Live data from Hacker News

Apollo – An open autonomous driving platform

github.com

31–40 of 80 posts

Re: Apollo – An open autonomous driving platform

#31
post #18

Earlier quoted context omitted.

Count me out on ever contributing to a project like this. I really don't want to be a test case.

Most open source projects have a disclaimer like this. You don't want to be responsible for the usage of whichever random programmer either.

I hope there is a difference in liability between a self-driving car library and it's left-pad dependency.

Re: Apollo – An open autonomous driving platform

#32
post #5

Autonomous driving based purely on machine learning from vision is scary. Machine learning is, after all, a statistical method. It's going to do really great most of the time, and really badly once in a while.

Humans also use statistics of course. You'll have to elaborate on other differences to be able to convey how scary machine learning really is.

Re: Apollo – An open autonomous driving platform

#34
post #19
post #8

Earlier quoted context omitted.

The exact same can be said about human drivers. Why do we demand that autonomous driving is 100% safe? I would be completely fine with autonomous cars that sometimes crash, but are still less likely to do so than with human drivers.

We accept that human drivers kill thousands upon thousands of people. We won't tolerate deaths from automation even in the single digits. That may not be rational but it's also unlikely to change.

I think it's a matter of sussing out liability.

Re: Apollo – An open autonomous driving platform

#35
post #33

I'd imagine this could work on a small fleet of miniature cars (rc size?). A model town/city could be built with various obstacles for training a large amount of the ai, I would think.

Why do that, when you could train in a computer simulation?

Cooler, cuter, funner.

Re: Apollo – An open autonomous driving platform

#37
post #19
post #8

Earlier quoted context omitted.

The exact same can be said about human drivers. Why do we demand that autonomous driving is 100% safe? I would be completely fine with autonomous cars that sometimes crash, but are still less likely to do so than with human drivers.

We accept that human drivers kill thousands upon thousands of people. We won't tolerate deaths from automation even in the single digits. That may not be rational but it's also unlikely to change.

The rational fear is of automation introducing systemic failures. Imagine a "flash crash"[1] type event applied to widely deployed self-driving vehicles.

[1]: https://en.wikipedia.org/wiki/2010_Flash_Crash

Re: Apollo – An open autonomous driving platform

#38
post #19
post #8

Earlier quoted context omitted.

The exact same can be said about human drivers. Why do we demand that autonomous driving is 100% safe? I would be completely fine with autonomous cars that sometimes crash, but are still less likely to do so than with human drivers.

We accept that human drivers kill thousands upon thousands of people. We won't tolerate deaths from automation even in the single digits. That may not be rational but it's also unlikely to change.

Yet, Robots and automated machines (not the autonomous vehicles) still killed/injured a few dozen people since 1984 according to U.S. Department of Labor. [1]

[1]: https://www.osha.gov/pls/imis/AccidentSearch.search?acc_keyw...

Worldwide stats will be far more than that. We've accepted it, haven't we?

Re: Apollo – An open autonomous driving platform

#39

I'd imagine this could work on a small fleet of miniature cars (rc size?). A model town/city could be built with various obstacles for training a large amount of the ai, I would think.

The autonomous racing league that meets in Oakland is doing exactly that. Typically the R/C car uses wifi to connect back to an AWS instance for the heavy lifting.

Re: Apollo – An open autonomous driving platform

#40

Awesome. Now to just get a couple LIDAR cameras...

Cheapest Lidar I know of with reasonable quality is a Neato vacuum cleaner. Go to Target, get a vacuum, give it a lidar-ectomy, throw away the sucky parts, and you still have the cheapest LIDAR in town. 350 pionts, 5HZ scan rate. Indoor range about 5m, outdoor 2+m with a sun shade, accuracy at 5m about +/- 2cm. Great for hobby robots. Totally not sufficient to drive a car at highway speeds.
Post reply on HN