March 5, 2024 • 25 mins
Today, on the Inside Track, we sit down with Gary Passon, the AI Racing Team Principal and leader for the University of Hawaii, to discuss how a school known for Oceanography became a leading name in robotics and engineering.
The small but mighty robotics and engineering team at the University of Hawaii Manoa is not only competing at Las Vegas Motor Speedway but is beating some of the most advanced engineering teams in the world. Gary gives us insight into why the IAC is so important for the future of racing and how his team put the University of Hawaii on the global stage.
The small but mighty robotics and engineering team at the University of Hawaii Manoa is not only competing at Las Vegas Motor Speedway but is beating some of the most advanced engineering teams in the world. Gary gives us insight into why the IAC is so important for the future of racing and how his team put the University of Hawaii on the global stage.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
You know, there's a general lack of appreciation of how
difficult and challenging these problems are. And to think that,
you know, by a snap of a finger or a
line of code, the work that we need to do
to develop these products to the level they need to
to be, to really bring a level of safety that's
measurable as sustainable and sufficient is you know, is really
(00:20):
a niativity on their health.
Speaker 2 (00:27):
The world around us is changing faster than ever before.
Ideas once only imagined in science fiction are becoming a reality.
Throughout the course of our amazing twenty three episode season,
we'll speak to some of the greatest minds in robotics
and artificial intelligence to discuss the groundbreaking work that's fueling
it all. I'm your host Ryan Marine, joined me and
my co host Paul Mitchell, the presidents of the Indian
(00:49):
Autonomous Challenge, and see why we call this the inside Track.
A few things come to mind when people picture the
Hawaiian Islands, stunning views, incredible surfing, deli coffee, and ocean
as far as the eye can see. But did you
know that a small robotics and engineering team at the
University of Hawaii MANOAH has been not only competing at
(01:11):
Las Vegas Motor Speedway, but also has been beating some
of the most advanced engineering teams in the world today.
On the Inside Track, we sit down with Gary Passon,
the AI Racing Tech Team, principal and leader for the
University of Hawaii to discuss how a school known for
students specializing in oceanography is now leading in robotics and engineering.
(01:35):
Gary gives us insight into what makes the IAC so
much more than great code and how this team put
the University of Hawaii on the global stage. I'm super
interested in hearing a little bit more about your background
and how you got into this role. You come at
(01:57):
this from a different perspective than most. But first I
should say congratulations. Another podium finished for you and the team.
You were second in the competition at Texas in November
of twenty twenty two, then in early twenty twenty three
third place in Las Vegas. The team seems to really
be hitting its stride.
Speaker 3 (02:16):
Yeah, Ran, we are.
Speaker 1 (02:17):
We're just very excited to continue to build on the
program that we started three years ago. We knew this
was going to be a long term program and that
we would need to evolve not only the team, but
the software stack to come up and be able to
compete with the Europeans. I mean, those guys are bringing
an amazing capability and experience to the table, and we're
(02:39):
proud to be on that podium with them two times
in a row now, but we still think there's some
place yet to go, and that would be at the
top of it, so we'll continue to continue to work
on that.
Speaker 2 (02:50):
You're quite right that the European universities have been at
the forefront of this to this point, and I'm prepared
to be told I'm wrong with this, but institutionally, it
does seem like the way that they go about their
universities and their research is perhaps a little bit different
than most approaches you see from American universities. Is that
correct and what kind of challenges come with that? If
(03:13):
my assumption is on point.
Speaker 3 (03:15):
Well, you're exactly right on.
Speaker 1 (03:17):
There's probably two main factors that to date have given
them the edge, but I think that edge is eroding.
The first is that they were the early participants in
a competition called Robocar YEP, which was an Nvidia and
otherwise sponsored organism program for fully autonomous race cars, primarily
(03:39):
situated out of Europe, although to do some racing here
in the United States, and that allowed them to get
ahead started developing their stack for the specific application.
Speaker 3 (03:50):
So that was.
Speaker 1 (03:50):
Certainly an advantage, but not an enduring advantage because obviously
we're closing the gap. The second is that the governments
in your tend to financially support educational institutes that are
that are specialists, and so in the cases for all
three cases, for the European teams, and no disrespect, I
(04:12):
wish we were lucky enough to get that same kind
of funding opportunities, but they get They get a continuing
annual budget which allows them to have a number of
PhDs pre and post am. I doing that, they're able
to continue to evolve over a long term. A very
(04:33):
difficult and sophisticated problem which this is, this is a
very very very challenging thing, but something that I think
is very important ultimately to the safety of autonomous vehicles
on the road.
Speaker 2 (04:44):
By contrast, what constitutes your team? What are you working with?
Speaker 1 (04:50):
So but you know, by contrast, we're and I think
this is somewhat true for most of the most of
the American teams, but not all of them. We are
actually an algemy of the University of Hawaii as the
prime university supported by University of California, San Diego, University
of California, Berkeley. This year we brought in Carnegie Mellon University.
(05:12):
So by being able to aggregate the graduate programs and
the undergraduate programs for all four universities, were able to
put together a group of not only the talent that
we're looking for, but also the manpower that we're looking
for to do these kinds of problems. It's not something
that a programmer in a back room.
Speaker 3 (05:31):
Is going to do. This is clearly a big, big,
big program.
Speaker 2 (05:35):
What has led you to this point? And specifically I
bring this up because I have to believe you come
to be a university professor in a much different way
than most. So let's go back to the beginning, because
I think the beginning explains a great deal of how
we got to where we are today.
Speaker 1 (05:54):
Yeah, I'm not sure which is the chicken and which
is the egg here, but background is having been born
and raised in Indiana, comes up through go karts, and
of course Indiana being the home of the Indianapolis five
hundred is the uh, you know, the most important arguably
race in the United States. Uh So everybody in Indiana
(06:17):
is a very has a lot of race in their blood.
Speaker 3 (06:20):
Uh.
Speaker 1 (06:21):
And I was very lucky to be successful in go karting,
move then into amateur racing, uh, then move into semi
professional racing and ultimately was a professional race car driver
for a short period of time. So I was very
lucky to be our unlucky depend on how you want
to look at it. Uh to be of the same
era as Bobby Unster junior, Al Unser junior, Mario Andretti junior,
(06:48):
Bobby ray Hall. All those drivers came up in the
same era of wielded wheel both USAK racing in those
days for the mini in these or Formula land Takes
or or other series and that stuff. So we ran
around nose to tail a lot during those years. And
and some of those guys went on to full Indianapolis careers.
(07:11):
My destiny was not that I was always uh. I
always had my soul uh uh an engineering interest. Uh
maybe me. Mark Donahue was probably you know for those
for those a little bit older people out there, Hey,
the Mark Donahue, give.
Speaker 2 (07:28):
Up on on on the youth here, I'm thirty four,
and I know about the unfair advantage in Mark Donahue.
Speaker 1 (07:33):
Very good, very good, but not not a lot of
thirty well sub thirties. Well, sure, sure, I'll give you
an Matt you know is familiar with Mark, and and
also Bruce McLaren. Uh So, you know, I always saw
the engineering sign maybe even a little bit more of
my level of interest than than even the driving side.
Speaker 2 (07:54):
Both of them met untimely ins in race cars. So
there there was a dangerous time to be doing a
particular profession, and perhaps pursuing the engineering had a bit
of a longer runway for you.
Speaker 3 (08:07):
I think it did.
Speaker 1 (08:09):
You know, I'm not sure that that was in the
top of the you know, because you shouldn't drive race
cars at the limit if you consider uh the you know,
what can happen, uh, you know too fully, it was
a combination I think of a recognition that those were
dangerous times and I did have a friend of mine
die in a race car people, and so that, you know,
that probably had an impact on me as well as
(08:34):
the just just the fact that I found that racing
against the families of Indianapolis car drivers meant that when
it came time to divide up sponsorships and those kinds
of things. They had a tendency to go with name
and brand, and maybe a little less with some upstart
(08:54):
that you know, was more interested in the role anti
roll bar rate at the rear the car then maybe
they were in you know, being the race car driver.
Speaker 2 (09:05):
How though, then, did these experiences earlier in your life
almost come full circle to put you in this position
where you're working on an indie lights indie, next mini indie,
whatever you want to call it, chassis and the technology
that you're honing has safety in mind, safety for the road,
safety for racing drivers as well. It seems to me
(09:26):
that all of your personal experiences have flowed into this
in an almost perfect manner.
Speaker 1 (09:33):
Yeah, I mean there is a serendipity here that's really
hard to deny. I have to agree with that. You know,
after I left the racing world and went on to
my you know, to my engineering life, I actually went
into focused on computer science, focused on artificial intelligence. Came
(09:54):
out of the graduate programs interested in being in computer science,
had an excellent career through most of most of my
life as in industry, managing running and building products based
on software, and then was blessed to be in a
financial and an opportunity to retire ten years ago early
(10:19):
and moved to Maui, which is how I ended up
at the University of Hawaii. But soon felt that I
was not getting the uh, the kind of life experience
I wanted in terms of continuing to build and do something,
you know. In recognizing that, I went to the university
and asked them to let me participate as a mentor,
(10:42):
as a lecturer, or in some way you know, bring
my interest to the table. And what we settled on,
amazingly enough, long before even the IC was announced, was
an autonomous vehicle technology program.
Speaker 2 (10:58):
UH.
Speaker 1 (10:59):
So again the Sarah depity of building off of my
interest in in racing in vehicles.
Speaker 3 (11:04):
So we built early.
Speaker 1 (11:07):
Early jetson scale racing ai uh little carts, and then
went to fifth scale UH. Then went to autonomous racing
go karts.
Speaker 3 (11:19):
UH.
Speaker 1 (11:19):
And it was while we were designing and building and
working with students in Hawaii on that program that I
announced this program.
Speaker 3 (11:29):
UH. The I the Autonomous Challenge for the full scale cards.
Speaker 2 (11:33):
UH.
Speaker 1 (11:33):
They put out a uh you know, sort of an RFP,
but they put out a call to universities to say, hey,
who wants to participate?
Speaker 3 (11:41):
Who can participate in something like.
Speaker 1 (11:43):
This, and went to the UH you know, went to
the dean, went to the engineering UH department heads and said,
you know, we're probably way over our head to even
talk about participating in this, but what the heck, you know,
it's a great giants, we're doing something with go karts.
The go kart race was in at Purdue in Indiana,
(12:06):
so therefore we were already scheduled to be a part
of moving developing the logistics to move back and forth
for that anyway, and they said, well, what the heck,
you know, it's your time if you want to if
you want to steer that way, go that way. So
we went that way, and in doing so, participating in
(12:28):
a year long simulation series of races for the for
the forty one universities, and we consistently finished in the
top six to eighth position, which again, I think, you know,
was probably a story that will repeat a few times,
hopefully for us, and that is that nobody expected us
(12:49):
to be a contender. Nobody expected, certainly the University of
Hawaii to be developing autonomous racing stacks at the kind
of level to compete with these same universe we compete
with today, uh.
Speaker 3 (13:01):
In a simulation series. And so when it came time to.
Speaker 1 (13:06):
Uh, you know, invite the top ten teams, we were
an obvious choice to be invited. So I went back
again and they said, and I said, well, we probably
are way in over our head, but the it's a
great experience and uh, you know, I think, you know,
we should we should try to do this and uh
(13:26):
again got support from the university. I want to thank
the folks at University of why they've been huge supporters
of us.
Speaker 3 (13:33):
Uh. You know, we certainly don't have the financial.
Speaker 1 (13:36):
Wherewithal of most of these other schools, but that's okay.
Speaker 3 (13:40):
What we have is a lot of heart and a lot.
Speaker 1 (13:42):
Of aloha and and a team that has really enjoys
the opportunity to be here.
Speaker 2 (13:47):
What has been the benefit for the university.
Speaker 1 (13:51):
Yeah, you know, the from what I hear from, you know,
from the folks that uh you know that take a
look at how do you evolve the strategic program for
University Hawaii. It has helped put University Hawaii into one
of the top ten or fifteen engineering possible schools for
(14:12):
people to take a look at so we were, you know,
to be honest, Universe WHOY was not considered a you know,
we were considered maybe a middle tier engineering school. I
think now some people look at what we're doing and
looking at not only this program. It turned out that
this was a perfect storm of some other programs at
(14:32):
the University Haaii, particularly in the marine area, as you
might guess, Yes, being stuck in the middle of the
Pacific Ocean, we have a lot of doctoral programs and
master's students that are very interested in autonomy above them
below the water. So we kind of we kind of
are the third leg of the autonomy stool for Universe
Hawaii and that's that's really helped put us on the map.
Speaker 2 (14:53):
Frankly, coming at this from traditional motorsports background, to what
extent has that informed your decisions in your role and
helped to propel this this program perhaps beyond the expectations
of many on the outside.
Speaker 1 (15:13):
Certainly, having been there, done that always is a huge
benefit for you know, for any program, and having you know,
been involved in professional racing and understanding the logistics, scheduling,
the the intensity of work, the you know, the meeting
is schedule is tough anyway under any circumstances, meeting is
schedule in a sports uh league of anything, whether it's football, basketball,
(15:37):
racing is a is a total another level, there is
no well, I'll get you that tomorrow when the event
is today. So I think that's been really a you know,
that's really helped us a lot, uh in that we
brought brought forth that discipline, recognized that educated students to
(15:58):
the absolute necessity of that, which I think really is
going to make them even better people out in the
industry because they're going to set you know, when schedules
get done and programs get developed and that kind of stuff.
They're going to understand their time and trade offs that
have to be made because we can't do everything. We
have to do just as much as we can and
(16:20):
the most important things that we can in the time
frame we've got, and that kind of discipline just doesn't
come historically to that point.
Speaker 2 (16:28):
My supposition is many of your students did not come
to this with a motorsports background or even a sporting
background necessarily. It was probably the AI component that initially
caught their attention. So I'm curious to what degree they
have a new appreciation for sports generally motorsports specifically, and
then to build on what you were saying before, the
(16:51):
lessons that maybe uniquely come from participating in such an environment,
how does that then prepare them in a way that, uh,
that that students who had not had this experience would
would be lacking.
Speaker 3 (17:05):
Well, I have a I have a great little story
on that. Just in the last few weeks.
Speaker 1 (17:10):
One of our vehicle dynamics engineers, who's who comes from
a background I had no background in racing and no
background just really wanted to but kind of likes autos,
you know, and wanted to do something with that. His
work and the paper that he's helped develop got him
(17:30):
an offer with Mercedes Formula One team.
Speaker 3 (17:35):
Wow. So I don't know if he's going to take it.
Speaker 1 (17:37):
So Mercedes, I can't tell you, but but I can
say that, you know, I mean, I think that just
demonstrates the you know, clearly, the fact that working on
these kinds of programs, uh, seeing the kind of the
way we run our pits, the way that uh, you know,
the telemetry requirements, data requirements, the kinds of things that
(17:59):
the real top level form racing teams need to have
to be successful, and that we've implemented on a smaller scale.
It just makes them makes these kids so excited, and
it's all obviously making them very marketable.
Speaker 2 (18:17):
Last angle of discussion that I'm interested in, again, coming
from your background, which is different, I'm sure you hear
some of the same things I do. Working in traditional motorsport,
there is some hesitancy about AI and autonomy when it
comes to racing as an entity as a sport. How
(18:37):
do you answer those critics? And do you think what
we've seen from the IAC so far has answered those
critics and perhaps also provided a glimpse into the degree
in which this technology you're working on hands on right
now could trickle down and benefit racing in a really
tangible way.
Speaker 1 (18:56):
Yeah. I think to the detractors of of you know,
autonomous racing and and and now and the sum for
autonomy onto the street, I think that what they, you know,
there's a general lack of appreciation of how difficult and
challenging these problems are. And to think that you know
(19:16):
by a snap of a finger or a line of code,
the work that we need to do to develop these
products to the level they need to to be to
really bring a level of safety that's measurable as sustainable
and and uh sufficient is uh you know, is really
a niativity on their health. We're on a long march.
(19:39):
If you go back to the Indianapolis five hundred, in
the very first days, they were excited about the fact
that somebody held a piece of broken mirror up and
realized that rear view mirrors.
Speaker 3 (19:48):
Was a good thing. So I'm sorry.
Speaker 1 (19:51):
You know, maybe it's going to take us a little
time to get to where we need to be, but
I think we're going to add value, uh, not only
to human driven racing, because I think that you know,
as a co pilot, some of the work that we're
doing can make racing safer because it can help take
control in certain circumstances where the human maybe isn't.
Speaker 3 (20:13):
Most the most capable solution space. And I know that's
going to happen on the street.
Speaker 1 (20:19):
I just you know, the fact that we have to
make decisions at two hundred miles an hour, that we
have to make decisions one hundred times a second, and
we have to make the right decision or you know,
or in fact, what we're doing can cause a lot
of physical damage to a race car. Maybe no humans
get involved, but it's not cheap and racing it gives
(20:39):
us the opportunity to test, test some ideas and uniquely
come bring some solutions to the table.
Speaker 2 (20:46):
Historically, racing has been the ultimate proving ground for automotive
We've seen less of that, it seems in recent decades.
Rules are very strict. There's not much innovation that is
that is permitted in many series globally. But this strikes
me as a return to the roots of the sport.
This is why people race cars in those early days.
Speaker 1 (21:07):
Right, you know, I think that's right, And to be honest,
that's that's one of the drivers that brought me out
of retirement and sends me around the United States multiple
times a year with you know, at ten or twenty
students working night and day to try to do something.
Is because I think we're with this is a new
era of innovation for racing and we're at the forefront
(21:30):
of it. And I tell that to the kids all
the time that you know you're going to be this
is one of those points in your life you're going
to look back to and say, you know, I was
part of that, and I've been lucky to have that
happened multiple times in my career and so I'm you know,
and I'm confident this is another one.
Speaker 2 (21:47):
So in the head to head competitions we're seeing now
in IAC, we talk a lot about each team has
a unique solution to the same problem. When you're encountering
this yourself, are there are there surprise that you see
at teams taking attack that you wouldn't have thought of before?
Are there frustrations that come along with this? What has
that experience been like as you see the eight to
(22:09):
nine teams branch off in their unique direction.
Speaker 1 (22:14):
Well, there's a lot of commonality and there's a lot
of differences in the commonality. You know, there's elements we
have to perceive, we have to plan, we have to
we have to control, So there's you know, at certain levels,
there's a lot of commonality. There may be different code
spaces to implement them, but but there's similarities. What I
(22:35):
find exciting about any time of innovation, and I happen
to be in the in the distributed Systems group that
helped develop the Ethernet program and and you know, help
bring some of the Internet to the to the current
current level it's at today. Times of great innovation are
times of great observation, and so yes, we observe those
(22:58):
other teams very closely and we as much as possible,
and it's still managing to do what we need to
do to be successful. Try to understand what we think
they're thinking and why they're thinking that, and then we
try to test our own decisions against those and if
something seems like it's a better solution or a better
direction or method, then we're headed that direction because we
(23:22):
want to build the best thing, the best tool for
this job that we can. So maybe as an example
of that would be in path planning, because that's a
very very sophisticated challenge, and that is you've now identified
not only your location, but you've identified the location of
the competitor's car. We call them bogies, and in doing so,
(23:45):
we have to now plan a path around them. Our
job is to pass them. It's not to follow them,
you know, it's to get around them. It's a competition.
That's why we're here for. And so there's been a
lot of really interesting innovation in that. One of the
one of the approaches we took initially. And I don't
think I'm giving away big secrets here, I hope not.
Speaker 3 (24:08):
You know was to.
Speaker 1 (24:09):
Create a large number of trajectory lines and then be
able to switch quickly between them and safely. And so
it's sort of a preprocessed view of the universe, which
on a racetrack is doable because it's.
Speaker 3 (24:24):
A finite it's a finite track.
Speaker 1 (24:28):
As we've evolved, we realized that we can we can
use less of those lines and make more of those
decisions in a dynamic space.
Speaker 2 (24:37):
Exactly what we were looking for. Thank you so much,
welcome or wish you a safe travels. Thank you and
hope to see you here down the road.
Speaker 3 (24:45):
All right, we'll see then.
Speaker 2 (24:51):
Thanks for joining us this week on the inside track.
That was Gary Passan, the AI Racing Tech Team principal
and leader at the University of Hawaii Noah, giving us
a behind the scenes look at the incredible levels of
innovation being explored way out in the middle of the
Pacific Ocean. I'm your host, Ryan Marine. Join us next
time to hear from Professor Sergio Savaesi of the Polytechnic
(25:14):
University of Milan. Sergio has authored over five hundred publications
focusing on autonomous technology and is considered by many to
be the grandfather of the Modern autonomous Revolution,
You know, there's a general lack of appreciation of how
difficult and challenging these problems are. And to think that,
you know, by a snap of a finger or a
line of code, the work that we need to do
to develop these products to the level they need to
to be, to really bring a level of safety that's
measurable as sustainable and sufficient is you know, is really
(00:20):
a niativity on their health.
Speaker 2 (00:27):
The world around us is changing faster than ever before.
Ideas once only imagined in science fiction are becoming a reality.
Throughout the course of our amazing twenty three episode season,
we'll speak to some of the greatest minds in robotics
and artificial intelligence to discuss the groundbreaking work that's fueling
it all. I'm your host Ryan Marine, joined me and
my co host Paul Mitchell, the presidents of the Indian
(00:49):
Autonomous Challenge, and see why we call this the inside Track.
A few things come to mind when people picture the
Hawaiian Islands, stunning views, incredible surfing, deli coffee, and ocean
as far as the eye can see. But did you
know that a small robotics and engineering team at the
University of Hawaii MANOAH has been not only competing at
(01:11):
Las Vegas Motor Speedway, but also has been beating some
of the most advanced engineering teams in the world today.
On the Inside Track, we sit down with Gary Passon,
the AI Racing Tech Team, principal and leader for the
University of Hawaii to discuss how a school known for
students specializing in oceanography is now leading in robotics and engineering.
(01:35):
Gary gives us insight into what makes the IAC so
much more than great code and how this team put
the University of Hawaii on the global stage. I'm super
interested in hearing a little bit more about your background
and how you got into this role. You come at
(01:57):
this from a different perspective than most. But first I
should say congratulations. Another podium finished for you and the team.
You were second in the competition at Texas in November
of twenty twenty two, then in early twenty twenty three
third place in Las Vegas. The team seems to really
be hitting its stride.
Speaker 3 (02:16):
Yeah, Ran, we are.
Speaker 1 (02:17):
We're just very excited to continue to build on the
program that we started three years ago. We knew this
was going to be a long term program and that
we would need to evolve not only the team, but
the software stack to come up and be able to
compete with the Europeans. I mean, those guys are bringing
an amazing capability and experience to the table, and we're
(02:39):
proud to be on that podium with them two times
in a row now, but we still think there's some
place yet to go, and that would be at the
top of it, so we'll continue to continue to work
on that.
Speaker 2 (02:50):
You're quite right that the European universities have been at
the forefront of this to this point, and I'm prepared
to be told I'm wrong with this, but institutionally, it
does seem like the way that they go about their
universities and their research is perhaps a little bit different
than most approaches you see from American universities. Is that
correct and what kind of challenges come with that? If
(03:13):
my assumption is on point.
Speaker 3 (03:15):
Well, you're exactly right on.
Speaker 1 (03:17):
There's probably two main factors that to date have given
them the edge, but I think that edge is eroding.
The first is that they were the early participants in
a competition called Robocar YEP, which was an Nvidia and
otherwise sponsored organism program for fully autonomous race cars, primarily
(03:39):
situated out of Europe, although to do some racing here
in the United States, and that allowed them to get
ahead started developing their stack for the specific application.
Speaker 3 (03:50):
So that was.
Speaker 1 (03:50):
Certainly an advantage, but not an enduring advantage because obviously
we're closing the gap. The second is that the governments
in your tend to financially support educational institutes that are
that are specialists, and so in the cases for all
three cases, for the European teams, and no disrespect, I
(04:12):
wish we were lucky enough to get that same kind
of funding opportunities, but they get They get a continuing
annual budget which allows them to have a number of
PhDs pre and post am. I doing that, they're able
to continue to evolve over a long term. A very
(04:33):
difficult and sophisticated problem which this is, this is a
very very very challenging thing, but something that I think
is very important ultimately to the safety of autonomous vehicles
on the road.
Speaker 2 (04:44):
By contrast, what constitutes your team? What are you working with?
Speaker 1 (04:50):
So but you know, by contrast, we're and I think
this is somewhat true for most of the most of
the American teams, but not all of them. We are
actually an algemy of the University of Hawaii as the
prime university supported by University of California, San Diego, University
of California, Berkeley. This year we brought in Carnegie Mellon University.
(05:12):
So by being able to aggregate the graduate programs and
the undergraduate programs for all four universities, were able to
put together a group of not only the talent that
we're looking for, but also the manpower that we're looking
for to do these kinds of problems. It's not something
that a programmer in a back room.
Speaker 3 (05:31):
Is going to do. This is clearly a big, big,
big program.
Speaker 2 (05:35):
What has led you to this point? And specifically I
bring this up because I have to believe you come
to be a university professor in a much different way
than most. So let's go back to the beginning, because
I think the beginning explains a great deal of how
we got to where we are today.
Speaker 1 (05:54):
Yeah, I'm not sure which is the chicken and which
is the egg here, but background is having been born
and raised in Indiana, comes up through go karts, and
of course Indiana being the home of the Indianapolis five
hundred is the uh, you know, the most important arguably
race in the United States. Uh So everybody in Indiana
(06:17):
is a very has a lot of race in their blood.
Speaker 3 (06:20):
Uh.
Speaker 1 (06:21):
And I was very lucky to be successful in go karting,
move then into amateur racing, uh, then move into semi
professional racing and ultimately was a professional race car driver
for a short period of time. So I was very
lucky to be our unlucky depend on how you want
to look at it. Uh to be of the same
era as Bobby Unster junior, Al Unser junior, Mario Andretti junior,
(06:48):
Bobby ray Hall. All those drivers came up in the
same era of wielded wheel both USAK racing in those
days for the mini in these or Formula land Takes
or or other series and that stuff. So we ran
around nose to tail a lot during those years. And
and some of those guys went on to full Indianapolis careers.
(07:11):
My destiny was not that I was always uh. I
always had my soul uh uh an engineering interest. Uh
maybe me. Mark Donahue was probably you know for those
for those a little bit older people out there, Hey,
the Mark Donahue, give.
Speaker 2 (07:28):
Up on on on the youth here, I'm thirty four,
and I know about the unfair advantage in Mark Donahue.
Speaker 1 (07:33):
Very good, very good, but not not a lot of
thirty well sub thirties. Well, sure, sure, I'll give you
an Matt you know is familiar with Mark, and and
also Bruce McLaren. Uh So, you know, I always saw
the engineering sign maybe even a little bit more of
my level of interest than than even the driving side.
Speaker 2 (07:54):
Both of them met untimely ins in race cars. So
there there was a dangerous time to be doing a
particular profession, and perhaps pursuing the engineering had a bit
of a longer runway for you.
Speaker 3 (08:07):
I think it did.
Speaker 1 (08:09):
You know, I'm not sure that that was in the
top of the you know, because you shouldn't drive race
cars at the limit if you consider uh the you know,
what can happen, uh, you know too fully, it was
a combination I think of a recognition that those were
dangerous times and I did have a friend of mine
die in a race car people, and so that, you know,
that probably had an impact on me as well as
(08:34):
the just just the fact that I found that racing
against the families of Indianapolis car drivers meant that when
it came time to divide up sponsorships and those kinds
of things. They had a tendency to go with name
and brand, and maybe a little less with some upstart
(08:54):
that you know, was more interested in the role anti
roll bar rate at the rear the car then maybe
they were in you know, being the race car driver.
Speaker 2 (09:05):
How though, then, did these experiences earlier in your life
almost come full circle to put you in this position
where you're working on an indie lights indie, next mini indie,
whatever you want to call it, chassis and the technology
that you're honing has safety in mind, safety for the road,
safety for racing drivers as well. It seems to me
(09:26):
that all of your personal experiences have flowed into this
in an almost perfect manner.
Speaker 1 (09:33):
Yeah, I mean there is a serendipity here that's really
hard to deny. I have to agree with that. You know,
after I left the racing world and went on to
my you know, to my engineering life, I actually went
into focused on computer science, focused on artificial intelligence. Came
(09:54):
out of the graduate programs interested in being in computer science,
had an excellent career through most of most of my
life as in industry, managing running and building products based
on software, and then was blessed to be in a
financial and an opportunity to retire ten years ago early
(10:19):
and moved to Maui, which is how I ended up
at the University of Hawaii. But soon felt that I
was not getting the uh, the kind of life experience
I wanted in terms of continuing to build and do something,
you know. In recognizing that, I went to the university
and asked them to let me participate as a mentor,
(10:42):
as a lecturer, or in some way you know, bring
my interest to the table. And what we settled on,
amazingly enough, long before even the IC was announced, was
an autonomous vehicle technology program.
Speaker 2 (10:58):
UH.
Speaker 1 (10:59):
So again the Sarah depity of building off of my
interest in in racing in vehicles.
Speaker 3 (11:04):
So we built early.
Speaker 1 (11:07):
Early jetson scale racing ai uh little carts, and then
went to fifth scale UH. Then went to autonomous racing
go karts.
Speaker 3 (11:19):
UH.
Speaker 1 (11:19):
And it was while we were designing and building and
working with students in Hawaii on that program that I
announced this program.
Speaker 3 (11:29):
UH. The I the Autonomous Challenge for the full scale cards.
Speaker 2 (11:33):
UH.
Speaker 1 (11:33):
They put out a uh you know, sort of an RFP,
but they put out a call to universities to say, hey,
who wants to participate?
Speaker 3 (11:41):
Who can participate in something like.
Speaker 1 (11:43):
This, and went to the UH you know, went to
the dean, went to the engineering UH department heads and said,
you know, we're probably way over our head to even
talk about participating in this, but what the heck, you know,
it's a great giants, we're doing something with go karts.
The go kart race was in at Purdue in Indiana,
(12:06):
so therefore we were already scheduled to be a part
of moving developing the logistics to move back and forth
for that anyway, and they said, well, what the heck,
you know, it's your time if you want to if
you want to steer that way, go that way. So
we went that way, and in doing so, participating in
(12:28):
a year long simulation series of races for the for
the forty one universities, and we consistently finished in the
top six to eighth position, which again, I think, you know,
was probably a story that will repeat a few times,
hopefully for us, and that is that nobody expected us
(12:49):
to be a contender. Nobody expected, certainly the University of
Hawaii to be developing autonomous racing stacks at the kind
of level to compete with these same universe we compete
with today, uh.
Speaker 3 (13:01):
In a simulation series. And so when it came time to.
Speaker 1 (13:06):
Uh, you know, invite the top ten teams, we were
an obvious choice to be invited. So I went back
again and they said, and I said, well, we probably
are way in over our head, but the it's a
great experience and uh, you know, I think, you know,
we should we should try to do this and uh
(13:26):
again got support from the university. I want to thank
the folks at University of why they've been huge supporters
of us.
Speaker 3 (13:33):
Uh. You know, we certainly don't have the financial.
Speaker 1 (13:36):
Wherewithal of most of these other schools, but that's okay.
Speaker 3 (13:40):
What we have is a lot of heart and a lot.
Speaker 1 (13:42):
Of aloha and and a team that has really enjoys
the opportunity to be here.
Speaker 2 (13:47):
What has been the benefit for the university.
Speaker 1 (13:51):
Yeah, you know, the from what I hear from, you know,
from the folks that uh you know that take a
look at how do you evolve the strategic program for
University Hawaii. It has helped put University Hawaii into one
of the top ten or fifteen engineering possible schools for
(14:12):
people to take a look at so we were, you know,
to be honest, Universe WHOY was not considered a you know,
we were considered maybe a middle tier engineering school. I
think now some people look at what we're doing and
looking at not only this program. It turned out that
this was a perfect storm of some other programs at
(14:32):
the University Haaii, particularly in the marine area, as you
might guess, Yes, being stuck in the middle of the
Pacific Ocean, we have a lot of doctoral programs and
master's students that are very interested in autonomy above them
below the water. So we kind of we kind of
are the third leg of the autonomy stool for Universe
Hawaii and that's that's really helped put us on the map.
Speaker 2 (14:53):
Frankly, coming at this from traditional motorsports background, to what
extent has that informed your decisions in your role and
helped to propel this this program perhaps beyond the expectations
of many on the outside.
Speaker 1 (15:13):
Certainly, having been there, done that always is a huge
benefit for you know, for any program, and having you know,
been involved in professional racing and understanding the logistics, scheduling,
the the intensity of work, the you know, the meeting
is schedule is tough anyway under any circumstances, meeting is
schedule in a sports uh league of anything, whether it's football, basketball,
(15:37):
racing is a is a total another level, there is
no well, I'll get you that tomorrow when the event
is today. So I think that's been really a you know,
that's really helped us a lot, uh in that we
brought brought forth that discipline, recognized that educated students to
(15:58):
the absolute necessity of that, which I think really is
going to make them even better people out in the
industry because they're going to set you know, when schedules
get done and programs get developed and that kind of stuff.
They're going to understand their time and trade offs that
have to be made because we can't do everything. We
have to do just as much as we can and
(16:20):
the most important things that we can in the time
frame we've got, and that kind of discipline just doesn't
come historically to that point.
Speaker 2 (16:28):
My supposition is many of your students did not come
to this with a motorsports background or even a sporting
background necessarily. It was probably the AI component that initially
caught their attention. So I'm curious to what degree they
have a new appreciation for sports generally motorsports specifically, and
then to build on what you were saying before, the
(16:51):
lessons that maybe uniquely come from participating in such an environment,
how does that then prepare them in a way that, uh,
that that students who had not had this experience would
would be lacking.
Speaker 3 (17:05):
Well, I have a I have a great little story
on that. Just in the last few weeks.
Speaker 1 (17:10):
One of our vehicle dynamics engineers, who's who comes from
a background I had no background in racing and no
background just really wanted to but kind of likes autos,
you know, and wanted to do something with that. His
work and the paper that he's helped develop got him
(17:30):
an offer with Mercedes Formula One team.
Speaker 3 (17:35):
Wow. So I don't know if he's going to take it.
Speaker 1 (17:37):
So Mercedes, I can't tell you, but but I can
say that, you know, I mean, I think that just
demonstrates the you know, clearly, the fact that working on
these kinds of programs, uh, seeing the kind of the
way we run our pits, the way that uh, you know,
the telemetry requirements, data requirements, the kinds of things that
(17:59):
the real top level form racing teams need to have
to be successful, and that we've implemented on a smaller scale.
It just makes them makes these kids so excited, and
it's all obviously making them very marketable.
Speaker 2 (18:17):
Last angle of discussion that I'm interested in, again, coming
from your background, which is different, I'm sure you hear
some of the same things I do. Working in traditional motorsport,
there is some hesitancy about AI and autonomy when it
comes to racing as an entity as a sport. How
(18:37):
do you answer those critics? And do you think what
we've seen from the IAC so far has answered those
critics and perhaps also provided a glimpse into the degree
in which this technology you're working on hands on right
now could trickle down and benefit racing in a really
tangible way.
Speaker 1 (18:56):
Yeah. I think to the detractors of of you know,
autonomous racing and and and now and the sum for
autonomy onto the street, I think that what they, you know,
there's a general lack of appreciation of how difficult and
challenging these problems are. And to think that you know
(19:16):
by a snap of a finger or a line of code,
the work that we need to do to develop these
products to the level they need to to be to
really bring a level of safety that's measurable as sustainable
and and uh sufficient is uh you know, is really
a niativity on their health. We're on a long march.
(19:39):
If you go back to the Indianapolis five hundred, in
the very first days, they were excited about the fact
that somebody held a piece of broken mirror up and
realized that rear view mirrors.
Speaker 3 (19:48):
Was a good thing. So I'm sorry.
Speaker 1 (19:51):
You know, maybe it's going to take us a little
time to get to where we need to be, but
I think we're going to add value, uh, not only
to human driven racing, because I think that you know,
as a co pilot, some of the work that we're
doing can make racing safer because it can help take
control in certain circumstances where the human maybe isn't.
Speaker 3 (20:13):
Most the most capable solution space. And I know that's
going to happen on the street.
Speaker 1 (20:19):
I just you know, the fact that we have to
make decisions at two hundred miles an hour, that we
have to make decisions one hundred times a second, and
we have to make the right decision or you know,
or in fact, what we're doing can cause a lot
of physical damage to a race car. Maybe no humans
get involved, but it's not cheap and racing it gives
(20:39):
us the opportunity to test, test some ideas and uniquely
come bring some solutions to the table.
Speaker 2 (20:46):
Historically, racing has been the ultimate proving ground for automotive
We've seen less of that, it seems in recent decades.
Rules are very strict. There's not much innovation that is
that is permitted in many series globally. But this strikes
me as a return to the roots of the sport.
This is why people race cars in those early days.
Speaker 1 (21:07):
Right, you know, I think that's right, And to be honest,
that's that's one of the drivers that brought me out
of retirement and sends me around the United States multiple
times a year with you know, at ten or twenty
students working night and day to try to do something.
Is because I think we're with this is a new
era of innovation for racing and we're at the forefront
(21:30):
of it. And I tell that to the kids all
the time that you know you're going to be this
is one of those points in your life you're going
to look back to and say, you know, I was
part of that, and I've been lucky to have that
happened multiple times in my career and so I'm you know,
and I'm confident this is another one.
Speaker 2 (21:47):
So in the head to head competitions we're seeing now
in IAC, we talk a lot about each team has
a unique solution to the same problem. When you're encountering
this yourself, are there are there surprise that you see
at teams taking attack that you wouldn't have thought of before?
Are there frustrations that come along with this? What has
that experience been like as you see the eight to
(22:09):
nine teams branch off in their unique direction.
Speaker 1 (22:14):
Well, there's a lot of commonality and there's a lot
of differences in the commonality. You know, there's elements we
have to perceive, we have to plan, we have to
we have to control, So there's you know, at certain levels,
there's a lot of commonality. There may be different code
spaces to implement them, but but there's similarities. What I
(22:35):
find exciting about any time of innovation, and I happen
to be in the in the distributed Systems group that
helped develop the Ethernet program and and you know, help
bring some of the Internet to the to the current
current level it's at today. Times of great innovation are
times of great observation, and so yes, we observe those
(22:58):
other teams very closely and we as much as possible,
and it's still managing to do what we need to
do to be successful. Try to understand what we think
they're thinking and why they're thinking that, and then we
try to test our own decisions against those and if
something seems like it's a better solution or a better
direction or method, then we're headed that direction because we
(23:22):
want to build the best thing, the best tool for
this job that we can. So maybe as an example
of that would be in path planning, because that's a
very very sophisticated challenge, and that is you've now identified
not only your location, but you've identified the location of
the competitor's car. We call them bogies, and in doing so,
(23:45):
we have to now plan a path around them. Our
job is to pass them. It's not to follow them,
you know, it's to get around them. It's a competition.
That's why we're here for. And so there's been a
lot of really interesting innovation in that. One of the
one of the approaches we took initially. And I don't
think I'm giving away big secrets here, I hope not.
Speaker 3 (24:08):
You know was to.
Speaker 1 (24:09):
Create a large number of trajectory lines and then be
able to switch quickly between them and safely. And so
it's sort of a preprocessed view of the universe, which
on a racetrack is doable because it's.
Speaker 3 (24:24):
A finite it's a finite track.
Speaker 1 (24:28):
As we've evolved, we realized that we can we can
use less of those lines and make more of those
decisions in a dynamic space.
Speaker 2 (24:37):
Exactly what we were looking for. Thank you so much,
welcome or wish you a safe travels. Thank you and
hope to see you here down the road.
Speaker 3 (24:45):
All right, we'll see then.
Speaker 2 (24:51):
Thanks for joining us this week on the inside track.
That was Gary Passan, the AI Racing Tech Team principal
and leader at the University of Hawaii Noah, giving us
a behind the scenes look at the incredible levels of
innovation being explored way out in the middle of the
Pacific Ocean. I'm your host, Ryan Marine. Join us next
time to hear from Professor Sergio Savaesi of the Polytechnic
(25:14):
University of Milan. Sergio has authored over five hundred publications
focusing on autonomous technology and is considered by many to
be the grandfather of the Modern autonomous Revolution,
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
My Favorite Murder with Karen Kilgariff and Georgia Hardstark
My Favorite Murder is a true crime comedy podcast hosted by Karen Kilgariff and Georgia Hardstark. Each week, Karen and Georgia share compelling true crimes and hometown stories from friends and listeners. Since MFM launched in January of 2016, Karen and Georgia have shared their lifelong interest in true crime and have covered stories of infamous serial killers like the Night Stalker, mysterious cold cases, captivating cults, incredible survivor stories and important events from history like the Tulsa race massacre of 1921. My Favorite Murder is part of the Exactly Right podcast network that provides a platform for bold, creative voices to bring to life provocative, entertaining and relatable stories for audiences everywhere. The Exactly Right roster of podcasts covers a variety of topics including historic true crime, comedic interviews and news, science, pop culture and more. Podcasts on the network include Buried Bones with Kate Winkler Dawson and Paul Holes, That's Messed Up: An SVU Podcast, This Podcast Will Kill You, Bananas and more.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.