this post was submitted on 03 Aug 2023
376 points (96.5% liked)
Technology
58303 readers
11 users here now
This is a most excellent place for technology news and articles.
Our Rules
- Follow the lemmy.world rules.
- Only tech related content.
- Be excellent to each another!
- 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, to ask if your bot can be added please contact us.
- Check for duplicates before posting, duplicates may be removed
Approved Bots
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
My understanding was that the challenge in making the next leap in self driving was not based in hardware (detecting objects with cameras vs LiDAR), but in software. As in, it isn't as difficult to detect the presence of objects as it is to make consistent and safe decisions based on that information.
But using LIDAR, you increase your data's accuracy and dimensionality, giving you more options to play with. It probably won't be a game changer, but it may be better than a camera only system.
Gathering more data, and being able to process it seems obvious as a way forward. How much better is this "new" LIDAR?
Edit: seems Tesla cars doesn't even use LIDAR...
They did. And every other competitor does. Musk believes since humans can drive with only two eyes that cars should be able to as well. Maybe someday, but nowhere in the near future. Cameras miss too much and are easily blinded.
It's also really stupid because the idea is to create a system that's better than humans. And let me tell you, people miss stuff all the time when driving. Tons and tons of accidents are caused by "negligent" drivers who looked both ways and missed someone due to a visual processing error or literally not being able to see something.
That's not necessarily true. What you get is two separate things inputting raw data into a system that both need to be parsed. Sometimes, one won't agree with the other and can cause issues with how the car thinks it should respond.
Nobody has a fully working system at this point, so it's premature to make claims about what hardware is and isn't needed. It may very well be that LIDAR is a requirement, but until somebody figures it out, we're all just speculating.
You can, today, download an app and go ride in a self-driving car around multiple US cities. All of those cars use LIDARs. Sensor disagreement is not a major issue because sensor fusion is a very well-understood topic.
Yes but they geofenced those cars into areas with the most optimal conditions for autonomous driving. What happens when you take the car on the freeway, a suburban neighborhood, or a mountain pass?