Hacker News new | past | comments | ask | show | jobs | submit login
AWS DeepRacer League (amazon.com)
145 points by jeffbarr 33 days ago | hide | past | web | favorite | 44 comments

I got a DeepRacer car at Reinvent, but three months later I still can't use it :(

Access to the DeepRacer section of the AWS console hasn't been made public, and I don't think there's any other way to program it.

It's cool that they are making a more packaged product. For a more hackable setup the original Donkey Car setup is pretty good and available now:


Of interest , the DeepRacer does reinforcement learning while donkeycar has a completely different approach (supervised learning). It would be interesting to see them compete

Donkeycar is open source and modular by design, so nothing prevents using reinforcement learning. People are going there already.

This is incredibly cool, and I'd love to enter, but despite having one of the races in Sydney, the actual DeepRacer hardware doesn't ship outside of the United States. Baffling from a company that has a local presence in Australia.

If anyone from AWS is reading, I'd love to buy one!

Any idea if you can enter your own homebrew system? I mean, the hardware listed isn't really all that special. You could probably take this:


Slap it on new-ish Samsung or Apple phone, coupled to a cheap-ish RC car...

That'd probably get you 80% of the way; also, check out NVidia's "End-to-End" self-driving vehicle paper:


I built a tensorflow model from that trained on a 750 TI for the Udacity Self-Driving Car Engineer Nanodegree (it was part of the course to build such a thing that ran on a virtual track), and after training it did quite well - at least on a virtual track.

A few tweaks to the model, maybe using the TensorFlowLite package or something on the phone - probably work fine for 99.9% of the runs...

I actually completed the Udacity SDC course as well and implemented the same Nvidia NN.

I looked into building my own implementation, but decided against it. I'm not particularly interested in the hardware, just the algorithms that drive it, and the amount of time spent getting the hardware working (not to mention a simulator ) would have been nontrivial.

Perhaps I'll grab a Donkey.

I'm not with Amazon and as much as I'd like to do the same, I suspect not.

The whole idea of the competition would be to use the same hardware so it's a level playing field from a hardware perspective which turns it into a software competition, right?

I assume it'll appear in the local Amazon instances soon enough. It doesn't ship until April 15th so plenty of time.

Is it hotlapping or a race? One is a lot easier than the other as evidenced by codemasters's valiant attempts to write AI over the last few years.

Also, it looks like they don't know the track beforehand but they seem to be taking absolutely bizarre lines through bends (Or is that an artifact of the model used?)

This is hotlapping, it seems. A race would be cooler.

you might be interested in marble racing, it is very exciting and uses much less energy

Correct. But you don’t get the RL/ML/DL experience out of it.

True, but once self-automated racing cars are a solved thing, marble racing will still be more interesting to watch.

Would be nice if there was a program to get these into high schools to bootstrap a more software oriented robotics option.

Also a decent quality five or six DOF arm option instead of the racer might be interesting. Just dont call it DeepGripper.

There is already FIRST robotics competitions for that. FIRST tech league in particular is pretty software oriented. There programs are a great way to get students into programming and learning about writing real software.

Wasn’t aware of tech league. My daughter tried FIRST a few years back and it was way more hardware/build oriented which kind of killed her interest.

They dont even ship to australia but they have a Sydney event?

There are a bunch of events before they actually ship. They gave a bunch of them out at this past year's re:Invent, so I guess only those people who got them can participate until then.

Udacity SDC alum here as well and this looks awesome. I'm tempted to buy one after reading about DIYRoboCars. I've been thinking about porting openpilot to my Jetson TX2 but have been reluctant since I have no way to test it. DeepRacer seems much funner and safer. I wonder how hackable the car will be. How hard would it be to set up a remote ROS nodes rather using AWS services?

Amazon gets to keep the winners models?

The models should be limited to within the confines of the s3 location you specify (and the physical deepracer device) unless you explicitly permit otherwise. Nothing that I see in the terms of use [0] cover access to the model.

[0] https://aws.amazon.com/deepracer/device-terms-of-use/

They already have them all.

Now that nvidia has released a cheaper jetson board, I'm really itching to get it and build a small RL car. Training on the device would be out of the question though.

Anyone know of any successful DIY small RL cars?

It's not clear to me why deep learning is required here. It seems like some dead simple lane following code would probably work just fine. This is generally how these competitions are done.

I got a deep racer at Re:invent but I'm still waiting to get into the preview to play with it. Anyone from AWS cruising the comments here who might be able to enable it for my account?

This is that weird future shit I was hoping would start to happen.

darn. the window for $249 deepracer is over? Still mentioned as $249 'introductory discount' at https://aws.amazon.com/deepracer/pricing/ but upon clicking through to Amazon.com it's $399 now.

My reading is that pre-order is more expensive than launch price. "Can" vs "will be"?

"You can pre-order your AWS DeepRacer car on Amazon.com for $399. For a limited time at launch AWS DeepRacer will be available at an introductory discount price of $249."

It was previously $249 on Amazon.com. It's now $399. Not sure how to make it any more clear.

I think he is confused by "Launch of pre-orders" vs "Launch of the product". It's an easy mistake to make.

Why is it so expensive?

Speaking for hobby RC car racers in Seattle, WA, average racer spend upto $1000 for a decent rig. That too when there are no returns on winning races.

are the hobby racers in Seattle switching to the deepracer platform since it's only $399?

Must the training be done in their cloud?

You require AWS to at least activate and set up DeepRacer [0].

You don't strictly speaking need the cloud for training a DeepRacer model, although cloud access (via the DeepRacer console) could be the path of least friction.

Much of the source code for training DeepRacer on AWS can be found on the amazon-sagemaker-examples github repo [1].

The sample training environment relies on Gazebo [2] which, while freely available, has a managed AWS offering [3]. But ultimately, you should just be able to deploy an RL Coach model to the device [4].

[0] https://docs.aws.amazon.com/deepracer/latest/developerguide/...

[1] https://github.com/awslabs/amazon-sagemaker-examples/tree/ma...

[2] http://gazebosim.org/

[3] https://docs.aws.amazon.com/robomaker/latest/dg/simulation-t...

[4] https://docs.aws.amazon.com/deepracer/latest/developerguide/...

Why is the age limit 18?

For some reason most robotics contests seem to be restricted to highschool students. It's bizarre. It's like "Hey kids get into STEM because robots are cool! OK now stop and go build accounting software."

Edit: Thinking more about it, it's even worse. "Hey kids, robotics is a toy for kids to play with."

I think a big part of it is because as an adult with the time, inclination, some money, and programming skills you're expected to just go off and build things.

I guess the same holds true for other adult things like paint-and-sips (you drink and paint a picture) though, so maybe you've come across an untapped market for bot-and-booze.

Yeah, generally that's true. I think I may be a different case (https://arxiv.org/abs/1807.08233).

Nice to see another high school student here!

Just wanted to compliment you on your self-driving car design. Also skimmed your blog and paper. My intended area of focus is generative modeling but I thought your project was awesome nonetheless.

> maybe you've come across an untapped market for bot-and-booze.

I am so down for this. Maybe I should organize something.

Because ultimately you will have to pay for AWS and they don't want a kid to charge her parent's credit card.

I'm more of a GCP fan since they provide so much in GCP credits. Unless you're doing hundreds of hours of compute, I don't think that getting a relatively high AWS bill to your card will be a problem.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact