Amazon DeepRacer is the fastest way to get started with machine learning (ML). You can use the 1/18 scale Autonomous Vehicle Training Intensive Learning (RL) model in a cloud-based virtual simulator and compete for prizes and honors in the Global Amazon DeepRacer League. Today, we will extend Amazon DeepRacer functionality by open-source Amazon DeepRacer device software to provide fun hands-on learning.
Why Open Source
Amazon DeepRacer virtual and offline races have been well received, but the developers now want their cars to go beyond racing leagues. Amazon DeepRacer is an Ubuntu-based wheeled computer car powered by our open-source robotic operating system (ROS) that allows developers with basic Linux coding skills to easily prototype interesting new uses for their cars. 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’ve compiled six sample projects from the Amazon DeepRacer team and members of the Global Amazon DeepRacer community to help you start exploring the endless possibilities open source can make possible. When developers share new projects using # DeepracerProject, we’ll highlight our favorites on the Amazon DeepRacer Robotics project page. Whether it’s using DeepBlaster to mount a Nerf cannon on a car, Mapping to create virtual effects in your home or office, or DeepDriver to come up with new ways to race with friends and colleagues, You can do all of this using open source code and sample projects. Documents can be found on GitHub and collaborate with thousands of community members on Amazon DeepRacer Slack channels. The only limit to Amazon DeepRacer’s potential is your imagination (and, of course, the laws of physics).
Let’s start experimenting
By open-source the Amazon DeepRacer device code, you can quickly and easily change the default behavior of the cars currently tracked at the track. Want to deploy countermeasures to stop other cars from overtaking? Want to deploy your own custom algorithm to increase the speed of A vehicle from point A to point B? You just have to think and code. We’d love to see your ideas, from new racing forms to new uses for Amazon DeepRacer.
Today, 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 one of your own. You can start by following the lead example project, which trains the car to detect and track an object. This is the fastest build and run project, and in the next section, we’ll demonstrate how easy it is to modify the default behavior of Amazon DeepRacer cars. To complete this setup, upgrade to the latest software version and then SSH into the car.
Upgrade version link:
Docs.aws.amazon.com/deepracer/l…
Download the Follow the Leader project
Connect to the car using SSH, switch to root, and create a working directory. Then clone the “Follow the leader” GitHub repository:
Sudo su mkdir -p ~/deepracer_ws CD ~/deepracer_ws git clone github.com/aws-deeprac…
The process of fully cloning the project repository to the car may take several minutes (depending on your network connection speed). The Follow the Leader project includes several installation scripts that help simplify getting up and running faster. In addition, if you’re more comfortable running shell-based commands or want to see the process in detail using documentation for each stage, you can do the following steps manually.
Download and transform the object detection model
First, we need to download and transform the object detection model. To do this, we can run the script that comes with 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 setup script will download and optimize the model, and then copy the optimization project to the model location. This process takes about 3-4 minutes to complete.
You can complete this phase manually using the detailed instructions for downloading and converting the object detection model:
Github.com/aws-deeprac…
If not initialized previously, rosdep is initialized
Rosdep facilitates the installation of dependency packages. Rosdep is initialized first if it has not been initialized previously on the device.
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
Upon 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. This process may take about 8-10 minutes to complete.
You can also complete this phase manually by following steps 1-10 in the download and Build section of the leader readme.md. The installation script will perform the same steps (save you some typing).
Launch the Follow the Leader app
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 open another SSH session for the vehicle using the command line interface (CLI) to enable “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 begin to follow! Isn’t it great?
Share your results
A: congratulations! You have completed your first sample project. Use the hashtag # DeepracerProject to share your experience with friends and family on social media so we can keep track of your work. As the community creates more Amazon DeepRacer projects, we’ll be adding them to the Amazon DeepRacer GitHub organization and detailing them in future blog posts so everyone can get inspired.
The resources
Amazon DeepRacer:
aws.amazon.com/deepracer/
Amazon DeepRacer Robotics Project:
Aws.amazon.com/deepracer/r…
DeepBlaster:
youtu.be/vP5-n_1Ze1M
The Mapping:
youtu.be/hMYKFOKqDWs
DeepDriver:
youtu.be/uGFHACot0e0
Making:
Github.com/aws-deeprac…
Amazon DeepRacer Slack channel:
deepracing.io/
Follow the leader:
youtu.be/8U9pSWYJt1E
Cross-country:
Github.com/aws-deeprac…
RoboCat:
youtu.be/_JzgC4b4xFA
Connect to the car using SSH:
Docs.aws.amazon.com/deepracer/l…
Follow the leader Readme.md:
Github.com/aws-deeprac…
This author
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 crashing), 3D printing, running (park runs), tinkering with code and spending time with his family.
[Amazon Cloud Technology China Summit 2021] Take you to feel the four-wheel brother era!
The summit will be held in Shanghai, Beijing and Shenzhen, bringing multi-dimensional perspective sharing around amazon cloud technology’s reliable technologies and services with many leading technology practitioners in the industry. The Amazon DeepRacer Driving Track was specially built in the exhibition area, allowing you to visit the front line of AI and get into the forefront of machine learning.
What are you waiting for? Sign up for this marathon with Amazon DeepRacer for a crazy and exciting auto race! Let’s pay tribute to the four-wheel drive boy with AI!