Everything You Need to Know About Control Theory
Вставка
- Опубліковано 1 чер 2024
- Control theory is a mathematical framework that gives us the tools to develop autonomous systems. Walk through all the different aspects of control theory that you need to know.
Some of the concepts that are covered include:
- The difference between open-loop and closed-loop control
- How we use system identification to develop a model of the system
- Why feedforward control is a straightforward method to control a system
- How feedback control affects system stability
- An overview of other control methods including adaptive control, optimal control, predictive control, and reinforcement learning
- Why path planning is an essential part of control design
- How statistical estimators like Kalman filters are used to observe system state
- Why modeling and simulation is required for almost all control engineering
Learn more:
- Control Tech Talks: bit.ly/3TaZ62T
- Resourcium journey of organized resources: Control Theory : bit.ly/3ftZ4nU
--------------------------------------------------------------------------------------------------------
Get a free product trial: goo.gl/ZHFb5u
Learn more about MATLAB: goo.gl/8QV7ZZ
Learn more about Simulink: goo.gl/nqnbLe
See what's new in MATLAB and Simulink: goo.gl/pgGtod
© 2022 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc.
See www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be trademarks or registered trademarks of their respective holders. - Наука та технологія
I took this subject 35 years ago as a Mechanical Engineering major. This seems like the first day of class. Followed by endless days of Laplace Transforms. On a related topic, most of my career has been in engineering project management. Control Theory is the single most useful class for future managers. A larger group of people in an organization subject to various external influences is just another system to optimize. Engineering control systems is more useful to management than most MBA curriculum
The economy is a closed loop control system with the delay in the feedback path causing oscillation in the output, the economy. That's what recessions are, the troughs in the oscillation.
I wonder how many government leaders and financial controllers actually understand that?
@@deang5622 Considering that I was a local government leader (utility management) and I’ve spent the past 15 years or so explaining what you just said to elected and appointed officials… I feel confident in answering your question. “I wonder how many …leaders…understand that?” Based on the looks on their faces when I talk about it…. I’d say “not many” or “none”. I doubt my NASA consultant college professor saw that application either.
@@deang5622 haha. I’m not sure what that means…but I’ll guess you’re from the U.K? No. I’m an engineer by education. Manager by occupation.
@@OhNoNotAgain42 apologies..I'm involved in responding to so many people, that post wasn't meant for you! I shall remove it. Apologies for the confusion.
I was responding to some other muppet that was trying to throw his weight around claiming superiority when he only has basic electrician's qualifications..and not realising I'm more highly qualified than him.
This is the problem with electricians, they think they are experts on electricity!
@@deang5622 government leaders can consult with economists
As a PhD, I am a TA for undergraduate control class. I will show them this on the very first day from no on.
That’s my master’s thesis in a nutshell applied to space robotics. My deepest gratitude to you Brian for always helping me in putting things into perspective. This is an excellent overview of what constitutes a control system design for anyone who is interested and just starting out. It’s easy to get lost in the weeds once you start going into the details and each of these blocks is a whole field in themselves but this video does a great job in putting everything together in an intuitive way. I have said this many times before and will say it again - you are an incredible teacher sir! Always a delight to learn from your videos.
Also, Resourcium is awesome! It’s an amazing tool to organize references. Thanks a lot for building it.
Thank you
link to the project?
How does one get into space robotics?
Thank you for this video. This is an insanely good overview of topics perfeclty laid out in a concise, rational and logical way for someone to grasp the ideas and start from there. Personally I took only basic classes on the subject but always felt I wanted more. Work has actually pushed me into another field which I also love, but this is motivating enough for me to take some time and study some of the concepts you've mentioned.
Watched it twice, it is that good! Thank you for putting everything in one place in a compact form. I hope this inspires more students to take on controls as their specialty.
Thanks a lot Brian for the video! It really clears and summarises control theory, really useful to people like me that study computer science as an amateur.
Thanks a lot Brian ! I am doing my masters in control and you have been a great mentor.
You are amazinggg!!!! Thanks for the overview. Helped anyone to gather where they are in their control theory journey. Best tutor ever.!!!!!!
:) economists use control theory. Unwanted disturbances like covid can hit the economy which is a dynamical system. Monetary and Fiscal policy, for example, are controls. Anyway...classical economists believe we don't need any controls. If there is a shock/disturbance, the economy will control itself. Keynesians believe the economy is not automous, it needs controls.🙂. This video actually better than my economic classes...lol.
Economics will be a science and not a religion the day it can build a successful predictive model and not a second before.
Simple and easy to understand. Thanks.
I always used to ask my colleagues about where does all this happen ? I can see vibration, physics, thermal transfer and all, but why control engineering ? Where does all of this apply in a machine ? This one video made me realize the importance and gave me everything that I needed to see how important it is. I can´t be more grateful Brian. You´re the man. Thanks a million.
Great to hear! Thanks!
Brian Douglas back on YT. Awesome :D Thank to you, I am really into controls and control theory. I study it by myself, but without your help I wouldnt be able to even start :)
goodluck man
@@BrianBDouglas Thanks for the video. The thing I didn't get correctly was that at 12:10, you said, We want to measure states. But measurements have their own noises. So we add noise. If this was the reason for adding the noise, then the "sum" block for the noise section should be added on the feedback path, not the main path between the input and output. Am I right? Again Thank you for the effort.
@@mhdls It's usually added where I put it (at least in these simplistic diagrams) because the output is often the measured state of the system which is subjected to noise. For example, with the car speed the cruise controller would use the noisy speed measurement that was fed back to the controller, however, if you as the driver just wanted to know the speed you'd be looking at the same noisy measurement. So, if you think of the entire diagram at 12:10 as a box with inputs and outputs, the inputs would be the things you as the driver do to the car (+ disturbances) and the output would be the things you as the driver can observe (noisy measurements since you can't know the true speed). Did that make sense?
@@BrianBDouglas Yes that makes sense. Thank you !
This is amazing video! such a perfect, concise, and organized overview of the field!
this is 'Sehr gut'. Its an excellent short video for people with no knowledge or full knowledge of the control system. Very beautiful perspective of the control system. It might induce interest in students in this field.
This is a great video I wish every student watch it at the beginning of any control-related course.
Thank you, sir. your videos helped me a lot in my career.
This is such a great explanation especially for someone like me trying to break into the field. Thank you so much for sharing. Keep shining ☀😎
Thank you brian for your in-depth and application oriented control system videos. This helped me to understand the concepts clearly as well as in my placements in the same company. Thank You!
I recently started with my masters Robotics at TU Delft, robots dynamics and control is one of my first courses. This video is a great refresher to understand the big picture basics again, I do too much math and programming that I’m zoned in the numbers, you forget what it’s actual about.
Very nice overview Brian, thanks!
So useful. Thank you Brian!
Brian Douglas's video really helped my study in control system. Have been waiting his videos
Thank you for sharing these interesting topics in a progression and step by step :)
Brian Douglas is the best in describing control theory in an understanding manner. I dusted off my knowledge very easily with him for teaching my own modules at my university.
Excellent introduction about control theory. My research area is Adaptive Filtering. I really enjoyed your video. Great job!
I've got a PhD in control theory, which is all about getting solution bounds of algebraic Lyapunov and Riccati matrix equations. Said equations play an important role in various control and engineering problems, especially stability analysis and linear quadratic optimal control of LTI systems.
Even though I have learned this before, I do like to watch Brian's video which is an enjoyable tech speech.
Wow! Glad I ran into this topic. This will help me understand personal development from a more logical standpoint. This topic makes so much sense when translating it to understand and change my individual psychology.
Thanks MATLAB for this overview. This is exactly what I need regarding the Controller introduction
Excellent overview, thank you!
man you are the best, you can even open a school for control just with the materials you have on the internet. I couldn't imagen what kind of job you that made you know all of that and can explain it like but I am sure this not an academic experience at all you feel like working at NASA
Absolutely goated video MATCHADS. Once again you have given me the gift of more knowledge. Keep doing what you do.
You make the complex look simple, only a genuis can do that . Tank you very much
Thanks so much!
I wish I knew this channel exist back when I was doing my control theory and it's lab in 2018.
This is by far the best video on the basics of the control theory. Thank you for this.
I appreciate that!
Great explanations Brian!
Very nicely summed up control theory
Thank you Mr. Douglas, I have been on the lookout for this.
Sometime in the last millennium I did a one semester university course on control systems, not because it was required for my degree, but because it sounded interesting (and it was). Since then, I have often wondered about naturally occurring control systems, such as how our bodies regulate insulin production and a host of other essential metabolic processes. It occurred to me that many natural phenomena could be explained by control systems that somehow evolved.
Also, in business, there are control systems that have been developed to keep a business on track. My degree was in actuarial studies, and there was a course devoted to an actuarial control cycle where new business premiums and reserves were adjusted at various frequencies to ensure that the business would remain solvent and effective. Of course, this was just something that a bunch of actuaries cooked up at some point, and I doubt that they used any formal mathematics relating to control systems in general. I expect that many businesses use some form of control system like this. These always seemed somewhat blunt to me, as they operated on quarterly cycles rather than instantaneously, and they measured only a few things which were really only weak proxies for what was actually going on. Also the predictive models used were not even as accurate as weather prediction these days.
In short, I have been imagining control systems everywhere as responsible for regulating the behaviour of all sorts of things - whether naturally occurring or somehow arising from trial and error. Of course, I imagine trial and error as a kind of control system in its own right :)
Sorry about the long ramble, but there are some questions I have. In a naturally occurring control system, there has been no planning, or at least none that has been revealed to us. Do you think that it is possible to work out from the behaviour of such a system what exactly is being implicitly modelled, measured and adjusted? If it were possible to work this out, perhaps it would then be possible to more effectively treat illnesses such as diabetes.
Assuming I am right, and that there are naturally occurring control systems, how do they arise? Are they somehow built into evolution?
Again, apologies for the long ramble. All the best, and thanks again for the video.
@@reverse9106 Yes definitely living things, but maybe also other things like societies, ecosystems and economies. At my age don't really have time to study everything, but the topic really interests me. I'm just not sure if any of these topics have been viewed through the lens of control systems.
It seems to me that if you look, control systems are everywhere. Or perhaps it's just my imagination :)
Your intuitions are spot on! Check out "Perceptual Control Theory (PCT) - Bill Powers"
@@earnestcarr6673 Thanks for the hint. Just found a whole lot of YT videos by or about him and PCT. Thanks, I will check this out.
Your intuition is on point. Please read "The Human Use of Human Beings" by Norbert Wiener. A good book on feedback systems throughout nature. He argues that intelligence, and in extension, any form of life, can be abstractly viewed as a feedback system.
@@daviskipchirchir1357 Thank you for that tip. I will have a look at that.
Thanks, great overview
Hey! Great video! I have a question regarding one of the statement you made, Brian. You mentioned that closed loop feedback is associated with possible instability but open loop control is not. Is there no way to have some feedforward bounded input to a signal that still yield instability. Perhaps some input that leads system resonance and instability?
I recognize this is centered around Systematic Engineering but I feel that this reflects how we can both prosper and tailor to our circumstances as individuals, or incidentally impede ourselves in a negative and or unintentional direction
Excellent resource. Thanks!
excellent and simple explanation...
Great Teacher!
It was very helpful . thank you
Brian i follow your work since 2019, you are amazing please never give up on bringing such an illuminating explanation of control theory.
Thanks so much!
I just had my first control system lecture today. As paar the comments seems like i have gotten a jackpot. In German they say This Modul is the hardest. Even physics is not as hard. Thats what everyone says in my Uni. But with help of someone like you,seem like there is light at the end of the turnel. Thanks alot
Thanks! Good luck with your controls course.
what a great video, thanks!
Fantastic video, thank you so much.
thanks for this introduction!
Amazing video, thank you!
Another masterpiece,thanks a lot. Brian, can you organize some workshops? I will be willing to spend my entire monthly salary on it if you do that. Any live sessions if you have , I would fancy participating in that as well.
Omg they recruited Brian to do a video for them, this is epic. I learned control from you!
You left out the best example of a feedback control system - the mouse and cursor. I actually had a manufacturer tell me that the position sensor in the mouse would have to have an accuracy of 1 part in 1024. He is assuming an open-loop control system. Real systems use a feedback system. The user moves the mouse until the cursor on the screen is where he wants it. That is feedback.
Brian Douglas is an artist!
Good explanation for a layperson like myself.
Thank you Brian for the contributions to our society of Control Engineers! 🎛️
thx very much man!!!! Great video!
Perfect video, can you talk about the application of the control theory on high altitude platforms of wireless communications
That is a wonderful video, love it!
Glad you liked it!
Awesome man!
Thank you, Devs.
Just Elegance 👏👏
So, whats the difference between Feedforward control and open loop control?
Please correct me if I am wrong. Feedforward has feedback in the system but doesn't use feedback as it slow, just uses the feedforward loop in case of any disturbances to just make it fast and not waiting for the entire system to get affected.
Thanks a lot
this was amazing
This is awesome.
Brian you are the best.
🤗
6:20 you actually _wrote_ "accelerate" and "brake", 4 times each! no copy-pasting.
dedication!
(There _is_ a way you would have done this with a text renderer that does human-like writing, yes)
Sometimes just writing it out is easier :)
Great video❤
Great Video.
Oh it's you actually. I watched your videos long time ago in your channel.
Wonderful
They should of lead with this in controls class instead of 1/1+Ts
Can you provide the tool that you are using for the presentation?
this is really cool
Nice video
so good ，nice！
i love automation thank you sir
Awesome
1:34 Disturbances are actually environmental. Not only wind and bumps, but also social - other vehicles in the road.
Thanks! Now I can understand why the missile knows where it is.
I hope you create a map for studying the control theory. :)
thanks
It's so good
Question: what software is this?
So this is how the missile knows where it is…
And where it isn’t
tried liking this video like 3 times already
Sounds very interesting. Thinking of ways to apply this to the immune system of humans in health and disease. But how did I get here???
*Effectivley the feed-forward is a "human" it understands the environment and therefore is effectively a feed-back. Goes back to my control systems classes in electrical engineering at uni, and yes lovely Laplace transformations. Feed-back again is the human who has knowledg of the system/car/env beforehand and during. Humans therefore, use ALL control methods interactively and to optimise actual control, unless you have a 18 year old.*
Good.
Which software are you using to create this video
I explain my process here: engineeringmedia.com/my-setup
Feedback control is not a "hack", it is a fundamental principle of both natural and engineered servo sytems. The behavior of a closed loop system is fundamentally different than the sum of its component parts. Failure to understand this leads to the hubris of presuming that everything can be reduced to a stream of ones and zeros.
1) Error signals are required (tee hee).
2) Adaptive (learning) control only works if the whole system responds ok. (You can realistically expect the parameters will eventually max out. It all starts over if the historical data are wiped out.).
3) Tune everything until you are bored to death.
4) Keep re-testing as often as possible.
4) "Real boats rock.", Bene Gesserit saying.
Love u Brian ❤️
❤❤
How does control theory apply to living organisms or their artificial counterparts?
There's a lot of control theory in living organisms. For example, when you eat carbs, your pancreas detects an increase in glucose in your blood and releases insulin to lower it. As the glucose increase slows, it releases less insulin. If glucose gets too low, it releases glucagon to raise it.
Thanks Ken, your example is clear - my refrigerator works same manner. Question: is neurons network involved to this control process?
I don't know, @@olegsheverda5167. Maybe in sensing, though the main control action is by signaling molecules in the blood.
Everything you need to know about Control Theory:
1. rub more PID on it until it works
2. if that doesn't work, blame the hardware and buy better hardware
Been in computerized controls since 1973.
Yeh, I agree that's what it comes down to after all.
Hey, whats this applications name?
you are beast
Supervisied learning is feedforward and Reinforcement learning is feedback controlled
A very good high level overview. However don’t claim this is all you need to know about control theory. It’s a vast topic. Better to do a university course in control and learn the topic properly. Good luck.
Yes, because you need a degree you pleb 🤭
God bless you .... i'm starting to learn how to tune Mpc and i really need helps .... can any one here help me ?.