Hands-on training of your own unmanned vehicle, this super service is now open source!

Amazon DeepRacer is the fastest way to start using machine learning (ML). You can train a reinforcement learning (RL) model with a 1/18 scale self-driving vehicle in a cloud-based virtual simulator, and compete for prizes and honors in the global Amazon DeepRacer League. Today, we will extend the functionality of Amazon DeepRacer by open-sourcing the Amazon DeepRacer device software to provide interesting hands-on learning.

should open source

Amazon DeepRacer virtual races and offline races have been well received, but developers now want their cars to surpass the racing league. Amazon DeepRacer is an Ubuntu-based computer wheeled car. The device is supported by our Robot Operating System (ROS), which can open source the code. It allows developers with basic Linux coding skills to easily make interesting new cars. Use for prototype development. Amazon DeepRacer device software is now publicly available, so anyone with a car and an idea can make new uses for their device a reality.

We have compiled 6 sample projects from the Amazon DeepRacer team and members of the global Amazon DeepRacer community to help you start exploring the endless possibilities that open source can achieve. When developers use #deepracerproject to share a new project, we will highlight our collection on the Amazon DeepRacer robot project page. Whether you use the DeepBlaster project to mount the Nerf cannon on the car, or use the Mapping project to create a virtual effect of your home or office, or use the DeepDriver project to propose a new way to race with friends and colleagues, you can use open source code and The sample project accomplishes all these tasks. Documentation is available on GitHub, and you can collaborate with thousands of community members in the Amazon DeepRacer Slack channel. The only limit to the potential of Amazon DeepRacer is your imagination (and of course the laws of physics).

Let's start experimenting

After the Amazon DeepRacer device code is open sourced, you can easily and quickly change the default behavior of the car on the track currently being tracked. Want to prevent other vehicles from overtaking by deploying countermeasures? Want to deploy your own custom algorithm to increase the speed of the vehicle from point A to point B? As long as you dare to think, then you can code. We very much hope to see your ideas, from new racing forms to new uses for Amazon DeepRacer.

From now on, you can choose from six projects (Follow the leader, Mapping, Off-Road (created by Amazon Web Services), RoboCat, DeepBlaster, and DeepDriver (created by the open source community), or create your own project. You can choose from Start with the lead sample project, which will train the car to detect and track an object. This is the fastest to build and run the project. In the next part, we will demonstrate how easy it is to modify the default behavior of the Amazon DeepRacer car. To accomplish this Settings, please upgrade to the latest software version, and then enter the car via SSH.

upgrade version link:

https://docs.aws.amazon.com/deepracer/latest/developerguide/deepracer-ubuntu-update-preparation.html

Download the "Follow the Leader" project

Use SSH to connect to the car, switch to the root user, and create a working directory. Then clone the "follow the leader" GitHub repository:

sudo su
mkdir -p ~/deepracer_ws
cd ~/deepracer_ws
git clone
https://github.com/aws-deepracer/aws-deepracer-follow-the-leader-sample-project.git

The process of fully cloning the project repository into the car may take several minutes (depending on the speed of your internet connection). The "Follow the Leader" project contains several installation scripts to help simplify the startup and operation process faster. In addition, if you are more accustomed to running shell-based commands or want to learn more about the process of using relevant documents at each stage, you can also complete the next steps manually.

Download and convert object detection models

First, we need to download and convert the object detection model. To do this, we can run the script attached to the "Follow the Leader" repository:

sudo su
cd ~/deepracer_ws/aws-deepracer-follow-the-leader-sample-project/installers
/usr/bin/bash install_object_detection_model.sh

The installer script will download and optimize the model first, and then copy the optimization project to the model location. This process takes approximately 3–4 minutes to complete.

You can complete this stage manually using the detailed instructions for downloading and converting the object detection model:

https://github.com/aws-deepracer/aws-deepracer-follow-the-leader-sample-project/blob/main/download-and-convert-object-detection-model.md

If not previously initialized, initialize rosdep

Rosdep helps to install dependency packages. If it has not been initialized on the device before, initialize rosdep first.

sudo rosdep init
sudo rosdep update

build the "follow the leader" package

Next, we need to extract the package dependencies required by the project and build them:

sudo su
cd ~/deepracer_ws/aws-deepracer-follow-the-leader-sample-project/installers
/usr/bin/bash build_and_install_ftl_application.sh

After success, you should be able to view a screen similar to the following:

The script will download and install the required package dependencies, and build the package. It may take approximately 8–10 minutes to complete this process.

You can also follow the steps 1–10 of the "Download and Build" section of the leader README.md to manually complete this stage. The installation script will perform the same steps (just save you some typing).

Start the "Follow the Leader" application

Now, we can run the "follow the leader" application:

sudo su
cd ~/deepracer_ws/aws-deepracer-follow-the-leader-sample-project/installers
/usr/bin/bash run_ftl_application.sh

enable "follow the leader" mode

Finally, we need to use the command line interface (CLI) to open another SSH session for the vehicle to enable the "follow the leader" mode:

sudo su
cd ~/deepracer_ws/aws-deepracer-follow-the-leader-sample-project/installers
/usr/bin/bash enable_ftl_mode.sh

Now, you or the volunteer (or an object) can move and watch the vehicle start to follow! Isn't it great?

Share your results

Congratulations! You have completed the first example project. Use the hashtag #deepracerproject to share your experience with friends and family on social media to let us know how you work. As the community creates more Amazon DeepRacer projects, we will add them to the Amazon DeepRacer GitHub organization and introduce them in detail in subsequent blog posts, so that everyone can get inspiration.

Reference

Amazon DeepRacer：

https://aws.amazon.com/deepracer/

Amazon DeepRacer Robot Project:

https://aws.amazon.com/deepracer/robotics-projects/

DeepBlaster：

https://youtu.be/vP5-n_1Ze1M

Mapping：

https://youtu.be/hMYKFOKqDWs

DeepDriver：

https://youtu.be/uGFHACot0e0

GitHub ：

https://github.com/aws-deepracer/

Amazon DeepRacer Slack channel:

https://deepracing.io/

Follow the leader:

https://youtu.be/8U9pSWYJt1E

Off-road:

https://github.com/aws-deepracer/aws-deepracer-offroad-sample-project

RoboCat：

https://youtu.be/_JzgC4b4xFA

Connect to the car using SSH:

https://docs.aws.amazon.com/deepracer/latest/developerguide/deepracer-manage-vehicle-settings.html

Follow the leader README.md:

https://github.com/aws-deepracer/aws-deepracer-follow-the-leader-sample-project/blob/main/README.md

Author of this article
David Smith
Amazon DeepRacer Senior Solution Architect

He is passionate about Amazon DeepRacer, technical support and learning. Outside of work, he likes Formula 1, drone flying (and crashes), 3D printing, running (park running), patching code and spending time with his family.

[Amazon Cloud Technology China Summit 2021] Take you to experience the era of four-wheel drive brothers!
The summit will be held in Shanghai, Beijing and Shenzhen. Together with many industry-leading technology practitioners, we will bring a multi-dimensional forward-looking perspective to share around the reliable technology and services of Amazon Cloud Technology; and a special "Amazon "DeepRacer driving the racing track" allows you to personally visit the front line of AI and go deep into the forefront of machine learning development.

For more summit information, please click to view https://aws.amazon.com/cn/about-aws/events/summit2021/index_dev/

what are you waiting for? Hurry up and sign up to join the marathon at the summit site and have a crazy and exciting autonomous car driving competition with Amazon DeepRacer! Let us use AI to pay tribute to the chasing four-wheel-drive boys!