Skip to content

Latest commit

 

History

History
35 lines (22 loc) · 1.97 KB

README.md

File metadata and controls

35 lines (22 loc) · 1.97 KB
Raw

DeepRacer Custom Track

This repository contains custom tracks for AWS DeepRacer. It's purpose is to be an extension to the default tracks in the DeepRacer Simapp, which contains the tracks that are also available in the AWS DeepRacer Console. See the Race Data Repository's Track List for all details about the official tracks.

Tracks can be:

  • A remix of the official tracks by adding/removing features like buildings, lights etc. - whilst still keeping the original track shape.
  • A 'cut-out' of an offical track to reduce the size.
  • A completely new track.

Motivation for creating a new track can be:

  • Increased variability to reduce risk of overfitting for physical racing.
  • Creating a smaller foot-print track to allow for physical racing at home.

See the full track list for details of the tracks.

Using the repository

The repository works by adding its files 'on top' of an existing Robomaker/Simapp Docker Image.

To use the repository run the following commands:

  1. make build to collect the relevant files into the build/ directory.
  2. make image TAG=<your_simapp_tag> to extend the awsdeepracercommunity/deepracer-simapp:<your_simapp_tag> image.

Once built the image can be used by altering DRfC's system.env by using the new tag <your_simapp_tag>-ext.

Other commands

Two other commands are available:

  • make copy-src TARGET=<your_simapp_path> - this will copy the files into the source of the Simapp, i.e from build/ to $TARGET/bundle.
  • make copy-install TARGET=<your_simapp_path> - this will copy the files into the built bundle in Simapp, i.e from build/ to $TARGET/install/deepracer_simulation_environment/share/deepracer_simulation_environment/.

Contributing

The repository is open for contributors' tracks. See CONTRIBUTING for details.