Tesla's Vice President Tao Lin issued a statement today, reaffirming the company's commitment to the development of autonomous driving technology. She mentioned that many years ago, company founder Elon Musk determined that true fully autonomous driving could only be achieved through pure visual technology for enhanced safety and intelligence.
She explained that since existing roads and traffic rules are designed based on human vision, the combination of cameras, visual neural networks, and autonomous driving hardware is the best way to mimic human observation, perception, and decision-making habits, potentially achieving or even surpassing human driving performance.
Tesla's autonomous driving technology utilizes an end-to-end large model that is capable of processing the entire process from photon input to decision output. The company believes that a fusion solution of cameras and LiDAR does not simply achieve additive effects; rather, if there is conflicting information between radar and cameras, it could create decision-making difficulties for the "brain" of the autonomous driving system. Additionally, the inclusion of LiDAR increases vehicle costs, even though its prices are decreasing. Tesla is committed to reducing unnecessary components to ensure that consumers' investments are focused on the most critical areas.
Tao Lin also announced that all models currently sold by Tesla are equipped with the latest autonomous driving hardware, the AI4 chip, which has five times the computing power of the previous generation, marking Tesla's readiness in hardware to achieve fully autonomous driving.
This progress demonstrates Tesla's deep commitment and continuous innovation in the field of autonomous driving technology, as the company consistently enhances the safety and intelligence of its autonomous vehicles.
