or... not. It depends what it does, what it is sensing. Flexion (even twenty in a certain amount of time) don't mean a dancer is reaching a nervous state. I think that looking for "states" is an interesting controller. More than any simple sensor. I think that the time factor will be very important. to control several sensors over time and to match a particular state. Mnm is interesting although I'd rather have a completely different approach. Instead of matching a figure I'd like to mach a state (wich is more od a specific set of dynamics).... Does this make sense to you?; ▶ Reply to This