Levandowski drives coast to coast without touching steering

[CarlK]

I recommend you to check all what he claims. I did. I checked my eyes to a 18MP camera. In the focused spot my eyes are equivalent to a 24MP camera. Tesla uses a 1 MP camera now. I compared my vision in low light conditions. I checked the colors at high angles (walk backwards and you can clearly tell what colors are coming up at the edges).
Lot of bs regarding this subject.

Not again please. Non-scientific observation is the reason why some still think the earth is flat.

 

[CarlK]

That's not 100%. Tesla's intention is 100% coast-to-coast, including stops for recharging.

Elon stated in the Q2 earnings call:.

Elon's definition of coast to coast is to drive between two points on either coast via any (not pre-planned) routes. At least that's my interpretation.

If anything, this is just further evidence that Tesla's no-LIDAR approach is viable. In my opinion, Tesla is doing the right thing by focusing on the general solution to self-driving, rather than special cases for theatrical effect. When Tesla is able to do a truly 100% coast-to-coast drive, it will be far more impressive because it means the entire fleet is close to that capability.

This is the first time someone follows Tesla's approach instead of Waymo's for FSD. Very ironic that it came from someone with his background. George Hotz did the same thing but he started about the same time as Tesla and his development was geared more toward driver assistance rather than full self driving.

None has the large machine learning fleet on the road like Tesla has though. With all the excitement nothing is the same as if you could buy a car and replicate the same thing. Tesla is still the only company that could provide you with whatever you want to say is ahead or behind.

 

[CarlK]

Interesting to note that Lewandowski uses low resolution cameras... Lower resolution than smart phones. I think the key is the neural net. How much resolution do you need to see a car or a person or a dog? (Don't need to see squirrels... Dangerous to stop for squirrels)

Good point. Ability to process the image is way more important than resolution of the image. Too many pixels beyond necessary will only slow down the process. Just how much resolution you need to recognize cars or objects around you? You would never be able to drive if you try to read every license plate or see how every driver looks.