Tesla’s choice using only cameras for its advanced driving assistance system has caused a lot of discussion. The American manufacturer, in fact, he does not intend to use any radar or LiDAR sensors on his car. According to the automaker, cameras, ultrasound sensors and neural networks are enough for the development of its Tesla Vision system. Although several manufacturers have been skeptical of this solution, Volkswagen reiterated the importance of using LiDAR, it seems that others want to follow the same path as the American brand.
Indeed, Woven PlanetToyota’s division that develops autonomous driving among other things, told Reuters it can use cameras alone to collect data and effectively train its autonomous driving system.
To enhance the autonomous driving capabilities, it is data collection is fundamental that allow to train and improve the artificial intelligence that manages the driving functions. According to Woven Planet, using simple cameras it is possible to collect all the data necessary to improve its platform.
This is an important “turning point” especially from an economic point of view as the cameras are 90% cheaper than the sensors used previously and can be integrated much more easily into the cars. Michael Benischvice president of Woven Planet with a background at Lyft, said:
We need a lot of data. It is not enough to have only a small amount collected from a small fleet of very expensive autonomous vehicles. Rather, we are trying to demonstrate the advantage that Toyota and the big automakers would have with access to a huge amount of information, albeit with lower fidelity.
Furthermore, according to the executive, the data collected only through the cameras gave improvements similar to those obtained with the data collected even from more complex and expensive sensors. For the moment, however, Toyota will continue to use radar and LiDAR for its future robotaxis and autonomous vehicles as it is the most reliable solution. For the future, however, the executive Woven Planet admits that technology based solely on cameras could surpass that based on the most advanced sensors.