Biao Ma, is the VP of Engineering and the Head of Autonomy at Cyngn. In this conversation, he’s going to give us an engineer’s insider look into the...
Transitioning Mines to Full Autonomy
In this conversation, we discuss the intersection of autonomous vehicles and mining, diving into its unique challenges and possibilities and exploring how AV technology development in this space is a far different animal.
LUKE RENNER: This is Advanced Autonomy. I'm Luke Renner. My guest today is a colleague of mine here at Cyngn, Tim Varecka. Tim is the director of engineering and has been developing technology solutions for his entire career. Before joining Cyngn he was in charge of new product development for Hexagon, an organization focused on digital transformation in the mining industry. Now at Cyngn, he takes his experience developing products even further by helping mining operations integrate advanced autonomous systems into the vehicles they already own.
In this conversation, we'll be talking about the intersection of autonomous vehicles and mining, diving into its unique challenges and possibilities, and exploring how AV technology development in this space is a far different animal.
Hi, Tim. Welcome.
TIM VARECKA: Hi, Luke. Thanks for having me.
LUKE RENNER: Yeah. So you are here to talk about mining. Um, and I thought we could kind of get started with you telling us a little bit about how long you've been in the mining space and what brought you to Cyngn.
TIM VARECKA: Yeah, I've been in mining since 2017, so fairly new to the industry, but like a lot of other industries they're going through digital transformation. And that's what brought me to the mining industry. I've done that in other business spaces, and it's an exciting time to get into mining. And at Cyngn, a lot of fortuitous circumstances happened and where I was at before was into autonomous vehicles and I've worked in transportation in the past where we were doing digital transformation, and it just was a logical, exciting place to be.
LUKE RENNER: Yeah, that's excellent. Okay. So let's get a rundown of the mining space. Where is the mining industry at with its autonomy journey?
TIM VARECKA: Yeah, it's just starting. I mean, we've been spending a lot of time working on autonomous. It's a logical extension of safety equipment and sensors that they've been adding to make vehicles safer, obviously, for drivers and personnel.
You know, if you've seen on TV or if you're familiar with those giant haul trucks that are bigger than a house, the line of sight on those is not the greatest, right? So seeing you know, what's around you is challenging. And so, there have been companies that have been making sensors to tell the vehicle what's near it while other vehicles are nearby.
And it's just a logical extension of that, taking into account vehicle intervention. So now, the vehicle will break if it detects a hazardous condition. Also, part of the digital transformation of mining and fleet management systems and production planning. It’s all a logical extension of that. Instead of just telling a human driver where to go, now we can just tell the vehicle where to go. But it’s been an incremental journey.
A lot of people in the autonomous industry thought things would be a lot farther along than they are. And you know, the same is true of autonomous vehicles in mining as well.
LUKE RENNER: What are some interesting projects that you can tell me are — that may already be underway in some of these mining operations.
TIM VARECKA: So, haul trucks that I mentioned before. Just taking the haul truck from the loader and depositing it to wherever it needs to dump that, or going from point a to point B and that’s just an easier problem to solve.
And now, we’re extending that to other vehicles as well. Dozers and everything you can imagine are getting involved now.
LUKE RENNER: So for those of us who aren’t entirely sure what a Hauler is, is that just like a dump truck? Is that how we can think of it?
TIM VARECKA: We can think of it as a giant dump truck that’s a big as a house. Yup.
LUKE RENNER: Okay.
TIM VARECKA: Two stories. Yep.
LUKE RENNER: It's pretty amazing to imagine that that thing could be driving itself.
TIM VARECKA: Yeah. Yeah. It's actually — you know, like at night or in different harsher conditions — it's actually safer.
LUKE RENNER: So I'd like to talk about mining as an ODD, as a unique environment where we may be bringing autonomy. So what are some of the unique challenges that this environment is facing? What can you tell us about that?
TIM VARECKA: Mines are typically in remote areas of the world where you don’t have good LTE or 4G coverage. They're also large spaces. So, you know, if they have a wifi network there, they're spread out and there are lots of dead areas in them. And then, in deeper parts of the mine, you don’t have to GNSS or GPS.
And, you know, a lot of the autonomous algorithms depend on GPS. That's how they're determining localization. And if you don't have that, how do you deploy autonomy? Right? So, you have to do it with mapping and point clouds. But, yeah, those are some of the challenges.
LUKE RENNER: I want to shift now and talk about where things are headed. So if you were to predict how the mining environment might look in five years, what could she say?
TIM VARECKA: Yeah, I think things would just continue extending the safety and sleep management that we've been doing in the past. I also think in five years, getting vehicles from point A to point B will probably be solved.
But you may be able to get a bulldozer from point A to B but the next challenge will be moving the components of the dozer. For example, dipping the blade down and making sure you’re efficiently moving the dirt. That’s going to be the next challenge — to create an autonomous vehicle that can do everything, and that’s where machine learning comes in. We can get a bunch of data of different drivers and then reduce the variation that would have with human drivers.
For example, you could have a new person that's new to driving a dozer — or if he’d give one to you or me. And in a day, you could train somebody to manipulate the controls and they could get dangerous. But they’re not going to be very efficient. And then, of course, there are the superstar operators who can move the earth very efficiently.
And that’s what we’re trying to do here: train the algorithms on how to be the superstar. So now, with AI, you have a full fleet of superstars, instead of a bell curve of some good drivers and some not so good drivers.
But before we get to that point, I think teleoperations will take over as well. So, then you’d have a dozer operator controlling several bulldozers, remotely, which would protect them from the potentially harsh environments and the vibration from vehicles.
LUKE RENNER: So I imagine bringing full autonomy to mining vehicles must be fairly difficult compared to regular vehicles. There are balance issues to manage. There’s a lot of safety issues. It seems like a bit more complex problem.
TIM VARECKA: It is, but, you know, again with machine learning and other AI techniques, I think it's a solvable problem. You know, in the freight industry, they’re already looking to go beyond just driving the semi-truck from Point A to Point B. But then, okay, once the truck gets to the dock, there’re robots or autonomous vehicles to unload the truck, right? So, how do we do that? How do we make that efficient?
So, you know, maybe the problem is a little bit different but it’s the same extension of AI and autonomous vehicles of all kinds can do that.
LUKE RENNER: Tell us a little bit about how autonomy is going to affect the planning and utilization of a mining environment.
TIM VARECKA: Yeah, so, at the end of the day, some reconciliation occurs. The mine engineers plan what’s going to happen for the day but now there’s variation because, you know, we’re all humans doing things. But with autonomous vehicles, that variation becomes a lot less. So now, you know exactly what’s going to happen and you’re going to get the efficiencies of mining. And this is going to save money. You’re going to save fuel. You’re going to get that oar a lot quicker than you would with just humans operating.
LUKE RENNER: You are our resident mining expert here at Cyngn. You’re having conversations with mining operations about potentially bringing autonomy to their work. So I'd like to ask you, what are some misconceptions that you're encountering as you're having these calls?
TIM VARECKA: Well, one is, you know, it's like the whole autonomous vehicle space. I think people assume things are a lot farther than they really are. So folks just need to have realistic expectations of what the limits are.
You're not going to have a dozer that’s going to completely drive itself this way or that. We’re just not there yet. But you are going to have safe and efficient vehicles that are moving from A to B and doing so in an efficient way.
The other thing is that people fear they may lose their jobs. If autonomous vehicles come in there, the person driving the vehicle may be doing something different. With all digital transformation, there’s going to be a shift in what people are currently doing. If you look at mines from a hundred years ago or older. I mean, there were people whose jobs it was to feed the animals, right? Or go around and refill people’s lamps with oil. Those jobs don’t exist anymore because we have electricity and cars.
LUKE RENNER: So I want to widen the scope a little bit. I know one of the things we've talked about a lot is the possibility of mining in space. For people who haven't been paying attention, what can you tell us is coming? And what, what have you heard people are working on?
TIM VARECKA: Yeah. I mean, that's an exciting development that a lot of people are looking into. For instance, The Colorado School of Mines just started a program on space mining or lunar mining.
One of the exciting things that have happened in the last couple of years is that they, I think there's a lot of water on the moon. With the water, you can create rocket fuel in space.
So you don't have to launch all the rocket fuel to get to the moon and back. It just needs to have enough fuel to get you into orbit. And then, you can refuel from fuel from the water to the moon.
So that's gonna reduce the total cost of mining. Right now, it’s cost-prohibitive but there are a lot of great minerals potentially on the moon, you know, rare metals to aluminum and iron — and other things that could potentially be on and beyond the moon.
There are the asteroids as well. So, I guess there’s an asteroid with so much gold, it would crash the economy.
LUKE RENNER: Yeah. I heard it had like 17 zeros after it or something. So much gold. Yeah. And so when, when do you think we'll be mining on the moon?
TIM VARECKA: I think it will happen in our lifetime. You know, I think especially if the model of water that we think is on the moon turns out to be true.
I think that'll happen. I think there’s also — which isn’t as sexy as space mining — but, you know, still under development is underwater mining. Autonomy is really going to be a big addition to space mining but it’s going to be necessary for underwater mining.
LUKE RENNER: Yeah. Excellent. All right, Tim. Well, I really appreciate the time. It's been interesting. Thanks a lot.
TIM VARECKA: Thank you.