Podcast: The Future of Traffic Flow with Eric Gannaway from Siemens ITS

traffic flow

There’s nothing more annoying than inching along the road in a city with heavy traffic. With today’s traffic lights, it’s not unusual to stop at a red light, travel 30 feet, then get stopped again. Rinse and repeat.

There are, however, innovations in the works to create intelligent traffic systems that collect data, analyze the patterns and make adjustments in real-time — ultimately making traffic systems more efficient than they are today. 

In this episode, we go beyond talking about just autonomous cars. Instead, we’re investigating the world of autonomous and connected cars, and the infrastructure changes that will transform the way we travel.

Eric Gannaway, from Siemens Intelligent Traffic Systems in Detroit, was my guest, one of a series technologists, researchers and policymakers who joined me to discuss new mobility topics last year for our podcast feed. (See one example in the blog I posted earlier this month about the episode with NAVYA’s Henri Coron.)

Here, Eric and I discuss the potential costs and benefits of connected vehicles, how intelligent traffic systems will improve traffic flow, and what the future of employment in the automotive industry might look like. 

A link to our interview is below, along with a lightly edited transcript of this episode. (You can hear the full unedited version of our conversation here.) For more conversations like this, see the Future Car podcast on iTunes, Spotify or wherever you subscribe to podcasts.

Show transcript

Eric Gannaway:  There’s a company that came out that does a valet parking service. They will turn your 2012 or newer vehicle into an autonomous vehicle. They can do that and it’ll park itself.

Ed Bernardon: Welcome back to The Future Car podcast. I’m your host, Ed Bernardon, and today, we’re talking about connected cars. However, connected cars are not to be confused with autonomous cars. They’re actually quite different. When we think about autonomous vehicles, self-driving cars, the reality of integrating them into our lives and onto our roads still has a bit of ground to cover.

Connected cars, on the other hand, are already working their way into our regular driving lifestyles. At the rate technology moves these days, the integration of connected cars is closer than we think. What exactly is a connected car?

Eric Gannaway: As we discover more and more what connected means, I think it means if you get in your car and you started up, and your navigation service comes on and tells you where you’re located, you’re connected. If you can get on your cellphone and lock your doors and your car, or start your car, you’re connected. I get a little message from the car dealership that I bought it that says, “Hey, we told you it’s time to get an oil change.” Connected vehicles… if you’ve got internet, your vehicle is connected. It’s gathering data, and that data is being stockpiled, right?

The manufacturer is using that data, the dealerships are using that data, it’s where it goes from there, but that’s a big portion of it. It’s all of the sensors and all of the technology that’s really in the vehicles we’re driving now that make them connected.

Ed Bernardon: That’s Eric Gannaway. He works in Detroit as part of the Siemens Intelligent Traffic Systems team. For the last 10 years, he’s worked on the transportation side of the industry. When it comes to cars, whether it’s self-driving, connected or an advanced combination of both, he has insight on how traffic patterns come into play once these vehicles are out on the road. Eric tells me that connected cars are all about collecting data.

When it comes to having your vehicle connected and they’re collecting data from it, in some ways, that’s a little scary, right? Everyone’s concerned these days, even like on social media, right? “Oh, what information are they collecting about me? How are they going to use it?” How do you balance that against, “What benefits am I getting as a driver or a passenger if my vehicle’s connected?” What are the trade-offs there?

Eric Gannaway: We believe we’re creating safer vehicles, and we’re moving towards a zero-death and zero-crash future, however far distant. That’s where we hope we’re going, and so I think that’s the upside for us, driving a connected vehicle, the convenience of knowing that your blind spot detection is working, and that maybe you can’t back into someone because you have park-assist or backup-assist. There’s a lot of features that cause us to be safer in general. I think the trade-offs, and that can be because of that data is being collected, spur conversations like, “What does that do to our car insurance? It should lower car insurance rates.” And it may, based on the amount of safety features you have, but being connected, how hard you hit the brakes and how often you hit the brakes, that information is also being collected.

How fast you go if you exceed the speed limit, if you’ve got a vehicle that tells you what the speed limit is, and then the little sign turns red when you go over the speed limit, that information is being collected. I think what we see is, is even though we’re being safer, I think car insurance companies, for example, can turn around and say, “Whoa, maybe we don’t charge you enough. It looks like you go into the city a little more than we thought you might go into city. You said you go 20 miles to work, but we didn’t realize you went 20 miles into a bad area. Maybe your premium should be a little higher.” I think that information can be used, we’ll say for the good, but, and no doubt, it could be a little more costly for us in the long run.

Ed Bernardon: Well, if you’re a good driver, though, it could be to your benefit, I suppose, right? They could see that you’re a good driver and you’re saying, “Maybe reduce my insurance rates.”

Eric Gannaway: Yeah, yeah, yeah, but the insurance companies make money on the good drivers and they lose money on the bad drivers, and you can be a good driver… It might be the minority, those that don’t go over the speed limit that don’t hit the brakes very hard. I haven’t had any accidents in a long, long time, but I do have some hit-the-brake-pretty-hard incidents, which may indicate those are near misses, right? Possibly, as far as the insurance company is concerned, that may be just as good as an accident. No doubt, somebody could benefit from the data, but it almost looks like the insurance company is not going to take a hit based on these connected vehicle features. We very well might be the ones.

Ed Bernardon: It comes as no surprise that when these kinds of vehicles become more common, the industry will change. Car insurance premiums are not the only change we’ll see though. The entire way we think about traveling, using, producing and owning cars is bound to shift. What do you think the future of employment in the automotive industry is going to be like? What will engineers be doing?

It sounds like they’re not only going to be engineering cars, they’ll probably be engineering the streets that these cars are connected to. What’s the future of employment in the automotive industry going to be like?

Eric Gannaway: It’s going to morph. We know that, and I think it’s still a little unknown. We’re probably going to lose 10%, maybe 5% of the workforce that’s driving vehicles, those people that drive for a living, and you’re probably not going to have the cabs that you had and some of the… even the big trucks that you have, so auto production is going to go down. I think as we begin to build vehicles for this autonomous market, the industry has a tendency to believe that even though we have autonomous vehicles, that you and I may choose to buy one because we only use our car 5% or 10% of the time anyway, and so the smart move would be for us to own one.

While we’re sitting here doing this, our vehicle is a shared ride now and it’s out making money for us. So you may pay $80,000 for that car, and it’ll turn around and turn a profit probably in less than a year for you if that’s what you want to do. On the flip side, because that vehicle is gone all the time, maintenance is still going to be there. Right now, autonomous vehicles are handled with kid gloves, right? Every time one comes in, it’s pampered, and it’s washed, and it’s cleaned, and it’s checked and it’s… everything’s updated, but vehicles aren’t going to stop for very long. We know some of the shuttle companies… they’ll run a dozen shuttles someplace because shuttles have to come in for maintenance while other shuttles are on the road, because they’re running all the time… So I think what we’re going to see is, I think, the importance of aftermarket maintenance. I think it’s going to stay strong.

I think the automotive companies are going to be building shared vehicles. You still have your mechanical engineers, you got your electrical engineers, you’ve got your computer scientists, for sure. Those are probably going to be the big jobs… But you’re still going to have production, probably not at the level that it is, but I think people are going to shift into different areas. I think we’re going to feel the pain a little bit in the beginning. But then, I think it’s going to settle in. I think those jobs are going to surface mainly because vehicles are being used so much more than they’re being used now… so I think it’s going to offer a lot of opportunity.

Ed Bernardon: In one way, the vehicles will last longer, so we have to produce fewer cars, but yet on the other hand, if autonomy especially, or even if connected technologies can make travel more affordable, you’ll need more vehicles running at a certain time. It seems to be a shift from production maybe to servicing, cleaning, making sure that they stay up and running. Does that seem to be the major shift then?

Eric Gannaway: I think so. But I know the big shift is going to be away from gas and diesel engines… it’s going to be to electric engines, so it’s going to be a different kind of maintenance. When a car runs 20 hours a day, I’m not sure that it lasts longer than a car that runs two hours a day, so I think this is some of the unknown. The thing that we’re not going to see is autonomous vehicles sitting in their driveways. I have three vehicles at my house my wife leases, and I’ve got 10,000 miles a year on her lease. We’ll turn it back in and she’ll still have 10,000 unused miles. You see that the way that we do this now might continue, but [make more sense] with an autonomous vehicle.

If I could make money sharing her vehicle right now and I felt comfortable doing it, believe me, that car would be on the road, so I think we’re going to put more miles on a vehicle. I think it’s going to have more wear parts and more maintenance. I think we’re going to see it because those vehicles aren’t going to sit around. They’re going to be moving all the time because there’s no reason for them not to.

Ed Bernardon: It sounds like the Airbnb of cars.

Eric Gannaway:  I think that’s what we’ll see, when people really get… I think the taxi driver and maybe the Uber driver buys his own car, and then he sits at home and watches the World Series while his car’s out making money. I think you’re going to see those kinds of things. You’ll see people buy fleets. Same way.

Ed Bernardon: We’ll be all like mini Ubers then?

Eric Gannaway: I think so.

Ed Bernardon: The possibility of cars on the road without us, without the owners and the drivers, you begin to wonder, “How is this all going to work? How can our current traffic systems handle this kind of technology? What’s being done to fit this kind of technology for a society with future cars?” The good news is plans are in the works. Here in Boston, they’ve been testing autonomous vehicles for quite some time, and that’s exciting, even though it may be a ways out before they’re accessible to everyone everywhere in Boston.

Just recently, I heard in the news that they’re very excited about synchronizing, using the latest technology — and I’m sure this comes from Intelligent Traffic Systems at Siemens as well. But synchronizing the traffic lights… there’s nothing more annoying than getting a green light and going one block and getting a red one, and back and forth and back and forth until it takes you forever to go through five, six city blocks. The whole idea of vehicles talking to traffic lights, talking to a master traffic control brain, that’s one of the key things that you get from an intelligent traffic system. Can you tell us a little bit of what Siemens has done in this area?

Eric Gannaway: Yeah, we’re very proud of the fact that we have a SCOOT system, and it is an adaptive system.

Ed Bernardon: What does SCOOT actually mean?

Eric Gannaway: It means Split Cycle and Offset Optimization, and those are the controls that traffic signal controllers use to control the pattern, the phasing, and the timing at intersections.

Ed Bernardon: Does this mean we’re going to have green lights wherever we go? Not exactly. The technology is a bit more complex than that. What does Siemens ITS do in this area of making that traffic flow better? I want to get rid of those red lights that are annoying me.

I’d love to have 10 green lights in a row. How does Intelligent Traffic Systems make that happen?

Eric Gannaway: Well, it happens really through a series of detection systems that are at intersections and it monitors the flow of traffic… so it will notify the central system, the speed the traffic’s moving at, the volume or the capacity of the intersection, and it can make adjustments to traffic timing on the fly. If it sees that it’s taking longer to clear traffic through an intersection, it can actually make some adjustments to extend the green time. It will also prevent, in a learned-type situation, it’ll learn over the course of a couple of minutes what traffic is doing, and it finds out we have more people that are making a left turn, and we’re not actually clearing the left turn lanes, it will begin to make some adjustments and hold traffic coming back from the opposing direction to try to clear that out. So it actually, it optimizes traffic by making real-time traffic adjustments, and it’s done through detection and automatic adjustments and the timing plan. There’s essentially a server that stores this information and begins to analyze it, and then makes those shifts and changes, so in the morning, if traffic is heavy in the northbound direct, the system will actually make changes to optimize northbound traffic, and if it finds out that in the evening in the southbound traffic is picking up, it’ll turn around and readjust the system to optimize traffic going the opposite direction.

Ed Bernardon: It sounds like there’s some sort of a master traffic brain or something that’s all-knowing of where all the cars are, and based on that, it makes adjustments.

Eric Gannaway: Yeah, and it does it based on the way… in the time that vehicles entered the intersection, that crossed that stop bar. You’ll see when you’re sitting in an intersection… they put a big, generally white line across there, and that’s kind of a measurement point if there’s something sitting there and it detects it. It will make the… It uses that as the point of counting vehicles as it goes across. It will actuate a signal if there’s somebody sitting there, telling them that there’s someone waiting… “Let’s give this person a turn…” but there is a software system.

It’s a matter of the controller and the cabinet, running the signal timing and the phasing, and then there is kind of that master brain that analyzes all this, and tries to make intelligent decisions as to who should get to go and who’s going to have to wait a little longer.

Ed Bernardon: I was a little surprised that a city like Boston is only now deploying a technology like this. Are there cities that are doing this right now? Is it commonplace? It sounds like a great technology. You would think everyone would be wanting technology like this in their city as soon as possible. There’s nothing more annoying than driving in a city with heavy traffic.

Eric Gannaway: Yeah there are lots of cities that are doing it. There’s probably a few thousand intersections in the United States, which sounds like a lot. There’s 330,000 traffic signals, but there’s a few thousand that have this type of technology. Some cities are so congested, they’re saturated, you can’t really do anything other than give people a turn, and so sometimes, an adaptive system doesn’t work in a very saturated situation like that. It’s in areas where they don’t want to pour more pavement.

Sometimes that’s the only way that you can ease up traffic, is to give people more lanes or take advantage of the pavement that you have and optimized traffic so people are moving at the right speeds. If we can keep them moving in platoons, you keep moving, you don’t create the carbon footprint that we do, we save a lot of time, we save a lot of fuel by doing that. Not every intersection or sometimes corridor is a candidate, but more and more are, and there’s a price that’s attached to it, and sometimes it takes a while for cities to decide what that return on their investment is going to be. But no doubt from the standpoint that you’re saving on fuel, and that’s big consideration, you’re saving on time, even changing the time by, on a corridor for six or seven minutes a day that you’ve increased someone’s commute time, you’re saving hours. You’re saving a week’s worth of time for someone at the end of the year, and that’s a valuable offset, and there’s a price that’s associated with that, so if they can justify that, you see it, and Boston has just been able to justify the cost, I believe.

Ed Bernardon: Eric, the benefits you just described seem tremendous, however, there’s 330,000 traffic lights in United States, but only a few thousand have actually been converted. With such great benefits, why is that the penetration of this technology so small?

Eric Gannaway: It’s the cost of implementing it is what it is. The government has made money available, a lot of money available for a community, a city to put a plan together, to show a justification for how much they believe that they can reduce congestion. There is a formula that does calculate hours of travel time reduced, that congestion, that carbon footprint, what the cost of that is, the cost of wasted fuel by sitting at traffic lights. If they can show a reduction, the government has money available for them to implement that, so sometimes it’s a co-op type. The government may pay 80%, you pay 20%, and over the last few years, many cities, municipalities, have had a hard time even coming up with the 20% because tax bases have dropped, jobs had disappeared.

We’re just starting to make a comeback, and so some communities are in a little better shape to do it, but it’s generally the money that prevents it from happening.

Ed Bernardon: One thing you hear about these days, speaking of money, is congestion charging, right? They do it in London, there’s talk of New York and other cities considering it. That could be one way of raising some money, right? How does congestion charging … Does that work together somehow with connected vehicles or other regulations?

Is there a way that you can combine some of these things to make the green wave occur sooner rather than later?

Eric Gannaway: That would help. I think initially, the congestion charging… those dollars were supposed to help improve the infrastructure because it’s really a tax or a toll. That was supposed to help make some of those adjustments.

Ed Bernardon: Did that not happen? Did that money go elsewhere?

Eric Gannaway: Well, you said in London. I’m not really sure where the money is going. I know in New York City, what they’re talking about doing. They’ll geofence in some areas and you’ll pay a toll when you come in and go out, and ideally, what it should do is it should mean you and I ride to work in the morning. Instead of taking two vehicles, we take one.

Ed Bernardon: What’s a geofence?

Eric Gannaway: I guess, it would be an internet or even a satellite radio type of area that’s designated… vehicles would be detected as they enter this area or they try to leave that area… that signals are tolling or certain information would be broadcast as you approached it or as you went through that area. That’s the best of my knowledge.

Ed Bernardon: In addition to thinking about the modes of traffic in our environment, we have to think about infrastructure as well. Let’s talk a little bit about vehicle-to-infrastructure communication. How does that help an autonomous car navigate? For instance, does it help it see around a corner or anticipate that pedestrians are going to be somewhere? How does that work to make an autonomous car safer and more reliable?

Eric Gannaway: Well, the autonomous vehicles have… they’re loaded with sensors. Again, the sensors are the connected part of the vehicle. Its ability to sense oncoming traffic around the corner, it can sense vehicles coming, it can also, from the detection capabilities in an intersection, be notified if a vehicle is coming at a high rate of speed and it doesn’t appear that it’s going to stop. It will receive dedicated short-range radio communication, DSRC from the intersection saying, “There could be a collision. There could be a red light runner coming.”

That’s based on the information of the vehicle coming, the speed and trajectory of your vehicle, and what’s happening at the intersection at the same time. Those vehicles will make a decision saying, “I don’t think this other guy can stop…” It will send you a notice and tell you to be on the lookout. It’s a combination of that sensory detection and the vehicle’s ability to communicate.

Ed Bernardon: Can it also see pedestrians that are crossing the street and anticipate that they’re going to jump from the curb onto the street, or jump out from between a couple of parked cars and let a car know that that may happen soon, so it keeps not only the driver and the car safer, but also the pedestrians?

Eric Gannaway: I think with the video camera algorithms that are being created now, vehicles have a tendency to see something and identify it, but there is external detection and algorithms that will look and notice a pedestrian. It can see that the pedestrian has his arm bent, make an assumption the person is on the phone, make the assumption he may not be paying attention based on the rate that he’s moving, and send the vehicle a notice. This would come from external detection and would tell a driver there’s a pedestrian that’s going to enter the intersection, or beware, or there is a pedestrian in the intersection. The system may see the pedestrian… again the sensing would happen externally. Now, the vehicle itself has the ability more and more through AI to be able to identify objects, like pedestrians, to know the difference between a person and an animal, for example. I think what happened, just for example, when the Uber vehicle hit… that unfortunate accident that hit someone and killed her… the vehicle was having a hard time determining whether it was an object, an animal, a person, and by the time it did, it wasn’t set up to notify the driver to take over and make a decision.

The vehicle was struggling with what was in the road. It couldn’t make that decision. External detection can be set up to make those types of decisions and get this information to the vehicle in enough time for it to make a decision.

Ed Bernardon: What are the things you’re starting to see in cities now in terms of multi-mode transportation or ability to choose different types of transportation? Multimodal transportation that could be a car or a shuttle or a scooter, or maybe a subway? Siemens has Intelligent Traffic Systems. We also… we build trains, subways. How does Intelligent Traffic Systems help in this multimodal world? If I wanted to take a ride on a bus and a train, how does all of that work together to get me from point A to B?

Eric Gannaway: Well, Siemens has all that technology and has the ability to schedule all of those modes of transportation from one app on a phone, and its ability to do that would be to call up an autonomous vehicle, would be to schedule a scooter or an Uber for that, as a matter of fact. If it’s an autonomous vehicle, the app would be able to call that particular vehicle. It would identify and register the rider when it got on the vehicle. You would have preprogrammed the destination, and of course, it would automatically take you to where you were going to go. I think it’s important from the congestion standpoint, that we were starting to talk about a little bit… lots of mobility options (including public transportation) really are different types of ride-sharing programs.

That’s where the reduction in congestion is going to come from. That’s where the ability for autonomous vehicles, I think is really going to come from… by removing a lot of the other types of traffic by clearing out the roadway itself, and that’ll happen with scooters and some of the autonomous shuttles. There’s supposed to be an air taxi company, I think in 2025. The city of Miami is already building rooftop landing zones. Whenever you do something like this, it takes more vehicles off the street, makes it easier for us to get around, and makes it easier for autonomous vehicles.

Ed Bernardon: I would love an app where it does all the thinking like, “Okay, Ed, you will start off with an Uber, and then you’re going to take the Red Line to this location, and the subway will drop you off, and you can pick up an electric bike there for that last 10 minutes to get you to your destination.” Do apps like that exist today?

Eric Gannaway: Yeah, they do, and we have an app with Bytemark that has the ability to do just that, and in fact, I believe that we’re participating in a smart city project in Columbus, Ohio. I think that we’re using that app. It will show all the registered transportation that’s available. The great thing is, just like you said, you put in where you want to go, it schedules the modes of transportation, you pay for it right there all at one time, then again, your phone is your registration to be able to take all those vehicles.

Ed Bernardon: Once this kind of technology is integrated into our lives and our streets, the possibilities become vast, the way we think about traveling and cars will change. For some, these changes will seem less extreme than others. Well, when we talk about technology, there’s always this focus on the software or the hardware, and sometimes people forget that in the end, it’s actually about the people, how technology and all of these innovations are going to make life better for people. The people that really matter are probably the ones that are going to be using this technology in the future, and I think you’ve had some interesting conversations recently, I believe with one of your grandchildren about what all this might be about when it becomes a reality sometime, oh, 10, 20 years from now.

Eric Gannaway: Yeah, and that’s the way I described it to him. I do have a grandson that has spina bifida, and he’s paralyzed from the waist down, so periodically, I will pick him up and take him someplace, but he spends most of his time when he’s traveling with somebody else taking him around. One day, I was taking him to school and I said, “Quinn, it’s not going to be that long before you’ll get on your phone and you’ll call a vehicle with an app, and it’ll pull up out front. You’ll get in, and it’ll take you wherever you want to go.” We had some conversations after that and I said to him, “What do you think about, how do you think about a vehicle pulls up, you get in, and you watch a movie or play a game and you go wherever you want to go?”

He says, “That’s what I do every day when I get in the car. I get in, and it takes me wherever I want to go.” It made me think of the generation that’s coming up. These kids spend all their time doing something else. They get in, and the vehicle takes them wherever they want to go, so it may not be that big of a stretch for them to call a vehicle on their app and have it come because somebody is driving them around all the time anyway.

Ed Bernardon: It almost seems like the future, like you’re saying, there’s not that big of a change for them, so it’s only going to impact the change. It’s only going to happen to the people that are used to driving, not the ones that are the passengers.

Eric Gannaway: No. I think that’s exactly right, and my thought started to head down the path of, “If we’re looking at vehicles and talking about the future and designing them, probably the ones that we should be talking to are those people that are going to be using them.” And it is the 10-year-old and the 8-year-old. “What do you want to see in a future vehicle?” They’re the ones that are going to get in, get on their phone, watch a movie, and ride someplace, and maybe never get a driver’s license.

Ed Bernardon: This is one last question, and it’s probably the most important question I’m going to ask you. When your grandson turns 21 or graduates from school, from college, he graduates from college and you’re going to give him an autonomous car for a gift, what would be the features in that car? What would you put in that car so that he’s the happiest 23-year-old on Earth? What would it be?

Eric Gannaway: It would be easy access for him, that would probably be the biggest … That’s probably one of the biggest obstacles he has now, is having to transfer from a wheelchair into a vehicle, and then being able to get into a seat. It’s very difficult, so probably the access, comfort, the ability to navigate his destination, probably on the big screen. I’d like him to be able to get in if he needs to and type in on the screen the destination that he’s going to, and then have whatever form of entertainment he wants while he’s making that trip. It would be comfort-driven, I think for the most part for him, something that he doesn’t get to experience as much probably as other people do.

For more on automated driving and new mobility from Siemens, see solutions from Siemens Digital Industries Software and Siemens Mobility.

If you enjoy podcasts, you might also check out Talking Digital IndustriesIs Your Company Ready for Industrial-Scale Additive Manufacturing? and Joining the Dots (on smart infrastructure).

Leave a Reply