this post was submitted on 08 May 2025
675 points (99.0% liked)
Technology
69867 readers
3031 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related news or articles.
- Be excellent to each other!
- Mod approved content bots can post up to 10 articles per day.
- Threads asking for personal tech support may be deleted.
- Politics threads may be removed.
- No memes allowed as posts, OK to post as comments.
- Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
- Check for duplicates before posting, duplicates may be removed
- Accounts 7 days and younger will have their posts automatically removed.
Approved Bots
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
FSD has lane-keeping in it. It’s not up for debate if they can do it or not, because they’ve been doing it for years.
Also I’m not sure what other technology you think they would use for lane-keeping other than cameras and “computer vision”? Things like Lidar don’t work for this because lidar can’t see lane markers. The only way to do it is with cameras.
Both. You can use both LiDAR and optical teaming, the technologies complement each other so you don’t fall for a Looney-Tunes ass painted wall, while the camera covers the one-dimensional recognition that LiDAR can’t.
Lidar does not help with seeing lane markers. At all. Radar can’t see painted lines on a road.
That looney tunes wall “test” was ridiculous and Rober was rightly raked over the coals and lost a lot of respect over it. It was basically a marketing stunt by his friends LiDAR company, and was full of dishonesty such as poorly photoshopped phones and lies such as not even using the self driving while smashing into the wall. These glaring flaws have been covered extensively.
Honestly, while not a scientifically rigerous test, it does demonstrate through absurdity the real risks of computer vision only for driver assist features.
Real world examples including of course the Tesla that plowed into a white truck on a foggy day because it mistook the truck for absolutely nothing among too many others
If only they didn’t fake it to get their desired result, then maybe it could have been useful.
I agree that LiDAR and other technologies should be used in conjunction with regular cameras. I don’t know why anyone would be against that unless they have vested interests. For various reasons though I understand that it isn’t always possible - price being a big one.