This Year, Autonomous Trucks Will Take to the Road with No One on Board

Post by cozzyd » Tue, Jan 12, 2021, 3:31 AM UTC

Erlich_Bachman wrote: post #98

BurningFrog wrote: post #95

Earlier quoted context omitted.

Don't a lot of truck drivers own their trucks? Anyway, the human is a witness which adds problems. He's also a potential murder victim, which drives up the potential legal consequences by a lot. I don't have much real world experience with heist execution, so take my speculation for what it is.

> Anyway, the human is a witness which adds problems. Because an automated truck will not have cameras all around it and also inside, and will phone home whenever some alert has been triggered?

Post by ljlolel » Tue, Jan 12, 2021, 11:36 AM UTC

joehartshorn wrote: post #81

ljlolel wrote: post #67

Earlier quoted context omitted.

That's right the limitation is Data. You can't effectively use the extra compute if you don't have enough data to train a bigger neural network. https://jperla.medium.com/why-tesla-autopilot-ought-to-be-aw...

Do you actually have data to back that up? I read your article, which (no undue disrespect intended) looked to be full of unsubstantiated qualitative data/graphs and opinion. Are you an "expert"? Or are you trying to get medium clicks on your article/ purchases of your book?

Post by ljlolel » Wed, Jan 13, 2021, 6:43 PM UTC

joehartshorn wrote: post #104

ljlolel wrote: post #103

Earlier quoted context omitted.

I built self driving cars (at Lyft), including behavioral planning algorithms. I have run these essays by engineers at Waymo, Lyft, Uber, Nuro, etc and they have verified all of the information (I've incorporated any feedback).

That's fair, although credentials != proof. Working for Waymo, Lyft, Uber, Nuro etc is a credential. The engineers you mentioned have verified that what you wrote fits with the method/ approach they think will work. There is a possibility that they're/ you're wrong. (This isn't intended as a personal attack)