Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Elon: "Already testing traffic lights, stop signs & roundabouts"

This site may earn commission on affiliate links.
I know what he said, but I’m not sure I believe it. The leaks we heard suggest Tesla tried gaming it and it still didn’t work out.
Well, you have to either believe the leaks or him.

With the amount of short interest in TSLA, its very, very difficult to figure out what damaging leak/information regarding Tesla is real and what is FUD to depress the stock price. We are talking about Billions of dollars and zero chance of anyone getting caught.
 
  • Like
Reactions: dhanson865
Thank you for the detailed answer.

I already drove AP1 for 4 years and with that hardware a roundabout was suicide.

But since i'm happy to get a Model 3 next week, I wanted to know how the progress is on that topic, but it seems it is not really better.

Where I live, they changed all traffic light crossroads to roundabouts. So this really matters, since Tesla talks about automatic driving on city streets later this year (text form the Tesla Model 3 Order Page):

"Coming later this year: Recognize and respond to traffic lights and stop signs. Automatic driving on city streets."

If they can't handle roundabouts till then, they better implement a safety net, like: A loud warning to take over and stopping the car before entering a roundabout, if no response from the driver!

The actual behavior, even what you described, would be very, very dangerous.

I'll hope for the best, but since they don't talk about roundabouts at Tesla, it probably is not on their radar.
Sorry to hear about all the intersection conversions to roundabouts. Be prepared to have your passenger front tire chewed up on a regular basis.
 
I'm a tall.
And my brief YouTube survey of "tesla autopilot roundabout" indicates AP does not support them currently.
When they do, they had better fix Google's picking of the wrong lane at State and Ellsworth in Ann Arbor, otherwise, there will be honking...
 
  • Funny
Reactions: am_dmd
In software development cycle "feature complete" means every piece of the software is in place and passed alpha testing. The software is ready for beta testing, either to selected people or everyone.
I've worked in software development my entire professional career (at a large company and now a much smaller one). The definition can vary depending on the group or company.

From what I've observed, usually devs use the term code complete, meaning they did complete all the feature work but it doesn't necessarily work very well and hasn't been necessarily been tested much at all except by maybe the developers themselves. At that point, it's almost never ready for beta testing to externals to the company or even those outside the team.
 
  • Informative
  • Like
Reactions: Richt and Earlian
When I was doing software product development it meant about what cwerdna said. All the code is there for minimal operation of all functions but it's messy and buggy and barely integrated. Alpha testing came later when the developers thought it had a reasonable chance to keep running and behave reasonably. Beta was when it could be put into friendly customers' hands knowing that is was still a bit shaky.

Personally, I'll be astonished if FSD is even close to feature complete within 3 years, let alone this year. To me "feature complete" means it can at least recognize and take reasonable action for most edge cases even if it can't do it in bad weather or if several happen at once, etc.
 
The case here is very different from traditional software development. In traditional software the final beta testing is for confirmation and bug fixing. In deep machine learning the neural net will still continue to develop by itself. Unlike traditional software development with a few testers and engineers Tesla's will have hundreds of thousands neural nets in the field to continue the work. This is a powerful tool and needed to solve this complex self driving problem. For edge cases the NN will solve it long as it has seen enough of them by all the cars. There will always be cases that are so rare but human drivers can't handle everything either.
 
  • Like
Reactions: EinSV