I used to want to study cybernetics, but it looks like the field of cybernetics is called "control theory" nowadays.
I'm interested in Automation, so I start by looking at various journals listed by SCI as related to Automation: Automatica and System Control Letters, for example.
These journals seem to have a lot of articles with advanced differential equations. Well, I guess that fits my intuition that dynamical systems are the basis of controlled systems; after all, one would expect to use differential equations to describe dynamical systems, right?
But the first snag is that differential equations can describe almost anything, and the second snag is that I'm not really an expert on differential equations. One can use them for mechanics, fluid dynamics, and other physics sub-specialties.
Beyond that, there seems to be a lot of scope for stochastic methods in control theory. I'm not an expert on stochastic stuff, either.
Automation is not concerned with all the reaches of control theory -- control theory can apply to LOTS of stuff -- economies, ecologies, animals, robots, etc.
When I try to find control theory as applied to automatic equipment, I run across articles like this:
http://www.springerlink.com/content/w2312110gl0243k8/
in a journal called "Real Time Systems." It's great, but when I read through it, I get the sense that to really understand it I need to be a specialist already. At very least, I need to be fully conversant with Queueing Theory.
Edit: One industrial engineer, on hearing a spoken description of this area, suggested that I should look at "Dynamic Programming":
http://en.wikipedia.org/wiki/Dynamic_programming
I guess there's no quick and easy way to get all the background, but I appreciate any suggestions on how to read and understand this kind of article.