Tesla’s Benz-baiting self-driving functions will soon receive their first upgrade. In a recent tweet, CEO Elon Musk outlined Autopilot 1.01 would have “curve speed adaption, controller smoothness, better lane holding on poor roads, improved fleet learning!” In our best Keanu Reeves impression: “Excellent!”
We tested Tesla’s beta Autopilot software using a P90D upon its over-the-air cellular release to all Model S owners with cars built after September 2014 (the only cars fitted with the ultrasonic sensors, radar, and visual camera required to perform Autopilot tasks). In short: It works pretty darn well, with the obvious limitation that it’s still imperfect enough that you’ll need to pay attention, especially around curves, as some Tesla owners have found out as they’ve let their $100,000 sedans lead them straight into guardrails. No one said well-heeled people were necessarily geniuses.
But those people—not including us, although our long-term P85D is equipped with the $4250 Autopilot package—will help improve the Autopilot software by letting Tesla spy on their sensors. That’s what Musk means by “fleet learning!” Similar to how mapping companies are relying on these same sensors to update high-definition maps for autonomous cars in real-time, so goes Tesla.
Most Model S customers enjoy being unpaid beta testers and don’t mind if Musk knows where they drove last night (which, in February 2013, Musk proved after he revealed detailed driving data from a New York Times reporter). In fact, judging by Musk’s social-media followers and product devotees ready to praise every corporate announcement, they just might love that he knows. We’ll keep an eye out for additional Autopilot updates and share our experiences with you, but we still can’t deny the exhilaration of engaging Insane/Ludicrous Mode with both hands firmly on the wheel.
This entry passed through the Full-Text RSS service - if this is your content and you're reading it on someone else's site, please read the FAQ at http://ift.tt/jcXqJW.
from Car and Driver Blog http://ift.tt/1Sa3SrT
via IFTTT
0 comments:
Post a Comment