> camera only navigation without sensor fusion with LIDAR will fail

I'm not so sure on this one. I think we'll see it this year. It will have embarrassing bugs (ie. running over cats which are hiding under the car) and we'll see lots of issues to begin with (ie. the car stops in the middle of a freeway because a camera got splattered with mud).

But I think they'll achieve the goal of something that can be deployed fairly widespread without public outrage causing it to be banned without lidar.

▲

brk 7 hours ago | parent | next [-]

It's been "coming this year" for almost a decade now. The bugs you describe are not embarrassing, they are critical issues that prevent it from being called FSD.

▲

londons_explore 5 hours ago | parent [-]

This is the first year that I personally think that it will come this year...

Actions speak louder than words, and the fact that a 'cybercab' production line is firing up this year is also a strong indicator - the fact they didn't do that 5 years ago means tesla leadership didn't think it was going to work back then. 'cybercab' wouldn't sell well as a 2 seater if it couldn't self drive. (although the actual mass production will be delayed till next year is my guess, but we'll see model 3 or y being used for a taxi service in the meantime)

	▲	brk 4 hours ago \| parent [-]
		It will be interesting to see. At this point, I think much of what is coming out of Tesla in both words and actions is stock price theater. They "fired up" a Cybertruck production line and we all see how that went.

▲

vimda 5 hours ago | parent | prev [-]

How are those "bugs" not immediately disqualifying? "Move fast and break things" is not an acceptable strategy for controlling 2 tonne bricks hurtling down the freeway