- Tesla detailed hardware specs for its fully-automated Autopilot system in a live feed on YouTube.
- The changes include a dual-CPU setup that Elon Musk promises is fail-safe.
- Tesla's self-driving technology has been controversial as some claim it has been implemented before it is completely safe.
Tesla’s fullest realization of Autopilot-a completely automated system it claims is fail-safe-fits in the same enclosure as its older Autopilot hardware and can potentially be retrofitted to older Tesla models, the company said today in a webcast.
Pete Bannon, an ex-Apple engineer and Autopilot director, said the new hardware fits in a small enclosure between the firewall and glovebox. “It does not take up half your trunk,” Bannon said, in reference to autonomous prototypes other automakers have developed over the past several years.
The new hardware has been installed in Model 3 vehicles since April, he said, and in the Model S and X since February. Tesla has been working on the system for three years since Bannon started at Tesla. There is no indication when Tesla will activate the new Autopilot with a software update.
Bannon emphasized a “tremendous amount of redundancy and overlap” that would enable Autopilot to function even with multiple on-board failures.
CEO Elon Musk, as usual, went further to describe it.
“Any part of this can fail, and the car will keep driving,” Musk said. “The probability of this computer failing is substantially lower than someone losing consciousness.”
More to come as the livestream continues.
('You Might Also Like',)