The Future Car Podcast | Accelerating Autonomy: Inside the Abu Dhabi Autonomous Racing League(A2RL) | Part 1

Autonomous racing is something we’ve seen before in events like the Indy Autonomous Challenge and Roborace, so what makes the Abu Dhabi Autonomous Racing League (A2RL) different from its predecessors?

Here to tell me all about it is the Executive Director at ASPIRE, Tom McCarthy. ASPIRE is the technology transition arm of the Advanced Technology Research Council, specializing in the creation of future transformative technologies, and the organization behind the A2RL. 

In Part 1 of this Future Car episode, host Ed Bernardon and Tom discuss topics like the creation of A2RL, the backgrounds and experience of the teams competing, and how cultural differences have shaped the way each team solves problems. Tune in to hear more!

Some Questions Asked

• What inspired the creation of the race? (3:38)
• Who do you think your biggest audience is going to be when it comes to age and interest (11:05)
• Will having an autonomous car race make people more comfortable switching that driver assist back on? (13:33)
• What did it take to qualify as a team and be part of this competition? (24:05)

In This Episode You Will Learn 

• What ASPIRE does and what the organization’s goals are (3:44)
• One of the biggest problems Tom has seen over the past few years when it comes to autonomous technology (12:59)
• How A2RL compares to Indy Autonomous Challenge and Roborace(16:46)
• About the nature of the A2RL race (22:11)

Connect with Tom McCarthy

LinkedIn

Connect with Ed Bernardon:

• LinkedIn
• Future Car: Driving a Lifestyle Revolution
• Motorsports is speeding the way to safer urban mobility

[00:00] Ed Bernardon: F1 has “Drive to Survive,” NASCAR has “Full Speed, and even IndyCar has its 100 Days to Indy. If Netflix was going to do a series on the ATRL, what would they call it? 

 

[00:16] Dr. Tom McCarthy: Race to Innovate.

 

[00:17] Ed Bernardon: Who’s the Guenther Steiner in your series? Who’s the big star?

 

[00:21] Dr. Tom McCarthy: We’d like to think that the car is going to be the star. But, obviously, we’re going to look at the personalities of people, and they will emerge. A lot of the teams have very interesting people and personalities behind them. One of the things we’re looking for is to see that as being the star, but there are a number of other things. Interestingly, we have launched on the Fortnite platform, and the avatars that have been developed on that are interesting. Depending on the age profile of the audience, I think you might find that the avatar has become the sustaining feature in the longer term.

 

[00:58] Ed Bernardon: Autonomous racing technology isn’t just for city streets; it’s finally making its way onto the race track. Abu Dhabi’s Yas Marina Circuit, home to Formula One, is now hosting the first A2RL (Abu Dhabi Autonomous Racing League) autonomous car race. The league, established by ASPIRE, the program management arm of Abu Dhabi’s Advanced Technology Research Council, is putting autonomous Super Formula Cars, which are just a step below those run in F1, onto the Yas Marina Circuit. So, why would you even want to create an autonomous racing league? What will the race and the spectator experience be like? What are ASPIRE’s future plans for autonomous racing? Well, today’s guest surely has all the answers. Take a listen to part one of two on the A2RL. 

 

[01:59] Ed Bernardon: Racing has a long history of pushing the boundaries of what cars can do, innovating the technologies that eventually make it into the cars that we all drive. In the past few years, racing technology development has even extended into the world of autonomous driving. Now, when we think of racing, we think of high-speed, daring passes and the human skill that’s required behind the wheel. Can an autonomous car race capture the same thrill? Or is it just about technology development? Has the time finally arrived for true autonomous car wheel-to-wheel racing, something that a few have tried but none have yet fully succeeded? Today, we’re thrilled to have Dr. Tom McCarthy. Tom is the Managing Director of ASPIRE, which is leading a venture called the Abu Dhabi Autonomous Racing League, or A2RL. What makes A2RL different? Is it going to be the first to truly succeed where others have seen limitations in autonomous car racing? Tom has the answers for us, and he’s ready to tell us all about it. Welcome to the Future Car Podcast. You’ve got a race on April 27th. You’re going to be hosting autonomous cars in a wheel-to-wheel competition at Abu Dhabi’s Yas Marina Circuit, $2,250,000 in prize money. Now, before we jump into the details of the competition, maybe you could tell us a little bit about what inspired the creation of the race. What are its goals? What do you hope to achieve with all this?

 

[03:44] Dr. Tom McCarthy: To understand what all of this is, where we are coming from, and what we are. The organization I work for is called ASPIRE. It’s part of the Abu Dhabi Advanced Technology Research Council. Our mission is to bring forward technological advances that will be the drivers of the diversification of the economy of Abu Dhabi. As most people know, the source of original wealth here is resource-based. What we want to see is this country developing in a diversified way that uses those resources to create industries as diverse as agriculture, transportation, and healthcare, where the driving force behind those sectors is the adoption and the development of technology. In that context, ASPIRE has been set up to push that development, to identify the use cases, and bring about the technology solutions to those through fostering research. One of the ways we do it is by Grand Challenges.

 

[04:42] Dr. Tom McCarthy: Now, Grand Challenges have a long and distinguished history. If you go back to the 18th century, I guess the original Grand Challenge was the so-called Longitude Prize, where, at that time, 20% of all shipping that went to sea foundered: people died, and cargo was lost. The reason is people weren’t able to tell the longitude at sea, and the reason they couldn’t do that is there wasn’t a clock that could keep time aboard a ship and be able to refer it back to the time on land. The British government set up a thing called the Longitude Act and set up a prize of £20,000 back in 1720 to start a competition to identify a solution so that you could measure longitude at sea. Now, it took a long time to do that. The first successful intervention took about 15 years. For 50 years, they kept developing it. But that approach to finding solutions to technology challenges is one that we’ve taken on. We said, “Look, there are many big problems that need solutions out there in the world. Those big problems that require technology to solve them.” Our big challenges are both doing that.

 

[05:47] Dr. Tom McCarthy: Where does A2RL come in? Well, the big challenge that we see out there at the moment is the fatalities and the accidents that occur on roads. We look at the average driver’s capability, and we look at the developments that have occurred in the automotive industry over the last century. We’ve got cars that are stronger, faster, and better at doing all sorts of things. But the average capability of the driver hasn’t really changed that much over the last 50 years. When we look at technology and the potential of technology, we believe in autonomous robotics and artificial intelligence; there is the potential to improve the capability of the driving experience of the average driver. However, most people are not yet prepared to accept a co-pilot, to accept that when they’re driving, they will share the driving tasks with an autonomous technology. So, what we set out to do is we set out to mount something that would demonstrate in real time what the potential of the technology was while, at the same time, involving the consumer so that they could see what could really happen on the track. We’re doing two things at one time: we’re bringing forward the demonstration of what the technology can do at the extreme and, at the same time, involving the public to see how it happens. The history of technology is that the potential of the technology oftentimes remains unused as you build the use case and the acceptance of consumers.

 

[07:20] Ed Bernardon: The whole idea of a challenge hundreds of years ago was to create a more accurate clock so you could navigate around the world. Now, this challenge is to navigate autonomously on city streets, ultimately. Maybe it’ll bring that technology. There’s definitely the technology part, but it is racing. When you think about racing, it has the human element in it. Humans are fragile; they’re controlling this complex machine. There’s the danger involved, and there’s the drama of the drivers, “Drive to Survive,” and all the drama that’s behind that. Do you think that you’ll have that thrill, you’ll have that attraction in autonomous car racing, just like you do in traditional racing with human drivers?

 

[08:07] Dr. Tom McCarthy: No. I don’t think that’s what we’re trying to do in replacing that. We’re trying to create something very different. What we see in motorsport is what we saw hundreds and hundreds of years ago in the forum in Rome, with the gladiators out against each other, where their lives were on the line, where you have a human being sitting inches from the ground, going at incredible speeds, using a physical and mental prowess and capacity over an hour of racing. That’s something that we all thrill to see. It’s something, in no way, that we want to replace. But what we’re trying to do in trying to create a spectacle for this is to go in a very different direction. That direction, we believe, really involves going online in the first instance. We want to bring you, the spectator, right into the cockpit. So, one of the things that we set ourselves a challenge for in the last year is how we build that online experience so that the audience that is looking at this is actually participating. So, we’ve built a VR experience.

 

[09:07] Ed Bernardon: Is it more about that online experience than actually sitting in the stands and watching the cars?

 

[09:14] Dr. Tom McCarthy: I think so. We will mount actual races, we will bring people to the event, and we will create a big event. But we don’t see that as the first motivation for doing it. We’re going to do that for the local audience, and we’re going to create fan zones around it with touch experiences that they can have. But we think in the long term, the real appeal of this racing will be through an online experience that will have you having that virtual reality experience. But we’ve also built an app. In this particular app, they can do everything from registering to actually physically being at the thing to looking at the broadcast, but also to be engaging in a game. So, what we’re going to see is, you’re going to see the ability to be able to hop in a ghost car and race against cars in real-time, not just a simulation race or historic race, but a real race that’s happening.

 

[10:01] Ed Bernardon: So, anyone can have their own personal race going during the actual race, in effect.

 

[10:06] Dr. Tom McCarthy: Precisely. We’re doing our first race here. In many ways, this is a proof of concept race; we’re showing that it can be done and what can be done. We can come back and talk about what we will do in terms of the exact race. But if I were to run this down the road and say, “Hey, what would my dream be?” We bring our cars to tracks all over the world, iconic tracks, and stage races, collect data on those races, and then invite the online audience to race with us. So, there you are, racing on iconic tracks; you can get on your app, you can get in the car, and you can race in real-time when a car is racing around one of these tracks so that you’re actually a participant observer now rather than a static observer looking at something happening in front of you. You’re actually doing it in real time as well.

 

[10:55] Ed Bernardon: I think people now like to experience things more than just watch things. What’s the demographic you’re targeting? Is it the under-30 crowd, the 30 to 50, everyone? Where do you think your biggest audience is going to be when it comes to age and interest?

 

[11:09] Dr. Tom McCarthy: With most things, you start, what you’re doing in your first event is you’re actually collecting data on where the interest comes across the demographic. But if you were to ask me to really guess where I see our demographic is coming from is people that are pre-teen right up into their 30s. That’s where we believe the hump of the demographic will be, and that’s where we’re pushing that online experience. One of the things that we are doing is we are doing a STEM race as well. We have races for teenagers, where they’re going to race 1/8-sized cars, cars that would normally be designed for radio control racing, but they will race them autonomously. That will happen on a kart track alongside the main track where we’re doing the thing.

 

[11:46] Ed Bernardon: How big are these cars? Are they a couple of feet long, like go-kart size?

 

[11:51] Dr. Tom McCarthy: These are actually 1/8th the size of a full-size car, so they’re three feet long, I think, at this stage. They’re not an open-wheel racer; they’re a saloon car. They’re cars that would normally be used in radio control. They have radio control features, but we switch on their autonomous features. What we’ve done is we’ve recruited schools, and those schools will have been, since December, receiving instruction and programming, and now they’re on track and practicing. Our long-term vision is that we want this to be a global event for STEM students so that each of the teams that will come to race at any time from all over the world would have a whole set of feeder schools into them, that would be competing at a STEM level that would then produce a pipeline of kids that would go on to seek to go to university to do engineering degrees, to become racers in this. But for those that don’t, they become aware of the technology; they become aware through hands-on learning of what autonomy can do. Then, they turn around and say, “Well, we’ve been doing this since we were pre-teens. Why would it be surprising that we would want to have these in our cars when we come to buy them?” One of the biggest problems that I have seen over the last few years is that, like most technologies, there’s been a technology bubble in autonomy, where people came around and said, “Hey, autonomy is the way of the future. Driverless cars are the way of the future.” There’s been a bit of a bubble on that. What you have seen now is you’re seeing consumers are rejecting this; we see major car manufacturers taking out autonomous parking capabilities because consumers aren’t using it; they don’t want it. We see consumers switching off Lane Assist because it makes them feel nervous when they’re driving.

 

[13:32] Ed Bernardon: How do you think that having an autonomous car race will make people more comfortable switching that driver’s assist back on?

 

[13:41] Dr. Tom McCarthy: I think this is where being a participant in the race is going to be very important. That’s why we’re focusing on younger people. So, if you’re online, if you’re wearing goggles and sitting in the car, going around the track in that VR experience—the one we’ve put together is amazing. You actually have to sit down; you get dizzy if you try to stand up, look at the car, be in the car, and go around the place. But it is an amazing experience. But then, if you are also a participant in the game, you’re hopping in your ghost car; you’re actually pitting yourself of person versus machine. That machine is racing there autonomously, and you’re pitting your skills against that.

 

[14:18] Ed Bernardon: You’re engaging with autonomous technology directly. That’s a great way to get familiar, I would think, and maybe feel a little safer around it, I would guess.

 

[14:28] Dr. Tom McCarthy: Absolutely. We can do all the demonstrations you want. We’re going to stage a demonstration at the event, which is a man versus machine. We’re bringing a Formula One driver in because we’ve also, in prototype, developed the human. We’ve taken a human-driven version of this car. We use that for shakedown initially for the autonomous car. Since we have it, we’re going to bring a Formula One driver along, get him to drive on the track, and get the autonomous car driving as well. 

 

[14:55] Ed Bernardon: Is this something you’ll have at that first race? 

 

[14:57] Dr. Tom McCarthy: Yes, on the main event race, we have two main races: we have Speed Trials, and we have Wheel-to-Wheel. But in between, we’re going to have that prototype demonstration where we do the Man Versus Machine race. We’ve already run this in prototype. There are a couple of surprises in there. There are some unique features of the Yas Marina Circuit that we’re going to explore to demonstrate the capability of autonomy and what we’ve already achieved with that.

 

[15:24] Ed Bernardon: It’s something that might give the autonomous racer a little bit of an advantage, you think, over a human driver?

 

[15:30] Dr. Tom McCarthy: No, actually, it’s one of the big things that would make it very difficult for an autonomous car to drive around. Maybe I can let you in on the secret here. To get into the pits at the Yas Marina Circuit, you have to go down through tunnels. 

 

[15:43] Ed Bernardon: Absolutely. It’s iconic. 

 

[15:46] Dr. Tom McCarthy: Yeah, that knocks out your ability to take in GPS signals. So, that makes it extremely difficult for an autonomous vehicle to navigate that pit tunnel. We’re not going to force the competing teams to do that this year. But we’re going to attempt to see if the autonomous car can negotiate the pit tunnel while racing against the human driver.

 

[16:06] Ed Bernardon: You raise actually a really interesting point because there have been other autonomous car events. For the most part, they’ve been on ovals—certainly, when there’s more than one car on the track, it’s been on ovals—open, where you can rely on GPS and all that. But you’re actually bringing up a great example here, where just to get on and off the track, you’re gonna have to say, “Hey, it’s not all about GPS,” but you have to start to look at all the sensors and utilize them. I’m sure you’ve heard of the Indy Autonomous Challenge, RoboRace. How is what you’re doing different from those series?

 

[16:46] Dr. Tom McCarthy: Well, first, I want to pay a huge tribute to the innovators that we’re building on the shoulders of. What those guys did in RoboRace, what they did in Indy, and then some of our colleagues here have participated in Indy already. I think the big difference for us is that we recognize you have to be in this for the long term. You have to create a whole ecosystem around this thing. And you have to commit because, as we said earlier, this is not going to be a solution that emerges in a year or two. We want to look back 10-20 years from now, where people are taking it as second nature, “Why wouldn’t you have a shared pilot in your car?” So, in order to get there, you have to build something that’s going to be sustained. So, we needed to put together a business model that could actually sustain itself over time. So, that was the first thing we did. So, we have committed not just to having a race this year but to having the same race repeated each year for the next four years with the same prize fund. We will then use that to build on races that occur at tracks all over the world. So, we said that this needs to be global, and it needs to be sustained over the long term.

 

[17:47] Dr. Tom McCarthy: The second thing is we wanted to build that as part of an ecosystem that was not just about cars, but it was about transportation in general and mobility in general. So, we will be announcing details at the end of April of the next platform for racing, which will be drone racing. We’re going to have autonomous drone racing come on as the next race after this. That will be followed by off-road buggy racing. We also intend to have water-based autonomous vehicle racing. Our focus is not just on autonomous mobility on the road but on air, land, sea, and off-road as well. We’re building the whole ecosystem that can deliver across all of these areas. We think that it’s very important to do that in order to make the gains that are required.

 

[18:36] Ed Bernardon: Will these be separate series, or will the same teams be competing with different types of vehicles, sort of a land, air, sea multi-competition?

 

[18:44] Dr. Tom McCarthy: Obviously, it will be open to teams to join, but we expect there to be multiple different teams. For example, when we come to drones, we believe that the use cases for autonomous drones may be in areas of the world that are less developed. We would hope to see teams emerging from, say, the African continent that we’re not seeing here at the moment. We have teams from Europe, North America, the Middle East, and Asia at the moment. In the car race, we want to add teams from South America and India. We don’t believe that we will see a car team emerging from Africa. But when we come to drone racing, we’re going to put a huge emphasis on getting participation from less developed areas of the world because we believe that the use cases surrounding autonomous drones will make the most impact initially in less developed areas. I talked about the use case for the cars, and it’s about making driving on the roads safer. When I look at drones, it’s about delivering medical intervention to remote areas without regular transportation. Therefore, we would hope to see participant teams coming from areas that are very dispersed. Also, because of the price point, say, something like drones being so much lower, we want to alert people who are not associated with a company or associated with a university. We want to get that kid who is in their basement and who has always dreamed about doing something. Maybe being able to enter into a race with very low costs and capital expenditure, maybe they emerge, and that be part of the development process that they can become involved in other elements of the autonomous series.

 

[20:20] Ed Bernardon: If you have a STEM competition and then you bring in younger people who are programming these real race cars, it’s like a ladder—step by step. They know it’s possible; they can be real autonomous racers. When you look at the Indy Autonomous Challenge and RoboRace, is there anything you saw in there where you said, “I saw it there; we’re gonna do it differently”? And is there anything that you saw there and you said, “Ah, they did that the right way. We’re going to do it the same way they did”? Can you give me an example of both of those?

 

[20:50] Dr. Tom McCarthy: I think the RoboRace was the one that served for me. The fact that they tried to remove too many things at one time, by which I mean they went with an EV and an autonomous vehicle at the same time.

 

[21:01] Ed Bernardon: Do you think it was a mistake to go with an EV?

 

[21:03] Dr. Tom McCarthy: I think it was, yeah. One of the things we talked about earlier was the gladiatorial nature of motor racing. 

 

[21:09] Ed Bernardon: The roar of the engine. Is that it?

 

[21:11] Dr. Tom McCarthy: It’s a feeling you get in the pit of your stomach when you hear the bang and the clatter of an engine, and we just didn’t want to pull that one out straight away. I think our learning there was one innovation at a time, one change at a time. That affected quite a number of the things that we’ve done. I think when we came to Indy, one of the things that we really learned from them was the approach to attack and defend. When you want to move from, say, speed trials to multiple racing, you actually have to start with two cars racing in an attack-defend format, and I liked the way they did it. But what we wanted to do was push that onto more generally being on a road track rather than on an oval. So, there’s been cautionary tales and good learnings. As I said, we’re building on the shoulders of innovators, and we really salute and celebrate them and hope we can see them prosper in the future, just as we do.

 

[22:01] Ed Bernardon: Why don’t you tell us a little bit about the nature of the race? Is there qualifying? Are there heat races? How long are the races? Is there going to be wheel-to-wheel?

 

[22:11] Dr. Tom McCarthy: The first thing where we’re really different is the team spent a lot of time here. We’ve had all the teams on-site here since the beginning of March. They started by being introduced to our cars, essentially getting to understand the different components of the cars, and then going out on a vehicle dynamic area to do basics around cones — make sure that they translate their theory into practice in the cars. Then, they were brought out onto the track. Now, they’re all housed in garages on the track. Each team is in the garage on the track. They’ll be here for a total of two months, and they’re practicing. We have eight teams. We have one team from the US, a team from Switzerland, a team from Germany, a team from Hungary, two teams from Italy, we have a team from Singapore, and we have a team that is a joint team between the UAE and China. So, eight teams from around the world. They are all housed. As I said, they’ve all got a garage, and they’ve all got a team cabin outside. The team cabin is full of guys on laptops; the garage is full of cars. We really put a lot of effort into that. One big feature of this is that it’s a one-design race. We’ve chosen the Super Formula car developed by Dallara for Japan Race Promotion, as the fastest, highest-performance car outside Formula One, but also as a car that’s a one-design car. Every team gets exactly the same car with exactly the same setup. That includes the setup of the autonomy perception components on that car. In addition, we provide a full mechanics team. So, the teams themselves can’t interfere in the setup; they will have mechanics assigned to them, and they will work alongside them. But those mechanics set the cars exactly the same. All the components in the cars are identical, so the only difference is the coding.

 

[24:02] Ed Bernardon: How did you pick the teams? What did it take to qualify as a team and be part of this competition?

 

[24:09] Dr. Tom McCarthy: That’s a really good question. In other competitions, what they did is they just issued a general invitation and asked people to submit white papers with their proposals. We knew some of the themes that were out there already, but we wanted to identify new teams that were out there. So, we actually went through a process that was, you might call it, a discussion, get-to-know-you process. We wanted to bring people who shared our vision for what they wanted to achieve, and we also wanted to be sure they had the capability and the capacity to do the programming but also to have that understanding of motorsport that could translate it to the track. So, we identified a long list of candidates through an open call just about 12 months ago. They actually announced this just 12 months ago. 

 

[24:54] Ed Bernardon: How long was the list? 

 

[24:55] Dr. Tom McCarthy: I think we spoke with approximately 2024 different potentials.

 

[25:01] Ed Bernardon: So, one-third made the cut. 

 

[25:03] Dr. Tom McCarthy: One-third made the cut, yeah. Some of them are coming on board for Year Two because we actually have a total of 20 cars that we’ve commissioned for this. So, we deliberately decided that we only wanted teams that we felt could pass the threshold of being able to make a good effort here on the track. One of the things that we really built into this is that there is science behind this. The way science works is all about collaboration and then competing to get the win out. If you’re a scientist, you are competing to get that paper published first, that idea published first. If you’re in A2RL, you’re competing to win that race and to get the prize. But to get out there, you have to collaborate, and you have to share. We wanted to make sure that they were all at a standard that could genuinely contribute to that joint learning. The selection process involved me, really, and months of getting to know these people, going to visit them, seeing what their facilities were, looking at them in the eye, and feeling, “Yep, these are the guys that can do it.” So, it was a long process, but I think it’s an important process to get the right team in place.

 

[26:09] Ed Bernardon: It looks like you’ve got teams from all over the world. You said something that’s really interesting: it’s not just about programming; there are also mechanical aspects to this. Although they don’t have to take care of the mechanical parts, they do need to know about racing. What would you say? Do all the teams have some kind of racing experience? What is the nature of that racing experience that complements what they have to know to program these cars effectively?

 

[26:35] Dr. Tom McCarthy: Well, all of the teams have, in some way or another, experience in some class of racing. Now, as two of the teams come from Italy, both of them come from Motor Valley, I think it’s bred into their DNA.

 

[26:46] Ed Bernardon: Just when they drive on the street, maybe, is enough.

 

[26:48] Dr. Tom McCarthy: I think when they inhale the air. One of the things I’ve learned—and this has been this amazing thing—we’ve had to partner up with the manufacturers and the suppliers. It’s not just a question of us buying components from them; we really had to get them to buy into the concept of what we’re trying to do. Imagine trying to persuade Dallara to let us go at their beautiful Super Formula car, take the halo off, put in a computer, and a whole bunch of computers. But what I learned there was what created that industry of Motor Valley. You see the same thing in Indy; you see the same sort of heritage that’s there. When we see those universities coming from there, we see that heritage in the universities; you go there, visit, you see how closely connected they are, but more dispersed. The same is true in Germany; you see exactly the same thing in Munich. You go to Hungary, and we see exactly the same thing there because the team from Hungary is part of the Hungarian Mobility Development Agency, which is a government entity that joins up with universities in Hungary and with the ZalaZONE testbed facility. So, that sort of heritage is built-in. Similarly, when you look at Constructor University coming from Switzerland, their connection with technology and motorsports and their past heritage with RoboRace. They’re all unique in their own little way, similarly, with COVID-19 from Indiana and their link to the Indy series in the past. They all have that motorsport heritage combined with their scientific nous.

 

[28:15] Ed Bernardon: How does that diverse background? As you said, you people from Indianapolis, you have people that are just down the street from Maranello, in Ferrari, you’ve got Hungary, China, the Emirates, from all over the world. How does that difference in culture and background, whatever you want to call it, how does that change how they approach this engineering problem? Do you see any differences?

 

[28:36] Dr. Tom McCarthy: It’s interesting. We have a daily debrief with all of the teams: “What was your experience at the track today?” All of the things we also look at, of course, is while we have laid out the racing and sporting regulations, we’re also creating a whole new set of sporting regulations here. I have a big motorsport team that I’ve built up; we’ve got our chief engineer, chief mechanic, and race coordinators. We have people who come from a non-autonomous racing background, and we’re sitting in a room here with all our autonomous coders. You’ve got our pure motorsport culture and approach to racing; you’ve got our autonomous coders and their approach. So, that’s an interesting blend of cultures in and of itself. Across the teams, you definitely have a difference in preference for risk. It’s not necessarily something you can say that one culture is different from another; it’s actually the personalities of the coders. I think that’s going to be kind of interesting. One of the things people said to me: “You’re going to use robotics, and you’re going to use AI. So, everybody’s going to be perfect, right?” No, that’s not where technology is, and that’s where technology has never been because personality always plays a part. So, if you think about how they’re gonna code these cars, there are a number of components. There’s the way in which the car just interacts with the ground that it sits on. How it gets feedback information from its performance as it moves along that ground. How does that team teach the car to learn about the track — that localization module? But then the other thing is the strategy and planning module: how do you actually set that car up to plan its attacks in and out of corners? That strategy is as much the personality of the coder being reflected in the coding. Some coders have a greater preference for risk than others. 

 

[30:29] Ed Bernardon: As drivers do. 

 

[30:31] Dr. Tom McCarthy: Exactly. You start with the personality of the coders being programmed in, and then you have the AI learning from that personality. So, each car will evolve over time, but it will never lose that human component that was put into it.

 

[30:48] Ed Bernardon: After all, they are human programmers. So, you do plan then to have wheel-to-wheel racing. How many cars on the track do you think you’ll have in this first race at the same time?

 

[31:01] Dr. Tom McCarthy: So, we have eight teams, and we want four teams to qualify for the final race. On race day, we’re going to have speed trials, and we’re also going to have four-car wheel-to-wheel racing. Now, on prior days, we’re going to do attacker-defender, two cars at a time. We’re going to run that because we’re going to do that in a round-robin series. We’re going to have multiple races on that. We’re going to record all of that and capture that so that can be played as part of the broadcast during the race day as well. We want to give teams plenty of time to do that attack-defend. That’s where you’re going to potentially have some crashes. So, we’ll give us a bit of time to do repairs to the cars for the final day. But it has never been the case before that more than two cars autonomously have raced on a track against each other. So, we believe the race to Basecamp this year is to get four cars racing against each other on the track. That’s the big ask of the teams.

 

[31:52] Ed Bernardon: You raise actually a really good point here. It’s one thing in a competition to have a passing competition where two cars are on the track, and you’ve set up a pass; you try and execute it properly. It’s something else to truly do wheel-to-wheel racing. So, if you have two human drivers going around a track, and the one that’s in second place is going to study the driver in front of them. They’re going to look at them and say, “Oh, they look weak in turn three or turn four. I’m going to go to the inside and pass on turn four, where they’re the weakest.” That’s a level of strategy that builds up overlap. Do you see that type of programming being done because that’s over and above what many autonomous cars, even on the street, have to do right now?

 

[32:38] Dr. Tom McCarthy: Absolutely. That’s why we have them here for two months. They’ve been focusing, during March, on getting their cars performing on the track during the localization module. In the last few weeks, from the 15th of April, we have taken exclusive use of the Yas Marina Circuit from 9 am in the morning to midnight. So, all teams will be there, and they will be required to go into multi-car racing on the track. They will be training and collecting data for two weeks in advance of race day. It’s not that they sit on their computers, and they program, and then they all turn up and race day to see if they can race against each other. They’ve been practicing against each other for weeks in advance, and that practice will enhance the learning and the training. 

 

[33:18] Ed Bernardon: And start to build it into the strategy. 

[33:21] Dr. Tom McCarthy: You start building it into the strategy, yeah.