Toyota Motor unit Woven Planet is following the lead of Tesla in trying to advance self-driving technology with low-cost cameras.
Woven Planet said it was able to use cameras to collect data and effectively train its self-driving system, a “breakthrough” that it hoped would help to drive down costs and scale up the technology.
Gathering diverse driving data using a massive fleet of cars is critical to developing a strong self-driving car system, but it is costly and not scalable to test autonomous vehicles with expensive sensors, it said.
Woven Planet still uses data collected from other sensors such as radars and lidars for training as well as long-term use.
Tesla has been betting on cameras to collect data from more than one million vehicles on the road to develop its automated driving technology, while Alphabet’s waymo and other self-driving car firms added expensive sensors such as lidars to a small number of vehicles.
“We need a lot of data. And it’s not sufficient to just 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 engineering at Woven Planet.
“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 corpus of data but with a much lower fidelity.”
Mr Benisch is a former engineering director at Lyft’s self-driving division, which Toyota acquired last year.
Woven Planet uses cameras that are 90 per cent cheaper than sensors that it used before and can be easily installed in fleets of passenger cars.
It said using most data from low-cost cameras increased its system’s performance to a level similar to when the system was trained exclusively on high-cost sensor data.
But Toyota would still use several sensors such as lidars and radars for robotaxis and other autonomous vehicles to be sent on the road, as this seemed to be the best, safest approach to developing robotaxis, Mr Benisch said.
Toyota is also in a partnership with Aurora in testing an autonomous ride-hailing fleet based on the Toyota Sienna minivans, equipped with lidars, radars and cameras.
“But in many, many years, it’s entirely possible that camera type technology can catch up and overtake some of the more advanced sensors,” Mr Benisch said.
“The question may be more about when and how long it will take to reach a level of safety and reliability. I don’t believe we know that yet.”
Tesla’s chief executive Elon Musk said it could achieve full autonomy with cameras this year after missing earlier targets several times.
