Episode 10. Shaking Up the System: Using disruptive technologies to create safer roads.

Bernie Fette (host) (00:15):
Welcome to Thinking Transportation — conversations about getting from point A to point B and the many things that can happen in between. I’m Bernie Fette with the Texas A&M Transportation Institute. When people say that something is disruptive, they typically don’t mean it as a compliment.

Bernie Fette (00:34):
A disruption is an interruption to an activity or a process. And we teach our children that it’s wrong to interrupt people when they’re speaking, don’t we? In the business world, on the other hand, disruption has been recognized for its positive attributes for more than 20 years. Much more recently, the idea of interrupting an activity or process, particularly in the service of saving lives, has gained traction in helping improve roadway safety. That’s the topic in our visit today with Sue Chrysler, senior research scientist at TTI. As an expert in understanding driver behavior, Sue’s work is focused on using disruptive technologies to make roadway travel safer for all of us. Welcome, Sue, and thank you for being here.

Sue Chrysler (guest) (01:24):
You’re welcome. I’m looking forward to it.

Bernie Fette (01:27):
All right. So, going straight to the name of the center that you’re involved with, whenever we say “disruptive” in the context of traffic safety, then disruptive is a good thing?

Sue Chrysler (01:40):
We think it could be a good thing. I mean, the old ways of dealing with traffic safety are still resulting in over 35,000 people dying in vehicle crashes every year. So disruptive technologies like automation and connectivity and new sources of traffic data could improve on that. And so, our center focuses on research on how those types of disruptive technologies can be used to improve safety.

Bernie Fette (02:07):
The examples that I heard you give didn’t involve having to build anything, which is what I think for quite a few years, what we thought about at least in part, whenever we talked about safety, particularly roadside safety. The approach that you’re taking by harnessing technologies in a more productive and less disruptive way — that would also save money then, right?

Sue Chrysler (02:29):
Yes. It could save money in the long run because the kind of infrastructure improvements that this technology needs are things like broadband internet access and other forms of connectivity through fiber optics or other items like that. And so there will be the need for some initial improvements, but the maintenance of those could potentially be less expensive compared to the more traditional guardrails and other.

Bernie Fette (02:56):
Steel and concrete?

Sue Chrysler (02:57):
Steel and concrete, right.

Bernie Fette (02:59):
Yeah. I’d like to come back to the broader infrastructure issue in a few minutes, but sticking for a moment with that contrast and comparison with the — I guess what we might call the more traditional approaches to safety — in that more traditional approach for quite a few years, agencies would install guardrails in a particular place where there’s a history of crashes, for instance. So there’s an assumption that there will be more crashes based on that pattern. Is the research of the variety that you’re doing based on certain assumptions too?

Sue Chrysler (03:35):
Yeah. I think that the trend in safety improvement just even in the past few years, has been toward a more, we call it systemic. So it’s a system-wide thing. So instead of putting a guardrail where there’s a history of crashes, an agency might in a more systematic manner, identify roadside features that warrant or merit consideration for treatment. So, something where there’s a culvert or a ditch of a certain angle that could pose a hazard. So there may not have ever been any crashes at that location, but experience has shown that, you know, a car going out of the lane in that location might be more likely to leave the road. So, if you take the disruptive technology piece to that in the future, we might have the guardrail have some sensor in it that could notify law enforcement or the agency that it’s been struck. So right now somebody might hit it and just drive away. And nobody even knows that it got hit. So things like that are one way to bring this disruptive technology to the more traditional safety features.

Bernie Fette (04:44):
The example you just gave makes me curious about the sensors, for instance, in a guardrail. Could those same sensors sense when a vehicle came really close to the guardrail, even if it didn’t hit it and therefore, you know, you’ve got a different type of data that might be useful to the people who are keeping track of conditions at that location?

Sue Chrysler (05:05):
Yeah, exactly. And that kind of data that’s for proximity for sort of a near miss can be really helpful to feed back into that systematic safety analysis, to know how frequent are people getting close to that. Even some really basic things about how frequently people go out of their lane — that underlies some calculations that the engineers have to do about whether or not to install a guardrail or how big, and it might, you know, a sensor might be able to tell what type of vehicle it was as well, that sensor could also be on the vehicle, not on the guardrail.

Bernie Fette (05:44):
Yeah, and you’re actually touching on a topic that we had in our first episode. I think it was when we spoke to Shawn Turner and Eva Shipp about near misses and how connected vehicle technology can help in a — in a predictive way — know more about where crashes might be likely to happen, even if they haven’t already started to happen. This sounds like it’s very closely related to some of the work that they’re doing.

Sue Chrysler (06:06):
Yeah. And actually our Safe-D University Transportation Center is providing funding for some of that work.

Bernie Fette (06:14):
Ah, okay. Another traditional safety thing that I know goes farther into your background, traffic signs, the focus of some of your earlier work. I wonder if there’s some people, maybe a lot of people who think, okay, it’s big, it’s red, it says “stop.” Could this be any simpler? But there’s a lot more than meets the eye with that. And you’re really thinking about human responses there, which is what are often not simple at all. How did some of the work that you were doing before influence or inform what you’re doing now?

Sue Chrysler (06:51):
I think things that we can do to improve sign visibility or sign legibility for humans will also benefit machine vision systems. So, a lot of new cars, even sort of as standard equipment, have traffic sign recognition software. So, the forward facing camera. So, if you see a little speed limit icon on your dashboard, it’s not getting that from a map. Often, it’s getting it from the system has spotted the traffic sign in this environment.

Bernie Fette (07:20):
Right there on the roadside.

Sue Chrysler (07:20):
Yeah. And it’s not perfect. And it misses some, I drove in this morning and I noticed it missed one, but things like having the sign big enough, having a border around the sign helps cameras identify that shape. But it also helps people identify that shape both daytime and especially at night. So that’s why there’s a white border on a red sign. ’cause at night that red isn’t very bright, but that white octagonal border, it stands out against the dark sky.

Bernie Fette (07:48):
Yeah, I get it. It’s always been about visibility in the work that you’ve been doing. It’s just that you’re concerned now about the visibility for cameras, not just the visibility for the human eye.

Sue Chrysler (07:58):
Right. It also points to the really increasing need for the uniformity of traffic signs, you know, across the country and possibly across the world to train these machine-vision systems, to recognize that that’s a stop sign. Stop signs always have to be red and octagonal.

Bernie Fette (08:14):
Right.

Sue Chrysler (08:14):
So, you know, you might go in some shopping center and they have some cutesy little stop sign. People don’t recognize that as a stop sign, you know, a machine-vision system certainly wouldn’t recognize it as a stop sign.

Bernie Fette (08:27):
And you know that because of the research you’ve done, right?

Sue Chrysler (08:29):
Right. And the research that our center has sponsored. I am not a machine-vision expert, but I have certainly talked to a lot of them.

Bernie Fette (08:37):
When we talk about human responses, which is part of the issue there with visibility. This is where your training as a cognitive psychologist really finds some direct application, I guess.

Sue Chrysler (08:48):
Yes.

Bernie Fette (08:48):
And that places you in a somewhat exclusive group, meaning that you’re not an engineer, like the majority of the people with whom you work. In fact, the profiles of people who work in traffic safety are a lot more varied than they were several years ago. Why would you say that that’s important? What does it make possible that otherwise might not be possible?

Sue Chrysler (09:09):
I think having input from people from behavioral science, writ large — psychology, sociology, public health — it helps contribute to understanding people’s behavior, motivation, how to possibly change that behavior. And I think the biggest thing is that it brings an appreciation for the wide variety of people. Often, I’ve seen it where someone with a more strict engineering background assumes that everyone’s going to behave a certain way and they’ll say, well, you know, how do we fix that? And I said, well, it depends. I say “it depends” a lot, because it’s going to depend on how the person is and what state that person is. A person can be a conservative, safe driver sometimes, but then if they’re impaired by alcohol or tiredness or screaming kids in the back seat, they might not behave so rationally anymore. And so I think trying to design systems that can accommodate that wider variety of roadway users, um, is what sort of this interdisciplinary approach brings.

Bernie Fette (10:20):
Because none of these problems are simple. And so the solutions aren’t necessarily simple, either.

Sue Chrysler (10:24):
Right.

Bernie Fette (10:24):
Along with multi-disciplinary teams, you are also, it seems, trying to start growing new professionals from a very young age. Can you talk a little about how Safe-D concentrates on workforce development and those varied professions that the transportation industries need?

Sue Chrysler (10:46):
Yeah. I think it’s not like there’s a straight path to it. You specialize in some discipline, then, and apply that knowledge to traffic safety. So even at the university level, trying to present some of these ideas to classes, not just in civil engineering, but in industrial engineering, mechanical engineering, trying to get people that are interested in automation and robots to think about applying that to vehicles.

Bernie Fette (11:12):
And cognitive psychology, too, right?

Sue Chrysler (11:14):
And cognitive psychology, too. So, we definitely work with those disciplines and public health, quite a bit, trying to get them to see that there are applications beyond. A lot of people with training in human factors like me end up applying that to medical devices or seating — you know, ergonomic type things. And so, trying to get industrial engineering, human factors people interested in traffic safety. And I think it can start even at an earlier age.

Bernie Fette (11:46):
Right, right, because aren’t you doing some work that is really focused on K through 12?

Sue Chrysler (11:53):
Right, so.

Bernie Fette (11:53):
Kindergarten? Really? What does that look like?

Sue Chrysler (11:56):
Well, my colleague Melisa Finley has done some great — truly for kindergarten. You know, it can be things as simple as coloring books that have traffic signs in them and talking about — you know, we used to — we would do tours of our visibility lab, and then the kids would come in and we would just hold up a card that had all the different colors of traffic signs on it and try to get them to name signs belong to that color.

Bernie Fette (12:23):
To that color, yeah.

Sue Chrysler (12:24):
And then as they get a little older, we’ve developed working with a science teacher, a sixth grade science teacher, we developed an actual curriculum that teachers can use in the classroom that follows the national standards for science curriculum. So, it checks all the boxes that are required by the core curriculum for the physics section about optics. And so, we talk about retroreflective traffic signs as a way of illustrating fundamental physics ideas of reflection and refraction and filtering colors. And so, they do little exercises with flashlights and tinfoil and marbles and make themselves some traffic signs. And we’ve had some schools even build little cars out of cardboard and put flashlights on the front like headlights so they can illustrate their traffic signs. So, that curriculum is available on the TTI website for free for any, any science teacher.

Bernie Fette (13:25):
This sounds like it might be one of the more enjoyable and entertaining parts of your job. That sounds kind of fascinating.

Sue Chrysler (13:31):
Yeah, the kids get so excited. Like I always imagined the kid going home that night telling their parents — ’cause I think the kid probably knows more about how traffic signs reflect light than the parent does at that point, because most people don’t know that; they’ve never really thought about why they can see traffic signs at night — but I can just see some excited 12-year-old at the dinner table explaining about the little tiny beads that are inside traffic signs.

Bernie Fette (13:55):
Yeah. And what you want to do is have them go home and tell their parents, “when I grow up, I want to be just like Sue Chrysler.”

Sue Chrysler (14:02):
Or at least, gee, I guess I ought to pay attention in science class, ’cause I could be doing something cool like that.

Bernie Fette (14:09):
There you go. I’m going to ask you to step back to the conversation we were having earlier about infrastructure. I mean, infrastructure is again kind of top-of-mind in terms of policies for the nation. The discussion has kind of been jump-started. How should we be linking roadway safety in the Safe-D context to the more expansive national discussion on infrastructure needs?

Sue Chrysler (14:36):
Thinking about infrastructure beyond just concrete and steel is really the first step. So that infrastructure can be data-handling systems. There’s just endless sources of data now from smartphones and location services and camera surveillance that can automatically classify vehicles. And so, the infrastructure that an agency needs to be able to store that data and process that data and turn that data into information is often overlooked. And so again, kind of going back to the workforce, people that are in computer science and other data science areas may not think about applying that to transportation. So yes, you know, we could wire a guardrail or put other sensors out there that type of digital infrastructure, but even sort of the processing of that data is part of the infrastructure discussion.

Bernie Fette (15:29):
And, as we were discussing earlier, that part of the infrastructure puzzle is in some ways more affordable than the infrastructure that’s that’s out there now.

Sue Chrysler (15:37):
It could be, yes. I think there’s going to need an initial investment to upgrade, especially in rural areas.

Bernie Fette (15:45):
Right. So it’s kind of front-loaded.

Sue Chrysler (15:46):
Right.

Bernie Fette (15:46):
You have to focus on the things that are immediately in front of you with your research program, that I know that you’re bound to be thinking about emerging needs, too. Maybe even things that we’re not yet talking about. What are the biggest questions lurking on the research horizon for you?

Sue Chrysler (16:05):
I mean, I think working in this area of disruption that the USDOT has, you know, provided funding for our University Transportation Center to look at these kind of be forward-looking technologies. I think the other area that our center touches on that we haven’t talked about much is transportation as a service. So things like ride shares, shared bicycle, dock-less scooters, things you see in urban areas. So, how do devices like that, how do we change the urban landscape to accommodate, be more pedestrian-friendly and have bikes and scooters and pedestrians interacting safely with vehicles of all sizes? I think that’s going to become more of an issue certainly as people return to the cities after the pandemic. But I think the future of vehicle ownership is in question, as people think more about shared vehicles or when way in the future, vehicles that you may not own, that you just summon and that they’re automated and they just show up like a taxi and then drop you off and go away and you don’t have to park. So when I … what do we do with all of these parking garages as we don’t need to park our cars anymore?

Bernie Fette (17:16):
Right. You have no shortage of things to work on, it sounds like.

Sue Chrysler (17:22):
Right. It’s not quite science fiction, but it’s fun to think about it.

Bernie Fette (17:26):
But it’s close enough?

Sue Chrysler (17:26):
It’s close enough, right. We aren’t to flying cars yet, but there are people thinking about that, too.

Bernie Fette (17:33):
Well, you know, we have to save something for a future episode.

Sue Chrysler (17:36):
Right, right.

Bernie Fette (17:36):
Last question. I don’t know how long ago you decided, oh, this is what I want to do. And I don’t know if that has any connection to the fact that your last name is “Chrysler.” I don’t know — you probably get a lot of teasing over that, but seriously, why do you do what you do?

Sue Chrysler (17:55):
I like to solve problems. And you know, I certainly entered school, never intending to work in traffic safety. And I somewhat stumbled into it my last year of graduate school. But having an effect on people’s well-being is rewarding. I like being able to point at something tangible and say, oh, I helped do that. One of the projects I was involved with at TTI 15 years ago was looking at installing pavement markings that were the color and shape of an interstate shield. So, right on the highway, you’re coming up say in Houston, there’s a great big interchange and lots of highways are meeting and you don’t know which lane to be in. And there’s six or seven lanes of traffic and the signs are overhead. And you’re not really looking at the signs because you’re too busy trying to look ahead of you. Well, what if we just slapped thermoplastic onto the road to help you? And it worked great. And I was in Houston recently and saw them and said, “oh yeah, they’re still there. I did that!”

Bernie Fette (18:57):
That’s cool. And we’re glad that you did that. Thank you for sharing your time. This was just as fascinating as I was expecting it to be. Thanks very much.

Sue Chrysler (19:05):
Thanks, Bernie.

Bernie Fette (19:06):
Sometimes solving a problem begins with the thing that caused it. When someone is bitten by a poisonous animal, the person can be treated with antivenom made from the actual venom of the animal in question. Roadway safety can work a little like that, too. Using a cell phone while driving can be distracting and potentially deadly, but the same technology that makes that communication possible can also help prevent vehicle crashes. It all depends on how the technology is applied.

Bernie Fette (19:40):
Thanks for listening. We hope you’ll be back next time for a conversation with Tim Lomax, a senior research engineer at TTI, and for more than 30 years, a co-author of the Urban Mobility Report, considered by many to be the most definitive periodic study of America’s growing traffic congestion problem. As Tim will explain, the pandemic helped us hit the pause button on roadway gridlock, but the traffic jam hiatus did not last long.

Bernie Fette (20:09):
Thinking Transportation is a production of the Texas A&M Transportation Institute, a member of the Texas A&M University System. The show is edited and produced by Chris Pourteau. I’m your writer and host Bernie Fette. Thanks again for listening. We’ll see you next time.