Engineer Spotlight: Brian Douglas of Control Systems Lectures on Education in the Age of YouTube

Engineering education, like everything else, is increasingly moving online. In this Engineer Spotlight, AAC's Hana Othman spoke with control systems engineer Brian Douglas about control engineering, YouTube education, and how autonomous vehicles and spaceships aren't so very different.

Brian Douglas is a controls engineer on a mission. He provides free, accessible engineering education materials online. His YouTube series, Control Systems Lectures, aims to educate engineers on control systems theory through engaging and creative video lectures.

So why should EEs be interested in control engineering? In his introductory video titled "Why Learn Control Theory", Douglas refers to "...control theory is the glue that stitches together all other engineering fields. And by understanding the fundamentals of controls, it opens up the door for you to understand and solve many different engineering problems—and not just as a controls engineer but as any engineer." 

AAC's Hana Othman spoke with Douglas about where control engineering differs from electrical engineering, and how modern online education works in conjunction with traditional education methods.

---

 

All About Circuits: To start out, can you give us a quick definition of what control engineering entails, especially as it relates to hardware design?

Brian Douglas: I’m usually pretty wordy when describing control engineering, so let me give this a shot. A short description of control engineering is getting a system to behave the way you desire by determining and adjusting the input signals into that system.

An easy way to think of this is in block diagram form. If you understand your system—this usually means having a mathematical model of it—and you know what you want the output to be, then you can use control theory to figure out what input creates that output. That’s really the crux of the problem: How can I add energy and mass into the system in a very particular way so that I get the system to do what I want?

 

Representation of a control system. Image courtesy of Brian Douglas.

 

Sometimes you can get away with a simple open loop design where you just select a set of predetermined input signals that play through the system regardless of how the system is reacting to them. This type of controller is typically reserved for simple processes that have well-defined input-to-output behaviors.

 

 

More often, you set up a controller that adjusts the input into the system based on the how the system is behaving. That is, your controller checks the output of the system, compares it to what you actually want, and tweaks the input accordingly. This is a closed loop, or feedback control design and it’s indispensable as a control tool.

 

 

But control engineering is more than just designing a controller and tuning it to get the performance you want. Over the course of a project, actually designing the controller might be a relatively small part of a control engineer’s day-to-day job. Depending on the size and phase of the project, our responsibilities and the groups we interface with vary greatly—building models and validating them, testing hardware, managing changes to the system, and writing and reviewing requirements, to name a few.

AAC: How does control design fit into the mosaic of systems, electrical, mechanical, and process engineering?

BD: I think control engineering is most closely related to systems engineering since the control system typically includes hardware, software, mechanical interfaces, and operations. Any change to the system, even ones that seem small, have the potential to affect your design and require you to assess its impact. Therefore, you typically have to be part of the process for any change to the system, whether it’s resistance change down at the board level or how you’re going to operate the product at the system level.

 

For comparison, the "mosaic" of engineering fields, according to a mechatronics. Image used courtesy of Team Chowbraka.

AAC: While the electronics industry’s constant state of evolution can be challenging, some of our electrical engineer readers joke that at least the rules that govern the behavior of electricity don’t change. What’s your take on how control engineering has changed as components and systems evolve?

BD: With size, weight, and power needs constantly shrinking, there are more components available for smaller and cheaper projects. Not only does this help out a DIYer at home, but it also allows professional products to have more distributed control systems.

In education, we like to use cruise control as the go-to example of a feedback controller in an automobile, but because of the ubiquity of really good, affordable components, modern cars have dozens of feedback controllers: lane assist, active air/fuel mixtures, rain-sensing windshield wipers, etc.

Additional feedback loops in a system means the complexity of the project overall has become more complicated. But that’s OK because design software is also evolving by abstracting away these complexities, making control theory more accessible, and allowing a single engineer to accomplish more than ever before. There is always the danger of the software doing all of the heavy lifting and a designer not truly understanding their design ("The software said it was correct!"), but with that being said, the foundation of control theory—classical and modern—is pretty well established. If you have a good understanding of the fundamentals, then I think access to more components and better software will only help you.

AAC: How do industry trends affect practicing control engineers?

BD: I think industry trends coming from some of the engineering fields that border control theory (like computer science, machine learning, and optimization) are impacting practicing control engineers the most. My background is in spacecraft control, and, as an example, there is always the desire to see the research in self-driving cars applied to spacecraft. After all, they are both autonomous vehicles so why not take advantage of the developments in Simultaneous Localization and Mapping (SLAM) for asteroid rendezvous missions?

 

Douglas's background in spacecraft control shining through in a Control Systems Lectures video on designing control systems.

 

Space systems haven’t been very quick to adopt new technology for a number of reasons but a main reason we’re slow is because of the huge cost associated with qualifying new technology, often to build what is just a few spacecraft. Therefore, the simple solutions are hard to beat when you balance state-of-the-art performance against cost, practicality, and risk. There’s a reason PID controllers are so prevalent.

AAC: Shifting gears, let’s talk about engineering education. As an engineer, what attracted you to the idea of developing educational videos and influenced you to establish the Control Systems Lectures YouTube channel?

BD: When I left college and entered the working world, it was hard for me to apply the theory I learned to a practical application. I found I was able to work through the math, but I didn’t have a good understanding or intuition about the problem, itself. This left me in a hole when the problem at work wasn’t stated exactly in the form that I was used to in college.

I ended up gaining that intuition through experience and from working with two fantastic mentors over the years and I wanted to share some of this perspective with others. YouTube seemed like a way to reach a lot of people and I was attracted to the visualizations that are possible in videos.

 

Douglas writes and draws on black backgrounds and photos to demonstrate engineering concepts. Image from "A real control system - how to start designing"

AAC: You provide informative videos on specific control concepts to thousands of viewers, many of them students. How do you think your style differs from traditional in-class instruction?

BD: I don’t think my videos provide anything that can’t be received from traditional in-class instruction. However, I think videos are a more efficient way of delivering some content.  My recommendation for how my videos fit into a balanced educational diet is as follows:

Videos are great for a fast introduction to a topic. They can tie multiple concepts together quickly and generate excitement for learning the subject. Videos alone, however, are not enough. In the classroom, students get to experience group dynamics and commiserate with other students who might be struggling just like them. There, the professor can guide their learning, answer questions, and challenge their knowledge.

Watching videos and being lectured to is also not enough; the students need to hone their skill through practice. This is done in two places: a lab where they get to work with hardware and experience the practical problems that arise from working with real systems, and at home where homework and reading the textbook provides focused, quiet contemplation.

I think using videos in this complementary sense, rather than the draconian approach of moving all education online, is a good way to get the most benefit from each style of teaching. 

AAC: Your experience as an engineering student and professional has clearly shaped the topics you choose to cover on your channel. Do you specifically cover topics that you think your audience may struggle with?

BD: Absolutely! Many of the topics I cover are the ones that took a long time to finally click for me. For me to learn, I read, listen to, and watch a lot of different explanations and then I form my own understanding. Often, I get that understanding from a single excellent source (e.g., book, professor, YouTube video)—but if one doesn’t exist to my satisfaction, then I create it. But the way people prefer to learn is very subjective. My videos don’t work for everyone, but they are another source people can go to for information.
 

“For me to learn, I read, listen to, and watch a lot of different explanations and then I form my own understanding. Often, I get that understanding from a single excellent source… but if one doesn’t exist to my satisfaction, then I create it.”

AAC: Many of your viewers have vocalized their frustration with university controls courses falling short of delivering practical instruction they can apply to real world scenarios. How do you think your videos address this issue?

BD: I don’t think the controls classes, themselves, do a bad job (there is a lot of good useful theory that is presented in these classes). I think the university experience for me and some of my viewers as a whole lacked the broad context of why I should learn these topics.

I found that if I didn’t understand how I was ultimately going to use some bit of knowledge later on, then I didn’t care to learn it as much as I probably should have.

I’ve already mentioned how I think videos can complement course work, and this is another one of those places. Videos can provide that context by describing real examples of how this knowledge is used in practice (or in some cases not used) and ignite that interest in the topic so that the student wants to spend the hours learning and practicing it. Professors can obviously cover this as well, but they have such a limited time in class with their students that I think it would be better for them to focus on the mechanics of solving the problems and correcting any misunderstandings the students have.

AAC: Breaking into a new media scene like YouTube sounds challenging. How do you feel your content differentiates compared to other education-focused channels?

BD: There are a lot of fantastic educational video channels online. But, for me, it wasn’t that hard to "break into" [YouTube] because I didn’t have an end goal in mind when I began. I started creating a few videos just to cover the topics that I had a good explanation for and they took off from there. I wasn’t creating all of this from scratch. My style is heavily influenced by other channels; the most notable is Khan Academy from which I borrowed the black background and colorful drawings.

 

Douglas drew inspiration from Khan Academy. Screenshot of ideal circuit elements courtesy of Khan Academy.

 

I think where my videos differ the most from other channels is that I spend a lot of time describing the problem in an intuitive way rather than mathematically. I think focusing heavily on why something works with words and pictures is a valuable counterpart to the mathematical proofs.

I let other channels and textbooks really get into the nitty gritty details and I get to stay above all of that and focus on the big pictures. This wasn’t always the case—in the beginning, I did spend a lot time working through the math. But I’ve been making videos for five years so my style has evolved as I learn what works for me and what resonates with the audience.

AAC: Would you recommend video content as a beneficial medium for the overall engineering community to use?

BD: I think the engineering community is already using videos to help them learn and solve problems. It’s unfortunate that YouTube isn’t set up really well for searching for and separating out these type of videos. I would love to see a StackExchange-style Q&A community that revolves around answering questions in video form.

AAC: You’ve written books, as well. How has the process of presenting information in a book differed from presenting information in video form?

BD: My book, which I’m still working on, came about because of my videos. I was receiving a lot of requests for PDFs of the drawings that I create in my videos. I thought that was a good idea, so I started compiling the PDFs. However, it became obvious that the drawings, themselves, don’t make a lot of sense on their own. So I began adding words between the pictures and voila—it’s the start of a book.  

I think a unique feature of my book is that I set up a system online where readers can submit bug tickets where they can point out errors, missing content, or confusing sections. I’ve also released it under a Creative Commons license so that it can be shared or used in letters and presentations. People can download it if they are a supporter of my Patreon account or I’ll send them a free copy if they email me. (Editor's note: You can find Brian's contact information on his about page.)

The main difference for me with the book is that I can correct errors in a book much more easily than in a video after it’s been published. I’ve made several mistakes in videos that bug me and I can’t go back and fix them. With the book, that’s not a problem!

AAC: What are your future goals for the channel?

BD: I will definitely continue making videos for my channel as time permits. There are a number of topics that I think I have a good explanation for that I would love to create still. Ultimately, I hope my channel and the style of video that I make inspires other students, professors, and working professionals to share their knowledge in a similar way.

AAC: One last question before we let you go: Do you consider yourself an educator, a practicing engineer, or both?

BD: As much as I enjoy being a YouTube educator, at heart, I’m still an engineer. I like solving problems and working with a team to design and build things. So, in addition to making videos and writing, I still consult for engineering companies. It keeps my mind fresh and learning new things. Hopefully, I’ll put things I learn today into a video that I can share with the world tomorrow.

 

Thank you for your time and insights, Brian! You can check out the full library of Control Systems Lectures videos here.