Woven Planet, a subsidiary of Toyota Motor (7203.T), has joined Tesla Inc (TSLA.O) in attempting to enhance self-driving technology without the need of pricey sensors such as lidars.
Woven Planet told the media that it can collect data and efficiently train its self-driving system using low-cost cameras, a “breakthrough” that it thinks would help lower costs and scale out the technology.
It added that gathering diverse driving data using a large fleet of cars is crucial to establishing a viable self-driving car system, but testing autonomous vehicles with pricey sensors is costly and not scalable.
While Alphabet’s (GOOGL.O) Waymo and other self-driving car companies added pricey sensors like lidars to a small number of vehicles, Tesla has been relying on cameras to collect data from over 1 million vehicles on the road to develop its autonomous driving technology.
“We’ll need a lot of information. And having a little quantity of data that can be obtained from a small fleet of extremely expensive autonomous vehicles isn’t enough “Woven Planet’s vice president of engineering, Michael Benisch, stated.
“Rather, we’re attempting to show that we can unlock the advantage that Toyota and a large manufacturer would have, which is access to a massive corpus of data with much lower fidelity,” said Benisch, a former engineering director at Lyft’s (LYFT.O) self-driving division, which Toyota purchased last year.
Woven Planet employs cameras that are 90% less expensive than the sensors it previously employed and can be simply put in passenger car fleets.
It claimed that incorporating a majority of data from low-cost cameras improved the system’s performance to the same level as when it was trained purely on high-cost sensor data.
He did say, though, that for robotaxis and other autonomous cars to be deployed on the road, Toyota will continue to use numerous sensors such as lidars and radars, since this appeared to be the best and safest strategy at the time.
“However, it’s completely possible that camera-type technology may catch up and overtake some of the more advanced sensors in many, many years,” he said.
“It’s possible that the concern is more about when and how long it will take to achieve a level of safety and reliability. That is something I don’t believe we are aware of yet.”
Elon Musk, the CEO of Tesla, stated that the company will be able to achieve full autonomy with cameras this year, despite missing his previous targets numerous times.