News for hackers. Stuff that matters.

A comment on The Next Leap in Self-Driving: Prediction. Parent comment

Re:The Next Leap in Self-Driving: Prediction by KKKKkkkk1 on Friday February 21, 2020 @01:05AM (#22381239)
It's categorically untrue that perception is solved, given that things like this happen: https://twitter.com/greentheonly/status/1228067903666348045. And that's considering that Tesla's perception team is run by one of the world's most respected experts in the field, and that their access to data is essentially unlimited.

Now what bothers me is that leaders in the industry say without batting an eyelid that perception is solved, given that they witness this type of events on a regular basis.

[ Parent | on HN ]
Re:The Next Leap in Self-Driving: Prediction by anonymousCar on Friday February 21, 2020 @01:36AM (#22381336)
There's a lot of assumptions in OPs post, but maybe we can assume they're talking in context of their sensor suite.

Tesla has made a name for over promising, under delivering on fsd, and refuses to consider lidar as a member of their sensor suite.

I don't think you can adequately compare two companies with vastly different approaches.

[ Parent | on HN ]
Re:The Next Leap in Self-Driving: Prediction by HellDunkel on Friday February 21, 2020 @01:59AM (#22381413)
Elon Musk recently claimed „correct vector-space representation“ is the hardest task of autonomous driving. If true or not for Tesla i dont know but i guess its a yet to be solved problem for everyone in the field. Would you agree?
[ Parent | on HN ]
Re:The Next Leap in Self-Driving: Prediction by Hydraulix989 on Friday February 21, 2020 @03:23AM (#22381697)
It looks like they are still using single-frame classifiers instead of temporally coherent models?!
[ Parent | on HN ]