Aviation Safety Certification - Part 2 of 2

Strap yourselves in for an electrifying journey with Lachlan Kilby from Javelin Aerospace, as we navigate the turbulent skies of aviation innovation and the certification conundrums of electric Vertical Take-Off and Landing (eVTOL) aircraft. Discover how the shift towards outcome-based certification processes is revolutionizing the industry, and why the flexibility of these pioneering aircraft is sending ripples through the world of regulations. We're tackling the tough questions: How do we balance the freedom to innovate with the imperatives of safety and compliance? And what does this mean for the future of flight?

In the cockpit with Kilby, we dissect the intricacies of emergency safety measures across aircraft types, pondering how eVTOL designs can incorporate the resilience of traditional planes and the agility of helicopters. The episode takes a critical look at the technological leaps necessary for eVTOLs to handle the unexpected, emphasizing the paramount importance of software in these advanced vehicles. As we chart the course through Australian skies, prepare to be enlightened by the strategies shaping the coexistence of autonomous and manned aircraft—a symphony of innovation and regulation that's redefining our airspace. Join us for this thought-provoking voyage into the next frontier of aviation.

Thank you for joining us on this episode of "Takeaway with the MEF."

We hope you found this discussion insightful and engaging. If you enjoyed today's episode, please consider subscribing, rating, and leaving a review on your favorite podcast platform.

Get in Touch:

• We love hearing from our listeners! Send us your feedback, questions, or suggestions at neeraj.chadee@mefsc.org.au

Stay Tuned:

• Don't miss our next episode where we'll dive into another intriguing topic. Until then, remember to keep seeking knowledge, staying curious, and finding your own takeaways.

"Takeaway with the MEF" is brought to you by Manufacturing Excellence Forum Sunshine Coast and Advertastic PTY Ltd. Thank you for your support!

Disclaimer: The views and opinions expressed on this podcast are solely those of the hosts and guests and do not necessarily represent the official position of the MEF or its affiliates.

Transcript

Speaker 1: 0:00

You knock a leg out of your dining room table. What's going to happen? Yeah, it's not going to be a good table anymore, is it, it's going? To tip over. Yeah Right, so that's a big consideration. Is that you really need to think about what happens when something fails?

Speaker 2: 0:15

Takeaway with MEF. This series follows a group of students and industry experts as they join forces to create a flying taxi or EV tour. I am your host, neeraj, and welcome to part two of my chat with Lachlan Kilby, co-founder and managing director at Javelin Aerospace. Make sure to check out part one. If you haven't already, let's take it away. We spoke about the traditional aircraft certification and that it's been around for such a long time, and so what about EVTOL now and emerging technologies? How does that all fit in into this certification framework? That's been quite old, in a sense that it's been around for a long time. Can you talk about that? How does that plug in there? Does it even plug in there, or does it need a new framework?

Speaker 1: 1:15

Look. The short answer to the question is there a cert framework for eVTOL aircraft at the moment is no, there's not, and that's probably the most challenging part about it all. The reason for that is, as we've spoken about, certification has grown, or at least the regulations have grown very organically in response to incidents on traditional aircraft and in that sense they've grown safer, but they've also grown more prescriptive as to what an aircraft in inverted commas needs to look like. A lot of that is probably down to economics as well. In certification, we have this concept of prescriptive versus outcome-based. Prescriptive will be something that tells you literally the boxes that you need to check, you need to have six seats, you need to have two doors for every two seats, so on and so forth. They're very easy to follow. Outcome-based, on the other hand, is really probably where regulations tend to start. So back in the 1930s it would have been outcome-based. Everyone needs to be able to exit the aircraft safely. It's a very broad kind of concept. If you were to test that or to try and show that that was able to occur, that if it had an incident everyone was able to evacuate safely, you'd probably have quite a hard time showing that. I mean, it's on you to come up with the test to demonstrate that, and it requires a lot of interpretation. The short of it is that interpretation in aviation is an expensive and time-consuming task, because you as the designer may see everybody evacuating safely, but the regulator who's looking at that may beg to differ, but because there's no rules, it can result in a lot of grey area. However, if you've got a prescriptive-based system, they'll say oh well, you met the 90 seconds, everybody's off and you had half the doors obscured. Check no more questions. So it becomes a lot more efficient. Check no more questions, so it becomes a lot more efficient. Your prescriptive based regs are a lot more efficient to work through, which is, I guess, one answer of why they've tended to go in that direction. As they evolve, they tend to get more prescriptive.

Speaker 1: 3:35

But now, with eVTOL, we're in this position where these aircraft that have really come about due to a pretty rapid advancement in technology being energy density of batteries and, obviously, light control systems becoming miniaturized and more electronic. We're essentially back to square one again. If you tried to apply traditional regulations to an eVTOL case, you wouldn't get anywhere. This is the challenge that pretty much every single company that's trying to develop, an eVTOL, is facing right now all around the world, is that they've had to really rethink the regulatory framework.

Speaker 1: 4:21

To their credit, the FAA, or the Federal Aviation Administration in the USA they've created another amendment to one of their most popular design frameworks for small aircraft or light aircraft, which is FAR 23. So they've recently released an update which did its very best to move from prescriptive to outcome-based aircraft in order to allow the development and the certification of they're calling it, advanced aerial mobility or non-traditional types of aircraft, which is great, it's fantastic. But now the onus is on the designers to show compliance with some very broad airy-fairy regulations that really there's no easy check boxes. You have to look at every single one and really show the regulator to the best of your ability that you can meet a requirement that is pretty subjective yeah.

Speaker 2: 5:23

so it's interesting the way that you frame that is that at the start of aviation, during the pioneering years, certification revolved around outcome based right and then, as it became more and more of a known entity, it went for this period of prescriptive base, which so, essentially, we're trading off flexibility for efficiency. So at first it was flexibility, less efficient, and then, as we got more of, we knew what was going to happen. You get less flexibility but more efficiency, so more prescriptive base. That's a good way of putting it. Yeah, well, you said it. And then now we're coming back to this pioneering age where we do need the flexibility again and we can't evolve in a frame that's too prescriptive.

Speaker 1: 6:08

And so that's why you're talking about the advanced air mobility framework, which is a lot more outcome-based, because we're in the pioneering phase and we need the flexibility, and what's happened, as you said, is that in order to get flexibility, we have traded efficiency, and that's really, in a nutshell, why the certification process for eVTOLs is so complicated and time-consuming and expensive and why we aren't flying around in eVTOLs at the moment. It's exactly that. Fact is that we're back to square one. We're back to, effectively, a blank sheet of paper. It's probably worse than a blank sheet of paper in many ways, because we do have a whole bunch.

Speaker 1: 6:50

We've got 100 years of aviation development right. That's just sitting in the background and it's very hard to know what we should be taking from that versus what we should be rethinking and learning again. So I guess we're in a phase of at least for certifiers and for people like the faAA and CASA, for example we're very much in this cherry-picking phase of really trying to learn as much as we can from the past, because it would be a real shame to discount 100 years of accidents and learnings, taking from that what we can in order to maximise safety, but also not trying to choke the innovation of what we're currently doing now.

Speaker 2: 7:30

And you spoke about a couple of different organizations there, the fa and I spoke about casa. Are they the only two involved in this evito?

Speaker 1: 7:41

I'm guessing they do things like yasa and all that european yeah, yeah, yeah, no, it's, it's not just um the faa and casa. You're, there's, obviously there's the European counterpart to the FAA, which is EASA European Aviation Safety Administration and obviously they're very active as well in that space. So really, people tend to defer to the FAA or to EASA in the first instance as the primary regulators in the aviation space around the world. And then you have the smaller states, like Australia and France and Canada, and now the UK, who've split since Brexit, so on and so forth, and they have their own competent aviation safety authorities as well, but typically all eyes are on the big guys to come up with a solution to this.

Speaker 2: 8:31

Yeah, okay, now if we get into the meat of things with regards to specific certs considerations for EV tools themselves, just thinking about what the technology is, how it flies, what are some of the main risks I guess main dangers and what are some of the specific considerations you yourself see for a vehicle of such a type?

Speaker 1: 8:58

Sure, when you're certifying an aircraft, the fundamental rules are the same and you can break down the certification into a number of different, I guess, technical areas, and the first one will be flight or flight performance. The next one will be structures, the next one will be design and construction, so how the thing is actually made. The next one will be the power plant. The next one will be the power plant, the next one will be equipment and then the final one will be sort of operational limitations and how that's communicated to the pilots, as in how do the pilots know how to fly the thing.

Speaker 1: 9:34

For the large part that won't change, but it gives us a framework to look at the certain things within that, and I think that in terms of eVTOL, there are some new challenges, things that we can't lean on tradition for. So the first one that I see it's going to be a major hurdle is probably what happens in the event of an emergency. So with the traditional type of aircraft, so with two wings, etc, then the assumption has always been you'll be able to glide out of a situation where you need to, so provided you have enough altitude. If you have an engine out, you have a mechanical failure of some description. Then you will be able to glide back down to the ground in a safe and controlled manner yeah, and why do you decide here?

Speaker 2: 10:21

because some people might be wondering okay, what happens if you're in a helicopter, in a rotorcraft.

Speaker 1: 10:26

Yep well, the same thing applies um rotorcraft in general. There's an assumption there that you will go into what's known as an auto rotation, which is a helicopter's way of gliding, for want of a better way of saying it and that's really handy as well. The there is an idea out there that once a helicopter loses its engine, then it will just drop like a rock. But that's not the case. They can actually glide, and quite well too. But moving to an eVTOL, that's a different story.

Speaker 1: 10:59

So if you're relying on, maybe, a traditional type of drone design where you've got the quad rotor type design so you've got two rotors at the front, two rotors at the back and you're sort of sitting in the middle if you lose one of those, if you lose power, then what happens in that scenario? So, based on a square geometry, you actually end up with an aerodynamically unstable. You're losing one of your legs at the table, for example, if you thought about it as a table. You knock a leg out of your dining room table. What's going to happen? Yeah, it's not going to be a good table anymore, is it?

Speaker 2: 11:35

It's going to tip over?

Speaker 1: 11:36

Yeah Right, so that's a big consideration is that you really need to think about what happens when something fails. It's not good enough, insert, to assume that nothing's going to fail. The more robust thing to do, and the acceptable thing to do, is to acknowledge that there will be failures. That's just the way the world works and you need to prove that your design is robust enough to get out of those situations safely. So that's one aspect is how will the aircraft conduct itself when there's a failure? So I'm talking about an electromechanical failure. How about software? So you know, software is going to be a massive part, probably the primary part, of the certification of any type of electronic vertical takeoff and landing type of vehicle, because for these vehicles, your average human is not capable of flying these things manually. So there is a lot of computer stabilization at play and we really are relying on that in order to keep the aircraft stable and flying appropriately and smoothly.

Speaker 2: 12:47

And we've seen what goes wrong when software goes wrong Exactly, or these glitches like the 737.

Speaker 1: 12:54

MAX, for example. Yes, and that's a much more simple case as well. You know that was an automatic control system that simply controlled one axis of rotation of the aircraft. So in an EV toll you've effectively given the computer primary control over all axes of your aircraft. So you need to make sure that the pilot is in control to some extent, in the sense that if he points in a direction or if he moves the joystick in a certain direction, the aircraft will, to the best of its ability, after keeping itself stable, will move in that direction.

Speaker 1: 13:28

However, if you have a malfunction from software side of things, then who knows, who knows what happens. It may think that it's upside down when in actual case it's not. It's the right way up. So software is going to be a crucial, crucial component to getting an electronic vehicle certified it always has been and especially moving between flight profiles as well. That's going to be a real challenge and from my discussions with people in the industry that have actually been trying that, that still remains. One of their primary challenges is taking a vertical flight regime and transitioning into a forward flight regime and keeping your software sane and safe while doing so, and that's the real challenge.

Speaker 2: 14:18

And then doing it the reverse, so once you're finishing your cruise, slowing down and going back to vertical, those sort of things. Yep, so we've spoken about software. What happens if an engine is out? We've spoken about the transition from upwards flight to forward flights. What about licensing of people actually going to fly the aircraft? What sort of experience do they need to have? We don't have any framework at all for any of those things right now. Yes, absolutely.

Speaker 1: 14:48

Even if we did get the aircraft in the sky certified, then there's a whole host of other challenges that we need to think about. So is our current licensing scheme available from CASA? Is that going to be suitable to allow your average Joe to suddenly be able to fly one of these EV tolls? Similarly with maintenance, if they need to be worked on, you can't just come off the street and start working on an aircraft. You need to be able to work through a licensing scheme there. So every single person that works as a maintainer of an aircraft has actually had many, many years of training to do so on particular aircraft types as well. So that's going to be a question is it's probably going to need a whole new training and maintenance schemes to come along with it again.

Speaker 1: 15:39

Then you have how these things are going to interact with air space, and that's another challenge as well. We do have provisions for Rotocraft to be able to interact with airspace, but the way that it's currently set up is that you have humans piloting these things. They're not autonomous in any sort of way. So there's been a lot of work that's been done, and I've been to a few AAOZ, which is the Australian Association for Unmanned Aerial Vehicles. There's a lot of very impressive work going into how we can allow autonomous vehicles into existing manned airspace. Very exciting, very complicated stuff. That goes beyond my head, but you know it's there, the work is happening there. But that is a huge challenge as well effectively allowing a computer to come into a manned airspace and do it efficiently as well. It's very easy just to shut down large sections of airspace and that's how you do it, but that's not really feasible, especially when you're around large airports and places like Sydney and the Gold Coast etc. It's not feasible to shut those things down.

Speaker 2: 16:54

Yeah, it's a good point because there's a massive aspect of we have this existing industry that needs to keep going still, and whatever new technology coming in needs to plug in to that existing infrastructure industry, which is under the design constraint, if you want. And you were talking about the different cert aspects, and it strikes me as there's a wide range of ways that these can be tackled. When you're talking about engine failures, for example, that could be something that you tackle in the design space. So maybe your eVTOL has got wings on it so it can glide, or maybe you have some parachutes on it, but then you have the licensing bit, which is more of an operational side of things that you still need to tackle. So it feels like it's going to bring together not only the design aspect but a wide range of other disciplines?

Speaker 1: 17:49

Yeah, definitely, and the operational aspect as well. We generally take the approach that you design out as much as you can. You design out failures, you design out these sorts of failure cases which can be quite catastrophic. You do the best you can in the design phase, but you can never design out 100% issues. So then you rely on operational mitigations. So there'll be a bit of that as well.