When I studied control theory at university many years ago we started almost immediately into the math of Nyquist stability theory. The point seemed to be that our lecturer was a clever dude and we undergrads were a bunch of idiots who should read more. Maybe he was right, but his job was to change the error signal between our level of knowledge and his own: a nice problem in control theory that he thoroughly failed to resolve. I got more from your 14 minute video than I did from hours of his instruction. Nicely done. Thank you.
Maybe his system was operating in open loop control, without a sensor to detect the error, or his controller wasnt tuned correctly to reach the reference value
I am in my third (!) semester of Signals + Systems / control system undergrad courses. I’ve gotten A+’s in the first two semesters, and I can 100% say I have never received this sort of explanation of a PID system. We have learned all of the math behind it, but I have never had it explained what P, I, D actually does. I didn’t know the integrator was integrating the error signal (seriously... I didn’t), it was just never explained. It was just referenced as 1/s, and we had some rules stating when you could ignore 1/s because it wasn’t needed - but never a physical explanation of what it was doing. I wish I found this channel 2 years ago.
@@kcmichaelm you can literally see at the definition that the integrator integrates the error signal lmao. u = K(e + 1/Tn intgrl(e) + Tv de/dt) = PID in time domain
Took a whole class in mechatronics as a computer engineer. You managed to explain a whole semester's worth of confusion in a half hour between these two videos!
hahaha I agree on that lool I searched for the information about PID and read lots of them, but still I felt like something's missing and this video is just.., after watching the video, I felt like I should pay hh
@@daesoolee1083 its a lot more complex than what is being explained, actually you need a strong knowledge in differential calculus to better understand it, there is something called Laplace transform hence the s symbols. Anyway good lecture.
I must say I normally don't comment on videos, but you have an amazing ability to take a complex idea and make it simpler and interesting with great real world type examples. This is how schools should be teaching a combined math science curriculum that actually keeps student interested. I'm an engineer and thus have made it through a rigorous math science education and I wish I had this back when I took control systems. Bravo Sir and keep it coming.
Yes! I'm taking a few weeks off for vacation but when I come back I'll be adding more controller design videos (PID included). I'm going to post a video series on how to design a control system for a real system from system identification to tuning and testing. Stay tuned!
I wish lectures at my uni were presented like this. I think I've learnt more from Brian's videos in a couple of nights than I learnt in my entire introductory control subject last semester.
I can't say I'm getting the whole thing, but in all fairness I'm not in university. This has greatly helped me understand what a PID is and what they do. Nice lecture, thank you!
I looked through all kinds of material and videos trying to get my head around these concepts with little success. Most people start out with a bunch of formulas and technical jargon which makes it very hard for somebody who is new to the field. It had me thinking the most clever people are horrible at explaining things in understandable terms. Thanks to you, I now know that isn't always true. Thanks for taking the time to make this incredible and easy to follow video. If you have a patreon or similar I would be happy to make a donation for your trouble.
you're totally awesome, I couldn't understand what my university lecturer is teaching and you basically summarized everything he taught in a short, concise and comprehensive video
These videos are nothing short of completely amazing...from your background and tie to the math, back out to the physical objectives by example. HELPS SOO MUCH! WISH you were my professor would have done so much better in this class. Thank you.
well...6 years passed, I play a submarine PC game and find this vedio. Thanks a lot to make this theory so clear explained I have ever watch. I expect to understand the math behind, but your given example is just too good and help me get rid of it. Thanks again, respect!
I wish I had proper youtube back in my child days (29 now). What would school have been so much more easier, resulting in learning even more in a shorter amount of time. No 'dry' text and pictures and brain wrecks. I'm a bit jealous and happy for my daughter the opportunities she have to educate herself. Thanks to people like you Brian!
In my opinion, teaching is much more than understanding theory, explaining it, a good teacher can use imagination and analogies to inspire intuition in to their listeners. You, sir, have a knack for being a teacher. Have a nice day.
This was extremely helpful. All these transfer functions and PID controllers in class and they never explain what this all really does and means. You did that just right
Thank you for the big complement! As for the P controller, you are correct in the general case. However, I applied a little control system trickery! The error did go to zero, so the output of the law was zero, hence no steady state error. The reason I was able to get away with this was because the output from the controller represented a change in pedal position and not an absolute pedal position. Once the velocity error went to zero, stop moving the pedal, and the velocity will remain constant.
If you are talking about adding the 1/s term at 05.43 in the video: When you go from velocity (xdot) to position (x), that is following the arrow (from left to right) in the block diagram, you multiply with 1/s (integrate). When you go from postion (x) to velocity (xdot), that is the opposite way of the arrow (from right to left), you divide by 1/s wich is the same at multiplying by s (differentiation).
I am attempting to tune a PID controller for 'closed loop boost control' on my car. This PID controller will regulate air pressure in the intake manifold of the car that a turbo charger is producing. The output of the PID controller drives a PWM solenoid that opens and closes a waste gate. The ideal PID controller for this application will let the turbo increase pressure as quickly as it can to the setpoint level and then quickly open the wastegate to avoid overshoot. Thereafter, it will modulate the PWM signal to the solenoid to actuate the wastegate and regulate pressure to the set point value. The PID controller is setup to not activate until within 30kpa of the set point. At which point, it starts actuating the wastegate. The pressure can either increase slowly or very, very rapidly depending on how quickly I mash the throttle, so keeping track of how quickly the pressure is increasing is very important to controlling pressure. Based on what I watched here, it seems that a PD controller with maybe a bit of I value, is best?
First let me say that your tutorials are clear and with good sound quality. And you're the first technical person presenting technical stuff with joy, my wife tells me ("all the others sound either dull or depressed"). If my wife tells me so, you should take that as a big compliment ;-). At 4:21 you state that with proportional controllers the steady state error will be zero. How can that be, since you need an error to have your controller producing an ouput (output=gain*error)
As An Engineering Technician who has 30 years working in the practical world: has been a US NAVY Aircraft Flight Controls Systems mechanic for 8 years and having a two year degree in Electrical / Electronics Engineering Technology, I am very interested in CONTROL THEORY, I have worked at LOCKHEED Environmental Testing Laboratory for 17 years doing ACOUSTICS, VIBRATIONS, THRMAL/VACUUM, EMI/EMC Radar Testing, Structural Testing = We DID PRACTICALLY TESTED EVERYTHING from STUCTURES to breadboarding building electronics sensors, set up instrumentations systems working with all types of Engineers from all the Engineering disciplines, now I am retired = I still have the passion to ANALYTICALLY REVERSE ENGINEER what I DID and learn advanced mathematics, DESIGNING DIFFERENTIAL EQUATIONS OR MODELS solving them with LAPLACE TRANSFORMS and implementing, and BUILDING Electronic Circuits to simulate anything in the Physical World. Microcontrollers, Mechatronics, Robotics, Remote Sensing.
Einstein said that if you cannot explain something simply, you don't fully understand it. You must be the next Einstein. That was sooo simple it was amazing. Thanks very much for sharing such a complicated subject, but in such a simple way. More text books should be written with your style - the world would be a better place
Hello Adwait, great catch! You are absolutely right. At 11:50 I should have increased theta for just a moment and then had it come back down to the same position. I'll make an annotation and add this correction to the errata in the description. Thanks for finding that.
I'm trying to learn this for a while now, but thanks to you I finally get the sence what I'm actualy doing with the calculations. The car example helped a lot to get things to make sense and I now finally understand what can go wrong with wrong gains. Thanks a lot.
Hello again Raed. Is it too hard to follow? I was concerned that it wasn't clear enough when I was finishing it up. I'm taking the next 2 weeks off but when I get back I'll see what I can do to make it easier to understand. Thanks for the comment.
Thanks for the reply! It was just that you seem to state that p-controller generally have no steady state error (and I always tell my colleagues otherwise). Position control using a stepper motor is another example where the process itself has integrating action (like a delta on the gas pedal), and hence p-control here does not have steady state error. Greetings from the Netherlands.
Hi Brian, thanks for the videos. Just a small comment. I liked your previous video on PIDs quite a lot, and you explained the theory quite well. However, with this one, I think the difference between the different controllers is a little more confusing since you change the problem each time. Perhaps it would be better to use a single example. Love the illustrations and your style of teaching. :)
Also, humans themselves are control systems. For example, if you try to walk from your kitchen to your bedroom, you accomplish this with feedback control. Your eyes are the sensors and your legs are the actuators. Your brain (controller) adjusts when to turn and how fast to walk using eyes as feedback. Also, your internal temperature control keeping you at 98.6 degrees is another controlled system in your body. There are tons of examples like this with humans.
Sincerely i enjoyed the instructions of the lecture but the example part which is this i still have issue please i can used steam plant for explainations
your work is wonderful. You presented the whole idea so easily and clearly . Frankly I have never seen any other teacher to make this topic so easy. Thank you sir.
Instead of replying to all your videos: thank you for these great videos! They are really helpful. You go pretty fast, but on the other hand if you already studied the material but still did not get enough 'ahah!' moments, then this is helpful. But hé, .. we also have a pauze button!
I'm a cs student working on a project relating physics-based simulated skeleton animation. Reading those robotics papers is really difficult for me since I dont have any knowledge mech engineering. Your vids help me alot!
you helped me out with a good resume and making everything clear. My teacher is going way to fast for all students in lecture. One down side... the real problem I have in the lectures are not in the videos (cascade regulator, position and speed regulator in a car, with accelerate and decelerate limits). But they help really much!
In the very last example at 11:40 isn't the P controller supposed to go UP first and then down? Assuming it controls the angle of the gas pedal in this example, and going up means accelerating (unless I misunderstood something in there). Thanks if someone answers.
Hey, hope u gonna see my comments it's kinda late but the book about control theory uv been writing it's more then great and helpful , shortly it's a must read for every control engineering student . Thank u
Thank you for the very visual and easily understandable explanation about the PID concept! I'll share it to my friends as the greatest explanaition I've seen so far.
Very nice and illustrative example of PID ! Just 1 remark though, with a Proportional only controller, I think you'll have a static error at the equilibrium and not 0 (if your process is stable and not integrator). That's why we generally add integral action to cancel this static error...
This might be hard in this comment box to explain .. but here it goes. 's' is a complex variable in the s-domain and is similar to how 't' represents time in the time domain. W0 is the frequency below which the low pass filter approximately passes the input straight through. Above this frequency the signals are attenuated (not fully but it's a good approximation). Check out my video Bode Plots by Hand: Real Poles or Zeros for a little better understanding.
This is probably the first time I'm commenting on a CZcams video: great job, it's so easy to understand. I hope you continue making more informational videos such as this!
Thank you so much for your videos Mr Brian... anyway, considering that some of them are very reach in contents and details, may I ask you to slow down a little bit your presentation, please? This would be very much appreciated by most of us that follow you with deep interest from all around the world...Thanks in advance for your consideration and to share your art with us...
Brilliant explanation but can you please prioritize making PID explanations for the next while ? I find it the easiest when you explain it and I have a lecturer who doesn't speak english as her first language and as you can imagine it get tough to know whats going on. All these control systems lectures have saved me so far.
At 8:30, whenyou show the derivative part of the controller response, it appears to be zero while there is a step in the error. I think that this is not right, although I understand that you wanted to keep it simple and the general idea is right, I think it's worth mentioning that the derivative part should give have a positive kick towords "infinity" since the slope of the step error is almost vertical. Am I right? Thank you for your awesome videos! I'm agreat fan!
Hi Brian! Thank you very much for the brilliant lectures! I do appreciate your work! But I think in this particular video it would be more accurate to say that the angle theta defines the reference velocity which is compared with the output velocity. The error velocity than is processed by the PID-controller.
Absolutely! Humans are very much the controller in some situations. And when you're designing a system that uses a human controller (like an airplane) you better make sure the system is "slow" enough that the human can respond fast enough to it, but not too slow that human can't control it at all. It's an interesting problem designing for humans. When you take the human out of the loop and place a mechanical or electrical controller, then it's an automatic control system.
I still don't understand why a P and PD-controller still have a steady state error. Is there a good video on this question? Thanks a lot for the videos by the way - they are great. You should become a professor. :)
This is such a nice example! it easily covers the exception which could arise with a solid understanding behind them. i am really Thankful to you for this Lecture.
In this video if 'phi' is the angle that gas pedal makes then after the chasing car catches the other car, the angle phi should settle at the same value if the speeds before the chase started and after catching-up are same but the video shows that phi after the lagging car catches leading car is more than what it was. Which means the car now needs more gas for exactly same speed. Did you mean something else by angle phi ?
What causes the integral term to hold the system at 0 error? In the instant before the system reached 0 error, the integrator was a large sum. So when the system reaches 0 error in the next instant, why doesn't the integrator just cause an overshoot? Is this some kind of balance between the I and D terms?
great work man...!! the way you are making understand with examples is superb. its like explaining very complex industrial problems but the way teachers do in kindergarten. really helpful. thank you.
When I studied control theory at university many years ago we started almost immediately into the math of Nyquist stability theory. The point seemed to be that our lecturer was a clever dude and we undergrads were a bunch of idiots who should read more. Maybe he was right, but his job was to change the error signal between our level of knowledge and his own: a nice problem in control theory that he thoroughly failed to resolve.
I got more from your 14 minute video than I did from hours of his instruction. Nicely done. Thank you.
Maybe his system was operating in open loop control, without a sensor to detect the error, or his controller wasnt tuned correctly to reach the reference value
I am in my third (!) semester of Signals + Systems / control system undergrad courses. I’ve gotten A+’s in the first two semesters, and I can 100% say I have never received this sort of explanation of a PID system. We have learned all of the math behind it, but I have never had it explained what P, I, D actually does. I didn’t know the integrator was integrating the error signal (seriously... I didn’t), it was just never explained. It was just referenced as 1/s, and we had some rules stating when you could ignore 1/s because it wasn’t needed - but never a physical explanation of what it was doing. I wish I found this channel 2 years ago.
@@kcmichaelm you can literally see at the definition that the integrator integrates the error signal lmao.
u = K(e + 1/Tn intgrl(e) + Tv de/dt) = PID in time domain
@@abcxyz4207 Was not taught any of that, but thanks for the reply
Being a good engineer or academic doesn't necessarily translate to being good at educating.
Took a whole class in mechatronics as a computer engineer. You managed to explain a whole semester's worth of confusion in a half hour between these two videos!
i want to be a mechatronics engineer
Go to CSU, Chico! We have a great Mechatronics program.
How can this possibly be a semester's worth of study material?
hahaha I agree on that lool I searched for the information about PID and read lots of them, but still I felt like something's missing and this video is just.., after watching the video, I felt like I should pay hh
@@daesoolee1083 its a lot more complex than what is being explained, actually you need a strong knowledge in differential calculus to better understand it, there is something called Laplace transform hence the s symbols. Anyway good lecture.
I must say I normally don't comment on videos, but you have an amazing ability to take a complex idea and make it simpler and interesting with great real world type examples. This is how schools should be teaching a combined math science curriculum that actually keeps student interested. I'm an engineer and thus have made it through a rigorous math science education and I wish I had this back when I took control systems. Bravo Sir and keep it coming.
this guy is good at drawing!
Hey Hi. I am not an engineer, I am building a quad and this video helped me a lot.
Just wanted to say thank you.
same here!
well, if you keep up with this kind of self instruction, after a while, you will find that you are an engineer.
@@maneki9neko seriously, these are hardcore stuff
Done with the quad?
on a serious note ! you are seriously talented man ! keep it up... im so glad that i found your videos... bravo sir.. you should be proud of your work
Yes! I'm taking a few weeks off for vacation but when I come back I'll be adding more controller design videos (PID included). I'm going to post a video series on how to design a control system for a real system from system identification to tuning and testing. Stay tuned!
wow..i was smiling through out the lecture...because i understood it.you teach so good.i wish i could become like you one day
I wish lectures at my uni were presented like this. I think I've learnt more from Brian's videos in a couple of nights than I learnt in my entire introductory control subject last semester.
I can't say I'm getting the whole thing, but in all fairness I'm not in university. This has greatly helped me understand what a PID is and what they do. Nice lecture, thank you!
Where were you when I needed you? I`ve graduated in 2011 and this was my last exam.
I looked through all kinds of material and videos trying to get my head around these concepts with little success. Most people start out with a bunch of formulas and technical jargon which makes it very hard for somebody who is new to the field. It had me thinking the most clever people are horrible at explaining things in understandable terms. Thanks to you, I now know that isn't always true. Thanks for taking the time to make this incredible and easy to follow video. If you have a patreon or similar I would be happy to make a donation for your trouble.
you're totally awesome, I couldn't understand what my university lecturer is teaching and you basically summarized everything he taught in a short, concise and comprehensive video
These videos are nothing short of completely amazing...from your background and tie to the math, back out to the physical objectives by example. HELPS SOO MUCH! WISH you were my professor would have done so much better in this class. Thank you.
One semester worth of classes in just a couple of phenomenal videos! Thank you Brian. I have my exams coming up soon and these videos are a lifesaver.
This is so good! university classes always focus too much on the underlying mathematics without properly reasoning why it exists
Great job sir. Very well explained. Anyone who thinks he's going to fast there's nothing stopping you from watching it again.
well...6 years passed, I play a submarine PC game and find this vedio. Thanks a lot to make this theory so clear explained I have ever watch. I expect to understand the math behind, but your given example is just too good and help me get rid of it. Thanks again, respect!
I wish I had proper youtube back in my child days (29 now). What would school have been so much more easier, resulting in learning even more in a shorter amount of time. No 'dry' text and pictures and brain wrecks. I'm a bit jealous and happy for my daughter the opportunities she have to educate herself. Thanks to people like you Brian!
i had to stop watching just so i can compliment u on how simple is and easy to understand this video is.
Thank you for your effort.
In my opinion, teaching is much more than understanding theory, explaining it, a good teacher can use imagination and analogies to inspire intuition in to their listeners. You, sir, have a knack for being a teacher. Have a nice day.
This is so beautiful!!! Excellent example.
My professor explained this and I was quite confused. I'm really glad I stumbled across your video. Thank you so much Brian.
Back to this again since last time when I was in 2nd grade in university. Such high quality
Nice work! I'm an industrial automation electrician, this is the best explanation I've ever seen. Thank you
I didn't have much hope to understand the PID systems until I saw your 2 videos ....I've actually understood them quite well...Thanks
I watched this for a robotics programming thingie and I ended up learning something insanely interesting, thanks!!
This was extremely helpful. All these transfer functions and PID controllers in class and they never explain what this all really does and means. You did that just right
Thank you for the big complement! As for the P controller, you are correct in the general case. However, I applied a little control system trickery! The error did go to zero, so the output of the law was zero, hence no steady state error. The reason I was able to get away with this was because the output from the controller represented a change in pedal position and not an absolute pedal position. Once the velocity error went to zero, stop moving the pedal, and the velocity will remain constant.
If you are talking about adding the 1/s term at 05.43 in the video: When you go from velocity (xdot) to position (x), that is following the arrow (from left to right) in the block diagram, you multiply with 1/s (integrate). When you go from postion (x) to velocity (xdot), that is the opposite way of the arrow (from right to left), you divide by 1/s wich is the same at multiplying by s (differentiation).
I am attempting to tune a PID controller for 'closed loop boost control' on my car. This PID controller will regulate air pressure in the intake manifold of the car that a turbo charger is producing. The output of the PID controller drives a PWM solenoid that opens and closes a waste gate. The ideal PID controller for this application will let the turbo increase pressure as quickly as it can to the setpoint level and then quickly open the wastegate to avoid overshoot. Thereafter, it will modulate the PWM signal to the solenoid to actuate the wastegate and regulate pressure to the set point value. The PID controller is setup to not activate until within 30kpa of the set point. At which point, it starts actuating the wastegate. The pressure can either increase slowly or very, very rapidly depending on how quickly I mash the throttle, so keeping track of how quickly the pressure is increasing is very important to controlling pressure. Based on what I watched here, it seems that a PD controller with maybe a bit of I value, is best?
Wow such a gifted teacher 12mins of your videos saved me 5hrs of studying painful lectuer notes. Thank you:)
First let me say that your tutorials are clear and with good sound quality. And you're the first technical person presenting technical stuff with joy, my wife tells me ("all the others sound either dull or depressed"). If my wife tells me so, you should take that as a big compliment ;-).
At 4:21 you state that with proportional controllers the steady state error will be zero. How can that be, since you need an error to have your controller producing an ouput (output=gain*error)
This might be the most beautiful analogy I've ever seen in my life
As An Engineering Technician who has 30 years working in the practical world: has been a US NAVY Aircraft Flight Controls Systems mechanic for 8 years and having a two year degree in Electrical / Electronics Engineering Technology, I am very interested in CONTROL THEORY, I have worked at LOCKHEED Environmental Testing Laboratory for 17 years doing ACOUSTICS, VIBRATIONS, THRMAL/VACUUM, EMI/EMC Radar Testing, Structural Testing = We DID PRACTICALLY TESTED EVERYTHING from STUCTURES to breadboarding building electronics sensors, set up instrumentations systems working with all types of Engineers from all the Engineering disciplines, now I am retired = I still have the passion to ANALYTICALLY REVERSE ENGINEER what I DID and learn advanced mathematics, DESIGNING DIFFERENTIAL EQUATIONS OR MODELS solving them with LAPLACE TRANSFORMS and implementing, and BUILDING Electronic Circuits to simulate anything in the Physical World. Microcontrollers, Mechatronics, Robotics, Remote Sensing.
better than a month and a half of engineering lectures. saved my ass on exam 1
AWESOME! You made me understand in 13 min what I couldn't in a 4 month course! I give you my best teacher award
Brian! If you were a stock, I would buy... You just can't beat this thorough explanation!
This was the best explanation of PID I've ever seen - it actually had relatable examples.
A whole semester of Control Theorie explained in a 13 min video. Great job!
Are you saying that you learned virtually nothing from a semester of control theory? Did you actually pass the subject?
Einstein said that if you cannot explain something simply, you don't fully understand it. You must be the next Einstein. That was sooo simple it was amazing. Thanks very much for sharing such a complicated subject, but in such a simple way. More text books should be written with your style - the world would be a better place
Hello Adwait, great catch! You are absolutely right. At 11:50 I should have increased theta for just a moment and then had it come back down to the same position. I'll make an annotation and add this correction to the errata in the description. Thanks for finding that.
2:07 Stuff like this needs to get phrased, we can all relate to but we never actually think about it this way, mind blown, amazing video.
You made the concept really easy for me to comprehend. I don't even need to go to classes, this class is way better. Appreciate it man!
this series of lectures deserves more views!
I'm trying to learn this for a while now, but thanks to you I finally get the sence what I'm actualy doing with the calculations. The car example helped a lot to get things to make sense and I now finally understand what can go wrong with wrong gains. Thanks a lot.
Hello again Raed. Is it too hard to follow? I was concerned that it wasn't clear enough when I was finishing it up. I'm taking the next 2 weeks off but when I get back I'll see what I can do to make it easier to understand. Thanks for the comment.
Thanks for the reply! It was just that you seem to state that p-controller generally have no steady state error (and I always tell my colleagues otherwise). Position control using a stepper motor is another example where the process itself has integrating action (like a delta on the gas pedal), and hence p-control here does not have steady state error. Greetings from the Netherlands.
Hi Brian, thanks for the videos.
Just a small comment. I liked your previous video on PIDs quite a lot, and you explained the theory quite well. However, with this one, I think the difference between the different controllers is a little more confusing since you change the problem each time. Perhaps it would be better to use a single example.
Love the illustrations and your style of teaching. :)
Also, humans themselves are control systems. For example, if you try to walk from your kitchen to your bedroom, you accomplish this with feedback control. Your eyes are the sensors and your legs are the actuators. Your brain (controller) adjusts when to turn and how fast to walk using eyes as feedback. Also, your internal temperature control keeping you at 98.6 degrees is another controlled system in your body. There are tons of examples like this with humans.
Sincerely i enjoyed the instructions of the lecture but the example part which is this i still have issue please i can used steam plant for explainations
Brian, my dude, you are saving this poor engineering student's skin. thank you!
Thank you so much. Best conceptual introduction to PID on the web
your work is wonderful. You presented the whole idea so easily and clearly . Frankly I have never seen any other teacher to make this topic so easy. Thank you sir.
This is the best explanation of PID control. Good job Brian !
These videos are some of the best on the internet. Thank you sir
Great video. Did anyone else notice that the traffic light patter was upside down? Made me chuckle.
+Alex Dlugosch those who are colorblind didn't
Instead of replying to all your videos: thank you for these great videos! They are really helpful. You go pretty fast, but on the other hand if you already studied the material but still did not get enough 'ahah!' moments, then this is helpful. But hé, .. we also have a pauze button!
I'm a cs student working on a project relating physics-based simulated skeleton animation. Reading those robotics papers is really difficult for me since I dont have any knowledge mech engineering. Your vids help me alot!
Great videos!! And good understandable spoken english for non-english speakers, thank you!!
you helped me out with a good resume and making everything clear. My teacher is going way to fast for all students in lecture. One down side... the real problem I have in the lectures are not in the videos (cascade regulator, position and speed regulator in a car, with accelerate and decelerate limits). But they help really much!
I am so glad that I am taking these courses after 2012.
Legend Explanation!! Give this guy a doctorate.
love how you linked the videos in the box for the next video in the series
Thanks for the video! I'm a first year mechatronics engineering student from Canada about to start some personal projects!
In the very last example at 11:40 isn't the P controller supposed to go UP first and then down? Assuming it controls the angle of the gas pedal in this example, and going up means accelerating (unless I misunderstood something in there). Thanks if someone answers.
You are the man, thanks by the way even 9 years ago your explanation is still the best
The two videos on PID are awesome and they helped me a lot as a student.
Hey, hope u gonna see my comments it's kinda late but the book about control theory uv been writing it's more then great and helpful , shortly it's a must read for every control engineering student .
Thank u
Thank you for the very visual and easily understandable explanation about the PID concept! I'll share it to my friends as the greatest explanaition I've seen so far.
Hello Brian,
Your lectures are more clearer than the university tuition's i had taken... Cheers to you
Im trying to learn as much about PID as possible. What a great start
Very nice and illustrative example of PID ! Just 1 remark though, with a Proportional only controller, I think you'll have a static error at the equilibrium and not 0 (if your process is stable and not integrator). That's why we generally add integral action to cancel this static error...
Dear Brian, Thank you for sharing your knowledge as it was informative
This might be hard in this comment box to explain .. but here it goes. 's' is a complex variable in the s-domain and is similar to how 't' represents time in the time domain. W0 is the frequency below which the low pass filter approximately passes the input straight through. Above this frequency the signals are attenuated (not fully but it's a good approximation). Check out my video Bode Plots by Hand: Real Poles or Zeros for a little better understanding.
I really liked the frequency and output example at around 2:00 . amazing!
This is probably the first time I'm commenting on a CZcams video: great job, it's so easy to understand. I hope you continue making more informational videos such as this!
Every second of this video helps! Thank you so much!
the amizing fact that you're good to choose the most efficient exemple
Thank you so much for your videos Mr Brian... anyway, considering that some of them are very reach in contents and details, may I ask you to slow down a little bit your presentation, please? This would be very much appreciated by most of us that follow you with deep interest from all around the world...Thanks in advance for your consideration and to share your art with us...
This was an amazing video,it proves to be really helpful for anyone trying to understand the basics of a PID controller ,thank you
Thank you for your work. It is so good. A good teacher can explain things simply. The work you put in your videos shows. Thanks
Brilliant explanation but can you please prioritize making PID explanations for the next while ? I find it the easiest when you explain it and I have a lecturer who doesn't speak english as her first language and as you can imagine it get tough to know whats going on. All these control systems lectures have saved me so far.
At 8:30, whenyou show the derivative part of the controller response, it appears to be zero while there is a step in the error. I think that this is not right, although I understand that you wanted to keep it simple and the general idea is right, I think it's worth mentioning that the derivative part should give have a positive kick towords "infinity" since the slope of the step error is almost vertical.
Am I right?
Thank you for your awesome videos! I'm agreat fan!
yea i had the same doubht
Hi Brian! Thank you very much for the brilliant lectures! I do appreciate your work! But I think in this particular video it would be more accurate to say that the angle theta defines the reference velocity which is compared with the output velocity. The error velocity than is processed by the PID-controller.
Very nice quick explanation on the concept of PID control. I appreciate your video
Nice! everything very well explained, very detailed and not boring at all to follow through!
Good job man!
wow !!! one of the most helpful and graphical educational video, i have ever seen !!! thank you so so much.
Absolutely! Humans are very much the controller in some situations. And when you're designing a system that uses a human controller (like an airplane) you better make sure the system is "slow" enough that the human can respond fast enough to it, but not too slow that human can't control it at all. It's an interesting problem designing for humans. When you take the human out of the loop and place a mechanical or electrical controller, then it's an automatic control system.
I still don't understand why a P and PD-controller still have a steady state error. Is there a good video on this question?
Thanks a lot for the videos by the way - they are great. You should become a professor. :)
GangiGangan X
Your lectures are fantastic. Thank you so much for making these!
This helped me out a lot, thanks! It's hard to find thorough explanations but you did a great job.
I love this videos because it all real time life explanations....
Thank you professor...I loved it
This is such a nice example! it easily covers the exception which could arise with a solid understanding behind them.
i am really Thankful to you for this Lecture.
Thank You!!! Never got a better explanation of the system!
In this video if 'phi' is the angle that gas pedal makes then after the chasing car catches the other car, the angle phi should settle at the same value if the speeds before the chase started and after catching-up are same but the video shows that phi after the lagging car catches leading car is more than what it was. Which means the car now needs more gas for exactly same speed. Did you mean something else by angle phi ?
Your videos helped me program a 3D video game of respectable complexity :D!
What causes the integral term to hold the system at 0 error? In the instant before the system reached 0 error, the integrator was a large sum. So when the system reaches 0 error in the next instant, why doesn't the integrator just cause an overshoot? Is this some kind of balance between the I and D terms?
This is incredibly awesome. Fortunately I am done since long ago with exams but it was amazingly fun to watch. Please keep it up! Cheers!
I have watched a few of your videos, and you do a great job explaining your examples. Thanks for your help.
great work man...!! the way you are making understand with examples is superb. its like explaining very complex industrial problems but the way teachers do in kindergarten. really helpful. thank you.