ITSA meetings to showcase transportation innovation in Detroit

The Intelligent Transportation Society of America (ITSA) convenes its annual conference in Detroit next week, so John Peracchio, a member of the Michigan Transportation Commission and a long-time participant in the ITSA, returns to the podcast to talk about what to expect.

In its mission statement, the organization “is a leading nonprofit uniting government, industry and academia to champion policies and investments that make our transportation systems safer, more innovative and more efficient. We work with agencies and industry leaders to advance technology-driven solutions that improve how people and goods move - enhancing quality of life, supporting a thriving economy and maintaining America’s global competitiveness.”

Peracchio talks about what that means and how ITSA's work can ultimately reduce crashes.

Transcript

Why Detroit Hosts ITS America

Jeff Cranson 0:07

Hello, welcome to the Talking Michigan Transportation Podcast. I'm Jeff Cranson. I spoke today with John Peracchio, who is a member of the State Transportation Commission here in Michigan, has been a frequent guest on the podcast, and I especially wanted to talk to him this week because of his years-long involvement in intelligent transportation systems and specifically the Intelligent Transportation Society of America, known as ITSA, which is convening its annual conference in Detroit next week. John has a lot of insight gathered because of a lot of curiosity and all things that involve transportation, but he's especially been plugged into what ITS does in terms of technology and innovation and everything to make it safer. The whole idea is that by getting people together from regional equipment manufacturers, the auto industry in general, researchers, academia, and certainly government agencies, that people can find out what's working and that can lead to better technologies and better outcomes. So John had a lot to say about all those things, where we're at with the development and deployment of connected and automated vehicles, and some examples of things that are working in terms of technology and that kind of communication and collaboration. So I hope you enjoy the conversation. John Peracchio, welcome back to the podcast. I always appreciate your insights. Um you're a frequent guest, but that's because you're into so many things.

John Peracchio 1:48

Um I appreciate that, Jeff. If we keep this up, I'm gonna get myself an agent.

Jeff Cranson 1:55

And uh maybe you'll get into the Five Timers Club because I think you've reached that milestone.

John Peracchio 2:00

So wow, wow. Is that like uh, you know, if you're a frequent host on Saturday Night Live, they do a special on you.

Jeff Cranson 2:08

I'm still working on a design for the robe. So look forward to that. Great. So I wanted to talk to you today because um your immense background and just curiosity about everything related to intelligent transportation systems and the fact that the Intelligent Transportation Society of America is convening their annual conference in

Progress And The 5.9 GHz Fight

Jeff Cranson 2:30

the motor city next week. And um I wanted first just to get the mission out there and then have you talk about that. And that's what they say is for 35 years, the Intelligent Transportation Society of America has been a leading nonprofit uniting government, industry, and academia to champion policies and investments that make our transportation system safer, more innovative, and more efficient. I think we can all get on board with that mission. Um how would you say uh I guess I tell me without critiquing it, um, how are we doing, not just ITSA, but all of us toward that goal?

John Peracchio 3:09

Uh so I I think we've made significant progress. Um we've had some major hiccups, for example, uh with dedicated short-range communications uh using the 5.9 gigahertz spectrum, where the uh Federal Communications Commission basically pulled a rug out from under us. Um and so we had to pivot uh into other ways to enhance vehicle connectivity. Uh so it's it's all about uh vehicles and and infrastructure communicating to uh enhance uh safety and to do what the ITS America mission uh really states, uh make our transportation system safer.

Jeff Cranson 3:49

We talked about that a lot at the time. In fact, I remember um having people on from the University of Michigan Transportation Research Institute to talk about the 5.9 gigahertz issue. Could you uh, you know, kind of briefly explain the fight there?

John Peracchio 4:03

Well, it was a fight. We were trying to maintain the use of that spectrum for traffic safety, uh, in short. And um the pressure, though, uh from the telecom industry to grab more and more spectrum because of the data needs that they have collided, uh, you know, no pun intended, with the traffic safety uh opportunity. And so, you know, the dollars were on the side of the telecoms, not on the side of traffic safety. Uh, hence the FCC took the position that um, you know, we should we should open that spectrum up to other uses. And to be fair to the telecoms, you know, we've had access to that 5.9 gigahertz spectrum for more than a decade. And really, except for General Motors and Cadillac, there was no broad deployment of dedicated short-range communications on the vehicle side. And we did have uh some deployment on the infrastructure side. Indeed, Michigan was a leading adopter of uh dedicated short-range communication roadside units, which were critical to having this all work. Uh, but you know, beyond that, uh it just kind of stagnated. And so, at that point, the FCC kind of threw up its hands and said, okay, telecoms, you can you can have uh really most of the spectrum.

Jeff Cranson 5:37

I don't think you'll ever lose a drinking game that involves uh every time someone says to be fair to the telecom industry.

John Peracchio 5:46

Well, okay. I mean, we we need them and uh you know they uh they know where we are at all times.

Jeff Cranson 5:53

Yeah. So talk about uh I want to get into how this relates to connected and automated vehicles and how um that's become some of the focus in recent years, but that's not all of what ITS does. But first, just talk about your personal involvement and what's provoked your interest and curiosity

John’s Path Into ITS

Jeff Cranson 6:10

in all of this.

John Peracchio 6:11

So um I worked for a uh major tier one automotive supplier uh for 10 years, um, uh Harman International Industries. And at the time we were just beginning to develop and deploy uh navigation systems using uh highly uh defined map data. And uh when I was at the uh at Harman, I I thought about this uh deeply. I thought, man, this is a huge opportunity to go beyond just navigation systems, but to also uh talk about other kinds of uh information that can flow between a vehicle and infrastructure. And ITS America attracted me. Indeed, I had Harman join ITS America at the time because it was the only organization that bridged that public-private um uh gap, if you will, in terms of uh promoting research, development, and most importantly, deployment of um uh technologies that could make the roadways safer and more efficient. Uh and then uh I started to think about other modes of travel, which you know became uh you know something of a um obsession for me to probe into things like maritime and uh and aerospace, because of course we're in a 3D world. And so transportation transcends all of those uh spectrums.

Jeff Cranson 7:45

So really it's just about everything that moves. It pretty much interests you.

John Peracchio 7:49

Indeed.

Jeff Cranson 7:50

Yeah. So I when I came to the department in the uh early 2010s, um, this was uh a nascent thing, the connected and automated vehicles thing. Um Michigan uh was sort of in a in a lead in terms of looking into it, in part because uh major university um had a center that was looking at it, and because the MDOT director at the time, Kirk Steidel, had a particular interest in it and was a speaker every year at the Center for Automotive Research Management Briefing Systems uh meetings in Traverse City. And I remember him joking at the time that when that session started, it was basically a phone booth, and over the years it grew into uh a big room. So I came to believe that that's what brought me to ITS. And while that's really important and there's

Michigan’s Framework For CAV Testing

Jeff Cranson 8:40

a lot we can talk about there, you made the point to me that that's that that shouldn't at all define everything ITS does.

John Peracchio 8:47

Well, but Michigan, I you know, uh to be fair to um uh Kirk Stoydle, uh, who along with um our legislature and governor at the time, Rick Snyder, uh created M City, which is what you're referring to at the University of Michigan, which was absolutely the leading edge of research and development on connected and automated vehicles. And um, and perhaps we'll get into this a little bit later, but I just want to point out, you know, Public Act 332 of 2016 created the uh old Council on Future Mobility here in Michigan, but more importantly, it created a framework for testing uh connected and highly automated vehicles on public roadways actually operating well correct and in it with very little restriction. Okay. Uh and and I thought, you know, at the time this was groundbreaking in terms of um what a state could do to promote the these technologies.

Jeff Cranson 9:49

Yeah, I remember having to correct uh many people, lawmakers and and media uh outlets at the time, because Matt Smith, who then headed up ITS within MDOT, was adamant that what made this groundbreaking legislation was actual operations.

John Peracchio 10:05

So well, and also with you know, back to the my point about very little restriction. I mean, you know, comparisons are odious, but if you look at how California handled this, you know, they, you know, for example, uh the historically the California DMV has required disengagement data from operators of highly automated vehicles on their roadways in in California. And frankly, it's like giving a gold watch to a monkey because they don't have the ability to analyze that data. All right, they just compile it, um, and it's actually not even uh publicly available because of non-disclosure agreements. And so that type of it to be honest, I think useless requirements uh we don't have here in Michigan.

Jeff Cranson 10:54

Thinking about that that pioneering legislation and everything that's gone on, and I use CAVs as an acronym for connected and automated vehicles, emphasis on automated, not autonomous. And I just like you, I bristle when people say autonomous, because that's very misleading for what we're talking about. Um but so all kinds of conferences, OEMs, academia, others, uh increased research into

Why Level 3 Is So Hard

Jeff Cranson 11:20

this. So, where do you think we stand? And and what are the many reasons we're not all driving highly automated vehicles? Is some thought we might be 15 years ago.

John Peracchio 11:28

You know, bluntly, the technology is not mature enough. Um, you know, there's promise there, uh, but it's on on both sides of the equation. On the vehicle side and the infrastructure side, we need to do a lot more to ensure safe operation of highly automated vehicles. And I think that's why you really don't see car companies uh heavily focused on uh SAE level three automation, where um, okay, the vehicle in some places can kind of drive itself, steer itself, uh, but at any moment you're going to need to take over. The system will tell you, oh, you need driver to human driver to take over. And um, and that's problematic because um, you know, people it takes uh people's uh time to adjust cognitively to the need to start driving when they were maybe reading a book. Okay. And so level three is very scary, and I'm not even gonna touch the potential product liability aspect of this, or if there's a traffic violation when in level three, is it the system, the vehicle itself, that's responsible, or is it the human driver? Those are issues that still need to be uh worked out before we're going to have uh deployment uh of highly automated at the higher SAE levels of automation at scale.

Jeff Cranson 13:05

I understand the marketing and branding decision, but I've always questioned whether it was really smart of Tesla to use the term autopilot right out of the gate. Or or full self-driving. Yeah, yeah. Because that applies, like you said, that you can read a book or take a nap, right?

John Peracchio 13:21

Right. And you could wind up in a ditch or wrapped around a telephone pole. So, I mean it's really that I think is uh highly

The Trolley Problem And Edge Cases

John Peracchio 13:30

prop problematic, uh, that type of marketing.

Jeff Cranson 13:33

That sparks a thought that uh is off of my show map, but because you and I have discussed this before and I think you and I love these philosophical dilemmas. Um talk about the trolley car uh dilemma and how that relates to trying to train machines to drive force.

John Peracchio 13:51

Right, right. I love the trolley um uh car example because you know, are you gonna and I'll you know change it a little bit. Are you going to uh you know hit a school bus full of children or your grandmother? Uh that's your choice. Uh and or someone else dear to you. And the system uh can only uh permit one course of action. And that's that is the the real dilemma here. Is it are humans uh good drivers? Well, the data doesn't seem to support that uh assertion much. Uh but are we ready to be able to do that? Exactly. But are we ready as a society to say, okay, um robot vehicles are are truly going to be safer? Um and you know, I follow people like um Professor Emeritus uh Koopman of Carnegie Mellon University, uh, who is very pragmatic, I think, in pointing out the deficiencies of highly automated vehicle operations. And then if you if you um uh listen to or read uh what uh has been published by Missy Cummings at George Mason University, she's very far uh into the criticism of car companies who are deploying even level two plus systems today without sufficient um uh research and development and testing.

Jeff Cranson 15:26

When when this concept first emerged, it was uh, you know, in in the olden days of trolley, and the idea was that an operator would have to make that decision between uh a crash that kills or injures many versus uh maybe just one. Um they weren't really contemplating AI then. So um in an AI world, you know, is there a way? How does somebody train a computer to make that kind of decision?

John Peracchio 15:54

Well, the the the challenge is what uh what the experts, and I'm certainly no expert, uh, call edge cases, right? It's those edge cases that you very difficult to program a system to handle. Um and uh you know AI has has its own challenges, indeed, demons, I think, where you know, um sometimes when when the AI system can't give you an answer that's grounded in fact and reality, it makes something up, so-called hallucinations. Now, do you want hallucinations driving your automated vehicle? I don't think so. So we have to we have to really take a step back and say AI can offer a lot of benefits, but it also requires uh monitoring and uh careful assessment before we deploy it at scale.

Michigan Hands-Free Law Reminder

John Peracchio 16:47

We will continue the conversation right after a quick break.

MDOT Message 16:50

In Michigan, safety comes first on the roads. To combat distracted driving, Michigan passed the hands-free law. The law makes holding or manually using a cell phone or other mobile electronic device while operating a vehicle a primary offense. This means an officer can stop and ticket you for violating the law. This can include, but not limited to, sending or receiving a call, sending, receiving, or reading a text or email, accessing, reading, or posting to social media sites, or entering locations into the phone GPS. With this law in place, drivers are encouraged to stay focused, keeping their hands on the wheel and their attention on the road. Drive smart, drive safe, drive hands-free. Michigan's hands-free law, making the road safer for everyone.

Collaboration Beats Transportation Silos

Jeff Cranson 17:41

Sounds like the whole idea of ITSA was so that everybody wasn't operating in a silo, that you've got um private industry and government and academia, um, sharing information, sharing, you know, I mean, I mean, obviously, with all the cautions about what's proprietary, but at least speaking the same language, you know, finding out. I I'm a big believer in borrowing from others that everything is derivative. And uh the first question anybody at MDOT will ask when something comes up is well, what are the what are the states doing? That's what I always want to know. Like we might not have to, you know, reinvent anything here. So that is probably the best of ITSA. Is that fair to say?

John Peracchio 18:23

It's a it's a completely accurate um uh summation. Uh, I mean, collaboration among private sector folks and public agencies uh is really what drives ITS America to um back to its mission, back to achieving the objectives that the organization has set out for itself.

Jeff Cranson 18:44

So talk about uh some examples. You mentioned something to me that uh the Ohio DOT is doing with Honda, which has a major presence in Ohio, obviously. Um talk about that and some others.

Ohio DOT And Honda Real-Time Data

John Peracchio 18:56

Right. So the really fascinating collaboration between uh Ohio Department of Transportation and Honda on using existing vehicle centers to uh um aggregate in real time road data. I mean actual uh information about potholes, about uh crashes, about uh really everything that's happening on the roadways, and uh using a software package to uh manage the flow of data and to aggregate it, uh assemble it in a useful way, and then disseminate it so that uh Ohio DT DOT can take uh action uh sometimes immediately if there's if there's a sufficient um uh emergency situation on a roadway. And I think that's a great example of how the private sector can collaborate with a public agency to enhance uh safety and improve transportation efficiency. Um and other examples are out there where uh state DOTs, tolling agencies, uh collaborate with the private sector uh to do the these the same thing.

Jeff Cranson 20:14

I mean, ultimately that's the goal here, right? To reduce crashes. So, if you're sharing that information, um I think you talked about even weather-related hazards. Um all of that is designed to reduce crashes. I always feel like as much as we develop new technology, we develop new ways to distract ourselves and still cause crashes. Um, so can we can we uh

Distracted Driving And Crash Avoidance

Jeff Cranson 20:40

can we win? Can we get on the positive side of that?

John Peracchio 20:43

Well, I think I think for the most part, law is on our side already. So, um most states have um you know distracted driving rules where you're not supposed to be able to, you're not supposed to text while driving enforcement is you know challenging. Uh it's hard to enforce those laws. I think that um uh the vehicles themselves are going to be um designed to detect distracted driving as we move forward. And that will help whether it's haptics or an alarm or something, uh, because we already have crash avoidance technology deployed in vehicles. And so I think that that is a trend we're going to see that that doesn't um necessarily require uh you know uh regulation, although it may come to that, because a lot of states are and the federal government uh right now is considering mandating crash avoidance systems. So um, you know, it could be come like seatbelts.

Jeff Cranson 21:47

I should look this up, but uh maybe you have some sense of this. But um I know that when we first started talking about this uh years ago, if I asked for a show of hands in a room about An automated vehicle, people would shake their heads and say, like, no way. And I've said, Are you okay with auto braking? And they'd say, Yes. Are you okay with rear assist? Yes. Are you okay with lane assist? Yes. Backing camera. Yes. So all those components are what ultimately make for an automated vehicle, right?

Level Two Plus And Privacy Reality

John Peracchio 22:19

Absolutely. Maybe another one you could mention in the same uh sentence is um location, uh, where you are. Uh and uh so people are very nervous for privacy reasons about uh having a vehicle that they're driving, um, you know, having uh the infrastructure know exactly where you are. But of course, you know, if you own a uh smartphone, uh they always know where you are. And if it's I mean if it's on. So it to me, it it's like uh we again the forest from the trees. To your point, you add all these technologies up, and it's very close to a highly automated vehicle. Uh and you know, we have now level two SAE level two plus. Yeah, explain what that means. So it's it it's basically um uh what's a good example? Maybe the uh GM cruise system, where for certain uh roads that have been um carefully analyzed by the um the system, uh you can take your hands off the wheel, okay? And this can go on for as long as you are those roads that are a part of the system, right? But once you get off those roads, you have to you know uh take over control of the vehicle. Um, but they're mapping more and more roads. And so the plus part is that this is an expanding universe of roadways where you can you know operate your vehicle without having hands on the steering wheel.

When Robot Cars Get Stuck

Jeff Cranson 24:01

So maybe to uh twist the cliche a little bit because it's been more incremental than we would have hoped 15 or 20 years ago, it's uh a half step forward, a quarter step back. I mean, we're making some progress, but then you know, you mentioned the story and I saw it, and I think anybody who's interested in this did of the Waymo's uh encircling a cul-de-sac near Atlanta. Um that's gotta be bad for PR.

John Peracchio 24:28

It's terrible. It was in an article in the Wall Street Journal. I mean, but also, you know, when they had a power outage in San Francisco, the vehicles just like stopped, right? You know, and it caused all kinds of uh problems. But the cul-de-sac um example is you know, it thankfully no one was hurt. So I think we could be mildly amused by it, but it really messed up a neighborhood for hours and hours on end. But that's why urban planners would say we shouldn't have cul-de-sacs. Well, there you there you go. That's right. And you know, I just uh conjured up uh roundabouts where the automated vehicles just go in circles forever, and there's no good explanation for that. Uh and um Waymo, uh, you know, to take a take a crack at them, they really have not been transparent about why that happened. And that's the kind of thing that Professor Koopman and Professor Cummings uh really latch on to as a as a failing. You know, we need more transparency about how these things uh function. And when they don't function, okay, please come up with a plausible explanation and don't just run and hide. And at the end of the day, NHTSA and the NTSB, they're going to demand that type of information. Uh, whether it ultimately becomes public, I don't know. But clearly uh they're doing themselves a disservice by not being completely transparent.

Jeff Cranson 25:59

So as you go into this conference and prepare to have conversations with lots of people from, you know, all over the continent, I guess, and beyond. Um, how do you feel? Do you feel uh cautiously optimistic or bullish? Or what's the future of technology and its ability to help reduce crashes, which is the ultimate

Accessibility Tech That Changes Streets

Jeff Cranson 26:20

goal?

John Peracchio 26:20

Yeah, so I am optimistic, and indeed, um uh sort of a precursor to the ITS America event in Detroit um next week was an ITS Canada annual conference and expo in Edmonton a couple of weeks ago. Uh Mishka DOT had a strong presence there. And uh what was fascinating was uh, among other things, is um technologies to keep uh disabled pedestrians safer. A really interesting set of systems uh that help uh folks who are visually and hearing impaired cross uh intersections safely using a range of technologies, indeed, using smartphone applications. Uh, and then there was another display of technology. Actually, our uh Michigan DOT director, Brad Wieferich, tried it out. It was a virtual uh planning tool to allow non-disabled folks to experience what it means to navigate a city street as a disabled person in a wheelchair. And uh it was it was absolutely fascinating. So, Director Wieferich um, you know, was and we could see what he could see, right? And he managed to crash into a couple of walls, he killed a couple of bicyclists while he was in his wheelchair. And I mean, it was humorous, but a huge opportunity in terms of planning what I call the curbscape to take care of um uh our most disadvantaged travelers who are disabled, uh, because if you can take care of them, you can take care of anyone. So it's a long-winded answer to your question about being optimistic. I'm optimistic if we can come up with those types of technologies to take care of um of those types of folks, uh I think we're on the right track.

Jeff Cranson 28:20

Well, that's really well said. And I know your friend and mine, David Bukowski, who directs the disability advocates of Kent County. Um, I've talked to him often about those kinds of simulations and things that you can do to, if not put us in their shoes, put us in in their wheels and help understand what that's like. And hopefully it gives you an appreciation as a driver, um, as an able-bodied driver or pedestrian, um, what hurdles they face. So, yeah.

John Peracchio 28:49

Absolutely. I mean, it's interesting. So, for the hearing impaired, I just give you one quick example. Uh hearing impaired people sometimes can hear certain sounds at the lower end of the spectrum. And so one of the technologies actually produces an alert at the lower end of the spectrum so that so that the uh the um uh hearing impaired person can actually hear something and know it's not safe to cross the street versus okay, it's fine now.

Closing Thanks And Where To Listen

Jeff Cranson 29:22

That's really interesting. Well, John, thank you as always. Um looking forward to uh seeing you in Detroit next week. Should be really interesting and uh be interested to hear your thoughts afterwards. Thank you very much for having me. I'd like to thank you once more for tuning in to Talking Michigan Transportation. You can find show notes and more on Apple Podcasts or Buzzsprout. I also want to acknowledge the talented people who help make this a reality each week, starting with Randy Debler, who skillfully edits the audio, Jesse Ball, who proofs the content, and Jacke Salinas, who posts the podcast to various platforms, and transcribes the audio to make it accessible to all the time.