When Amazon debuted AWS DeepRacer in 2018, it was meant as a fun way to help developers learn machine learning. While it has evolved since and incorporated DeepRacer competitions, today the company announced it was adding a new wrinkle. It's open sourcing the software the company created to run these miniature cars.
At its core, the DeepRacer car is a mini computer running Ubuntu Linux and Robot Operating System (ROS), both open source components. The company believes that by opening up the device software to developers, it will encourage more creative uses of the car by enabling them to change the car's default behavior.
"With the open sourcing of the AWS DeepRacer device code you can quickly and easily change the default behavior of your currently track-obsessed race car. Want to block other cars from overtaking it by deploying countermeasures? Want to deploy your own custom algorithm to make the car go faster from point A to B? You just need to dream it and code it," the company wrote in a blog post announcing the open source project.
After introducing the cars in 2018, the company has developed in person DeepRacer leagues, and more recently virtual races. In fact, the company reorganized the leagues last month to encourage new people to get involved with the technology. Adding an open source component could increase interest further as developers get a chance to make this their own, and really add new layers of usage to the cars that haven't been possible up until now.
The idea behind all of this to teach developers the basics of machine learning, as AWS' Marcia Villalba wrote in a blog post last month:
“AWS DeepRacer is an autonomous 1/18th scale race car designed to test [reinforcement learning] models by racing virtually in the AWS DeepRacer console or physically on a track at AWS and customer events. AWS DeepRacer is for developers of all skill levels, even if you don’t have any ML experience. When learning RL using AWS DeepRacer, you can take part in the AWS DeepRacer League where you get experience with machine learning in a fun and competitive environment.”
If you want to get involved customizing your car's software, the project documentation is available on GitHub and on the AWS DeepRacer Open Source page, where you can get started with six sample projects.