Page 40 - EE Times Europe Magazine - June 2025

40 EE|Times EUROPE

Imec’s Vision for Automotive Over the Long Haul

Example configuration of an automotive sensor suite supporting advanced driver assistance (Source: imec)

thermo-mechanical simulation vehicles and sensor-laden dummy chips frequency-modulated continuous-wave LiDAR that removes the
to predict failures. “We want them to fail,” Placklé said, “because every mechanical weaknesses of today’s systems.
delamination, every ball shear, tells us where the weak points are—and According to Placklé, no single company can make chiplets and
how to improve the design for reliability.” sensors work in automotive alone. A coordinated ecosystem is needed.
That’s the premise behind imec’s Standardization and Automotive
BUILDING AN ECOSYSTEM FOR SMART, SUSTAINABLE MOBILITY Reuse (STAR) initiative, which brings together OEMs, Tier 1s, and
Compute alone doesn’t make a car smart; sensors are also needed. semiconductor players. “We need standards, not silos,” he said. “If one
Imec’s SENSAI program is advancing next-gen modalities, including vendor’s chiplet can’t talk to another’s, the whole model breaks down.”
CMOS cameras, shortwave infrared imaging, and fully solid-state sili- Through workshops, forums, and technical focus groups, STAR is
con photonics LiDAR. “In today’s perception stacks, it’s the Wild West,” building consensus around interfaces, protocols, and interoperability
Placklé said. “Every OEM is trying different sensor combinations, with layers. “We are creating the foundation for economies of scale,” Placklé
no standard on what works best.” said. “We’re tapping into the collective wisdom of the ecosystem.”
Imec’s approach is to build a simulation framework—essentially As vehicles evolve into software-defined platforms, long-term
digital twins of sensor architectures—to test configurations virtually. hardware headroom becomes critical. “Software-defined vehicles are a
“We can simulate how a new sensor affects accident rates or corner joke without compute headroom,” Placklé said. “You need to design for
cases without physically building it,” Placklé said. “This saves cost and updates, for added functionality five or 10 years down the road.”
accelerates development.” But long-term vision must also account for sustainability. Automo-
Imec is also running focused research and development projects, tive players are well-versed in measuring the CO 2 impact of mechanical
including work on a wide virtual-aperture radar system composed systems—but not electronics. Imec is working to quantify the silicon
of multiple coherent modules for greater accuracy, and a solid-state lifecycle footprint and design lower-power architectures to reduce
operational emissions. “We’re guiding the
industry toward both performance and sus-
tainability,” Placklé said. “Low-power design
directly reduces CO 2 , whether you’re pulling
from a battery or burning fuel.”
With everything from modular chiplets to
virtual radar arrays, imec is reshaping how the
automotive industry approaches electronic
systems. “It’s a long game, with a goal of OEM
adoption around 2030,” Placklé said. “But the
clock is ticking. Our A-sample platforms need
to be on the bench by 2027. That gives OEMs
time to validate, integrate, and scale.”
In this convergence of silicon and steel,
imec plays a vital role—not just as a tech
provider, but as a catalyst for a smarter, safer,
and more sustainable automotive future. As
Placklé put it, “We’re bringing semiconductor
Comparing the resolution of 79-GHz and 140-GHz radar at a 30-meter range (with a innovation to wheels—and making sure it
radar aperture of 15 × 15 cm) (Source: imec) lasts the whole ride.” ■

JUNE 2025 | www.eetimes.eu