H Infinity and Mu Synthesis | Robust Control, Part 5

This video walks through a controller design for an active suspension system. Actually, we design two controllers. For the first, we use H infinity synthesis to design a controller for a nominal plant model that will guarantee performance but not necessarily be robust to variation in the system. Then we build an uncertain model like we did in the last video and design a robust controller using mu synthesis.
Watch the first videos in this series:
Robust Control, Part 1: What Is Robust Control? - • What Is Robust Control...
Robust Control, Part 2: Understanding Disk Margin - • Understanding Disk Mar...
Robust Control, Part 3: Disk Margins for MIMO Systems - • Disk Margins for MIMO ...
Robust Control, Part 4: Working with Parameter Uncertainty - • Working with Parameter...
Check out these other references:
Robust Control of an Active Suspension: bit.ly/3bt8VCE
--------------------------------------------------------------------------------------------------------
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
© 2020 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.

[@40NoNameFound-100-years-ago]

I have been waiting for this demonstration for H infinity control for long time

[@mohammadokour7969]

Mr.Brian Douglas, I really can't find a way to thank you for your huge help and support in control field, I have finished my Bachelor's degree in mechanical engineering and I am almost done with my Master degree in mechatronics system and you were the best reference for all control topics, nobody can explain these topics in such nice and clear way direct to the point and also giving examples using MATLAB to proof the concept of the topic. I would appreciate it if you could do something related to adaptive control and autotune because these topics are not clearly defined among youtube. I wish you to have more success in your future life.

[@camilotorres9835]

You are the real mvp in the control field, it is just so amazingly easy to understand once you explain it

[@mansoorzahoorqadri9553]

Beautifully explained.

[@GeneralKiller1337]

Wow, nice explanation! Thank you so much.

[@MrEdnz]

I had a course on the control of MIMO systems. This stuff is amazing

[@deepak_nigwal]

are there any resources for design of MIMO systems for full car?

[@engammar1794]

made it clear as sun shine. Thanks

[@gretarmark]

This is so good video about Robust Control, now I understand the whole concept I have been struggling with. Thanks alot!

[@avinashmalik9441]

First off good work -- we usually teach this in 4th year EE. FUTURE IDEA: Describe sliding mode control design, its stability analysis and non-chattering implementation via stateflow. Next, go all the way and describe stochastic hybrid control with wiener processes (unbounded stochasticity) and its convergence/stability analysis. A lot of biological processes (including human organ) control is via stochastic hybrid systems -- experience in medical device design/control for cardiac arrhythmias and GI motility disorders. Just one caveat, stateflow/simulink will choke when simulating hybrid stochastic Wiener processes via Euler-Maruyama.

[@mohamednihalnadvisaleem2237]

Great Video. Can anybody explain how the parameters are chosen for Handling Target and Comfort Targets?

[@imp00ssible]

Can anyone suggest a good book for Robust Control (Mainly Related to H-infinity) and Also Great Explanation

[@luisruvalcaba9159]

My dream come true

[@kamelbenyattou5044]

how we can fo r example compute and do some simulation of H_infinity Control of Two-Dimensional Systems with matlab

[@hakemass]

thank you very much. I have a question regarding Wact, you have said that putting a high pass filter on the actuator is going to penalize high frequency controller commands more than low frequency, isn't it the opposite ? or i didnt understand it

[@kwinvdv]

It might be worth mentioning that H-infinity does have the downside that the order of the generated controller can be quite high (I believe it is equal to the order of the plant plus the orders of the weighting functions).

[@BrianBDouglas]

Good point! I had a section in there about order reduction for the resulting high-order controller but cut it for time. Thanks for the comment!

[@rezah336]

why is high order bad?

[@SamFekri]

@@BrianBDouglas Thanks Brian. It just happen to surf here. You are great! :) Just a quick comment: In addition to reduction of higher-order robust controllers, it is also possible to use the so-called fixed-order or structured H-inf optimization methods (i.e. Hinfstruct, HIFOO, etc) where designer can specify in advance what type of controller(s) you wish to design (say PID, Lead-lag etc). I have used all of these methods for numerous industrial apps. Well, it is very difficult to say they always work! This pretty much depends on the plant dynamics, in my opinion. But, what I can confirm here is that the more complex plant (i.e. with more unstable poles and/or lightly-damped modes), the higher order for controller will be required (for the satisfactory robust-stability and performance), and the worse results one shall expect from such reduced-order/structured Hinf optimization, similar to what one can see after reducing the order of high-order controllers. :)

[@SamFekri]

@@rezah336 A high-order controller means: more expensive (both software and hardware), more bulky (bigger size - more heavy - not suitable for application etc), very time-consuming in re-tuning and fine-tuning for industrial apps, low reliability in operation, ... Are these enough? ;-) Good luck

[@MrMaximal96]

Hello, a very nice demonstration of the H_infinity control design. How does the mu synthesis perform compared to a robust controller where you model out the uncertainty (Delta) and use H_infity and the small gain theorem to provide robustness.

[@SamFekri]

H_infinity optimization does not consider the uncertainty (Delta) in a structured format but deals [with Delta] like a full MIMO matrix! This turns out that the results of the controller robustness (both stability and performance) could be very conservative! Of course, this conservatism depends on the application. But, mu (as seen from the name: Structured Singular value) does this in clearly specified format, including real, complex or a combination of real-complex (mixed) uncertainties. :)

[@sayandey02]

how to understand mathematically? Do you have good source for H-infinity control theory? please recommend thanks in advance

[@Choice7890]

Can someone tell what’s the benefit of the generalized plant representation? Couldn’t find any intuitive explanations

[@chiro6865]

Can anyone help me. Which value record the value gain K of H-inf ?

[@paolzpk]

Can you suggest some references where to study about H infinity and mu synthesis?

[@kwinvdv]

Multivariable feedback control: analysis and design from Skogestad and Postlethwaite contains sections on both.

[@paolzpk]

@@kwinvdv that looks really nice, thanks!

[@sulielroli]

Gracias!!

[@deepak_nigwal]

any direct paper for H-Infinity controller on full car model?

[@lyessaidi501]

you can check a Ph.D thesis done this year about merging Hinf and MPC under the name of charles Philippe directed by lounis Adouane page 46,49

[@lyessaidi501]

@@deepak_nigwal never mind

[@haligali4360]

Hi! It's a really great video! But it's so fast, particularly for nonnative speakers. Of course, if you're familiar with this topic you can swallow it and digest it, otherwise choking...

[@omkarkudalkar4621]

Can I get your channel, good explained

[@sangluuhuu457]

the video image is too poor, you need to fix it more

[@nyquist_control]

it's you, not the video. turn on 1080p.

[@leolorenzeen6055]

「目標を指定してお金が必要」、

[@XxXproobenXxX]

Toes