Like Tesla, Toyota is developing self-driving technology with low-cost cameras

0

San Francisco, April 6 (Reuters) – Toyota Motor Corp’s Woven Planet unit (7203.T) is following Tesla Inc’s (TSLA.O) lead in trying to advance self-driving technology with low-power cameras cost.

Woven Planet told Reuters it was able to use cameras to collect data and effectively train its self-driving system, a “breakthrough” it hopes will help cut costs and expand technology. technology.

Collecting diverse driving data using a massive fleet of cars is key to developing a robust autonomous car system, but it is expensive and not scalable to test autonomous vehicles with expensive sensors alone, he said. -he declares.

Join now for FREE unlimited access to Reuters.com

Register

Woven Planet still uses data collected from other sensors like radar and lidar for training as well as long-term deployment.

Tesla has relied on cameras to collect data from more than a million vehicles on the road to develop its automated driving technology, while Alphabet’s Waymo (GOOGL.O) and other self-driving car makers have added expensive sensors like lidars to a small number of vehicles. .

“We need a lot of data. And it’s not enough to have a small amount of data that can be collected from a small fleet of very expensive autonomous vehicles,” said Michael Benisch, vice president of the engineering at Woven Planet, in an interview. with Reuters.

“Rather, we’re trying to demonstrate that we can unlock the advantage that Toyota and a large automaker would have, which is access to a huge body of data, but with much lower fidelity,” said Benisch, a former director of engineering at Lyft’s self-driving division (LYFT.O), which Toyota acquired last year.

Woven Planet uses cameras that are 90% cheaper than the sensors it used before and can be easily installed in passenger car fleets.

He said using a majority of data from low-cost cameras increased his system’s performance to a level similar to when the system was trained exclusively on high-cost sensor data.

He said, however, that Toyota would still use multiple sensors such as lidars and radars for robotaxis and other autonomous vehicles to deploy on the road, as it currently seemed like the best and safest approach to developing robotaxis.

Toyota is also partnering with Aurora to test a fleet of autonomous vehicles based on Toyota Sienna minivans, equipped with lidars, radars and cameras.

“But many years from now, it’s entirely possible that camera-like technology could catch up and overtake some of the more advanced sensors,” he said.

“The question is perhaps more when and how long it will take to reach a level of security and reliability. I don’t think we know that yet.”

Tesla CEO Elon Musk said he could achieve full autonomy with cameras this year after repeatedly missing his previous goals.

Join now for FREE unlimited access to Reuters.com

Register

Reporting by Hyunjoo Jin Editing by Stephen Coates, Alexandra Hudson

Our standards: The Thomson Reuters Trust Principles.

Share.

About Author

Comments are closed.