AWS DeepRacer Robomaker Image

This repository contains the extracts from the AWS DeepRacer Robomaker simapp; including

• the environment and tracks
• the robotics agent code ('markov')
• Dockerfile

Available pre-built versions

In most cases it will be sufficient to use one of our pre-built images which use optimized Tensorflow binaries.

• CPU images support training using the CPU for inference.
• GPI images support training using CUDA GPU for inference.

Both containers support OpenGL acceleration.

Built images are available via docker pull awsdeepracercommunity/deepracer-robomaker:<tag>.

Version 5.2

Version 5.2 is the current development version for 2024, and it is receiving both new tracks as well as functionality updates as they are released by AWS.

5.2 introduces an upgrade to Ubuntu 20.04, ROS Noetic, Gazebo 11 and Tensorflow 2.

Version	Comment	CPU	GPU
5.2.0-dev	Initial test version	5.2.0-dev-cpu, 5.2.0-dev-cpu-avx2	5.2.0-dev-gpu
5.2.1	Release	5.2.1-cpu	5.2.1-gpu
5.2.2	Patches, and Telegraf Metrics	5.2.2-cpu	5.2.2-gpu

Version 5.1

Version 5.1 was the development version for 2023. It is now deprecated.

Version	Comment	AVX	AVX2	GPU
5.1.0	Initial release	5.1.0-cpu-avx	5.1.0-cpu-avx2	5.1.0-gpu
5.1.1	CUDA 11.4.4 + misc	5.1.1-cpu-avx	5.1.1-cpu-avx2	5.1.1-gpu
5.1.2	Node management + misc.	5.1.2-cpu-avx	5.1.2-cpu-avx2	5.1.2-gpu

Building the image

A build script is available as build.sh. By default both images will be built (cpu and gpu for version 5.3). Use the -a switch to limit number of images.

Tensorflow build

As of Version 5.2 the standard PyPI Tensorflow image is being used, which supports CPU and GPU inference.

Development build

To get a folder compatible with DRfCs DR_ROBOMAKER_MOUNT_SIMAPP_DIR use the build-dev-bundle.sh script. Through using a build container it will create a ./install directory that can be mounted in Robomaker.