Listen Now:      

Transcript:

Luke Renner: This is Advanced Autonomy. I'm Luke Renner.

My guest today is a colleague of mine here at Cyngn, Ben Landen. Ben is the Senior Director of Product and Partnerships, and if we're being honest, my boss. Ben has been in the self-driving space for most of his career. He was the head of product and business at DeepScale, a startup that built computer vision solutions using deep learning. DeepScale was acquired by Tesla in 2019. 

Prior to that, Ben ran an automotive semiconductor product line at Maxim Integrated. Hundreds of millions of his products are embedded in vehicles all over the world. In this conversation, we'll be talking about the changes that have come to the autonomous vehicle sector in recent years, particularly last year, as COVID transformed the way organizations all over the world do business. 

Hi, Ben.

Ben Landen: Hello, good to be here.

Luke Renner: This is, of course, our inaugural interview. So just to get us started, can you tell me a little bit about what you do here at Cyngn?

BEN LANDEN: Yeah, so I've been at Cyngn almost two years now, since the acquisition of deep scale by Tesla. What I do is basically everything. I head up everything that's not finance or engineering. And within a startup, you know, you typically do a little bit of everything, and we definitely get exposure to that here at Cyngn.

Luke Renner: Yeah, absolutely. Okay. I want to talk about the frenzy that happened a few years ago. You know, Elon Musk famously said, we will all be using self-driving cars by 2017. I think a lot of that is cooled off now. And we sort of entered a new phase. So how do you think about all the shifts that have happened?

Ben Landen: So, we really are in this phase of an increase in pragmatism in terms of starting to find ways to generate some returns on the huge investments that have gone into AV while still having these far-reaching goals that are way out on the horizon, that are now I'd say, looked at with more scrutiny than they were in the frenzy when folks were saying that, that cars would be driving themselves all over the place.

What do you think explains why we're not there yet?

Ben Landen: Bottom line? It's, it's just a really complicated problem. I mean, the reason that I have dedicated my career to autonomous driving like I have, is that I genuinely think that it's one of the most technically challenging and, therefore, fulfilling and interesting subjects that I'll have the opportunity to work on in my lifetime.

Luke Renner: Why do you think the sector got this all wrong? Did they know they were off-base or was it an honest mistake?

Ben Landen: That's, that's a good question. The part of me that wants to believe that we all intend to do good and that we share our true beliefs, wants to say that it was an honest mistake. And that it's just one of those situations where as you peel back the onion, once you see that something is working, let's say in a certain environment or on a certain vehicle, and you try to replicate that you learn firsthand that it's just much more difficult than you ever thought that it would be, and that's a contributor.

But the cynic in me says that, with the right type of announcements and noise, a lot of people benefited significantly off of the hype that is autonomous vehicles in the 2015 to 2020 timeframe, where a lot of investment, a lot of monetary success was created without much delivery of end product quite yet.

Ben Landen: Yeah. So I know that a lot of companies in this space have really focused on what I know you have described as kind of the robo-taxi version of this problem. Maybe you can tell us a little bit about what the other sub-sectors in this space are? What are the other areas of interest that people are really doing more work on?

Ben Landen: Yeah, so robo-taxi is, is definitely a focus, and especially in the media for a couple of reasons. One, it is, it's the biggest market opportunity in terms of what's estimated to be the total disruption, an opportunity that's available to self-driving vehicles. So the robo-taxis make up on the order of three to $4 trillion, is the typical estimate of what folks think is going to be a $7 trillion disruption across all industries. So that's why the focus is there. And it's obviously one of the most complex problems we’re facing. So going back to work on very challenging problems you have, you have the biggest, the biggest pot of gold at the end of the rainbow there and you also have the most enticing area for engineers to work.

But you have all of these smaller verticals, which even a small vertical in autonomous driving tends to be like $100 billion-plus market opportunity that range all the way down into warehouse robotics into mining into varied applications of similar things that robotaxis do where you narrow it down to just the last mile, just the middle mile. And then you can create all sorts of operational design domain (ODD in the jargon) constraints that help you sort of parcel down that enormous problem space of robotaxis into something that's more manageable, which then correlates to a smaller market. But as far as we can see, any one of these markets is big enough to essentially be its own industry.

Luke Renner: Yeah, absolutely. And, you know, obviously, because I work here, I understand that Cyngn is really focused on those more narrow odd spaces. So I wanted to ask you, as you've been talking to people, you know, in warehouse yards, companies that do mining, what has shifted about the conversation over these last few years?

Ben Landen: What I've seen has been the really major shift is you don't need to go into a first conversation or a meeting and convince people that they want autonomy or that autonomy is coming soon. That is generally agreed upon, which is fantastic, right? Because we're no longer making the case that a disruption is coming and trying to help people see the future. That's largely agreed upon at this point, which just means that we can talk about actual solutions.

COVID has really affected everything and really changed the way people all over the world are, are living and conducting business. How has COVID just changed how Cyngn as a company operates?

COVID was definitely—still is—a double-edged sword. We saw the type of pullback that you would expect when there are fears of recessions when folks can't go to work due to the constraints of the pandemic. And the type of retraction that that creates in the willingness of again, those customers who had come to the conclusion that they want to be working on and developing autonomous solutions with partners and deploying autonomous solutions.

They said, in spite of that, we're going to have to wait and all the way to the flip side of that, though, where you'd have people that say, Hey, we had a realization that our operations can actually continue in spite of this pandemic if we can take humans out of the loop in certain scenarios. And we now have, call it a COVID Solution Task Force, where we need machines that can automate elements of our operation so that they're safer and so that we can keep shipping products so that we can keep accepting shipments from our suppliers. And so we saw both of those. And unfortunately, I would say the more pronounced impact was the negative one.

Luke Renner: You mentioned during our pre-interview that mining has become an area of interest for Cyngn. So my question is, what can you say about the particular challenges mining operations face? That autonomy would be good for?

Ben Landen: Quite a few come to mind. There are a variety of different mining column types of odd or sites, which each have we each have their own implications on the technology from, from open-pit mines to underground mines. I mean, you can just envision how different those are to navigate for an autonomous solution. It has major implications on the perception system on localization. Beyond that, though, you as you dig into what really how these mines operate.

Definitely, for me, a huge insight that I had never thought of coming from the traditional automotive industry was underground mines are obviously very safety-oriented. And there is this constant mining and excavation, further excavation activity that needs to be balanced.

So when you're excavating and potentially blasting underground, basically no humans are allowed on the site because it's very dangerous in these blast scenarios, you literally have to shut down the operation for hours like every human must leave the site only the blasting or excavating may occur, then they tend to wait for a period after that to make sure everything has settled down. And some of these mines are remote so it might actually be a pretty long distance between the mine and wherever everybody would need to go to leave the mine. So these mines that want to operate 24/7 for years and years and years, need to go down for hours at a time to enable this further excavation. And that like, that's massive, right? That's huge.

You have these operations that are 24/7, and everybody's excited about percents' worth of efficiency. And here we have, like massive chunks of, of efficiency that are just being thrown away by nature of how they do things today.

Luke Renner: So, you know, there's no doubt that the rise of autonomous vehicles is going to lead to a shift in how people do work. It could lead to displacement, it could lead to layoffs. And I just wanted to get your thoughts about this transition that you're working on every day to help build?

Ben Landen: The automation that we're targeting, first and foremost, is, frankly, what people are not particularly good at, or don't want to do. So for example, graveyard shifts of very repetitive tasks, not really an attractive job. Now, we can augment that. And maybe now a person can, instead of sitting in a vehicle that is moving cargo around out in the cold in the middle of the night, can sit in an office and watch that vehicle and multiple vehicles on a screen. He can be more comfortable, be more willing to do that job, and develop skills that are probably more transferable to having job mobility and being able to do to switch to other various similar jobs, which would be computer facing user interface, human-machine interfaces, and the types of skills that are more translatable to the type of technology developments that we're seeing come out of the leading companies today.

As opposed to saying, “I can drive this type of machine, I can I can do this type of task”  And, and to me that that's exciting, and is going to create, create upside.

Luke Renner: So when you're in conversations with business leaders and organizations are, you know, autonomous vehicle curious, what are some common misconceptions that you find yourself having to clear up over and over?

BEN LANDEN: Autonomy is not a product that you can just click order and buy. Buy your autonomous vehicle and then plug it in and it works! We're not there yet. It requires a partnership to get value out of autonomous vehicles today.

Luke Renner: Okay, so we're almost out of time. And I just want to finish up by asking you, what are you excited about in 2021? How do you think this space will evolve? What are you looking forward to?

Ben Landen: I do think that we're starting to see some light at the end of the pandemic tunnel, which, I think, obviously, is a positive for all of us. And in the case of business and the potential, that means we do have opportunities that maybe took a step back last year, coming back around and saying, Hey, I'm ready to work on this now. And I and we're going to be seeing not just from us, but from other great technology developers in various autonomous verticals.

We're, we're seeing more and more small scale deployments, you're hearing about them or you're hearing about older more traditional companies starting down the journey to autonomy, whether it's by adopting teleoperation solutions first or whether it's by installing various sensors or, or etc, use that enabled telemetry that they never had before. And I think we're just going to start kind of chipping away at this big scary problem of autonomy, and making people realize that we can make our way to a gradually, either by scale or by how we scope the problem, and that we're basically just going to see a lot more autonomous solutions rolling out. It's going up from here.

Luke Renner: That's fantastic. All right, Ben, thanks so much for the time. It's been fun. 

Ben Landen: Yeah, definitely. Thank you.