The Garage by Sonatus

In this podcast episode recorded at CES 2026, Jeremy Vayssettes, Chief Technology Officer for Tire Digital Twin at Michelin, discusses how the company is expanding beyond traditional tire manufacturing into AI-powered software solutions. Jeremy explains Michelin's innovative tire digital twin technology, including SmartWear and SmartLoad algorithms that monitor tire condition in real-time using existing vehicle sensors without additional hardware. These models enable predictive maintenance, helping fleets optimize tire replacement timing, reduce costs, and improve safety. The conversation covers how the technology integrates with ADAS systems to provide better grip prediction and vehicle performance optimization throughout the tire's lifecycle. Jeremy also discusses Michelin's partnership with Sonatus, which uses the Sonatus AI Director infrastructure to scale these solutions across different vehicle manufacturers. The episode highlights the sustainability benefits of the technology, and how AI can help extend tire life while maintaining safety and performance.

Chapters:
0:00 Introduction to Michelin's Innovations
0:47 Introducing Jeremy Vayssettes
2:42 Understanding Michelin's Broader Mission
3:55 The Role of Software in Tire Technology
4:48 AI Innovations in Tire Management
7:54 The Benefits of Smart Sensors vs. AI
9:34 Developing a Comprehensive Tire Digital Twin
12:37 Sustainability in Tire Management
13:50 Collaboration with Sonatus
18:29 Future Prospects and Closing Thoughts

Creators and Guests

Host
John Heinlein, Ph.D.
An experienced technology and marketing leader, John brings his background from startups and established companies to Sonatus. He worked for 14 years at Arm, most recently leading Automotive Partnerships for North America where he engaged OEMs, Tier-1s, and others to deploy Arm-based solutions into automotive applications, including autonomous vehicles. His team was integral to launching the SOAFEE industry initiative for software-defined vehicles of which Sonatus is a member. Earlier he served as VP and Chief of Staff to the CEO and led a group responsible for competitive strategy. For three years, he was VP of Corporate Marketing where he led centralized outbound marketing, spanning marketing campaigns, press, events, web, digital marketing, ecosystem programs, and working closely with investor relations. Prior to Arm, John had an 11-year tenure at microprocessor startup Transmeta where he held several senior roles spanning business development, marketing, and customer success, among others. John earned his B.S. in Computer Engineering from Carnegie-Mellon University and an M.S. and Ph.D., both in Electrical Engineering, from Stanford University.

What is The Garage by Sonatus?

The Garage Podcast from Sonatus brings you conversations with thought leaders from around the vehicle technology space discussing far-reaching topics about software innovation in vehicles.

These episodes will include industry experts, whether from Sonatus, or from our amazing partners across the industry, all of whom will share their Ideas and their outlook for the most important topics in vehicle technology. Episodes span from more technical topics, to business evolution, and market trends. Join us and learn about the future of vehicle in The Garage.

Today in The Garage, our
guest is Jeremy Vayssettes,

who is Chief Technology Officer
for Tire Digital Twin at

Michelin.

In today's conversation, we
discuss all about Michelin's

innovation in algorithm
and software products.

While you may know about
Michelin as an incredible tire

innovator, there's much
more to it that they do.

They're bringing their know
how to make tires safer,

make them more sustainable,
better predictive maintenance,

and more advanced capabilities
for ADAS and performance.

It's a wide ranging
conversation showing what's

possible for AI in
new applications.

Let's go!

Welcome to The Garage.

I'm John Heinlein, Chief
Marketing Officer with Sonatus.

We're recording today live at
the Sonatus booth at CES 2026.

We're excited to welcome
Jeremy from Michelin.

Jeremy, welcome to The Garage.

Yeah.

Thank you, John.

So start by telling us about
you and your background.

So I have a background in automatic
control and signal processing.

I did engineering school and
I have a PhD in this domain.

And then I joined
Michelin as engineer

to develop tire models and
now I'm developing software

products for connected mobility.

It's quite a range and you had
also in the background in in

aerospace engineering.

What was that right?

I started my career as a
engineer in aeronautics

field in Toulouse
in south of France.

Doing two years as engineer,

then I switched to a PhD.

I did my three
years of PhD also in

this field, and then two
more years in research in

different labs before joining
Michelin ten years ago.

That's great diverse background.

You'll have to tell us a fun fact
about you to get to know you.

So I'm married. I have two sons.

And each year, we like
with my family to go to do

road trips around
Europe with a small van,

minimal stuff, no
technology at all.

No technology!?! Going around.

That's a very big commitment!

I guess my fun fact
back to you, right,

would be that my my family
likes to my my parents like to

to do that as well.

They like to travel and several
months of the year they travel

oftentimes in the south of the
US, go avoid the wintertime.

So it's it's a lot of people
are are love to do that,

but I it's so exciting for
you to to go away from technology.

That's a big commitment.

Yeah.

It's kind of a break every
year before coming back

to to to work with a
stronger commitment.

I think that's a that's a
great idea. Good for you.

So Michelin is a famous company and
many people may know the company.

But so tell us a little bit
about the company in general

and then about what
you do for the company.

Yeah.

So, yes, people know
Michelin usually

a bit because of tires.

But actually, Michelin
do much more than tire.

Michelin DNA is first more

about mobility in general.

And

also now Michelin is
diversifying activities

to go around tires
and beyond tires,

leveraging our skills and
expertise in high-tech

materials to develop high-tech
materials to new domains

like hydrogen, medical
and other many things.

And on my side, I'm
developing software products.

So embedded Tire Digital
Twin to be able to

provide insight the
vehicle information

regarding tires so that
the vehicle are aware of

the tire state
performance all the time.

Let's talk more about that.

I'm so interested to hear
because the average person

might assume, you know, they're
just a just a tire company,

but the reality is, as you
say, you're doing so much more.

So how does software in the
vehicle and and the software

defined vehicle change that
we we often talk about,

How does that give you
opportunities to do new things

with software in
addition to tires?

So with the shift
from hardware-centric

to more software-centric
in automotive industry,

It's actually a
great opportunity

for us, for Michelin,
companies like Michelin to

provide much more
than only hardware and

to also be able to
innovate more by

providing more smart
tires by connecting

them to their vehicles.

That's really
interesting to hear.

Can you help us
understand how [your]

innovations can benefit tires?

What are some of the algorithms you
have and kind of how do they work?

Among the product we have
we have SmartWear and SmartLoad

for instance, a product that we
developed to monitor the tire

wear, tire load during
the usage phase.

And thanks to that, we
are able to provide useful

insights to the
fleet, to the vehicle,

to the user about
the tire state.

And when the tire will have
to be changed, for instance.

So that you can and
fleets can have a

predictive maintenance
to reduce their costs,

reduce downtime and
have much more efficient

maintenance than
when it's reactive

or scheduled maintenance.

You're right.

So that's a that's a good point
because I think as one driver

and another driver might drive
the the same vehicle in very

different ways that might
cause different tire wear.

For example, aggressive acceleration
caused the tires to wear sooner and

perhaps driving in hot
conditions or cold conditions

could cause different wear.

So by by kind of monitoring
and studying how the tire was

actually used versus
some generic schedule,

you can be more accurate.

Yeah. Better safety,
maybe some cost savings.

All of the above, I
guess. Is that right?

Yes.

Yeah. And so let's talk about
these two different models.

You mentioned SmartWear
and SmartLoad.

So let's start with SmartLoad.

Can you explain what that does and
some of the inputs that go into that?

Yeah, sure.

So those models
take as inputs the

data coming from the vehicle,

from the sensor that are
equipping already the vehicle.

So it means it's only
software without specific

additional sensor that we
put in the tire for instance.

And from those data, we use
the tire physics and what

we know about the tire, the
interaction with the road, the vehicle

and some advanced AI
techniques and signal

processing to convert
those data into a

useful insight to

know what is the tire wear,

what is the tire load
in real time so that the

vehicle can adapt to the change
in the tire wear for instance.

But also the driver can

know what is the wear
level of the tire.

He can be warned when the
tire have to be changed.

And

also the fleet managers can
also get all those data from

all their vehicles to
know how they need to

plan the maintenance to
optimize it and to be sure that

they will change the
tire at the right time.

So not too soon
but not too late.

So many interesting
things you said there.

So I wanna I wanna touch
on a couple of ideas.

First, there's a lot of focus and
there's been some conversation

in the years past with putting
smart sensors in vehicles, in tires.

Of course, every tire in many
cases required by law has a

TPMS sensor, tire pressure
monitoring system sensor.

But there's been a conversation
in years past about adding an

an extra smart sensor that
does additional diagnostics.

But I think what you're saying,

and I think one of the key
innovations of this AI is you

get the benefits of those smart
sensors without the additional cost,

without the additional complexity
of those of those sensors.

So you can use, you know, any
kind of tire or, you know,

any kind of tire to get
these kind of benefits.

Is that right?

Yes. Completely.

And then the second thing
that was so interesting is

how it can really
benefit in multiple ways.

Because, of course,

there's a there's a benefit to
if your maintenance is is proper.

You you don't waste money
replacing the tires sooner than

necessary, but also not later.

Because you know when it's
the right time to replace it,

which provides better safety.

Right?

Yes. Exactly.

And then I think that the last
part that really struck me with

what you said is using your
knowledge of the materials and

using your knowledge
of the tires,

you can create you didn't say,

but kind of a digital twin
of the of the tire And

that you understand how that
specific tire or that specific

set of tires has been used.

And and it's really smart that
to use your unique know how to

be kind of the world's
expert in your tires and to

create this value
added solution on top.

Yes, exactly.

The ambition is not to
only develop two products

like SmartWear and
SmartLoad but a full tire

digital twin that will provide a full
package of insight regarding tires.

To

provide useful
insights to the driver,

so the the consumer.

To the vehicle so that
the ADAS system can

adapt also or to the
fleets so that they can

optimize their maintenance
and reduce their costs.

And this full digital
twin would come with SmartWear,

SmartLoad that provides useful
useful information regarding

the tire state because
during the usage phase,

the tire is changing a lot.

For instance, it's wearing
out and it affects a lot the

performance of the tire
and all the performance.

So when we talk about
grip, for instance,

it plays a potential role in
vehicle safety and performance.

Right.

Once we know in real time
what is the wear level of the

tire, then we can also provide
useful insight regarding

grip and the grip prediction.

So that the vehicle,

the braking system for instance
or the steering system can know

what is the maximum level
of grip the tire can provide

This instant, anytime,
anywhere so that the

system can leverage the optimal
level of tire performance

all along the usage phase.

It's a really really great
point because of course when

your tires are new,
your vehicle performs better,

your braking is more responsive,

you stop in shorter time.

So as the ADAS system
is able to be informed

that the tire is more worn,

it can provide automatic
emergency braking and other

kind of braking system tuning more
proactively to provide better safety.

So actually, the driver
receives not just at the tire

replacement time, but
throughout the life of the

vehicle, they get a
better experience,

better performance experience,
better safety experience.

And then of course later
on they get a better

knowledge of when to
replace your tire.

That's fantastic. Yes.

By doing this, we are able to

make the tire last longer
because people won't

change the tire too soon.

Right.

But also leveraging the
really the performance

potential of the tire.

So if we do better tires,

the user will benefit
more this all along

the tire life because
the vehicle will be able

to leverage those
performance better than what

it's done actually.

Well, you mentioned that because the
tires don't get replaced prematurely,

I'm assuming that provides a
great sustainability benefit as

well because we're not wasting
materials and we're not

obsoleting tires that have
perfectly good life left in them.

Is that right?

I know that sustainability is a
big focus for you as well, right?

Yeah, yeah.

It's one of the big
focus and the added value

because the automotive industry

has a lot of pressure to
reduce the environmental

costs and to

be more sustainable.

And for instance, when we talk

about tire wear,

the average the tire is

removed on average at
when there is still

3.5 millimeters
[worldwide average]

of tread When the legal
limit is 1.6 millimeter.

And when a tire in
general starts between six

and seven millimeters.

Right.

So it's more than
twenty percent.

Of life left over.

Of, the,
yeah.

The material that is wasted.

That's great. Well, that's
a that's a great benefit.

And I'm so excited to have have
had this conversation with you

that I'm even learning some
additional benefits along the

way that I hadn't thought of.

But we have to shift over to
talking about our fantastic

Sonatus and Michelin, who had
a fantastic collaboration.

And just as we're
recording this,

just right outside the
the podcast studio,

we have our demonstration
with you where we have your

algorithms running on our
Sonatus Director infrastructure.

I'd love you to talk about why
did you partner with Sonatus

and what are some of the
benefits that our Sonatus AI

Director deployment
infrastructure was able to

bring to Michelin as you deploy
your algorithms into vehicles?

Yes.

So first, we are
very complementary.

We

work and focus on the tire part.

And

with our expertise, we are
able to provide and to develop

those software that

could provide all the
benefits we mentioned before.

Sonatus, on your side,
you are developing the

vehicle infrastructure,

the software part and
also the ability to

manage quite efficiently
and properly the AI

models and deploying.

And thanks to [Collector AI] ,

we have a way to

calibrate quite efficiently
our models because they

need to be calibrated for specific
vehicles before being deployed.

So we can do this
efficiently and then

deploy to the vehicle --

thanks to AI

Director -- our models.

So it's a very efficient
way to be able to

provide and deploy
our models at scale

to many constructors
and carmakers.

We're so grateful for
your collaboration.

And first thing as you say,

it's absolutely complimentary
because we're providing we're

not trying to replace your
expertise and nor could we

possibly be an expert in the tire
physics and materials that you have.

But the infrastructure we
provide allows vendors like you

to bring that intelligence
you've generated into the

vehicle more easily.

So it's fantastic collaboration.

The second thing I think
is really interesting that you

mentioned is when
sometimes people think

about AI models
and they say, oh,

you train the AI model and now
it's trained and that's it.

We're done.

Well, first thing, of course,

AI models improve
over time in general.

There's no there's no finished.

It's we learn new things,
we improve the AI models.

But even more importantly
than that, as you mentioned,

and I learned this in in
working with you that the

specific model when deployed
to a specific vehicle still has

some calibration parameters that need
to be set for the

exact tires, the exact
load, the exact weight,

the specific vehicle.

So that then after some amount of
driving on that particular vehicle,

then the model is is more
accurate and ready to be fully

operational with that vehicle.

So we're excited to be able to
deploy not only our AI Director

deployment and AI management
platform for in vehicle edge

AI, but also the ability to
help you get that data for that

vehicle tuning more easily.

So it's a great multi-way
collaboration I think.

Yes.

And actually

having a tire digital
twin and products like

SmartWear or SmartLoad
is not only about having

good algorithms or models to

make good predictions
and accurate predictions.

It's also a matter of
being able to deploy

it to at scale, to the carmakers

in an easy and efficient way,

taking into account
all the constraints

we can face when
we want to do that.

And yeah, as we mentioned,

we developed those
products for any tires.

So it's tire-agnostic, not
only for Michelin tires,

for any vehicles.

But the "one size fits all"

does not exist or at least it will
be at the cost of loss of accuracy.

So we need to be a bit specific.

And the choice we made is
to have this specificity

at the vehicle model level so
that when a carmaker want to

adopt the solution, we have a
setup for the vehicle model.

And all the vehicles of this
model have the same setup so we

can deploy in any
of these vehicles.

Fabulous.

Well, it's so exciting
to see what's possible.

I've learned a lot in our
collaboration with you.

I think that the ability to
help accelerate the deployment

of such exciting technologies like
yours is very inspiring for us.

And we're we're thrilled for your
partnership and there's been incredible.

I hope you've had experience.

We've had incredible
interest in the demo nonstop.

People have been watching
it outside the the hall,

the booth all day
today and yesterday.

So thank you for coming to visit

with us, and thank you
for joining us on the podcast,

and thank you for
your partnership.

Yeah. Thank you very much.

If you like what you're
seeing on this episode,

please like and subscribe
to see more like it,

both from CES and other
events around the world.

We look forward to seeing
you in another episode of The

Garage very soon.