February 27, 2024 • 21 mins
This time, on the inside track, Andy Saba joins us to discuss his role in leading a mostly undergraduate student team for the IAC. Andy Saba is a master’s student of engineering at the Carnegie Mellon University Robotics Institute and has firmly rooted himself in the world of autonomous race cars. Ryan and Paul get the inside scoop into how Andy and his team went from a few lines of code to racing an autonomous car at 150+ miles per hour.
This is a two-part series.
This is a two-part series.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:00):
For me, like, I just genuinely love this stuff. I
love like working on the race cars, I love like robotics.
And you know, if I see someone else is like
struggling with a problem, I'm not gonna just ignore it
because they're from another team. I think like, if every
team can solve the million dumb problems, it's right, the
million small problems and really focus on the hard ones.
(00:23):
We together all advanced.
Speaker 2 (00:25):
Right, 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
(00:46):
it all. I'm your host Ryan Marine. Joined me and
my co host Paul Mitchell, the presidents of the Indian
Autonomous Challenge and see why we call this the inside Track.
This week, we're diving back into our conversation with Andy Saba,
leader and mentor for the mid pitt Autonomous Racing team
from the University of Pittsburgh at Massachusetts Institute of Technology.
(01:07):
We'll hear about how working with undergraduates allows him to
educate his team about the rationale behind specific engineering choices
rather than simply rushing to meet deadlines. And we'll learn
why mentoring undergraduates is a sure way to inject fresh
ideas and genuine enthusiasm for enhancing the safety and advancement
of the road cars that we all use. Here is
(01:30):
the rest of our conversation with Andy Saba.
Speaker 3 (01:37):
We're talking about the technology and the hardware that comes
together in these in these race cars in this unique,
unique form factor, which is its own engineering and packaging challenge.
But as you as you move at higher speeds, just
to push back a little bit on the because I
(01:58):
agree with you, there's clearly similarity and building blocks in
autonomous mobility that are relevant whether you're going twenty miles
an hour around a cul de sac, worried about a
kid running in front of the car or not. But
in particular, you hear a lot of discussion about, you know,
can you do a lot of the navigation and decision
(02:19):
making using optical sensors or cameras? Right, they go back
to the Tesla debate, can you do all self driving
with cameras only? And yet in our competition we clearly
see that teams tend to favor light ar as a sensor,
And is that kick in in some part because of
(02:41):
the speeds, because of the closing distances. And when you
think about passenger cars going seventy eighty miles an hour,
maybe one hundred and twenty miles an hour on the
German Autobahn, you know, encountering another car, does it start
to extend beyond the capabilities of the of the urban
(03:02):
suburban railbo taxi hardware software stack.
Speaker 1 (03:06):
So you bring up this really good point of like,
you know, light our versus camera, right, And the big
argument that Tesla makes for camera is that it's a
lot cheaper and it's a lot easier to scale up
to you know, millions and millions of cars, right, And
you know, Luminar will will make the counter argument that, well,
then let's just make better light our sensors, right that
(03:26):
can that can also do the same thing. And I
think I think that's actually one of the unique things
about about the I A C Is that you know,
these these sensors are getting used in you know, vehicles
that you and I can go buy. Right, So the
AB twenty three will have the luminar hydra or the
(03:47):
sorry the luminar irises. Right. You know these cameras there
there there's nothing really particularly special about them. I mean,
they're great cameras, but there's nothing particular about them too,
you know, on miss racing. So I think that is
kind of one of the nice things about this this
competition is that we are making we're making strides and
(04:10):
showing that it is possible with you know, without specialty hardware, right,
And I think that that is that is a case
then for okay, if these sensors can survive you know,
going one hundred and eighty around around an oval or
around around a road course, then maybe then they can
survive you know, two hundred thousand miles, right of like
(04:32):
driving around uh in you know, downtown Indianapolis, right. I
actually I don't want to make this stoke about Indianapolis,
but in Pittsburgh there's a lot of pod bowls, right,
and so if you if you want to make an
av so, if you want to make an AP that
last you need to be able to handle those riggers. So, yes,
there there are there are extremes in in atom as racing,
(04:56):
but I would still stand that I think the fact
we can handle those extremes mean that we're better equipped
to handle the every day, the everyday stuff regarding regarding
the this is like kind of like a tangent regarding
the whole like light our light our camera debate. One
(05:17):
of the things that that our team kind of like
focus on pretty early on was, you know, we wanted
to make sure that we had solutions for all of
the sensors, right, cameras, radars, and light r We invested
a lot of time in developing advanced solutions for especially
for the light ar and for the camera. We're working
(05:40):
on something new right now. And part of that is
because each sensor brings its strength and its weaknesses, and
the same applies for commercial vehicles, the same applies for
autonomous racing. And one of the one of the really
nice things about cameras is that they can just see
further than like our scan right, Like the luminar can see,
(06:02):
you know, two hundred meters, the hydro the iris could
see a little bit further. But the cameras you have,
you know, significantly more resolution, significantly more range. Really that's
really helpful when you're when you're racing. They can also
run at a much faster rate the camera can sample
up to seventy five frames a second, right versus the
(06:24):
light R twenty, So you're talking, you know, almost three
to four times faster. So it each comes with its drawbacks, and.
Speaker 3 (06:33):
It's really an argument. Yeah, it's really an argument for
the value of sensor fusion and how ICE is giving
you a platform to advance sensor fusion because you may
be using the camera to see a vehicle further out
and know that it's there, but then as you approach
it and you're thinking about and overtake, you're relying more
(06:56):
on light R and being able to either toggle between
those sensors in terms of what your algorithm is based
on or the fusion of those. Yeah, that's that's a
really you know, another contribution that's being made that I
know is also relevant to the commercial space.
Speaker 2 (07:15):
I'm really interested in kind of going back to what
Paul was talking about earlier, being on the pioneering edge
of this technology and the choices that you're making in
a compact amount of time because of the nature of
this competition, is there any responsibility knowing that, in all
likelihood the next couple of generations are going to be
(07:36):
building upon your work here? Today in the same way
that you are on previous generations. Do you feel the
responsibility to when you make some of these decisions, position
not just yourselves but our posterity to be in a
better position to make better choices down the road.
Speaker 1 (07:54):
Definitely, definitely that that's something that I always think about.
And it kind of goes in into two like there
there's kind of two two sides to this. The one
side is is really thinking about the long term impact
of decisions. You know, there's definitely a lot of parts
of our stack that are there's a lot of tech debt, right, Like,
(08:14):
you know, whatever you're developing fast and for a competition,
you make, you cut some corners where you can, right,
And that's that's just that's just the nature of the beast.
If you're always chasing a perfect solution, you'll never have
a solution. And so yes, you always have to you
always have to kind of make the best decisions now.
(08:35):
But at the same time, it's it's also it is
also very difficult to predict what the problems are going
to be down the road. So for me, my philosophy
has always been about flexibility. How can I what is
the solution that's going to provide the most flexibility down
the road, right. So like, for example, there are parts
of our stack where I know we could shave like
(08:56):
a millisecond of latency, right, But the flexibility that like
certain abstractions provide is worth that like that cost, right,
and then we say, okay, how can we save that elsewhere? Right?
How can we make up for that? Or you know,
is that even a problem? Right? And that that goes
to the flip side of the coin, which is I
(09:18):
think more important than making the decisions is the philosophy
behind the decisions, Like why did you choose the decisions
you made? And that's something that I really try to
instill a lot in our team is really thinking about, Okay,
why did we why did we choose the light our
detection algorithm that we did, Why did we choose, you know,
(09:40):
to design our staff this way? You know, why did
we choose a controller that we did? Right, really getting
into the nitty gritty of like the reasonings behind it,
because it's kind of like that old age, like the
old saying of like you know, you teach you teach
someone to fish, you know, the feed them for life,
give them a fish, the feed beat for a day.
(10:00):
If you give if you give young engineers the correct
not even correct. If you give young engineers a philosophy
or like a reasoning behind why decisions were made, they
can then go and make their own decisions.
Speaker 2 (10:15):
Right.
Speaker 1 (10:15):
They understand, okay, this these are the drawbacks, this is
the things you considered. Now I can go and reevaluate
and say, okay, does that still make sense or do
I need to do something differently? Right? And I think
that is the more important like responsibility as a as
as as chief engineer on on our project, on our team,
(10:39):
is really instilling that mentality.
Speaker 3 (10:42):
You talk about your guys team and your philosophy, and
you know, I always appreciate that the mid pit r
W team really does take the the concept of being
a full team very seriously. You look at how you
guys prepare, the kind of pre meetings you have, the
(11:03):
post meetings you have after races, and yet yours is
the only team that is entirely student led, that that
is predominantly undergraduate for for those that aren't aware, right,
the majority of the of the Paddock UH and an
India Autonomous Challenge Event UH and the teams are our
(11:27):
PhD students certainly some master students and very few, if
any undergrads. I think Technical University of Munich at one
point had ten or more PhD students that were, you know,
contributing to their team. So you've got you you've got
to overcome the let's say that the deficit of just
(11:50):
let's say, pure experience or hours in the classroom or
you know, collective research experience sort of. What is it
about the team led approach that, you know, how do
you make it work? Because I think from the out
from an outsider's perspective, the assumption would be that, well,
(12:10):
these guys will never figure this out. There they're all
a bunch of students. There's no there's no grounding and
the faculty institution, they're they're not PhD students yet, and
yet here you are, you know, competing at the highest
level just within a couple of years.
Speaker 1 (12:26):
Yeah. Uh, it definitely is a challenge, and I think,
uh more more than the experience, I mean experiences. I
don't want a discount experience at all. Experience is extremely
important and understanding you know, how to spend your time
and where to spend your time, I think is the
most valuable thing that like by the time you're a
(12:49):
pH student. Is a skill you should have learned is
time management and understanding how I should spend my time
and where the problems are. But just from a pure
like manpower perspective, and undergrad is significantly less than than
a dedicated PhD because a dedicated PhD this is their
full time job, right And for undergrads, you know, you
(13:12):
have other classes, you have other things to worry about.
So yeah, we we have to overcome a deficit of
of experience. We have overcome a deficit of also just
pure time. And so there's there's it's a it's a
chicken and egg problem, right. You need the time to
get the more experience, but you can't get the more
experience because you are all the time. So for us,
(13:34):
we we really try to Okay, we know that people
have a deficit of time. We don't they have a
deficit experience. Where can we focus people's attention and people's
time to get the greatest value out for the team, right,
And so we we really focus a lot on making
sure there's proper you know, project management, and that people
(13:57):
understand Okay, this is what I'm working on. And more importantly,
we don't just give people task. We don't just say okay,
work on this whatever. We really try to pair people
up with other people they've been on the team for
a little bit longer. Because one, when you get stuck
on a problem, someone who was working on the same
(14:17):
problem a month ago can be like, oh yeah, yeah,
do this and then that that like that'll help you
along faster. Right, So by having a more interconnectedness between
the members, you get to share that experience is a
lot faster, and then you get people up to speed
a lot faster. So that's like point number one. Point
(14:37):
number two is that it's also really important for people
to understand how their piece fits into the bigger the
bigger thing right early on, when we you know, okay,
we're going to do some project management stuff. Guys, let's
do some press management stuff. You tell people, okay, do this,
do this, do this, then we're gonna do a sprint
blah blah blah. It's very easy for an undergrad who
(15:00):
has you know, midterms and other things going on, to
just say, Okay, this task like I don't understand why
it's important. I don't understand why like it's given to me,
I don't really care, right, and they're not going to produce.
But if you meet with someone and you have a
one on one and you say like, hey, like, how's
it going. What are you working on? Oh, you're working
(15:21):
on this? Okay, Like yeah, you know. The really cool
thing is like, once you have that, then we can
do you know, once you have that new feature in
the controller, then we'll be able to like do this
turn at Manzo a little bit faster. And once we
can do that, then we can test you know, this
other thing. Right, And once someone understands how their pieces
kind of fit in, they get more motivated and they
(15:44):
they better understand Okay, this this, this is the impact
of my work. And then the third most important piece
and this feeds into the other two things I'm talking about.
It's really comes down to building a good team culture.
At the end of the day, people don't want to
disappoint each other. People all want to work together and
(16:05):
achieve something incredible and by really grounding. Okay, here's how
your piece fits into everything else. Here's you guys meeting
one on one, getting to know each other beyond just
a screen on a face. Right. Here's you know how
the team as a whole is going to achieve their goals.
(16:26):
That I think that culture of we are a team,
we we're starting from behind, you know, we're the underdogs here.
That really gets people motivated, That gets people moving, and
then you're able to you know, you're able to get
a lot more value in a lot less time.
Speaker 3 (16:45):
You know. One of the things that strikes me is
the the collegial, kind of collaborative environment that we have
on the track acurely during test days and shakedowns, maybe
less so the day of an actual competition, but in general,
(17:08):
the relationships that are built among the teams are more
about kind of working together to advance this concept of
autonomous racing. So what does that meant to you? The
relationships that you've developed, not just within your team, but
(17:30):
among the different teams. And you know, if you think
into the future, do you see, you know, a network
that will emerge from this that as you all move
into careers, either starting companies or joining companies, or staying
in academia and becoming becoming researchers or faculty. How do
(17:53):
you see the relationships and the network that's been developed
evolving over time?
Speaker 2 (18:01):
Yeah?
Speaker 3 (18:01):
Or how do you hope it evolves over time, maybe
I guess.
Speaker 1 (18:03):
Yeah, yeah. I mean for me, one of the most,
if not the most valuable pieces out of out of
I A C. That that have gotten out of it has
been has been the relationships and and all the incredible
people have gotten to meet, both on our team and
and on other teams. You know, some of my some
of my best friends now are are people I would
(18:26):
have never met if it weren't for the I A C.
And that that applies to both our team and other teams.
And a lot of a lot of what I've learned
is is not you know, is a lot of what
I've learned has come from from just seeing what the
other teams are doing and just learning from them and
like asking questions and and and helping each other the
(18:50):
For me, like I I just I just genuinely just
just genuinely love this stuff. I love like working on
the race cars, I love like robotics. And you know,
if I see someone else is like struggling with a problem,
I'm not gonna just ignore it because they're they're from
another team. I think, like you know, if if if
every team can solve the million dumb problems, right, the
(19:14):
million small problems and really focus on the hard ones.
We together all advanced, right because the solutions to the
heart problems, you know, no one, no one has has
the right answer, right, There's always there's gonna be dozens
of different answers and someone coming to that solution, and
you know, learning from each other. That's like where we
(19:37):
all grow together, we all advance as a field. So
for me, collaboration is is such an essential part of
no matter what the circumstances are for this because at
the end of the day, we're all trying to we're
all trying to do something that's never been done before.
And how I how I hope that it continues to go.
(20:00):
I hope it continues to.
Speaker 3 (20:01):
Be like this.
Speaker 1 (20:01):
I hope like all the other teams you know, are
continuing to have that collaborative nature. You know, whoever whoever
takes takes the mantle from me on the met PET team.
I mean, it's it's gonna be really important for us
to uh to continue that collaboration and and to in
that collaborative nature because the competition as a whole will
(20:26):
suffer and the field as a whole will suffer. So yeah,
I hope it continues I really I really do enjoy
working with everyone, and uh I hope I get to
continue to work on stuff with them in the future.
Speaker 2 (20:41):
It's been a fascinating conversation. Really really enjoyed your insights,
and we wish you the best of luck at Manza
the new challenge that's coming up here very soon.
Speaker 1 (20:50):
Thank you.
Speaker 2 (20:55):
That was Andy Saba explaining his personal and working philosophy
and why he finds us mentoring undergraduates so vital to
the development of autonomous vehicles and the next generation of engineers.
The next time on the Inside Track, we'll be speaking
with Gary Passon, the faculty lead for the AI Racing
Tech team at the University of Hawaii, on how a
(21:16):
school known for its oceanic and astrological studies is contending
with some of the most prominent engineering schools in the world.
For me, like, I just genuinely love this stuff. I
love like working on the race cars, I love like robotics.
And you know, if I see someone else is like
struggling with a problem, I'm not gonna just ignore it
because they're from another team. I think like, if every
team can solve the million dumb problems, it's right, the
million small problems and really focus on the hard ones.
(00:23):
We together all advanced.
Speaker 2 (00:25):
Right, 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
(00:46):
it all. I'm your host Ryan Marine. Joined me and
my co host Paul Mitchell, the presidents of the Indian
Autonomous Challenge and see why we call this the inside Track.
This week, we're diving back into our conversation with Andy Saba,
leader and mentor for the mid pitt Autonomous Racing team
from the University of Pittsburgh at Massachusetts Institute of Technology.
(01:07):
We'll hear about how working with undergraduates allows him to
educate his team about the rationale behind specific engineering choices
rather than simply rushing to meet deadlines. And we'll learn
why mentoring undergraduates is a sure way to inject fresh
ideas and genuine enthusiasm for enhancing the safety and advancement
of the road cars that we all use. Here is
(01:30):
the rest of our conversation with Andy Saba.
Speaker 3 (01:37):
We're talking about the technology and the hardware that comes
together in these in these race cars in this unique,
unique form factor, which is its own engineering and packaging challenge.
But as you as you move at higher speeds, just
to push back a little bit on the because I
(01:58):
agree with you, there's clearly similarity and building blocks in
autonomous mobility that are relevant whether you're going twenty miles
an hour around a cul de sac, worried about a
kid running in front of the car or not. But
in particular, you hear a lot of discussion about, you know,
can you do a lot of the navigation and decision
(02:19):
making using optical sensors or cameras? Right, they go back
to the Tesla debate, can you do all self driving
with cameras only? And yet in our competition we clearly
see that teams tend to favor light ar as a sensor,
And is that kick in in some part because of
(02:41):
the speeds, because of the closing distances. And when you
think about passenger cars going seventy eighty miles an hour,
maybe one hundred and twenty miles an hour on the
German Autobahn, you know, encountering another car, does it start
to extend beyond the capabilities of the of the urban
(03:02):
suburban railbo taxi hardware software stack.
Speaker 1 (03:06):
So you bring up this really good point of like,
you know, light our versus camera, right, And the big
argument that Tesla makes for camera is that it's a
lot cheaper and it's a lot easier to scale up
to you know, millions and millions of cars, right, And
you know, Luminar will will make the counter argument that, well,
then let's just make better light our sensors, right that
(03:26):
can that can also do the same thing. And I
think I think that's actually one of the unique things
about about the I A C Is that you know,
these these sensors are getting used in you know, vehicles
that you and I can go buy. Right, So the
AB twenty three will have the luminar hydra or the
(03:47):
sorry the luminar irises. Right. You know these cameras there
there there's nothing really particularly special about them. I mean,
they're great cameras, but there's nothing particular about them too,
you know, on miss racing. So I think that is
kind of one of the nice things about this this
competition is that we are making we're making strides and
(04:10):
showing that it is possible with you know, without specialty hardware, right,
And I think that that is that is a case
then for okay, if these sensors can survive you know,
going one hundred and eighty around around an oval or
around around a road course, then maybe then they can
survive you know, two hundred thousand miles, right of like
(04:32):
driving around uh in you know, downtown Indianapolis, right. I
actually I don't want to make this stoke about Indianapolis,
but in Pittsburgh there's a lot of pod bowls, right,
and so if you if you want to make an
av so, if you want to make an AP that
last you need to be able to handle those riggers. So, yes,
there there are there are extremes in in atom as racing,
(04:56):
but I would still stand that I think the fact
we can handle those extremes mean that we're better equipped
to handle the every day, the everyday stuff regarding regarding
the this is like kind of like a tangent regarding
the whole like light our light our camera debate. One
(05:17):
of the things that that our team kind of like
focus on pretty early on was, you know, we wanted
to make sure that we had solutions for all of
the sensors, right, cameras, radars, and light r We invested
a lot of time in developing advanced solutions for especially
for the light ar and for the camera. We're working
(05:40):
on something new right now. And part of that is
because each sensor brings its strength and its weaknesses, and
the same applies for commercial vehicles, the same applies for
autonomous racing. And one of the one of the really
nice things about cameras is that they can just see
further than like our scan right, Like the luminar can see,
(06:02):
you know, two hundred meters, the hydro the iris could
see a little bit further. But the cameras you have,
you know, significantly more resolution, significantly more range. Really that's
really helpful when you're when you're racing. They can also
run at a much faster rate the camera can sample
up to seventy five frames a second, right versus the
(06:24):
light R twenty, So you're talking, you know, almost three
to four times faster. So it each comes with its drawbacks, and.
Speaker 3 (06:33):
It's really an argument. Yeah, it's really an argument for
the value of sensor fusion and how ICE is giving
you a platform to advance sensor fusion because you may
be using the camera to see a vehicle further out
and know that it's there, but then as you approach
it and you're thinking about and overtake, you're relying more
(06:56):
on light R and being able to either toggle between
those sensors in terms of what your algorithm is based
on or the fusion of those. Yeah, that's that's a
really you know, another contribution that's being made that I
know is also relevant to the commercial space.
Speaker 2 (07:15):
I'm really interested in kind of going back to what
Paul was talking about earlier, being on the pioneering edge
of this technology and the choices that you're making in
a compact amount of time because of the nature of
this competition, is there any responsibility knowing that, in all
likelihood the next couple of generations are going to be
(07:36):
building upon your work here? Today in the same way
that you are on previous generations. Do you feel the
responsibility to when you make some of these decisions, position
not just yourselves but our posterity to be in a
better position to make better choices down the road.
Speaker 1 (07:54):
Definitely, definitely that that's something that I always think about.
And it kind of goes in into two like there
there's kind of two two sides to this. The one
side is is really thinking about the long term impact
of decisions. You know, there's definitely a lot of parts
of our stack that are there's a lot of tech debt, right, Like,
(08:14):
you know, whatever you're developing fast and for a competition,
you make, you cut some corners where you can, right,
And that's that's just that's just the nature of the beast.
If you're always chasing a perfect solution, you'll never have
a solution. And so yes, you always have to you
always have to kind of make the best decisions now.
(08:35):
But at the same time, it's it's also it is
also very difficult to predict what the problems are going
to be down the road. So for me, my philosophy
has always been about flexibility. How can I what is
the solution that's going to provide the most flexibility down
the road, right. So like, for example, there are parts
of our stack where I know we could shave like
(08:56):
a millisecond of latency, right, But the flexibility that like
certain abstractions provide is worth that like that cost, right,
and then we say, okay, how can we save that elsewhere? Right?
How can we make up for that? Or you know,
is that even a problem? Right? And that that goes
to the flip side of the coin, which is I
(09:18):
think more important than making the decisions is the philosophy
behind the decisions, Like why did you choose the decisions
you made? And that's something that I really try to
instill a lot in our team is really thinking about, Okay,
why did we why did we choose the light our
detection algorithm that we did, Why did we choose, you know,
(09:40):
to design our staff this way? You know, why did
we choose a controller that we did? Right, really getting
into the nitty gritty of like the reasonings behind it,
because it's kind of like that old age, like the
old saying of like you know, you teach you teach
someone to fish, you know, the feed them for life,
give them a fish, the feed beat for a day.
(10:00):
If you give if you give young engineers the correct
not even correct. If you give young engineers a philosophy
or like a reasoning behind why decisions were made, they
can then go and make their own decisions.
Speaker 2 (10:15):
Right.
Speaker 1 (10:15):
They understand, okay, this these are the drawbacks, this is
the things you considered. Now I can go and reevaluate
and say, okay, does that still make sense or do
I need to do something differently? Right? And I think
that is the more important like responsibility as a as
as as chief engineer on on our project, on our team,
(10:39):
is really instilling that mentality.
Speaker 3 (10:42):
You talk about your guys team and your philosophy, and
you know, I always appreciate that the mid pit r
W team really does take the the concept of being
a full team very seriously. You look at how you
guys prepare, the kind of pre meetings you have, the
(11:03):
post meetings you have after races, and yet yours is
the only team that is entirely student led, that that
is predominantly undergraduate for for those that aren't aware, right,
the majority of the of the Paddock UH and an
India Autonomous Challenge Event UH and the teams are our
(11:27):
PhD students certainly some master students and very few, if
any undergrads. I think Technical University of Munich at one
point had ten or more PhD students that were, you know,
contributing to their team. So you've got you you've got
to overcome the let's say that the deficit of just
(11:50):
let's say, pure experience or hours in the classroom or
you know, collective research experience sort of. What is it
about the team led approach that, you know, how do
you make it work? Because I think from the out
from an outsider's perspective, the assumption would be that, well,
(12:10):
these guys will never figure this out. There they're all
a bunch of students. There's no there's no grounding and
the faculty institution, they're they're not PhD students yet, and
yet here you are, you know, competing at the highest
level just within a couple of years.
Speaker 1 (12:26):
Yeah. Uh, it definitely is a challenge, and I think,
uh more more than the experience, I mean experiences. I
don't want a discount experience at all. Experience is extremely
important and understanding you know, how to spend your time
and where to spend your time, I think is the
most valuable thing that like by the time you're a
(12:49):
pH student. Is a skill you should have learned is
time management and understanding how I should spend my time
and where the problems are. But just from a pure
like manpower perspective, and undergrad is significantly less than than
a dedicated PhD because a dedicated PhD this is their
full time job, right And for undergrads, you know, you
(13:12):
have other classes, you have other things to worry about.
So yeah, we we have to overcome a deficit of
of experience. We have overcome a deficit of also just
pure time. And so there's there's it's a it's a
chicken and egg problem, right. You need the time to
get the more experience, but you can't get the more
experience because you are all the time. So for us,
(13:34):
we we really try to Okay, we know that people
have a deficit of time. We don't they have a
deficit experience. Where can we focus people's attention and people's
time to get the greatest value out for the team, right,
And so we we really focus a lot on making
sure there's proper you know, project management, and that people
(13:57):
understand Okay, this is what I'm working on. And more importantly,
we don't just give people task. We don't just say okay,
work on this whatever. We really try to pair people
up with other people they've been on the team for
a little bit longer. Because one, when you get stuck
on a problem, someone who was working on the same
(14:17):
problem a month ago can be like, oh yeah, yeah,
do this and then that that like that'll help you
along faster. Right, So by having a more interconnectedness between
the members, you get to share that experience is a
lot faster, and then you get people up to speed
a lot faster. So that's like point number one. Point
(14:37):
number two is that it's also really important for people
to understand how their piece fits into the bigger the
bigger thing right early on, when we you know, okay,
we're going to do some project management stuff. Guys, let's
do some press management stuff. You tell people, okay, do this,
do this, do this, then we're gonna do a sprint
blah blah blah. It's very easy for an undergrad who
(15:00):
has you know, midterms and other things going on, to
just say, Okay, this task like I don't understand why
it's important. I don't understand why like it's given to me,
I don't really care, right, and they're not going to produce.
But if you meet with someone and you have a
one on one and you say like, hey, like, how's
it going. What are you working on? Oh, you're working
(15:21):
on this? Okay, Like yeah, you know. The really cool
thing is like, once you have that, then we can
do you know, once you have that new feature in
the controller, then we'll be able to like do this
turn at Manzo a little bit faster. And once we
can do that, then we can test you know, this
other thing. Right, And once someone understands how their pieces
kind of fit in, they get more motivated and they
(15:44):
they better understand Okay, this this, this is the impact
of my work. And then the third most important piece
and this feeds into the other two things I'm talking about.
It's really comes down to building a good team culture.
At the end of the day, people don't want to
disappoint each other. People all want to work together and
(16:05):
achieve something incredible and by really grounding. Okay, here's how
your piece fits into everything else. Here's you guys meeting
one on one, getting to know each other beyond just
a screen on a face. Right. Here's you know how
the team as a whole is going to achieve their goals.
(16:26):
That I think that culture of we are a team,
we we're starting from behind, you know, we're the underdogs here.
That really gets people motivated, That gets people moving, and
then you're able to you know, you're able to get
a lot more value in a lot less time.
Speaker 3 (16:45):
You know. One of the things that strikes me is
the the collegial, kind of collaborative environment that we have
on the track acurely during test days and shakedowns, maybe
less so the day of an actual competition, but in general,
(17:08):
the relationships that are built among the teams are more
about kind of working together to advance this concept of
autonomous racing. So what does that meant to you? The
relationships that you've developed, not just within your team, but
(17:30):
among the different teams. And you know, if you think
into the future, do you see, you know, a network
that will emerge from this that as you all move
into careers, either starting companies or joining companies, or staying
in academia and becoming becoming researchers or faculty. How do
(17:53):
you see the relationships and the network that's been developed
evolving over time?
Speaker 2 (18:01):
Yeah?
Speaker 3 (18:01):
Or how do you hope it evolves over time, maybe
I guess.
Speaker 1 (18:03):
Yeah, yeah. I mean for me, one of the most,
if not the most valuable pieces out of out of
I A C. That that have gotten out of it has
been has been the relationships and and all the incredible
people have gotten to meet, both on our team and
and on other teams. You know, some of my some
of my best friends now are are people I would
(18:26):
have never met if it weren't for the I A C.
And that that applies to both our team and other teams.
And a lot of a lot of what I've learned
is is not you know, is a lot of what
I've learned has come from from just seeing what the
other teams are doing and just learning from them and
like asking questions and and and helping each other the
(18:50):
For me, like I I just I just genuinely just
just genuinely love this stuff. I love like working on
the race cars, I love like robotics. And you know,
if I see someone else is like struggling with a problem,
I'm not gonna just ignore it because they're they're from
another team. I think, like you know, if if if
every team can solve the million dumb problems, right, the
(19:14):
million small problems and really focus on the hard ones.
We together all advanced, right because the solutions to the
heart problems, you know, no one, no one has has
the right answer, right, There's always there's gonna be dozens
of different answers and someone coming to that solution, and
you know, learning from each other. That's like where we
(19:37):
all grow together, we all advance as a field. So
for me, collaboration is is such an essential part of
no matter what the circumstances are for this because at
the end of the day, we're all trying to we're
all trying to do something that's never been done before.
And how I how I hope that it continues to go.
(20:00):
I hope it continues to.
Speaker 3 (20:01):
Be like this.
Speaker 1 (20:01):
I hope like all the other teams you know, are
continuing to have that collaborative nature. You know, whoever whoever
takes takes the mantle from me on the met PET team.
I mean, it's it's gonna be really important for us
to uh to continue that collaboration and and to in
that collaborative nature because the competition as a whole will
(20:26):
suffer and the field as a whole will suffer. So yeah,
I hope it continues I really I really do enjoy
working with everyone, and uh I hope I get to
continue to work on stuff with them in the future.
Speaker 2 (20:41):
It's been a fascinating conversation. Really really enjoyed your insights,
and we wish you the best of luck at Manza
the new challenge that's coming up here very soon.
Speaker 1 (20:50):
Thank you.
Speaker 2 (20:55):
That was Andy Saba explaining his personal and working philosophy
and why he finds us mentoring undergraduates so vital to
the development of autonomous vehicles and the next generation of engineers.
The next time on the Inside Track, we'll be speaking
with Gary Passon, the faculty lead for the AI Racing
Tech team at the University of Hawaii, on how a
(21:16):
school known for its oceanic and astrological studies is contending
with some of the most prominent engineering schools in the world.
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.