Ep.50 Driving towards better mobility through autonomous transport

Maria Rampa: Hi, I’m Maria Rampa and welcome to this episode of Engineering Reimagined.

When you were learning to ride a bicycle as a child, were you ever brave enough to try it with no hands? Did you ever dream that one day, that same feeling of freedom could be achieved behind the wheel of a car?

Technological advances in transport have come a long way since I was learning to ride a bike and we’re now living in the era of the Connected Automated Vehicle.

Modern cars include safety features that can control speed, predict upcoming hazards, help us to park and keep us in our lane, and soon they won’t require a driver at all.

But it’s not just vehicles that will need to adjust to the connected future; engineers are working on intricate transport strategies to reshape urban and digital infrastructure to accommodate these new automated vehicles. And while the end goal is very tantalising, it’s a hugely complex exercise. Should we be adapting all existing infrastructure or forcing the makers of Connected Automated Vehicles (or CAVs as they are affectionately known) to create vehicles that can operate in a world that already exists? And who should be responsible for covering the cost of developing these vehicles – industry or government? Plus of course equity and cybersecurity issues abound.

Let’s begin today’s episode with Aurecon's Director of Future Transport, Terry Lee-Williams, speaking with Megan Sharkey, Acting Director, Future Mobility, at Transport for NSW about the challenges and strategies she’s considering as we adapt to the future of automated transport.

+++++

Terry Lee-Williams: Megan, thanks for joining us on Engineering Reimagined. I'm sure most people listening to this will understand what a Connected and Automated Vehicle is. But just for clarity’s sake, how do you define them?

Megan Sharkey: At its simplest, automated vehicles use technology to assist or replace the driver of a vehicle. For example, your vehicle helps to keep you in the centre of the lane, that's lane assist technology, which is autonomous. Connected vehicles use something called Cooperative Intelligent Transport Systems or CITS, or 4 or 5G to communicate with other vehicles, road users or roadside infrastructure. This is the vehicle to everything component, for example, if the vehicle is approaching a traffic jam, thing connectivity could support four-sided driving. So, like seeing things that you can't actually see. It's the exchange of data to inform about upcoming obstacles. So when we talk about CAVs, CAVs can be automated, they could be connected, or they could be both.

Terry Lee-Williams: I always drive with my Waze on, which tells me where the traffic is, helps me to avoid it. And my car is providing me constant feedback on my atrocious driving, keeping me in the lane, stopping me catching up with traffic, stopping me exceeding speed limits. We're getting there by stealth.

Megan Sharkey: We are and most people don't realise we already have levels of automation in our vehicles now. Cruise control was first generation automation. And now we have adaptive cruise control or lane assist or emergency braking. So, we actually have lower levels of automation already in our cars helping us become safer.

Terry Lee-Williams: There are calls always for safer and smarter and more integrated transportation. What do you think is the role of engineers and designers and advisors in the CAV space to achieve this?

Megan Sharkey: A couple of things. It’s integration and adaptability. Engineers, designers and advisors really have a role in trying to drive changes rather than building new. But CAVs have to actually be fully integrated into our existing infrastructure, or some of our changing infrastructures. In order to do that, engineers, designers and advisers have to get better at looking at what we already have, and how that can be used differently, rather than trying to build new stuff.

Terry Lee-Williams: We know we can get CAV faster if we adapt or change our infrastructure to support the CAV. But placemakers have an opinion, that the vehicle should adapt to the place because we're essentially responding to 70 years of shaping our societies around cars. And that's had some unintended and some quite significant consequences. Where do you see that balance sits between adapt everything or just say to the vendors, live within the world that exists, from city centre to metro to region to rural.

Megan Sharkey: I think it'll be both, but primarily adapt in the early years. So, the technology should cope with existing because it's very, very expensive to try and adapt it all immediately in the short term. However, we do need to adapt infrastructure to get the most out of our technology for our people or businesses to use it effectively and efficiently. So, an example of this might be, do we need to update the lane markings for all hundreds of 1000s of kilometres of roads that we have to have perfect edge markings? Well, the cost to do that now would be a lot. However, the cost of updating that in our normal asset and maintenance plan over time, is much more realistic and already built into that asset replacement strategy.

Terry Lee-Williams: Can you take us through what the future transport strategy and the CAV strategy is saying because we see the pundits on television shows talking about the next six months, we're all going to be in fully automated vehicles, we're not going to own them anymore. But we know it's a much more complicated set of issues to solve than that. So, what's the government's perspective on this?

Megan Sharkey: Until there's a national legislative framework, we are really about learning and understanding the impacts on our network, impacts on our customers, impacts on the movement around the city. We're about to have a Dubbo ute regional autonomous vehicle trial. We have another one that's been announced, which is a full-size bus autonomous vehicle trial. There's a lot that's happening here and now. It's just whether or not expectations of, will it be everywhere, anywhere, anyplace, anytime. And that's much more far off. That's probably closer to 40 years away, to have an AV go anyplace, any environment, anywhere. Certain levels of automation are already on the road or being trialled on the road. And if you confine them to certain areas, so an example might be a dedicated corridor, then it'll be much sooner because you can have dedicated transit lanes.

Terry Lee-Williams: It raises a few issues. The first one that pops into my head is the equity issue.

Megan Sharkey: So, my team and I have had a lot of conversations about the equity of CAV and this time horizon in terms of the transition of the urban to regional and rural and what's available. This is where our role as the government needs to enable and deliver a transport system that's equitable. So hypothetically, assuming that CAVs are legal, regulated and capable, it could play out a few ways. In urban areas, cities and towns, CAVs at some point will likely service all areas but then you have to question, will it be affordable? So, if they're commercially owned services, then it becomes about the business model for CAVs. If they're really expensive, then only a small number of users can use it, it becomes like a helicopter ride, out of touch for most. If the price drops to the current taxi or Uber prices, then more can. If it drops further to some of the carpool or car sharing, then even more could afford it. If we get to subscriptions that become like annual public transport cost, then even more could, or if they become actually a government or government-private-public partnership. Now in terms of rural areas, there’s a number of companies that are actually working on lane assist technology without lane markings and some of them are starting to do pretty well, particularly for exactly that rural problem - can they identify non lane marking roads and what that might mean for the rural area?

Terry Lee-Williams: So, what's the fallback if we're relying on all our vehicles to be connected? And there are disruptions to the network or in fact there's a cybersecurity issue? How do we navigate that?

Megan Sharkey: When you're looking at the national legislation, these are definitely conversations that are being had around the cybersecurity requirements. So, we do have to look at that. But not all vehicles do need connectivity. So, some have opted precisely not to go down that route, can we automate without needing to be connected? We've seen over the last few months, there's been a number of increasing incidences of cybersecurity attacks. With that, the more open a system is, the more it can be hacked. With connectivity that will most certainly be the case. Now, that doesn't mean that the whole vehicle can be hacked. So, someone might think of those movies where someone can hack into the car, and then they move. A lot of the technology is isolating some of its connectivity versus its automation. So, if you could hack into that car software, doesn't mean you could actually take over the car, because they're treating them as separate systems within the vehicle itself, as a way to overcome potentially some of those cybersecurity issues.

Terry Lee-Williams: How do you work out those issues of connectivity versus cost versus outcome? So, if people can travel more, do they travel more? What does that mean for bus services? Is the technology still too young to be having those conversations? Or do you have to have those conversations now to shape your response to the technology?

Megan Sharkey: Absolutely, you need to have those conversations now. CAVs have the potential to fundamentally reorder cities, that could be more space for parklets, more space for housing, that could be more space if you reduce private vehicle ownership. So what does that look like? And this is what the future transport vision is about, we have a goal to be sustainable, we have a goal to be carbon neutral, we have a goal for everyone to be able to use whichever transport they need efficiently.

Terry Lee-Williams: Very keen to understand the potential of freight platooning. And whether that's real, and particularly for the use case, between major cities and ports where we have daily routes of up to 900 kilometres. And it's very hard to manage driver fatigue on those links, so the concept everybody wants to explore is, can the driver sleep while it's on a national highway, which is a very well-marked, well-connected piece of infrastructure, and then potentially just take the controls for once they get off the major network to do the local delivery, and thus overcome all of those driver shortage issues we have and long haul times because drivers have to sleep as well as drive. Is that a realistic scenario yet?

Megan Sharkey: It's actually a realistic scenario now, depending on the level of automation, so level three and level four require something called a fallback ready user, someone who can take back over the vehicle. You can imagine automated freight being driven during night, for example, with less traffic, reducing congestion, or on select highway and motorways, and we're seeing this already. So in the US state of Texas, they're launching freight AV trials. The other area where freight is emerging in different ways is micro mobility. So electric freight, but on a smaller scale, all the white van deliveries. So, we know that online shopping and changes brought on from COVID have accelerated deliveries. And there is a potential. You're seeing this now with small delivery robots in UK. This is an area, and why freight or areas of public transport are actually accelerating rapidly in this space, because some of the constraints of the private vehicle can actually be removed through that fallback ready user, as commercial. So we're seeing it now, you'll definitely see it on some corridors, there's a whole lot of trials that are already happening, different Asian countries, are doing platooning trials, and having quite a bit of success. It's just whether or not that success is realised into whatever commercial benefits they feel that that platooning might deliver. From a government point of view, we'd be interested in, how does it help from a safety point of view? Is it helping us to drive down fatalities?

Terry Lee-Williams: One of the biggest areas of contention is the transition from automation, right through to today, where we have a lack of automation and how we keep people responsible in vehicles, which are largely automated but not fully automated. And how does that automation deal with cyclists riding around? There could be a 30-year and potentially a 50-year transition between automated vehicles and non-automated vehicles. Is that something that the policies you've been working on looks to accelerate, so the transition is shorter? Or is it just something that you have to live with, because the social licence to force the issue doesn't exist?

Megan Sharkey: I think it's probably something that you have to live with to some extent. I will say though, there is a massive education issue. There's a lot of people using this technology that don't actually understand what it means. In terms of the partially automated well, I have a very strong view of this, because my kids ride their bikes, they scooter, they walk to school. So, I would query the necessity of speeding in the school zones. So, if CAVs can enforce speed control, well then that seems like it should be automated. Now how you deal with these partially automated vehicles, and not automated and fully automated, the automated vehicles will have to recognise those non-automated vehicles, and they do this now. At the future mobility centre, we actually do cycling testing, we do children, adult dummies, and a whole host of stuff. None of those are smart technologies. But what we look at is, how can the automated vehicle understand what's in front of it and adapt?

Terry Lee-Williams: I went away to Europe, came back, bought a very small Japanese car that has all of these driver aids in it. And it’s the most relaxing thing I've ever driven in my life and the first question that popped into my head was, why doesn't this thing, because it knows what the speed limit is, just not let me exceed the speed limit? Are we on the cusp of more people understanding that they may as well just let the vehicle do the stuff that stops them being a danger to others? Or are you predicting that's quite a big social shift?

Megan Sharkey: So, my PhD research was looking at socio technical transitions, what is the social and technical requirements to create these transport or future transitions? That does require a large social and cultural change, because it's been ingrained over many generations around personal accountability and what that means. Whether or not it, should be automated? Well, as I mentioned, with my kids who ride their bikes, well, you know, I don't want them speeding around schools, that for me is a perfect area where there should be an enforcement. So then there aren't those injuries. Europe is already looking into this. I think in terms of the broader automation, if I were to ask some of my family members, or some of my neighbours, they don't know much about CAVs, they don't know much about its impact, or its limitations, and the technology is moving rapid. We might not think as much because of big promises that were made. But we're making significant strides in a very short period of time. In 2010, 12 years ago, we were barely talking about this. And now, it's really accelerating. I think it needs to be a much larger conversation. At what point, if automation is going to be in every part of our lives, which things are we willing to give up to have the best safety and sustainable outcomes for society? And that's the conversation we all have to have together. Not just one person.

Terry Lee-Williams: Megan Sharkey, thank you so much for such an informative conversation, and educating me as well as our listeners.

Megan Sharkey: Thank you for having me, Terry.

+++++

Maria Rampa: Ireland is one of the many countries around the world where owning and driving a car surpasses the use of public transport. In part two of today’s episode, Terry Lee-Williams speaks with Cathal Masterson, Director, Commercial Operations, at Transport Infrastructure Ireland, about changes to their infrastructure network and how to shift the attitudes and opinions of the public towards Connected Automated Vehicles.

+++++

Terry Lee-Williams: Cathal, thanks for joining us on the podcast today. I would say that you're renowned for your long-term thinking about future networks. Is there anything you'd like to add to that description?

Cathal Masterson: Thanks for having me. I would say my view and experience in the transport and mobility sector has drastically changed over my 30 years. I'm always learning. The big thing for me is just trying to understand how the services that we provide, or will provide in the future, how that can really help our societies and help our environment to really deal with the challenges we have.

Terry Lee-Williams: So, I worked with you in Ireland, and it's a very car dominated country. More than three quarters of the trips in the country are done by car. Do you think Connected and Automated Vehicles can help shift what's been described by Professor Brian Caulfield as ‘forced car ownership’ in Ireland?

Cathal Masterson: Probably not. No, and certainly not on the road, it's not something I think that we can drop into that mix that we have today that will really kind of make that massive difference. In Ireland today, we're really starting to focus on the concept of sustainability, and sustainable mobility and active travel, well all over the country, but particularly in urban areas where it can actually make a difference in terms of mode shift and shifting people out of their cars. The other big investment we're trying to make is in public transport infrastructure, which takes time, but the delivery of transport infrastructure schemes takes I'd say twice as long as it did when I started, for a myriad of reasons, so it's not an easy fix to deliver extra capacity, particularly in urban areas.

Terry Lee-Williams: There's a real public benefit to the introduction of these vehicles on the network. This is a commercial technology. There are literally billions being invested in its development. But is that something that then you as a public sector need to pick up? Or is it something that belongs with the industry that is selling this into the community?

Cathal Masterson: There's two opposing views on that, at either ends of the spectrum, depending on where you're working. We’re a transport agency, in TII but we also work with the road agencies across Europe and a lot of the road agencies, including ourselves, are a bit puzzled as to the ask of the road authority in this regard, and the ask seems to be quite significant to upgrade the infrastructure to be able to interact with the vehicles themselves. We're still trying to assess whether the benefits that will be derived from that will be significant enough to develop a business case around this to equip a whole network with the types of technology required to interface with the vehicles.

Terry Lee-Williams: There's this complexity around what is the ask. What do you think the offer is? What do you think the offer that CAV brings, that is relevant to you as a transport agency?

Cathal Masterson: When the topic came up in Ireland, I'd say maybe five or six years ago, the reason that the government were interested was largely around almost ancillary benefits. The technology industry in Ireland is quite strong. And it was being proposed that, if we get ahead of this in Ireland, we can develop a strength, an expertise, an industry around this. There is a hub in Shannon, where there's a lot of activity going on. And that's leading the way in Ireland, that type of incubator idea. But the benefits that will be ultimately delivered here, we're not really clear on those yet. I think perhaps we need to broaden it out a little bit beyond the passenger car as well, the heavy freight side of things, the port operations side of things, you could see the benefits. Our new metro link that we're trying to push ahead with in Dublin will be automated. So it'll be driverless. Maybe we need to bridge across a little bit to other modes of mobility and see what can work there as well. So, I think the jury's out, but we're certainly willing to go on the journey, try and explore it more. I don't think we have a choice actually on that.

Terry Lee-Williams: There is always the very contested matter of the transition state. What's the social licence to mix robots in with humans on publicly owned and operated networks, because as an administrator as a regulator, ultimately, you can't delegate that responsibility, even with a commercial agreement or a legal agreement? Ultimately, government is the one that either gives permission or withholds permission. How do you see that evolving in Ireland?

Cathal Masterson: And sometimes it's perhaps not a case of government giving permission or withholding permission. Sometimes it's a case of governments being nudged into a certain situation by the market and having to respond, because they don't have regulation. And we see that in lots of different ways in many cities, particularly in transport and mobility. It's changing so fast that cities are going whoa, what's that? Now we need to maybe think about regulating, even something simple like bike sharing, a lot of cities are behind the curve on that. It's interesting on this artificial intelligence piece, or the technology piece that, we know the current system is not perfect. There's still a lot of fatalities on our road network in Ireland. Far, far too many. A lot of those are caused by human error. A couple of weeks ago, the Portuguese Minister for Transport was speaking. And he was the first guy I had heard actually talking about the price of the transition, the cost of the transition. Now he was talking more about the decarbonisation journey. But it really struck me because we're all saying, Yeah, we want to decarbonise transport, we all get that. And we all actually really are trying as hard as we can, but there will be a cost or a price during that transition. It's the same for this. But what is that? We don't know. I think the basic level is it can't get any worse than things are now in terms of the basic metrics in transport on accidents and fatalities. If it does, we've got to be prepared to stop, and say, Hang on the technology is not mature enough yet to continue to the next phase of this transition.

Terry Lee-Williams: There is another transition, which is very relevant just for Ireland in Europe, we might be cheeky and include the United Kingdom in Europe because it's there geographically, even if not in spirit. So, you're a small country, right-hand drive market. In the fight for decarbonisation, you're being punished because you’re right-hand drive. You're therefore at the end of a very stretched supply chain. Imagine if all vehicles are robots, the vehicles don't care what side of the road they drive on. That is just a software adaptation that could be built in. Is that something that would be encouraging for the government to consider?

Cathal Masterson: One of the challenges of being a smaller economy is really having the funding required to kind of develop something like this and roll it out. There's perhaps benefits that we haven't really thought about enough right now. We go back to the decarbonisation question and the behaviour of drivers as well, and how people physically engage in driving. It's not very efficient. We know it's not safe. Perhaps there's decarbonisation benefits and congestion management benefits for the urban, peri urban areas, that by syncing up the vehicles and communicating together, maybe they can be rerouted or rerouted in advance, before they hit the congestion, rather than hit the congestion and slow down. So the technology we're using now is trying to smooth the flow on the motorway with variable speeds and smart motorway type concepts. But maybe they shouldn't be there in the first place. And if you're going to get in your vehicle in the morning, and your vehicle says no, no you're not getting me today, because it's far too expensive in terms of congestion, or time lost. Go get the bus, please. Now, that'd be a smart vehicle, wouldn't it?

Terry Lee-Williams: I'd be very interested to see the public reaction to that in Ireland. Because it would be true to say that individualism and the personal agency is a huge topic. So, autonomy has been evolving since the beginning of the automobile. People had particular trouble coordinating their bodies to be able to change gears in early cars. And if you've ever tried to drive a car from the early 1900s, you'll understand that there are four pedals and two levers to operate. And you've still got to steer. So straightaway, we tried to reduce the complexity for the human by automating the robot in some sense. This technology is developing continuously. So the vehicle that you drive today has automatic braking regulation on it so that it doesn't skid, and any new vehicle that you buy in Europe now is going to have automatic speed control capability and distance capability on the radar, that's in the front grille, so that it won't run into the car in front of it. What parts of those automations, that will occur before we get full automation, are things that you think the public would accept as being a requirement in new vehicles to be activated? So, for example, could you imagine a government saying, the vehicle will drive to the speed limit, and you are not allowed to override that?

Cathal Masterson: I think perhaps we could evolve to that situation for certain stretches of the network, because in a way, you're talking about the vehicle becoming smarter and you're right, I learned to drive my granddad's tractor, which actually was quite easy in that they had made it simple and you had a little accelerator on the side of the wheel and off you went, but in terms of making the vehicle smarter, we're also trying to make the infrastructure currently smarter and there will be certain parts of the infrastructure that you're driving on today that actually are almost doing that to say, no, no, if you breach the speed limit here, you will be penalised. That's happening now. We still get a lot of speeding. We have single carriageway networking in Ireland, and the behaviour on those roads can be shocking at times. You talked about personal agency, that is very true, but also I suppose not really focused on their own personal safety, or the safety of other road users, it's just something in a cohort of our population, that they feel that their own behaviour is more important and the risks they take are their own risks, but actually, the truth is the risks they're taking can impact on other people. And quite often do pretty horrendously at times. So, I think we could get there, but I would say that's more of a piecemeal approach, if we try to run a referendum in Ireland to say, your vehicles will limit everybody to the speed limit, that would certainly not pass.

Terry Lee-Williams: Right now, you're looking at introducing e-scooters into your traffic mix, which has proved hugely challenging globally. And one of the control mechanisms that proved to be very effective is geofencing on those scooters, then a lot of scooters in built up areas are restricted to 15 kilometres per hour. When you think about cities and urban populations are still the majority of the Irish population, 20% of the population is in the capital. Is it possible that you could put a universal speed limit in the city and that the cars obey the speed limit there, because there's no real advantage to going fast because you're just running into traffic, right? Do you think there's more social acceptability in that sense?

Cathal Masterson: Say with the scooters, you've a good opportunity with a new mode like that, as you're bringing it into the city to set those boundaries around that. And people will accept that without really overthinking it, you know. I was trying on the scooters in Lisbon a couple of weeks ago. And because I was a new user, it had a speed limit on me, and I was perfectly fine with that. So, I do think there's potential on that. And I think as the new vehicles roll out, it's the trigger point, which will generate opposition is a mandatory operating rule on this. So, if your new vehicles have a setting to say, Would you like me to make sure you obey the speed limit at all times, so you don't get fined? Click, I’ve got to drive at that, and the car can then do the thinking. And in a city environment, you would probably get used to that pretty quickly, because as you said, if you are exceeding the speed limit, there's no real particular time benefit to that because you get caught at the next junction anyway.

Terry Lee-Williams: So, we know the world's most popularised automated vehicle is the Tesla, although it's not actually certified to drive automatically in any place in the world. But it does. And there have been some fatalities. A jeep was hacked and driven off the road by people trying to demonstrate the lack of cybersecurity in vehicle systems now. Where's your confidence level with the progression of the technology as you contemplate it in your networks?

Cathal Masterson: I think we're quite away from convincing the regular professionals in transport and mobility that this is actually a serious trend. Now, that might sound a bit odd, because I know the people working in the CAV and connected, autonomous driving industry are very, very serious about what they do and doing very good work. But generally speaking, I think people are sceptical of that. There's all these myths around it. The public haven't really got to grips with the topic of connected vehicles really, in Ireland anyway. But it's also become more difficult to do certain things, around that idea of data protection, we now have to do a lot of work to get something launched, be it a new product or service. Cybersecurity is another one, I remember joking to somebody, I think around 2013 that we should all become cybersecurity experts, because that's where the money will be. The risks are certainly real. And there are hurdles to my mind, so there’s a lot more for the CAV industry to have to do today than they would have had to do 15 or 20 years ago if they wanted to bring stuff to market.

Terry Lee-Williams: If you could use this podcast as a platform to speak to the CAV industry, the people who are spending billions of dollars developing this technology, hoping that a lovely person such as yourself, sees a benefit in it and would promote it to a government as a worthwhile investment. What’s your burning question?

Cathal Masterson: There's some big challenges for the transport sector, we're still almost completely dependent on fossil fuels, we know we need to go on that pathway to decarbonise and not just road transport, aviation, maritime, to a much lesser extent rail, because they've got the technology figured out. My challenge is to think about that journey and that transition and see how the industry can help us with that. Also, more generally speaking, the need to reduce the materials we're using and principles of circular economy, which is a big buzzword, but actually to think about that, as we go forward on the production, manufacturing, kind of introduction of new vehicles into the mix. Please don't just view these as status symbols that can be sold to individuals and marketed as lifestyles. We've had enough of that, that's what we've had for the past 30 or 40 years from the car industry. There's a time for a change. And that's what I would ask for. And if you can do that, and demonstrate that, you will find that the agencies and the governments will collaborate much more effectively. People are suspicious of the car industry, to be blunt, and I think for good reason. And we need to just deal with that openly and honestly, rather than trying to shy away from it, and particularly the CAV work that's going on now is firmly embedded within the car industry sector. And we'll continue to have that conversation.

Terry Lee-Williams: Cathal Masterson, thank you very much for your time, your very deep insights. And we look forward to witnessing your conversations with the CAV sector over time.

Cathal Masterson: Thank you very much, Terry

+++++

Maria Rampa: We hope you enjoyed this episode of Engineering Reimagined.

The automated vehicle revolution has come a long way in a very short space of time. And while there is still a long way to go, it’s exciting to consider how our transport networks and personal car use will change over the coming years. I for one can’t wait for the day where the stress of finding a park is no longer needed!

If you enjoyed this episode, hit subscribe on Apple, Google Podcasts or Spotify, leave us a review, and don’t forget to follow Aurecon on your favourite social media platform to stay up-to-date and join the conversation. Thanks for listening.