Re:Apollo – An open autonomous driving platform
by rottyguy on Tuesday July 18, 2017 @05:31PM
(#14800224)
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.
[ Parent
| on HN ]
Re:Apollo – An open autonomous driving platform
by dbcurtis on Tuesday July 18, 2017 @07:55PM
(#14801045)
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
by jnbiche on Tuesday July 18, 2017 @09:17PM
(#14801409)
> Typically the R/C car uses wifi to connect back to an AWS instance for the heavy lifting.
Wow, and the latency is manageable for the task?
Re:Apollo – An open autonomous driving platform
by koffiezet on Wednesday July 19, 2017 @05:29AM
(#14802997)
Well, the average human reaction time is 215ms, which would be considered unacceptable in online-gaming, where 50ms would already be on the high side of things, I normally see 30ms or so.
Let's assume 50ms or 20 updates per second, and a car driving at 40m/s or about 145kmh / 90mph, the reaction time of the of the computer is after the car moved 2 meters (about 6.5 feet). A human would only react after about 8.5 meters (about 28 feet) - quite the difference.
Re:Apollo – An open autonomous driving platform
by chplushsieh on Tuesday July 18, 2017 @08:01PM
(#14801069)
Related:
http://duckietown.mit.edu/
Re:Apollo – An open autonomous driving platform
by cr0sh on Tuesday July 18, 2017 @06:32PM
(#14800562)
Why do that, when you could train in a computer simulation?
Re:Apollo – An open autonomous driving platform
by thealexbaron on Tuesday July 18, 2017 @06:45PM
(#14800641)
Cooler, cuter, funner.