Is Control Systems Engineering based?

[Anonymous No.12295475]

[Anonymous No.12295476]

no

[Anonymous No.12295482]

>>12295475
No. Computer Science and Data science are.

[Anonymous No.12295514]

>>12295475
yeah

[Anonymous No.12295543]

>>12295476
>>12295482
Average CSeething

[Anonymous No.12296246]

Yes, read Norbert Wiener's "Cybernetics"

[Anonymous No.12297206]

>>12296246

Thank you I will look into reading that one

[Anonymous No.12297373]

>>12295475
It’s crazy lucrative and low stress, so yes.

[Anonymous No.12297378]

No it’s hard as fuck. I’m in a control systems class this semester and it’s kicking my ass.

[Anonymous No.12297387]

Do any of you have a good book on mathematical systems theory?
I picked up Kampis "Self-modifying systems in biology and cognitive science" and it's way over my head -- he just assumes a lot of background knowledge on the ways a mathematical model corresponds to reality, and gets really autistic about formalizing some aspects of models that I assume to be pedantic but end up important later on.
It's not like the math is hard, I think I just want a primer on the philosophy of mathematical modeling.

[Anonymous No.12297411]

>>12297378
This. Probably the hardest class I took for my ME degree.

[Anonymous No.12298242]

>>12297378
>>12297411
How? Is ME control different from what EEs do? I’m 3/4 of the way through my controls class and it’s literally a mix of signal processing, diff eq, and linear algebra concepts. There’s only been a few wholly new concepts like the state space representation, root locus, and some performance metrics. Not sure if my class is shit though, and this is just the first controls class you can take which is basically linear SISO systems. Really underwhelming at the moment, expected something more interesting, and although I’m sure the grad level courses are, I was hoping this would be at least somewhat more interesting than
>model the system using known dynamics
>define input and output parameters
>define control performance requirements
>adjust controller to meet said metrics either analytically or in simulation

[Anonymous No.12298247]

>>12298242
ME students typically have build projects in the lab component of their controls class. EEs probably just write some dumb code in their labs lol

[Anonymous No.12298306]

>>12295475
Yeah it's cool af look what you can do

https://www.youtube.com/watch?v=Tnv186IFovQ&ab_channel=DynamicsUci

[Anonymous No.12298517]

>>12298242

Depending on the classes you take, Controls can be quite difficult, especially when the mathematical backgrounds are presented rigorously. Most engineers don't have the prerequisits available and therefore the math can get a bit overwhelming.

[Anonymous No.12298744]

>>12295475
I'm mech e and I hate this class. The fact it's online only is making it worse.

[Anonymous No.12298745]

>>12298517
We are already doing laplace transforms and differential equations up to the 2nd order in controls class, we haven't done those topics in math class yet and won't until 2nd semester.

[Anonymous No.12298764]

>>12298247
EEs also do a ton of physical lab work, because the real world will constantly fuck with you and you need to learn that lesson via cutting your teeth on shit that "should" work but doesn't. CEs do nuts and bolts lab work too, but much less of it after year 2.
CS does almost no physical lab work and they rarely have a clue about anything that isn't in code. Frequently they don't have a clue about things in code either. The good ones are math autists, and are very respectable but pretty rare. Most are retards who heard programming means $, or just want to make video games. They really are intolerable.

[Anonymous No.12298766]

>>12298247
Get a load of this retard, control engineering was first done by EEs, you have no idea what you are talking about.

[Anonymous No.12298771]

>>12298517
I'm having a really hard time understanding mpc as written by rawlings book, does it get easier?

[Anonymous No.12299321]

>>12298247
>EEs probably just write some dumb code. . .
You don't know what an EE is, do you

[Anonymous No.12299784]

>>12298745
Differential equations isn’t a prereq for your controls class? Seems pretty backwards since it’s pretty much all about differential equations.

[Anonymous No.12299823]

>>12299784

Here in Germany only the Laplace Transformation is really needed for the first controls course, since all diffeqs are linearized anyways. The focus is more on the analysis and design of SISO systems as was stated here earlier. The mathematical prerequisits are just explained 15 minutes beforehand and that's it.

But I can totaly see how further control courses require really heavy maths, which seems like I will have to learn for myself...

[Anonymous No.12299843]

>>12299823
Siso systems are introductory bachelor stuff, you'll have to learn about advances state space methods which requires knowledge of differential equations

[Anonymous No.12301396]

>>12298744

Online sucks ass, I shaved down to part-time so I'm only doing subjects that are more-or-less completely self-taught anyway this semester.

t. mechatronics student

[Anonymous No.12301403]

>>12297387
I don’t know that one, but Try some easier/other books. I’ve heard that Bayesian cognitive modeling book with the red cube guys on the cover is easy. Trappenbergs foundations of comp neuro is the easiest intro to the neuro side that I know. Dayan is harder (though the most dated). Gerstner and Paninski is harder still but has lectures.

Maybe read the rescorla Wagner papers / ratcliffs Brownian motion models (DDM) and stuff like that if you don’t wanna do a book?

[Anonymous No.12301784]

>>12295476
WRONG!

[Anonymous No.12301901]

>>12295475
Yes. Doing EE with control systems specialization, and I'm not even sure I want to work in controls. It is, however, giving me a really solid quantitative skillset (math/stats/cs), which I am after. I took some extra cs and math electives, and have gotten internship offers in data science, finance, EE, and software engineering.

[Anonymous No.12301912]

>>12301901
Based, any particular skills (e.g. particular branch of mathematics or physics) or tools (e.g. particular software/program/lang/etc.) to focus on to prep for control systems when I start uni?

[Anonymous No.12302155]

>>12295475
If you’re an ME major, RUN. DROP OUT. GET A CS DEGREE. ALL THE ME JOBS ARE BULLSHIT. THERE IS NO FUCKING MONEY ALL THE MONEY IS IN SOFTWARE AND IT. BSME IS THE BIGGEST REGRET OF MY LIFE ALL I DO IS FUCK AROUND ON AUTOCAD AND RANDOMLY GET RETARDED BOOMER PROJECTS REEEEEEE

[Anonymous No.12302378]

>>12299784
>>12298745
This happens in a lot of unis recently, especially in CS/CE departaments. Having prerequisite classes after the classes they depend on is a standard.
Some unis try to push back math as far down the curriculum as possible, so many students have diff eq, or even calc 2 only when they start their ME. The assumption is, it will scare away too many students, and that it's not so crucial for most of the things they'll be doing after college anyway. Which is partially true, because yes, there is a lot of work for programmers which won't ever need those things, but there's also a lot of work for programmers who'll need it like air. The problem is, some classes they have during their BE still relies on having those classes as a prereq, so the lecturers have to improvise and it's not pretty.

[Anonymous No.12302597]

>>12301912
Make sure you have pre-calc understood. Going in with calculus is helpful but many don't. Your first two years will be pretty much getting you up to speed. If your uni is anything like mine, you won't take classes in your specialization until junior year. Control theory is dense, requires a lot of prerequisite knowledge, and is hard to apply on your own so it is a hard topic to self-study.

One thing I'd highly recommend is getting familiar with is the basics of programing. The people who came in with knowledge excelled easily in programming courses, but those who knew nothing never really understood it that well even after classes. I'd reccomend learning python and/or C. Python is useful in all fields and transitions really well to MATLAB, which you will likely use in your courses. C is just good to know and will be used for microcontroller programming, and transitions really well to Java/C#/C++ which are used for writing more serious software.

[Anonymous No.12304484]

>>12298242
At my uni ME, CE and EE all have the same control class as a core unit.

EEs have a better time because they were already exposed to some of the ideas in their Signals and Systems unit.

[Anonymous No.12304496]

>>12302597

Programming is something I've been putting off for years now (in my early 20s), I've tried to learn C but it's definitely not intuitive to me. I feel like I could and should have started years ago. I'll be starting again next semester, not giving up.

[Anonymous No.12304553]

>>12304496
Don't worry about it, starting now is better than starting later and that's all you can control.
C lacks a lot of modern programming niceties. It is autistic about memory management, and requires you to be very explicit with how much memory you allocate to each variable. It hinders your ability to make more complex programs while you are learning.
Python handles all that for you, so it's nicer to learn on. Starting out, I'd reccomend downloading spyder as it comes with all the packages you will want.

[Anonymous No.12304730]

>>12304553
Not the person you replied to, would you suggest jumping straight into C++? That's what my first year course did and I just completely blanked out and barely passed and forgot about it, I'm dreading the day I return to it.