Does AP actually pay attention to engagements and learn/send to Tesla? Perhaps a better question is if “shadow mode” is actually useful data?
good pedestrian test. Note the star next to pedestrian ID - that means car senses it as an obstacle so on real AP would not try to drive through it. Also there's radar return the whole time too. Conti radar FTW? also note how distance jumps back and forth - IC jitters reason.pic.twitter.com/NDUGNp0Xqn
-
-
-
*disengagements
End of conversation
New conversation -
-
-
I love these videos you post. Thanks!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
NN itself does not have any memory. Every frame is from a blank state. There's some state outside the NN, and that's how object IDs are persistent from frame to frame. But it does not last too long, if an object disappears from view, it'll come as a new ID should it reappear.
- Show replies
-
-
-
Their neural net is so impressive
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I saw this article that states that Continental radar identifies pedestrians. My post here:https://teslamotorsclub.com/tmc/threads/hw2-5-capabilities.95278/page-6#post-2267058 …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
thank you again for sharing these.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks for this! Has been impressive watching my AP 2 S learn this last 2 yrs. appreciate your info. Fascinating!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
FWIW, that jogger is crazy. Should run facing the traffic, and with clear distance from the traffic lanes. Just asking to get into trouble.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.