Welcome to tech Stuff, a production from I Heart Radio. Hey there, and welcome to tech Stuff. I'm your host Jonathan Strickland. I'm an executive producer with I Heart Radio. And how the tech are you? You know, we've been talking about driverless cars for quite a few years now, Even on this show and in previous episodes of tech Stuff, I've talked about the various levels of autonomy, which range from zero, which means the vehicle in question really has no driver assist features at all, and the human behind the wheel is responsible for everything, you know, barring some sort of mechanical issue at any rate, and then you go all the way up to level five autonomy, which is a car that's so smart you could enroll it in Harvard. No. But seriously, a level five autonomous vehicle would have no need for human accessible controls. It would just transport people and our cargo wherever, under any driving conditions, at any time of day or night. Where currently hovering around levels two and three at the moment, mostly there are pushes to get to level four, but as it has turned out, the technology necessary to create truly autonomous vehicles that can operate anywhere at any time is harder than some of us myself included, initially thought and most of our discussions about driverless cars treat the vehicles as you know, self contained computing systems that just happened to be on the go. So by that, I mean the car mostly is relying on its own sensors to understand what is going on around it. The car's decision making systems are fully contained within the vehicle itself. And you can think of a highway filled with those types of autonomous vehicles being kind of similar to a room filled with computers that are not connected to each other, so they are not network. They're all individ jewel computers. Each machine is sophisticated and it's capable, but it's also independent of all the others. Other visions of a driverless future involved cars that can communicate with one another directly or even potentially with roads and infrastructure. These cars would depend not just on their own sensors, but those are the vehicles around it. They would remain in constant communication to provide the smoothest, safest, and most efficient ride from point A to point B, so they would be kind of part of a mesh network. And to get to that future, we would need to agree upon a standardized protocol for these cars to follow. Otherwise you could have a world where Toyotas wouldn't talk to Fords, they would only talk to other Toyota's, and that kind of thing. That would be a nightmare, that would not be helpful. Anyway, That's not the direction that most companies are going in right now, though I suspect some companies that are more focusing on creating autonomous transportation vehicles for stuff like cargo, or at least looking at this from a fleetwide perspective that could serve a similar purpose, but specifically for this particular you know, business model. But today I wanted to talk about a slightly different approach. Uh And I've talked about this a little bit in the past when I've talked about autonomous cars. So this one has had several different incarnations over the years, and it's it's a cool idea on the surface, but it's probably the least practical for driver lest vehicles or autonomous vehicles insofar as the amount of work we would need to do on an infrastructure level and the amount of consensus we would have to have across all parties in the space but this is the concept of the automated highway system. With this kind of system, you would have embedded networks that are along or under or above highways, and these systems would provide guidance to the vehicles on the road, perhaps even to the extent of controlling all the vehicles on the road. So and that kind of a system, the road would be the one doing the driving. In other words, it would send commands to all the vehicles traveling on the lanes, which would follow those commands. And it would be kind of like an automated air traffic control system, with the automated highway guiding vehicles on and off the highway seamlessly in a way to avoid traffic congestion and to avoid accidents. And it's a neat idea. It's just I won't I won't say impossible, but about as close to impossible as you can get without it being impossible to do. But that doesn't mean that folks haven't tried. The idea itself is actually coming up on being a century old. I mean, okay, we still have seventeen years to go, and yees, seventeen years for for many of you will seem like a really long time. But the older I get, the less the less long. Seventeen years sounds to me, but we are getting close to a century of following alone this idea. So back in nine GM sponsored a pavilion at the World's Fair in New York. I've talked about this in past Tech Stuff episodes too, but it's one of my favorite historical examples of futuristic thinking. So GM had a pavilion that was called the Highways and Horizons Pavilion, and the centerpiece of this pavilion was an exhibit called Futurama. Bum bum bum bum bum. That's all I can hum of that before I get a d m C A strike. Anyway, Futurama, in this case was this massive model of what a landscape of the future could look like, and it included, you know, city scapes and rural areas, included farms and hydro electric facilities like it was expansive. It was huge and incredible and included some moving components to A theatrical designer named Norman Bell get Is masterminded this exhibit and get his he was a real forward thinker. You can actually watch a full film that details the Futurama model on YouTube. It's about five minutes long. And you can even tell that this film inspired lots of later attractions, particularly the ones that opened with the debut of EPCOT way back when at Walt Disney World. If you've ever been on any of those old EPCOT rides, or you watched videos of those old epp cut rides, they owe a lot to this particular film of the Futurama exhibit from way back in. Anyway, the film covers a lot of different bits, but the part we're interested in is this idea of automated highways. Now, keep in mind this, this blue sky automated highway concept came at a time when highways themselves, we're really new. In the United States, the US was in the process of laying out highways in nineteen thirty nine to connect major cities together, even as get as his exhibit suggested a revolutionary approach. Also, I want to add a note here. While the view of the future was very forward thinking, if you watch this film, you're like, wow, this was, you know, really imaginative, the social concepts definitely were not forward thinking. The film. Um that's about the Futurama exhibit. It pretty much outright says that men are behind all the progress. It just uses men as the default, and clearly that ignores the fact that countless others besides men made significant contributions up to ninety nine and and since and uh yeah, I just want to call that out because you know we should. So it's It's certainly not a perfect film by any stretch of the imagination. However, the exhibit gives us a view of the far off future of nineteen sixty. Keep in mind that was twenty years away when this exhibit was being shown, and the motorways in this exhibit allow cars to travel at designated speeds. The accompanying film says that those speeds are fifty miles per hour and they will maintain proper distance from other cars through quote unquote radio control. So presumably the motorway itself is the transmitter and the cars are receivers. Alternatively, the cars could be both transmitters and receivers and be communicating with each other. It's not entirely clear. Now, you could argue that the nine exhibit was marketing and it was just overly optimistic projections about the future for the purposes of marketing for GM, And I think you would have a pretty solid argument if you said that, I'm not or anyone involved really believed that this particular vision of the future was what was going to happen by nineteen sixty. But then it's always hard to protect the future, even if it's just a short ways out. Also, the US of nineteen nine was entering into very uncertain territory. It was recovering from the Great Depression, which was finally coming to an end after a decade of economic turmoil, and the US had not yet entered into World War Two, And obviously the experience of World War Two would affect the American point of view significantly. It would really boost xenophobia for one thing. But let's skip on ahead while get us his vision of nineteen sixty did not manifest in reality. We did not get these automated highways by nineteen sixty. There were folks who were working on the concept of automated highways, at least researching ways where perhaps it could be possible, and that was mostly focusing on mechanical systems and radio control systems. And over time, as computers began to evolve and become more commonplace, there was a transition toward looking at computer controlled systems, the hope that maybe a computer would be sophisticated enough to handle the complex operations that would be required to have a truly automated highway system. But computers themselves were really large and limited early on. As they evolved, there was also another big shift, and I'll talk about that after we come back from this quick break. Okay, before the break, I was talking about how as computer systems were starting to get more sophisticated and smaller, mentorization being a big part of this, we started to see a shift, and that shift would end up being a really significant one as far as the concept and plausibility of automated highway systems are concerned. And that was that car companies were starting to design and integrate driver's safety features in vehicles, and these features bordered or sometimes overlapped the autonomous car concept. So, for example, cruise control. The actual idea of cruise control is almost as old as the car itself. It's essentially a way of having a car maintain a set speed. In some cases, it was a set speed that the driver couldn't choose, like it was either full throttle or not. But then later you had more sophisticated systems that let you set cruise control at a speed of your choosing, and your vehicle would just maintain that speed. Now, that was just one tiny step toward creating systems that would augment a driver's ability to navigate the roads safely and more over, service kind of building block toward autonomous cars. Obviously, a lot more systems would be needed in order to even remotely consider a car to be autonomous, but these were the steps that we're leading in that direction, and we would see that trend continue over time. Cars would get more systems that would make them safer and be able to take over certain operations. You know, cars would get integrated cameras, which would help when you're backing out of the space for example. They'd get collision sensors, they get lane detection sensors, and much more so. The cars were evolving into smart machines that could operate on essentially a dumb network. Right. You can think of the roads and highway systems as a network, and the network itself is fairly robust. I mean, if you drive in Atlanta, you might question that with all the potholes, but you get what I'm saying. It's a robust system, but it's dumb. It does not have any real intelligent components itself. And the cars became the smart art components, smart devices that could navigate dumb systems. Meanwhile, in research organizations such as the Partners for Advanced Transit and Highways or PATH, which originated out of the University of California, you had researchers who were looking into various approaches for autonomous driving, which included auto highway technologies. So not just you know, how can we make an autonomous car, but would it be possible to make a highway system that itself was intelligent and remove the requirement of having intelligent vehicles instead. Now, this was in the nineteen eighties, and it started to take up some momentum because you know, in the nineteen eighties we certainly were miles, no pun intended miles away from the technologies that would allow a truly autonomous vehicle. So by n things got shifted into a higher gear. To use another pun, I'm just gonna keep doing those I guess I don't to, I honestly don't. But the US government passed the Intermodal Surface Transportation Efficiency Act or is T and that's often a question we ask here in Atlanta, Georgia, when someone hands us a drink, we point to and say, is t anyway? This Act laid the groundwork for early testing of automated vehicles and roadways. So the U. S Department of Transportation would then create the National Automated Highway System Research Program or in a h s RP. This program's main purpose was to establish specifications for an automated highway system. Again, because with so many different players in the space, like all the different law agencies across you know, federal and state and regional systems. You have all the different car companies that are involved, Like there are a lot of different players who are uh deeply concerned with things like highway technologies. You have to establish standards so that there is interoperability between the system and all the vehicles. Right. Again, if you have a system that only communicates with one type of vehicle, then it's not really a good system, right for anything other than that vehicle, it's a terrible system, or it might as well not exist. In the Department of Transportation created the National Automated Highway System Consortium, So this was kind of a think tank that was made up of more around like a hundred different members across nine major categories. Those categories included things like research organizations, government agencies again at different levels like federal, state, and local. It included academic organizations like the University of California, and of course representatives from the transportation industry. So the hope was that by including all of these groups in the conversation that everyone would be able to arrive at a consensus that would make the most viable, efficient, safe, and practical solution for automated highways. Um also hopefully most affordable, because you would ultimately be turning to the US taxpayer to foot the bill for these kind of things. And I think that was a wise move in some ways, because it's always good to get stakeholders involved. One just to make sure that the solutions you are designing actually solve real world problems. There are a lot of times where we talk about an approach or a technology and we refer to it as a solution that's looking for a problem. That's not ideal. Right, if you are if you have identified what the problem is, then you can work to solve it, but you have to make sure that your solution actually does address the problem them. That's one reason why you want to have lots of different stakeholders there. You also want to have buy in across the entire industry or else Again, interoperability becomes a problem. However, when you include all these different voices, there's always the danger that projects descend into a designed by committee approach where it feels like all the really progressive concepts have been sanded down until they aren't really interesting anymore, or worse, you fail to find any common ground and nothing gets done. So between and researchers poured some serious effort into conceptualizing and automated highway system. So the benefits of such a system, if it worked, were obvious on the surface. Like, everyone could agree this is a good idea insofar as what the goal is because you would reduce traffic congestion, which would improve traveled times. It would also reduce fuel consumptions that was an added benefit, and it would improve the quality of life from motorists across the nation. They'd be spending less time in transit and more time doing whatever it was they wanted to do. So like, those are pretty obvious benefits of such a system. The reduction of accidents would also be incredibly important. It could save tens of thousands of lives every year. It is impossible to overstate how important that is. The impact of deaths due to traffic accidents is gargantuane. There's the personal impact on all the friends and family of the people who are lost, there's the financial impact. There's the societal impact. You know, we no longer have those people contributing to society. So yeah, that's one of those things where not only is it great just the idea of saving these lives, we all stand to benefit from that, right, we all stand to benefit from the fact that these people are still around and contributing to society. Not to mention, you save a lot of money and property in the process of reducing accidents. So the goal was clear. How to get to the goal was the matter of debate. That debate largely ended in why, well, I mean, ultimately the easy answer is the money fell out. So in part the issue was that all those stakeholders were in fact having trouble agreeing upon the right approach and the components that would have to go into an automated highway as well as into the vehicles that would be compatible with such highways. You know, the whole goal was to build consensus, but over the years that just became impossible to do. You had too many different voices with different agendas, and there was no common ground to really solidify a path forward. So nothing was happening, and since the consortium was making very little headway, it was pretty hard to justify continuing to fund it. So in nineties seven, the US government passed another act. This one was the T twenty one Act t E A one, and that pulled financial support from the Automated Highway Research programs to put it in other areas. The d O T essentially made the decision to focus more on safety oriented technologies like the aforementioned lane detection technologies, and essentially what the government was leaning toward was where the automotive industry was already headed, which would mean you would have these independent autonomous vehicles or increasingly autonomous vehicles that would navigate within an environment rather than an intelligent environment that would guide vehicles to their respective destinations. Now, some say the technologies we're seeing evolving vehicles in general and autonomous vehicles in particular, could be a stepping stone toward automated highways in the future. I'll talk more about that after we take this quick break. So could autonomous vehicles be a step toward intelligent highways? You know, maybe we will see autonomous vehicle technologies converge with intelligent highways and that the true future of motoring will be a combination of both intelligent infrastructure and intelligent vehicles working together. You could imagine, you know, highways that are able to meter traffic onto them with autonomous cars so that there are no bottlenecks on entrances and exits to the highway for example. Right, that's a possibility. However, for all of that to work out, we again have to create an agree upon standards that work across all parties, all the intelligent highway systems whenever we would design and implement those, and all the different vehicles, or else you have cars that won't know how to talk to highways, or cars that don't know how to talk to each other, and so on, and you just end up again with individual smart vehicles and the same effect of an overall dumb system. Also, we have to consider some of the massive challenges that come along with building and implementing an automated highway system. So, first of all, doing so would require massive investment in infrastructure. That is always a tough thing to get going in the United States for multiple reasons. Like we did see a national infrastructure bill this year, but it did not go easily, and it turns out like folks are not super jazzed about having to pay more in taxes, which is kind of understandable. Um, And without massive investment, it would not be possible to build out a smart infrastructure, even if we had had one that was designed and ready to go and everyone had agreed upon this is the right approach. Even if that were the case, and it's not, it would still be really hard to sell that to the American public and say this is worth paying for. So that's one hurdle just getting the money to pay for the rollout of such a system, let alone the continued maintenance and operation of that system. Also building out that kind of a system would take a huge amount of time. You get into a lot of real world issues with this one. I mean, like like finding the right contractors to do the work is it's kind of a boring thing to talk about, but it's realistically one of the big challenges. I mean, hopefully the government ends up finding contractors who don't end up having suspicious links to the people who hold the purse strings, because that happens all the time and it's never good. Um do. You also have to deal with unexpected problems as you're deploying systems. This happens in any system, doesn't matter how big or how complicated it is. It will happen, and with something this huge, it will happen a lot. Uh. Also, just building out the system so it reaches all the areas necessary. I mean, the broadband initiatives in this country show that's very easy to underserve or even not serve at all certain regions. That would be a huge disadvantage to people who live, in work in those regions. So this is a ton of work, and it's it's again, it's not impossible, but it is very hard, and the harder it is, the less likely we're going to see it happen. The actual technical requirements would be pretty darn daunting too. You would need a system capable of detecting, controlling, and coordinating every vehicle on the road, and the system would have to have an exceptionally high reliability factor because Obviously, mistakes could lead to accidents, which could mean loss of property and obviously much much worse. In some ways, the system could be simpler than what you see with autonomous vehicles, at least as long as you know conditions are normal. But then we all know that driving conditions are paradoxically rarely normal. All it takes is an animal running across the lane of traffic or pop up heavy thunderstorm to turn a normal drive into a dangerous one, and automated highway system would need to be able to detect and cope with these sort of things in a safe and reliable way that wouldn't affect the overall system operations. Right, So, let's say you've got a big old pop up thunderstorm along one stretch of highway. While you would have to have the highway deal with this all down the pathway of the road so that you didn't have congestion and traffic jams, because while the vehicles moving through the very heavy storm might have to reduce their speed, that means well, three miles back where there is no rain, you're gonna have to deal with that too. You're gonna have to reduce speed there or else you're going to start encountering traffic congestions, so like, these are really complicated systems. Right, then there are privacy concerns. So presumably an automated highway system would have to differentiate against all the different vehicles traveling on the lanes. Right. If you didn't differentiate all the vehicles, that would be disastrous because you wouldn't know that vehicle one and vehicle two are in fact two different vehicles. So maybe the system would assign a unique identifier to each vehicle as it entered the highway. Maybe manufacturers would include a unique identifier within a car's systems itself, and the highway would detect that. But either way, the highway will quote unquote no where each car is going when it got on the highway, when it got off the highway. That means that each and every driver on that highway, at least every driver who has a compatible system. More on that in a second, will be tracked as long as they are on that highway. That is a major privacy concern, So that's another issue that has to be solved. Then that brings us to the other big hurdle, which is adoption. By that I mean cars won't magically be able to communicate with an automatic highway system. Let's let's imagine that you know, we've designed and built out an incredible intelligent highway system. Uh down this one particular highway. Well, you have to have cars that can interact with that system, right, I mean there's no there's no magical connection there. So the cars themselves have to have required systems in order to interoperate with the highways. Now, if you had taken a very collaborative approach along with like car manufacturing companies and you worked with them, well, potentially those car manufacturers might have started to include compatible systems in their vehicles in the years leading up to the deployment of the intelligent system on the highway. So in other words, maybe you've got two or three years worth of vehicle models that already have those systems in place. They just you know, weren't connecting to anything until the highway was ready to go. That would help a little bit, right, People who had newer cars would be able to interact with this intelligent highway. But it's not like you're going to convince every single person to trade in their vehicle for a new one, and that means you're automated highway system will still have human operated vehicles on it. It will only be communicating with certain cars on the highway, and everybody else will still be regular or humans driving on on the roads. That adds in uncertainty, right, That's an uncontrolled variable in the system. So even if the system is working flawlessly, these uncontrolled variables can cause problems. That's very dangerous and there's not really an easy solution to that. In fact, you know, even if you have a system that works, what you're likely going to see is that the only the people who are buying the most expensive cars will initially be able to interact with it, which will create haves and have nuts on the highway. And that doesn't sound like it's a really good system either. Now, one way you could potentially get around this, it's not a practical solution, but it's it's kind of like a pie in the sky sort of solution, is to incorporate something similar to one of the many variations we've seen on the hyper loop concept, namely the sled. So one of Elon Musk's hyper loop concepts involved driving your vehicle onto a sled that in turn could navigate through the hyper loop system. So with this concept, drivers would drive up and park their car or other vehicle on a sled, and the sled would then navigate onto a track of some sort like a highway, So the car remains parked, but the sleds zooms off to the destination, where upon the driver then pulls off of the sled in their their vehicle and they drive the last mile or whatever. An automated highway system that used something similar to this could get around the fact that not all the vehicles on the road are compatible with the system. In fact, none of the vehicles would need to be compatible. It's the sleds that are compatible, not the vehicles. But in order to do that, you would have to build out these sled based highway systems like you'd essentially have to recreate all the highways that exist or slowly convert them over while also allowing for traditional traffic to pass through them. That is why it doesn't sound like a practical solution to me. It just sounds like it's too big of a problem to tackle. Even with an idealized implementation, you still have concerns. The concept covers highways but not other types of roads, which means whenever you're not on the highway, the vehicle will presumably be under human control. If it's an autonomous vehicle, well then the question is, well, why do we need an automated highway? If all the vehicles are autonomous anyway, then we don't need the automated highway. If the vehicle isn't autonomous, well then you still have issues. When you're talking about entrances and exits to the highway, those could become bottlenecks and congestion points. Sure, traffic along the highway itself is smooth, but getting onto the highway or getting off the highway, that could be a very different situation. So it could create a new kind of headache. And there are a lot of other issues as well. I think people are more in favor of car companies doing all the R and D work and the implementation of technologies to make cars smarter and safer. Then it's just up to the individual customer to decide if they want to pay for a vehicle that has those features, instead of collectively supporting a national effort to transform the actual highway system. Like when you leave it up to the individual, it's more palatable to people in the United States, because here in the US there's always been more of an inclination toward supporting individual freedom over buying into a larger program that, if it works, could benefit everyone. That's not a judgment on my part. I mean it kind of is a little bit, because you know, I do tend toward the more um collaborative spectrum. But we have plenty of examples of government projects that failed somewhere along the way, So I wouldn't say that the inclination toward individual freedom is entirely selfish. I think part of it is somewhat practical, simply because there are these examples of big government programs that failed which show a massive sunk cost, and it's possible that on made highways could be in that category. So yeah, I think that's important to to point out. Now there has really been a shift more towards the concept of intelligent highways, which is sort of it's really just a horse of a different color. It's the same same animal, just as a different name. The intelligent highways largely talk about embedded technology along highways that can communicate and interact on at least some level with the vehicles that are traveling along the highways. In some cases we're talking about things like a dedicated lane for autonomous vehicles, But again, the still brings up the challenge of having to create a standard that everyone agrees to play by so that the technologies are compatible. Otherwise it ends up working for just a tiny slice of the population and represents a massive, uh inefficient approach to trying to solve a very difficult problem. So I find the idea of the automated highway system really attractive from the end result, But the problem is getting to the end result requires so many different pieces falling into place that it is, like I said, close to impossible. Uh you know, if you had buy in everywhere, then maybe, like I think, if we lived in a very different world, which would not necessarily be a world I'd want to live in, but if we lived in a very different world where you had state run everything like state essentially you're talking about like a truly socialist or maybe even communist kind of society, then you could have a system where everything works together. But we've seen pure socialist and communist societies that you know, when you throw people into a system, it doesn't work as perfectly as the concept necessarily made it out to be, so I don't think it's practical. I wish that it were, because I definitely see issues with the way we are going. We're relying on the smart vehicle and the dumb infrastructure. I see problems with that as well. It's not like all the problems disappear. It's just that they become a little more manageable, a little more contained than we see when we try to address the overall system. So there's no perfect approach I don't think. I do think that while I would love to see a more integrated system that allows for rich communication between the infrastructure and vehicles, because I think that that would greatly boost effectiveness, I don't see it as being something that we can realistically expect any time in the foreseeable future. Still, maybe that vision from way back in ninety nine, maybe it's still on the horizon, just like the pavilion was called. Maybe that horizon is still there. Maybe we're still headed toward it. It's just that our road is a little more long and winding than we first anticipated. I hope you enjoyed this episode. If you have suggestions for future topics, please reach out to me. You can do that by downloading the i Heart Radio app, navigating over to the tech Stuff podcast section, click on that little microphone icon. You can leave me a voice message up to thirty seconds in length, or you can reach out to me on Twitter. The handle for the show is tech Stuff h s W. I look forward to hearing from you, and I'll talk to you again really soon. Text Stuff is an I heart Radio production. For more podcasts from my heart Radio, visit the i heart Radio app, Apple Podcasts, or wherever you listen to your favorite shows

Welcome to tech Stuff, a production from I Heart Radio. Hey there, and welcome to tech Stuff. I'm your host Jonathan Strickland. I'm an executive producer with I Heart Radio. And how the tech are you? You know, we've been talking about driverless cars for quite a few years now, Even on this show and in previous episodes of tech Stuff, I've talked about the various levels of autonomy, which range from zero, which means the vehicle in question really has no driver assist features at all, and the human behind the wheel is responsible for everything, you know, barring some sort of mechanical issue at any rate, and then you go all the way up to level five autonomy, which is a car that's so smart you could enroll it in Harvard. No. But seriously, a level five autonomous vehicle would have no need for human accessible controls. It would just transport people and our cargo wherever, under any driving conditions, at any time of day or night. Where currently hovering around levels two and three at the moment, mostly there are pushes to get to level four, but as it has turned out, the technology necessary to create truly autonomous vehicles that can operate anywhere at any time is harder than some of us myself included, initially thought and most of our discussions about driverless cars treat the vehicles as you know, self contained computing systems that just happened to be on the go. So by that, I mean the car mostly is relying on its own sensors to understand what is going on around it. The car's decision making systems are fully contained within the vehicle itself. And you can think of a highway filled with those types of autonomous vehicles being kind of similar to a room filled with computers that are not connected to each other, so they are not network. They're all individ jewel computers. Each machine is sophisticated and it's capable, but it's also independent of all the others. Other visions of a driverless future involved cars that can communicate with one another directly or even potentially with roads and infrastructure. These cars would depend not just on their own sensors, but those are the vehicles around it. They would remain in constant communication to provide the smoothest, safest, and most efficient ride from point A to point B, so they would be kind of part of a mesh network. And to get to that future, we would need to agree upon a standardized protocol for these cars to follow. Otherwise you could have a world where Toyotas wouldn't talk to Fords, they would only talk to other Toyota's, and that kind of thing. That would be a nightmare, that would not be helpful. Anyway, That's not the direction that most companies are going in right now, though I suspect some companies that are more focusing on creating autonomous transportation vehicles for stuff like cargo, or at least looking at this from a fleetwide perspective that could serve a similar purpose, but specifically for this particular you know, business model. But today I wanted to talk about a slightly different approach. Uh And I've talked about this a little bit in the past when I've talked about autonomous cars. So this one has had several different incarnations over the years, and it's it's a cool idea on the surface, but it's probably the least practical for driver lest vehicles or autonomous vehicles insofar as the amount of work we would need to do on an infrastructure level and the amount of consensus we would have to have across all parties in the space but this is the concept of the automated highway system. With this kind of system, you would have embedded networks that are along or under or above highways, and these systems would provide guidance to the vehicles on the road, perhaps even to the extent of controlling all the vehicles on the road. So and that kind of a system, the road would be the one doing the driving. In other words, it would send commands to all the vehicles traveling on the lanes, which would follow those commands. And it would be kind of like an automated air traffic control system, with the automated highway guiding vehicles on and off the highway seamlessly in a way to avoid traffic congestion and to avoid accidents. And it's a neat idea. It's just I won't I won't say impossible, but about as close to impossible as you can get without it being impossible to do. But that doesn't mean that folks haven't tried. The idea itself is actually coming up on being a century old. I mean, okay, we still have seventeen years to go, and yees, seventeen years for for many of you will seem like a really long time. But the older I get, the less the less long. Seventeen years sounds to me, but we are getting close to a century of following alone this idea. So back in nine GM sponsored a pavilion at the World's Fair in New York. I've talked about this in past Tech Stuff episodes too, but it's one of my favorite historical examples of futuristic thinking. So GM had a pavilion that was called the Highways and Horizons Pavilion, and the centerpiece of this pavilion was an exhibit called Futurama. Bum bum bum bum bum. That's all I can hum of that before I get a d m C A strike. Anyway, Futurama, in this case was this massive model of what a landscape of the future could look like, and it included, you know, city scapes and rural areas, included farms and hydro electric facilities like it was expansive. It was huge and incredible and included some moving components to A theatrical designer named Norman Bell get Is masterminded this exhibit and get his he was a real forward thinker. You can actually watch a full film that details the Futurama model on YouTube. It's about five minutes long. And you can even tell that this film inspired lots of later attractions, particularly the ones that opened with the debut of EPCOT way back when at Walt Disney World. If you've ever been on any of those old EPCOT rides, or you watched videos of those old epp cut rides, they owe a lot to this particular film of the Futurama exhibit from way back in. Anyway, the film covers a lot of different bits, but the part we're interested in is this idea of automated highways. Now, keep in mind this, this blue sky automated highway concept came at a time when highways themselves, we're really new. In the United States, the US was in the process of laying out highways in nineteen thirty nine to connect major cities together, even as get as his exhibit suggested a revolutionary approach. Also, I want to add a note here. While the view of the future was very forward thinking, if you watch this film, you're like, wow, this was, you know, really imaginative, the social concepts definitely were not forward thinking. The film. Um that's about the Futurama exhibit. It pretty much outright says that men are behind all the progress. It just uses men as the default, and clearly that ignores the fact that countless others besides men made significant contributions up to ninety nine and and since and uh yeah, I just want to call that out because you know we should. So it's It's certainly not a perfect film by any stretch of the imagination. However, the exhibit gives us a view of the far off future of nineteen sixty. Keep in mind that was twenty years away when this exhibit was being shown, and the motorways in this exhibit allow cars to travel at designated speeds. The accompanying film says that those speeds are fifty miles per hour and they will maintain proper distance from other cars through quote unquote radio control. So presumably the motorway itself is the transmitter and the cars are receivers. Alternatively, the cars could be both transmitters and receivers and be communicating with each other. It's not entirely clear. Now, you could argue that the nine exhibit was marketing and it was just overly optimistic projections about the future for the purposes of marketing for GM, And I think you would have a pretty solid argument if you said that, I'm not or anyone involved really believed that this particular vision of the future was what was going to happen by nineteen sixty. But then it's always hard to protect the future, even if it's just a short ways out. Also, the US of nineteen nine was entering into very uncertain territory. It was recovering from the Great Depression, which was finally coming to an end after a decade of economic turmoil, and the US had not yet entered into World War Two, And obviously the experience of World War Two would affect the American point of view significantly. It would really boost xenophobia for one thing. But let's skip on ahead while get us his vision of nineteen sixty did not manifest in reality. We did not get these automated highways by nineteen sixty. There were folks who were working on the concept of automated highways, at least researching ways where perhaps it could be possible, and that was mostly focusing on mechanical systems and radio control systems. And over time, as computers began to evolve and become more commonplace, there was a transition toward looking at computer controlled systems, the hope that maybe a computer would be sophisticated enough to handle the complex operations that would be required to have a truly automated highway system. But computers themselves were really large and limited early on. As they evolved, there was also another big shift, and I'll talk about that after we come back from this quick break. Okay, before the break, I was talking about how as computer systems were starting to get more sophisticated and smaller, mentorization being a big part of this, we started to see a shift, and that shift would end up being a really significant one as far as the concept and plausibility of automated highway systems are concerned. And that was that car companies were starting to design and integrate driver's safety features in vehicles, and these features bordered or sometimes overlapped the autonomous car concept. So, for example, cruise control. The actual idea of cruise control is almost as old as the car itself. It's essentially a way of having a car maintain a set speed. In some cases, it was a set speed that the driver couldn't choose, like it was either full throttle or not. But then later you had more sophisticated systems that let you set cruise control at a speed of your choosing, and your vehicle would just maintain that speed. Now, that was just one tiny step toward creating systems that would augment a driver's ability to navigate the roads safely and more over, service kind of building block toward autonomous cars. Obviously, a lot more systems would be needed in order to even remotely consider a car to be autonomous, but these were the steps that we're leading in that direction, and we would see that trend continue over time. Cars would get more systems that would make them safer and be able to take over certain operations. You know, cars would get integrated cameras, which would help when you're backing out of the space for example. They'd get collision sensors, they get lane detection sensors, and much more so. The cars were evolving into smart machines that could operate on essentially a dumb network. Right. You can think of the roads and highway systems as a network, and the network itself is fairly robust. I mean, if you drive in Atlanta, you might question that with all the potholes, but you get what I'm saying. It's a robust system, but it's dumb. It does not have any real intelligent components itself. And the cars became the smart art components, smart devices that could navigate dumb systems. Meanwhile, in research organizations such as the Partners for Advanced Transit and Highways or PATH, which originated out of the University of California, you had researchers who were looking into various approaches for autonomous driving, which included auto highway technologies. So not just you know, how can we make an autonomous car, but would it be possible to make a highway system that itself was intelligent and remove the requirement of having intelligent vehicles instead. Now, this was in the nineteen eighties, and it started to take up some momentum because you know, in the nineteen eighties we certainly were miles, no pun intended miles away from the technologies that would allow a truly autonomous vehicle. So by n things got shifted into a higher gear. To use another pun, I'm just gonna keep doing those I guess I don't to, I honestly don't. But the US government passed the Intermodal Surface Transportation Efficiency Act or is T and that's often a question we ask here in Atlanta, Georgia, when someone hands us a drink, we point to and say, is t anyway? This Act laid the groundwork for early testing of automated vehicles and roadways. So the U. S Department of Transportation would then create the National Automated Highway System Research Program or in a h s RP. This program's main purpose was to establish specifications for an automated highway system. Again, because with so many different players in the space, like all the different law agencies across you know, federal and state and regional systems. You have all the different car companies that are involved, Like there are a lot of different players who are uh deeply concerned with things like highway technologies. You have to establish standards so that there is interoperability between the system and all the vehicles. Right. Again, if you have a system that only communicates with one type of vehicle, then it's not really a good system, right for anything other than that vehicle, it's a terrible system, or it might as well not exist. In the Department of Transportation created the National Automated Highway System Consortium, So this was kind of a think tank that was made up of more around like a hundred different members across nine major categories. Those categories included things like research organizations, government agencies again at different levels like federal, state, and local. It included academic organizations like the University of California, and of course representatives from the transportation industry. So the hope was that by including all of these groups in the conversation that everyone would be able to arrive at a consensus that would make the most viable, efficient, safe, and practical solution for automated highways. Um also hopefully most affordable, because you would ultimately be turning to the US taxpayer to foot the bill for these kind of things. And I think that was a wise move in some ways, because it's always good to get stakeholders involved. One just to make sure that the solutions you are designing actually solve real world problems. There are a lot of times where we talk about an approach or a technology and we refer to it as a solution that's looking for a problem. That's not ideal. Right, if you are if you have identified what the problem is, then you can work to solve it, but you have to make sure that your solution actually does address the problem them. That's one reason why you want to have lots of different stakeholders there. You also want to have buy in across the entire industry or else Again, interoperability becomes a problem. However, when you include all these different voices, there's always the danger that projects descend into a designed by committee approach where it feels like all the really progressive concepts have been sanded down until they aren't really interesting anymore, or worse, you fail to find any common ground and nothing gets done. So between and researchers poured some serious effort into conceptualizing and automated highway system. So the benefits of such a system, if it worked, were obvious on the surface. Like, everyone could agree this is a good idea insofar as what the goal is because you would reduce traffic congestion, which would improve traveled times. It would also reduce fuel consumptions that was an added benefit, and it would improve the quality of life from motorists across the nation. They'd be spending less time in transit and more time doing whatever it was they wanted to do. So like, those are pretty obvious benefits of such a system. The reduction of accidents would also be incredibly important. It could save tens of thousands of lives every year. It is impossible to overstate how important that is. The impact of deaths due to traffic accidents is gargantuane. There's the personal impact on all the friends and family of the people who are lost, there's the financial impact. There's the societal impact. You know, we no longer have those people contributing to society. So yeah, that's one of those things where not only is it great just the idea of saving these lives, we all stand to benefit from that, right, we all stand to benefit from the fact that these people are still around and contributing to society. Not to mention, you save a lot of money and property in the process of reducing accidents. So the goal was clear. How to get to the goal was the matter of debate. That debate largely ended in why, well, I mean, ultimately the easy answer is the money fell out. So in part the issue was that all those stakeholders were in fact having trouble agreeing upon the right approach and the components that would have to go into an automated highway as well as into the vehicles that would be compatible with such highways. You know, the whole goal was to build consensus, but over the years that just became impossible to do. You had too many different voices with different agendas, and there was no common ground to really solidify a path forward. So nothing was happening, and since the consortium was making very little headway, it was pretty hard to justify continuing to fund it. So in nineties seven, the US government passed another act. This one was the T twenty one Act t E A one, and that pulled financial support from the Automated Highway Research programs to put it in other areas. The d O T essentially made the decision to focus more on safety oriented technologies like the aforementioned lane detection technologies, and essentially what the government was leaning toward was where the automotive industry was already headed, which would mean you would have these independent autonomous vehicles or increasingly autonomous vehicles that would navigate within an environment rather than an intelligent environment that would guide vehicles to their respective destinations. Now, some say the technologies we're seeing evolving vehicles in general and autonomous vehicles in particular, could be a stepping stone toward automated highways in the future. I'll talk more about that after we take this quick break. So could autonomous vehicles be a step toward intelligent highways? You know, maybe we will see autonomous vehicle technologies converge with intelligent highways and that the true future of motoring will be a combination of both intelligent infrastructure and intelligent vehicles working together. You could imagine, you know, highways that are able to meter traffic onto them with autonomous cars so that there are no bottlenecks on entrances and exits to the highway for example. Right, that's a possibility. However, for all of that to work out, we again have to create an agree upon standards that work across all parties, all the intelligent highway systems whenever we would design and implement those, and all the different vehicles, or else you have cars that won't know how to talk to highways, or cars that don't know how to talk to each other, and so on, and you just end up again with individual smart vehicles and the same effect of an overall dumb system. Also, we have to consider some of the massive challenges that come along with building and implementing an automated highway system. So, first of all, doing so would require massive investment in infrastructure. That is always a tough thing to get going in the United States for multiple reasons. Like we did see a national infrastructure bill this year, but it did not go easily, and it turns out like folks are not super jazzed about having to pay more in taxes, which is kind of understandable. Um, And without massive investment, it would not be possible to build out a smart infrastructure, even if we had had one that was designed and ready to go and everyone had agreed upon this is the right approach. Even if that were the case, and it's not, it would still be really hard to sell that to the American public and say this is worth paying for. So that's one hurdle just getting the money to pay for the rollout of such a system, let alone the continued maintenance and operation of that system. Also building out that kind of a system would take a huge amount of time. You get into a lot of real world issues with this one. I mean, like like finding the right contractors to do the work is it's kind of a boring thing to talk about, but it's realistically one of the big challenges. I mean, hopefully the government ends up finding contractors who don't end up having suspicious links to the people who hold the purse strings, because that happens all the time and it's never good. Um do. You also have to deal with unexpected problems as you're deploying systems. This happens in any system, doesn't matter how big or how complicated it is. It will happen, and with something this huge, it will happen a lot. Uh. Also, just building out the system so it reaches all the areas necessary. I mean, the broadband initiatives in this country show that's very easy to underserve or even not serve at all certain regions. That would be a huge disadvantage to people who live, in work in those regions. So this is a ton of work, and it's it's again, it's not impossible, but it is very hard, and the harder it is, the less likely we're going to see it happen. The actual technical requirements would be pretty darn daunting too. You would need a system capable of detecting, controlling, and coordinating every vehicle on the road, and the system would have to have an exceptionally high reliability factor because Obviously, mistakes could lead to accidents, which could mean loss of property and obviously much much worse. In some ways, the system could be simpler than what you see with autonomous vehicles, at least as long as you know conditions are normal. But then we all know that driving conditions are paradoxically rarely normal. All it takes is an animal running across the lane of traffic or pop up heavy thunderstorm to turn a normal drive into a dangerous one, and automated highway system would need to be able to detect and cope with these sort of things in a safe and reliable way that wouldn't affect the overall system operations. Right, So, let's say you've got a big old pop up thunderstorm along one stretch of highway. While you would have to have the highway deal with this all down the pathway of the road so that you didn't have congestion and traffic jams, because while the vehicles moving through the very heavy storm might have to reduce their speed, that means well, three miles back where there is no rain, you're gonna have to deal with that too. You're gonna have to reduce speed there or else you're going to start encountering traffic congestions, so like, these are really complicated systems. Right, then there are privacy concerns. So presumably an automated highway system would have to differentiate against all the different vehicles traveling on the lanes. Right. If you didn't differentiate all the vehicles, that would be disastrous because you wouldn't know that vehicle one and vehicle two are in fact two different vehicles. So maybe the system would assign a unique identifier to each vehicle as it entered the highway. Maybe manufacturers would include a unique identifier within a car's systems itself, and the highway would detect that. But either way, the highway will quote unquote no where each car is going when it got on the highway, when it got off the highway. That means that each and every driver on that highway, at least every driver who has a compatible system. More on that in a second, will be tracked as long as they are on that highway. That is a major privacy concern, So that's another issue that has to be solved. Then that brings us to the other big hurdle, which is adoption. By that I mean cars won't magically be able to communicate with an automatic highway system. Let's let's imagine that you know, we've designed and built out an incredible intelligent highway system. Uh down this one particular highway. Well, you have to have cars that can interact with that system, right, I mean there's no there's no magical connection there. So the cars themselves have to have required systems in order to interoperate with the highways. Now, if you had taken a very collaborative approach along with like car manufacturing companies and you worked with them, well, potentially those car manufacturers might have started to include compatible systems in their vehicles in the years leading up to the deployment of the intelligent system on the highway. So in other words, maybe you've got two or three years worth of vehicle models that already have those systems in place. They just you know, weren't connecting to anything until the highway was ready to go. That would help a little bit, right, People who had newer cars would be able to interact with this intelligent highway. But it's not like you're going to convince every single person to trade in their vehicle for a new one, and that means you're automated highway system will still have human operated vehicles on it. It will only be communicating with certain cars on the highway, and everybody else will still be regular or humans driving on on the roads. That adds in uncertainty, right, That's an uncontrolled variable in the system. So even if the system is working flawlessly, these uncontrolled variables can cause problems. That's very dangerous and there's not really an easy solution to that. In fact, you know, even if you have a system that works, what you're likely going to see is that the only the people who are buying the most expensive cars will initially be able to interact with it, which will create haves and have nuts on the highway. And that doesn't sound like it's a really good system either. Now, one way you could potentially get around this, it's not a practical solution, but it's it's kind of like a pie in the sky sort of solution, is to incorporate something similar to one of the many variations we've seen on the hyper loop concept, namely the sled. So one of Elon Musk's hyper loop concepts involved driving your vehicle onto a sled that in turn could navigate through the hyper loop system. So with this concept, drivers would drive up and park their car or other vehicle on a sled, and the sled would then navigate onto a track of some sort like a highway, So the car remains parked, but the sleds zooms off to the destination, where upon the driver then pulls off of the sled in their their vehicle and they drive the last mile or whatever. An automated highway system that used something similar to this could get around the fact that not all the vehicles on the road are compatible with the system. In fact, none of the vehicles would need to be compatible. It's the sleds that are compatible, not the vehicles. But in order to do that, you would have to build out these sled based highway systems like you'd essentially have to recreate all the highways that exist or slowly convert them over while also allowing for traditional traffic to pass through them. That is why it doesn't sound like a practical solution to me. It just sounds like it's too big of a problem to tackle. Even with an idealized implementation, you still have concerns. The concept covers highways but not other types of roads, which means whenever you're not on the highway, the vehicle will presumably be under human control. If it's an autonomous vehicle, well then the question is, well, why do we need an automated highway? If all the vehicles are autonomous anyway, then we don't need the automated highway. If the vehicle isn't autonomous, well then you still have issues. When you're talking about entrances and exits to the highway, those could become bottlenecks and congestion points. Sure, traffic along the highway itself is smooth, but getting onto the highway or getting off the highway, that could be a very different situation. So it could create a new kind of headache. And there are a lot of other issues as well. I think people are more in favor of car companies doing all the R and D work and the implementation of technologies to make cars smarter and safer. Then it's just up to the individual customer to decide if they want to pay for a vehicle that has those features, instead of collectively supporting a national effort to transform the actual highway system. Like when you leave it up to the individual, it's more palatable to people in the United States, because here in the US there's always been more of an inclination toward supporting individual freedom over buying into a larger program that, if it works, could benefit everyone. That's not a judgment on my part. I mean it kind of is a little bit, because you know, I do tend toward the more um collaborative spectrum. But we have plenty of examples of government projects that failed somewhere along the way, So I wouldn't say that the inclination toward individual freedom is entirely selfish. I think part of it is somewhat practical, simply because there are these examples of big government programs that failed which show a massive sunk cost, and it's possible that on made highways could be in that category. So yeah, I think that's important to to point out. Now there has really been a shift more towards the concept of intelligent highways, which is sort of it's really just a horse of a different color. It's the same same animal, just as a different name. The intelligent highways largely talk about embedded technology along highways that can communicate and interact on at least some level with the vehicles that are traveling along the highways. In some cases we're talking about things like a dedicated lane for autonomous vehicles, But again, the still brings up the challenge of having to create a standard that everyone agrees to play by so that the technologies are compatible. Otherwise it ends up working for just a tiny slice of the population and represents a massive, uh inefficient approach to trying to solve a very difficult problem. So I find the idea of the automated highway system really attractive from the end result, But the problem is getting to the end result requires so many different pieces falling into place that it is, like I said, close to impossible. Uh you know, if you had buy in everywhere, then maybe, like I think, if we lived in a very different world, which would not necessarily be a world I'd want to live in, but if we lived in a very different world where you had state run everything like state essentially you're talking about like a truly socialist or maybe even communist kind of society, then you could have a system where everything works together. But we've seen pure socialist and communist societies that you know, when you throw people into a system, it doesn't work as perfectly as the concept necessarily made it out to be, so I don't think it's practical. I wish that it were, because I definitely see issues with the way we are going. We're relying on the smart vehicle and the dumb infrastructure. I see problems with that as well. It's not like all the problems disappear. It's just that they become a little more manageable, a little more contained than we see when we try to address the overall system. So there's no perfect approach I don't think. I do think that while I would love to see a more integrated system that allows for rich communication between the infrastructure and vehicles, because I think that that would greatly boost effectiveness, I don't see it as being something that we can realistically expect any time in the foreseeable future. Still, maybe that vision from way back in ninety nine, maybe it's still on the horizon, just like the pavilion was called. Maybe that horizon is still there. Maybe we're still headed toward it. It's just that our road is a little more long and winding than we first anticipated. I hope you enjoyed this episode. If you have suggestions for future topics, please reach out to me. You can do that by downloading the i Heart Radio app, navigating over to the tech Stuff podcast section, click on that little microphone icon. You can leave me a voice message up to thirty seconds in length, or you can reach out to me on Twitter. The handle for the show is tech Stuff h s W. I look forward to hearing from you, and I'll talk to you again really soon. Text Stuff is an I heart Radio production. For more podcasts from my heart Radio, visit the i heart Radio app, Apple Podcasts, or wherever you listen to your favorite shows