TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Pixelopolis, a self-driving car demo from Google I/O built with TensorFlow-Lite

149 点作者 baylearn将近 5 年前

11 条评论

axegon_将近 5 年前
Damn it... Just before the pandemic I had started piling up hardware to start building something similar(I was intending to use a raspberry pi). And since I couldn't go and look for parts in person(I'm not that crafty with my hands so I genuinely need to see in person the part in order to be able to tell if it's going to work) so I postponed it. And today I see this... Feeling kind of sucks, even though I can't really explain why.
评论 #23829214 未加载
评论 #23829228 未加载
评论 #23835473 未加载
Abishek_Muthian将近 5 年前
Amazon has been selling it&#x27;s self-driving AWS Deep Racer car[1] and has been promoting race tournaments for a while now.<p>Has anyone got that? Is it worth the price to test self-driving algorithms in the hopes of making it to the full sized vehicle one day or is it best we make our own self-driving toy card with Nvidia Jetson?<p>[1]<a href="https:&#x2F;&#x2F;aws.amazon.com&#x2F;deepracer&#x2F;" rel="nofollow">https:&#x2F;&#x2F;aws.amazon.com&#x2F;deepracer&#x2F;</a>
评论 #23830784 未加载
评论 #23834296 未加载
评论 #23832120 未加载
Animats将近 5 年前
They were trying very hard to use Tensorflow. It&#x27;s amusing that their first pre-training operation failed because they trained on straight roads. Direct camera to steering is possible but not really the best approach.<p>Micromouse competition, 2019.[1] The first run, it&#x27;s learning the maze. The second run is a speed run.<p>[1] <a href="https:&#x2F;&#x2F;youtu.be&#x2F;sKFIBQ64_zs" rel="nofollow">https:&#x2F;&#x2F;youtu.be&#x2F;sKFIBQ64_zs</a>
评论 #23832015 未加载
hendi_将近 5 年前
105.41 MB transferred<p>Maybe think twice before putting a 40MB GIF on a website?
评论 #23830473 未加载
praveen9920将近 5 年前
Wonderful. Are there any other &quot;robotics&quot; projects which use android phones instead of rasberry-pi or microcontrollers?<p>Imagine in future there are generic robots. Rather than trusting its custom AI&#x2F;processing, I can connect my phone to the robot and my phone gets hardware&#x2F;robotic extension and is personalised for me.<p>Its like ironman suit for my phone :)
评论 #23858905 未加载
sandGorgon将近 5 年前
does anyone how the android phone is interfacing with the rest of the car hardware ?<p>i did not know that android phones could do realtime i&#x2F;o like this. Is this the &quot;Android Things&quot; or &quot;Android Accessory Protocol&quot; ?
评论 #23829791 未加载
评论 #23829725 未加载
评论 #23829562 未加载
评论 #23829539 未加载
quadrifoliate将近 5 年前
One thing I have always failed to understand about self-driving tech -- what&#x27;s the motivation for each car to do its own self-driving computation in busy metropolitan areas?<p>Are there technologies that integrate with some kind of <i>existing</i> infrastructure like beacons, etc. that would just <i>tell</i> the car where the streets are? If not, why not?
评论 #23834350 未加载
评论 #23833128 未加载
评论 #23832918 未加载
评论 #23833520 未加载
评论 #23833547 未加载
评论 #23833607 未加载
domovoykot将近 5 年前
There is a definite resemblance to Duckietown[1], a robotics class&#x2F;competition done developed at MIT.<p>[1] <a href="https:&#x2F;&#x2F;www.duckietown.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.duckietown.org&#x2F;</a>
评论 #23838676 未加载
woile将近 5 年前
Can this be scaled into building many small cities for self-driving cars training? Could it be used for real scale? Kind of like GTA V is used for training.
pselbert将近 5 年前
“Please excuse the crudity of this model. I didn’t have time to paint it or build it to scale” — Doc Emmett Brown (Decidedly not Google)
numpad0将近 5 年前
Interesting that they went for expensive microcontroller based servos, I guess they consider Realtek chip RC cars to be just toys.
评论 #23830796 未加载
评论 #23832062 未加载