TetraVue, a Vista company that’s developed a 3D sensor system to help self-driving vehicles better identify what’s around them, said Thursday that it has raised $10 million in venture capital funding.
Robert Bosch Venture Capital and Nautilus Venture Partners led the funding round. The Bosch Group is a supplier of sensors and other products to the automotive industry.
They were joined by the Samsung Catalyst Fund and Foxconn, the Taiwanese contract manufacturer that builds Apple’s iPhones.
Founded in 2008, TetraVue had been funded until now by angel investors and nearly $10 million in government grants to demonstrate its technology, said founder and President Paul Banks.
“After a lot of pain trying to convince people that 3D is important and doing it right is invaluable, it is good validation that some heavy hitters in several industries understood what we are trying to do and are going to join us,” he said of the new funding round.
An engineer and physicist who previously worked at Lawrence Livermore National Laboratory and General Atomics, Banks said TetraVue’s solid state, flash LIDAR technology improves current LIDAR technology for self-driving vehicles.
LIDAR, or Light Detection and Ranging, works like radar. It pinpoints objects by measuring light bouncing off nearby cars and other objects.
LIDAR scanners are about the size of pickle jars. They’re typically mounted on roofs or near side-view mirrors of today’s prototype self-driving vehicles. Though prices are falling, they can cost as much as $70,000.
While LIDAR is good at measuring distances between objects, it can struggle to identify what those objects are – particularly if it’s something unexpected in the road.
TetraVue takes a different approach. It taps digital camera image sensors to deliver 2 million simultaneous distance measurements and ultra-high definition, 3D images surrounding a self-driving car.
“Instead of this collection of spots that is being scanned around, you get video with all the distance information at the same time,” said Banks.
The imagery is expected to make it easier for algorithms to identify cars, pedestrians and other things that pop up in an autonomous car’s path.
“Imagine a car being able to instantly determine if a black spot is a rock or a plastic bag,” said Connie Sheng, founding managing director of Nautilus Venture Partners in a statement. “TetraVue’s LIDAR gives cars perfect vision and allow them to make better decisions.”
The 10-employee company has built prototypes of its LIDAR system, which today is about the size of a large camcorder, said Banks. The new funding will help shrink the technology and reduce its cost, he said.
“Right now we have built a few of a kind, so they are expensive,” said Banks. “That is one of the things we have targeted with the money we have raised, to make the investment in the engineering to reduce the costs to at least match the lower performing LIDARs that are out there.”