Teradar emerges with $150M backing to replace radar and LiDAR in autonomous vehicles

“We went to the very last unused practical band—the terahertz band—and turned it into a sensor,” Carey said. “Long enough wavelength to bend around raindrops and snowflakes like radar, short enough to give you insane angular resolution.”

The technology performs consistently in adverse weather conditions, maintaining image quality through fog, rain and snow. “I can flood this entire scene with dense fog, heavy rain or snow, and our image stays exactly the same. Zero degradation. That is not an incremental improvement. That is a new category,” Carey said.

During the interview, Carey shared a technical demonstration comparing Teradar’s capabilities with a Continental ARS540 radar, a high-end sensor used in Mercedes S-Class vehicles and heavy trucks. The comparison showed Teradar producing approximately 20 times more data points than the radar, with sufficient detail to distinguish human limb movements and precise environmental features.

For automotive applications, Teradar’s technology particularly enhances Level 3 automation (hands-off, eyes-off driving). “Where we move the needle is true L3 at mass-market prices,” Carey said. The architecture allows performance tuning for specific vehicle segments: trucks receive long-range versions reaching 350–400 meters, while passenger cars might use versions optimized for 200–250 meters with denser point clouds for urban driving scenarios.

The company has prioritized supply-chain resilience with a manufacturing strategy focused on U.S. and European Union foundries. “No China sourcing for the critical stuff, and we have full Taiwan contingency plans,” Carey said. “Automotive OEMs demand supply-chain resilience; we were built that way from the beginning.”

Regarding future prospects, Teradar plans to showcase its production-intent silicon (B-sample) at the upcoming Consumer Electronics Show (CES). The company projects sensor deliveries to customers in 2027, with vehicles utilizing the technology on roads by 2028.

Teradar’s potential market position remains flexible, with some OEMs indicating they might replace radar with Teradar while retaining LiDAR, others planning the opposite approach, and a “growing camp” considering a streamlined perception suite of cameras and Teradar only.

“Our goal is to be good enough to replace both,” Carey said. “We don’t dictate architecture—we just give them a sensor so good they can choose whichever path makes sense for their brand and their market.”