But Tesla will gladly keep charging a lot of money for FSD.
this post was submitted on 02 Nov 2024
940 points (96.5% liked)
Technology
59626 readers
2814 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
Anyone knowledgeable about city planning? Why did we never put some type of signal in our roads? (I don't know. Passive RFID every few feet?) It would only cost what, ten, twenty thousand on top of each million spent paving every mile?
Seems it would be better baseline navigation than self driving cars and occasionally map apps. The cars would still have to do obstacle avoidance, of course.
I'm not particularly knowledgeable about self driving tech or city planning. But if interstates are replaced every 10 years, and highways every 20, and Musk first made these claims in 2013? Then we'd have the base tech for every auto manufacturer to do moderately reliable self driving on interstates and a lot of our highways already.
Or maybe that large view pathfinding is the relatively easy part? That's why I'm asking. I'm sure there's something more obvious from an informed viewpoint that I don't know.
I imagine power is the tricky part. Badge readers and the like that use RFID also use wireless electricity to "power" the card. The range of that is limited without massive coils. You may be able to harness power from heat in asphalt (from traffic or sunlight beating on it), but I'd think that'd also be very limiting.
Better would be low power RF beacons set up at every transformer or every N utility poles. Something like BLE, maybe a little bit beefier. Power is readily available. They don't require data. All they need to do is broadcast their exact location and time (which they can get from GPS receivers).
Y'know GPS didn't even enter my mind. Hell, depending on GPS 3 accuracy (isn't it supposed to be in the centimeters?) my talk of signals is completely moot. That measured against a map of roads on a server somewhere would probably let you download an entire map of nodes toward your destination. Along the way the car just measures against its current location and does the math for obstacles. Great point. This is why I ponder shit out loud. Thanks.
It is way more experience. Has to withstand great pressure and sheer forces. Updates as road changes. Heat of building asphalt and later of summer. Horrible road maintenance. And what does it provide exactly to a car that actually helps? How long until most roads are updated with it?
Yay for OpenPilot
Good. I don't trust them dern robuts.
"Fuck that particular con, I have a president in my pocket now."
I realized self-driving on roads is impossible for so-called when someone pointed out what human drivers do when there's like a flock of geese camped out in the middle of the road.
We know that we should slowly move forward until they get out of the way, including bonking then with the car (gently). Do we want cars deciding that some obstruction in the road is "ok" to hit? I don't. So what's the solution? Something other than pure autonomous self driving.
We can probably have some very high level driver assist. Maybe.
All the issues with self-driving could be solved if they actually gave a shit about making it work. You don't let the machine choose. You give it hard fucking rules to follow. It doesn't need to identify geese, human, ball, dog, child to react differently to each; it should see an obstruction and stop to avoid damaging the fucking object and car, regardless of what it is. They are making it way more complicated than it really has to be.
load more comments
(5 replies)
load more comments
(1 replies)