Vehicle Perception for my Self Driving Car
HTML-код
- Опубликовано: 20 дек 2019
- In this video I explain what I've been up to for the past 3 months, working on developing a "self driving car" for my senior capstone. This video I go over the Vehicle Perception system I built to collect data from a video feed of the road. Make sure to like, share and subscribe, stay tuned for more updates on this project as well as new projects soon to come.
Twitter: / greerviau
Support me on Patreon: / greerviau
Songs by Blue Wednesday:
Cabin Crew ft. Taiyo Ky
Deep In Love (feat. Taiyo Ky)
Middle School (feat. Magnus Klausen) Наука
Cs student myself.
At like 5 mins in my brain goes: "he should cache that and avera... there he goes yup".
It's 12.38 a.m .
I should really sleep.
Yet I'm gonna head out and watch a neural network play snake, it looks way better than my bed.
You’re officially at code bullet level man.
Love your channel love your video style.
Lol he’s above code bullet
Good luck with the project!
Love the videos man good luck on your capstone
Greer, Fantastic... beyond words. Watching this with Paula. - Uncle Chuck
I love this channel.
Quality. Good luck bro
At 05:00 its like: "I'm out!"
Ya thats what happens when theres no lane lines on the road
my blood pressure rises slightly every time the clip replays and I see that black honda ridgeline cut you off, then proceed to pass everyone on the right...
When you say you're using the average of the last 5 frames, do you mean median average or mean average? Because in the event of a huge miscalculation, the mean average could still give you a totally fucked value, while a median value would be able to ignore one or two fucked values without a noticable hickup.
I'm not sure what you mean by "mean average" and "median average", mean and average are the same thing and there is no such thing as median average. Also taking the median of the cached values wouldn't work, since my cache is simply a queue, if you take the median of a queue and insert it back into the queue, then the median of the queue will stay the same and the queue will fill up with that median value and no new values will be added. Taking the mean doesn't entirely prevent large variance it is simply meant to greatly reduce it but you need some variance to allow for the curvature calculation to be changed normally when turning.
If you cannot explain it to a 5year old you haven't learn it quite enough. But really you have a done a great work.
Uhh I wouldn’t consider that as a basis on what you learn, I highly doubt a kindergartner can grasp quantum physics
@@elizabethpavelka2998 with all due respect from where I come 5year olds are not in a kindergarten. But you made your point. Now I would like to make mine. If you explain entanglement to a 5 year old I think he would be fascinated,if you just left out the details and the maths out of it.
@@nirbhay_raghav where tf are u from that u think 5 year old will understand this??
you could -use comma ai-
if you need help editing videos, I could help you. I'll do it free, if I can cite you as source of experience
Can I get the code of this plss..
It should be on my GitHub as GlorifiedCruiseControl and it would be on the gcc-v1 branch
sentence
Somehow I find the background voice of this vedio like Gilfoyle from silicon valley, hhh
Pid control