Episode 223: From Crawling to Flying: Bio-Inspired Robotics with Heiko Kabutz

Published 2024-05-30

Episode notes

What can we learn from cockroaches and spiders?

Hello Smart Firefighting Community!

Welcome to another episode of covering real world innovations via interviews with fire service and technology industry experts that empower YOU to develop your very own Smart Firefighting strategy!

Welcome back to this 20-episode Texas Robotics Summit Mini Series! This 4 day event (March 4-7, 2024) was hosted at the Reveille Peak Ranch and by our friends at Granite Defense , and showcased the use of robotics (air/water/ground) in public safety via keynote presentations, training scenario stations, roundtable conversations, and more.

In this episode:

Insights into bio-inspired robotics and practical applications

The synergy between biology and robotics for innovation

The future of untethered robots in various fields

Find out from Heiko Kabutz - an Insect Robotics PhD student at the University of Colorado Boulder.

Heiko talks about his research on bio-inspired robotics, specifically studying cockroaches and spiders. He explains the importance of making research relevant and applying nature's concepts to robot design.

Heiko also explores the difference between bio-inspiration and biomimicry. He shares his future plans and challenges in developing untethered robots and using them for practical applications. And lastly, Heiko emphasizes the value of learning from field experience and using robotics to understand the world around us.

So click play now to hear what he has to share! And stay tuned for the next Mini Series episode!

Head to www.smartfirefighting.com to discover how SFF accelerates innovation for emergency responders, to find out when our next event is, or review our curated resources!

Facebook | Instagram | Twitter | LinkedIn

Chapters

0:00 Kevin's intro: the Texas Robotics Summit mini-series
0:45 Heiko Kabutz joins: from South Africa to CU Boulder
1:25 Heiko's experience at the Texas Robotics Summit
2:24 Making research relevant
3:14 Studying cockroaches and spiders in the lab
4:19 Watching and observing nature
6:53 Biomimicry versus bio-inspiration
9:31 How insects climb, crawl and dig
10:12 Getting bio-inspired robots into the field
12:20 Heiko's plans and the challenge of untethered robots
13:10 The four-legged prototype
14:13 The value of feedback from field experience
15:46 Parting thought: use robots to understand biology
16:43 Closing thanks and outro

Questions answered in this episode

What animals does Heiko Kabutz study for his robotics research?

Heiko Kabutz says his lab studies cockroaches and spiders. They let spiders run and observe how they get through environments, noting that these animals can get through really small crevices and spaces that are typically inaccessible.

Hear it at 3:15

What is the difference between biomimicry and bio-inspiration in robotics?

Heiko Kabutz explains that biomimicry is copying exactly what is in nature, while bio-inspiration focuses on using concepts from nature rather than exact methods. He says engineering is still far behind biology, so they look at how to get closest to natural designs with current capabilities rather than copying complete systems.

Hear it at 7:46

Why does the robotic insect prototype have four legs instead of six or eight?

Four legs are easier to control than six or eight legs. Heiko Kabutz says he is working toward an eight-legged system within two years because more legs provide redundancy. If one leg fails, the robot can still keep going, similar to how spiders can still run with missing legs.

Hear it at 13:27

What is the biggest challenge for the insect robot right now?

The robot is still tethered with cables, which makes it pretty useless for deployment. Heiko Kabutz says getting the robot untethered is the biggest challenge, requiring development of electronics, communication, and a backpack to carry weight. He hopes to have a version without cables within the next year.

Hear it at 12:21

What has Heiko Kabutz learned from the Texas Robotics Summit?

The highest value has been getting feedback from people with field experience on their needs and first impressions. Someone suggested considering environmental challenges like dust, which he had not necessarily thought about. These different viewpoints will help make improvements when he returns to the lab.

Hear it at 14:41

Transcript

Machine transcript of the published audio. Names may be misspelled and some sections are missing.

Ep 223 transcript

From Crawling to Flying: Bio-Inspired Robotics with Heiko Kabutz

Speakers: Kevin Sofen, Heiko Kabutz

Source: Whisper medium.en machine transcription of the published Podbean audio (no original transcript found in Drive), lightly cleaned. Speakers were separated by voice clustering and then checked and assigned by hand at every speaker change. Gaps where the model returned no usable text are marked [inaudible m:ss].

[0:00] Kevin Sofen: Welcome to a 20-episode mini-series at the Texas Robotics Summit hosted by our friends at Granite Defense. We dive into the evolving world of drones and robotics in public safety aviation. Through 20 fantastic conversations with leaders across public safety, from the fire service, law enforcement, EMS operations, wildlife management, and more, we explore the challenges, investments in the critical role of technology in enhancing emergency services operations. A special thanks to the sponsor of the Smart Firefighting podcast, W.S. Darley & Company, for making this podcast possible. Enjoy the show. We're rolling here at day three at the Texas Robotics Summit, sitting alongside Heiko Kabutz, who is a PhD student at University of Colorado Boulder. Heiko, how are you doing? Did I say your name completely wrong or correctly?

[0:58] Heiko Kabutz: Oh, that's correct.

[0:59] Kevin Sofen: How would you say it in your native tongue?

[1:01] Heiko Kabutz: Heiko Kabutz.

[1:02] Kevin Sofen: OK, that's a lot better than I can ever say it. Where are you from originally?

[1:06] Heiko Kabutz: Originally from South Africa.

[1:07] Kevin Sofen: OK.

[1:08] Heiko Kabutz: Yeah, and came here for a PhD.

[1:09] Kevin Sofen: And how did you go from South Africa to University of Colorado Boulder?

[1:13] Heiko Kabutz: Specifically lining up just research interest and just trying to find a connection between a lab that's doing biology and robot development and being relevant to the environment.

[1:25] Kevin Sofen: And now we're sitting here in a place called Burnet, Texas. From South Africa to Colorado to now here, Burnet, Texas, how did you get involved? And what's your experience been like here at the Texas Robotics Summit?

[1:39] Heiko Kabutz: Yeah, so the experience has been really great. The main motivation for that has been to come to a place where we can actually not just sit in our lab with our robots, but actually get feedback from people in the field. What would they want in the next 10 years? How can we build everything we do in research to make it relevant to the environment? That we're not sitting in our lab developing some new ideas, which actually no one cares about, no one wants. That might be interest on some science question, but doesn't really help anyone in the longterm. This has specifically been for us getting feedback on what improvements should we do? What changes should we do? How can we progress the field in the direction where it's really helping the first responders in the next 10 years or something.

[2:24] Kevin Sofen: I love that phrase, making research relevant. And it's very important to be able to take a lot of the academic foundation and the importance of academia and apply it to the field with different program PhD institutions to a lot of the stuff that NIST and PSCR is pioneering. And I was very intrigued when I first met you, as well as I'm sitting here looking at some of your work. We're in Texas, and I've had a range of different bugs and wasps and bees and terrifying looking insects that I'm like, oh my gosh, insects terrify me. And just the idea of something really small that like, you know, it's a safety brief and they're like, yeah, black widows are pretty prevalent here. I'm like, that just terrifies me. Snakes, scorpions, and spiders. But here on the table, I'm looking at somewhat of a looking evolution of a robotic insect. What is this?

[3:15] Heiko Kabutz: So to give a bit of background, In our lab, we're studying, specifically, cockroaches and spiders. So we have a bunch of spiders which we let run and observe how they get through environments. And the observation of that has been that these animals start getting through really small crevices and small spaces and have access to environments we typically can't get access to. From a robotics side, we've been very interested in, okay, how can we bring this compliance and also the small scale of robots into an environment of like, let's have something robot which we can control that gets into spaces where all the animals, all these irritating bugs can also get into. And how does that open our stores into new environments? And then this system, this is actually like double the size of our newest one that we've gotten like even smaller with the intention of having a currently four legged system also possible with six legs or so that can walk around with control of every leg and where we can put sensors on top to then navigate into small spaces.

[4:19] Kevin Sofen: I personally love the whole science of just watching and observing. There's actually a book that I'm reading, I'm blanking on the author's name, but it's called Peregrine. And it's this one gentleman who lived in the English countryside. He literally was in the English countryside every day for two years studying the Peregrine Falcon and looking at its interactions of where it was bathing, where it was flying. And oftentimes he wouldn't observe necessarily the Peregrine Falcon, he observed all the other birds, the impact of its presence on everything else. And he just kind of, the whole book is fascinating because it goes from being the third person perspective and watching it, to then first person perspective where he literally becomes the bird. And it's kind of giving you this kind of objective of really going deep into what it's thinking, how it's operating. And there's a lot of just really good lessons in terms of customer exploration, just really end user experience, empathy awareness. But that's sort of what you're doing with the dynamic of cockroaches and spiders of like, you are studying and trying to almost become a cockroach and spider. Is that fair to assume? Or in terms of how you're then applying it to robots, of course, but what are your thoughts on that?

[5:27] Heiko Kabutz: Yeah, like becoming, I don't want to become a spider or cockroach, but yeah, that concept is very much of like, at least in the robotics environment typically, like going to a robotics conference or something, it's typically, everyone just wants to build robots and progress the engineering side. But our focus has been very much on like, let's see what's happening in biology and start interacting with it. Of even just feeding and handling spiders every day, you start interacting and appreciating them very differently. And that makes us develop our robots with a very different mindset of seeing, okay, this works, that doesn't work. This seems to be important, but we don't actually understand it. And we never thought about this concept. And specifically with the spiders now walking over logs or walking over uneven terrain. And that's been a big inspiration on how can we use some of these concepts into building robots. And that's where this robot, for example, comes in that it actually has a compliant body that the body can actually move, which is a very untypical thing for robots. Most of the ones even here around are all one fixed body. And this system is just like, how does the environment change our robot design? And how can our robot interact with the environment? Sort of that connection between nature and robot, between biology and robot, and closing it that we learn from what's around and don't have to redevelop everything.

[6:53] Kevin Sofen: So one concept that I personally love, and there's so many good examples, and I'm sure you're aware of it and I'd love to talk about it, is the concept of biomimicry, of how can we copy nature and apply it to different aspects of human ingenuity, from how we figured out how to fly a plane, build a plane, then fly a plane, and the idea of Velcro and burs to looking at whale fins and applying that to wind turbines and I forget the exact bird, but the bird that dives into the water at like 100 miles an hour and uses his beak to break the water and then it flies in seamlessly. And I know in Japan, when they were redoing their train system, they redesigned their train to have a little bit of a nose to break the sound barrier so it wasn't making this big sound explosion. Stuff like that fascinates me. What are your thoughts in biomimicry and what can we learn from nature and apply that to the field of robotics?

[7:46] Heiko Kabutz: So yeah, biomimicry is very much like also copying what's exactly in nature. So our focus has been a lot more on bio-inspiration because we've realized that, at least on the engineering side, we're so far behind biology and many of the things that we could copy one aspect of it, but we can't really copy the whole system. And often like in nature, we see a whole system very much working together with like all the different aspects of it. And with bio mimicry, it's of big advantage if we can copy it as close as possible to the original system. And much of our engineering is still lagging behind very much. So we're more on the inspiration side of now like using, Okay, we see the beak of the bird, for example, breaking something, and now we're more looking at the legs and observing, it's like, okay, how can we, with our current capabilities in engineering, get the closest to it and start using the concepts before using the exact method? So that's like a lot of the, for example, our legs here currently are still like pretty far away from actual design of a real insect, but it's getting towards that design. Okay, how can we change our leg designs because we see in insects, they use small legs at the front, but really powerful ones at the back. How can we start implementing these concepts and really letting the innovation be done by what we see in nature?

[9:08] Kevin Sofen: Dude, bioinspiration, that is a really good term. I agree with his bio mimicry in the sense of, yeah, we wanna copy it, but it's like, well, Dave and I just mentioned like the whale fin. I mean, we're not copying the whale, we're copying the arm, the fin of the whale and applying it, so it really is copying that aspect and then being inspired about how we can use that to apply to some problem that we have.

[9:28] Heiko Kabutz: Yeah, I think that's a good summary of that.

[9:30] Kevin Sofen: Yeah, no, that's what I'll say. And then you had mentioned too, just like different dynamics of how a particular insect, let's say, for example, cockroach and spiders are doing some act, whether that's climbing, whether it's crawling, whether it's digging, and it's getting to new places. You said something that was kind of intriguing of, this is important, but we don't know why. And I find like maybe that's sort of an interesting thing within the field of academia, where then hopefully that translates into ideas, insights, research, to practical implementation. solutions, that kind of, this is important, don't know why research journey kind of starting there. I'd love to hear you just riff on that. Like, what are your thoughts on that and how does that kind of move closer towards getting this in the field someday?

[10:14] Heiko Kabutz: Yeah, so the exciting part in this is like, there's the two aspects. We could easily be just stuck or sitting in a lab exploring all these nature things. There's so much we can explore and just learning more and more. But the real exciting part is actually seeing what we develop come out somewhere and being useful. From a science aspect, there's so much in biology we don't understand yet where we can use robots to start understanding biology better and biology to use to build better robots and in the same sense, again, building better robotic systems to start simplifying our understanding of biology. But the real exciting part is then using everything we've learned from that study to actually implemented now to first responders or in that direction, that they can start using it for their purpose and actually simplifying life in some aspect or improving life by those who really like are out there in the field. I think with this specific robot where we've sort of been aiming towards a specific getting because a small scale getting swarms out of them, like building 50 of them, someone can and just deploy them, and hopefully with some sensors, crawl onto different aspects, for example, inspecting if structures are safe, if something is underneath it, crawling around all the spaces where typically you can't get in, instead of getting a map out of that's sort of the direction where we're trying to get to.

[11:41] Kevin Sofen: I love that, using robots to understand biology, and using biology to understand robots.

[11:47] Heiko Kabutz: Yeah, two connections.

[11:48] Kevin Sofen: Yeah, I've learned a lot from all the podcasts, This is number 16, but I'm taking a lot away from learning from you, man. That's one of my favorite things about podcasting, selfishly, is just being able to talk with people like yourself and learning about the cool stuff you're working on. So it's March 2024, what's the next, like, how much longer do you have left in school? What's your kind of roadmap for this prototype here? You know, what else are you kind of working on? You know, tell me about kind of the next nine months, 12, you know, and beyond with stuff in the works.

[12:21] Heiko Kabutz: Yeah. So for me currently, I have about another one to two years of studies, my PhD in research, and probably staying in a research environment and trying to develop these robots. The very first thing in our robot development is this. We have cables on our robots. They still tethered, which makes them pretty useless. So we need to get this thing untethered. It needs to be fully deployed. And that brings a whole new challenge of electronics, of communication, and of carrying weight. we need to have pretty much a backpack for the robot. And developing all of that is sort of where we've currently been focusing for the past year already, and then also what we're sort of working towards in the next year. So hopefully within the next year, we'll actually have, you'll see a version with this out cables. So that's like the biggest challenge.

[13:10] Kevin Sofen: And so this looks like a prototype more for the, I guess, is that more of a cockroach or a spider?

[13:16] Heiko Kabutz: It's like neither of the two. It's like four legs, but longer legs.

[13:21] Kevin Sofen: I was about to say, I was like, I think it's a car garage, but I also could be a spider because yeah, they're obviously different, but yeah.

[13:27] Heiko Kabutz: The main reason why four currently is just because four legs are a lot easier to control than six legs or eight legs. While I'm currently working towards, within the next two years, is hopefully a eight-legged system because there's a lot more redundancy. If one of the legs fails, the robustness of that, one leg fails, hopefully the robot can still keep on going and really being robust to anything that could happen, like being just out in nature, anything could happen. And then even looking at animals, there's like often a spider running past with only seven or six legs, and it's still running away. You still can't catch it. The other thing is also, currently the legs are like two degrees of freedom. They have quite a bit of controlled motion, but giving that into a full three degree of freedom to move anywhere in space, like at once in the mechanics.

[14:13] Kevin Sofen: Yeah. So just kind of bring it back to where we're at here in Burnet, Texas. And luckily, we've got the shade up now. We've both got our hats on, sun safe, which is great. It's been a lot of really cool conversations with public safety members, touch and feel on drones, and everything at the ranch here, and just everything that Granite Defense has organized has been awesome. What have maybe been a takeaway or two? What's something that you've learned from this experience so far?

[14:41] Heiko Kabutz: The highest value has been interacting with a lot of people here who have so much field experience and getting their feedback on their needs and their first impressions. Because there's many first impressions after working on this for multiple years, which we don't think about. And someone just coming like, yeah, this won't work, or have you considered this direction? Someone suggested yesterday of just like, yeah, have you considered the amount of dust? is these other environment challenges which one doesn't necessarily think about. And getting that new perspective from people who somewhat understand but have a very broad, different viewpoint and background and knowledge. And that's like, will be very interesting for us going back to life and making new improvements based on all the different viewpoints and opinions we've seen here.

[15:30] Kevin Sofen: That's great. Yeah, I mean, that's why we come to these kind of events to be able to surround yourself with, yes, like-minded people, but also very different thinkers to round out of, have you thought about this, have you done that, you know, that sucks, that's great.

[15:43] Heiko Kabutz: Yeah, and that's been like, I guess, the highest value in coming here.

[15:46] Kevin Sofen: Yeah, no, that's great, well, kudos to the event. And so I've been asking everyone on this to kind of close this out here, as far as somewhat of a mic drop moment of sorts, but just kind of curious of like, and this doesn't have to necessarily be even robotics related, but I know you spend a lot of time with that of, if you have a, I would love to hear you a favorite quote, a question, a challenge, or some sort of statement or anything that you could leave us with here today?

[16:11] Heiko Kabutz: Well, that's a tough one.

[16:13] Kevin Sofen: I'm throwing you a wild card.

[16:15] Heiko Kabutz: Yeah. I guess like on this field specific, it's like what I also mentioned before, it's like we see most people use robots just as tools to use them somewhere, but how can we use robotics also to understand what's around us? Like the understanding robotics or using robotics to understand biology in the same sense, we also use biology to start improving robotics.

[16:43] Kevin Sofen: Well said, and I've learned a lot because I mentioned the biomimicry aspect, but I needed to start talking more about bio-inspiration and I really appreciate the work you're doing. It's critically important to the future of all things public safety, and also I'm sure it gets a lot of just environmental science research, so excited to stay in touch and keep up the good work.

[17:00] Heiko Kabutz: Yeah, thank you.

[17:00] Kevin Sofen: Thank you. Thanks for tuning in to this mini-series created at the Texas Robotics Summit hosted by Granite Defense. Make sure to put this amazing event on your calendar for 2025. For this podcast episode, what do you like? What do you think? Drop us a comment on social media and let us know your favorite quote. Thanks for all that you do and have a great day ahead.