AWS DeepRacer's Projects
Simulation artifacts, core components and configuration files to integrate AWS DeepRacer device with ROS Navigation stack.
The DeepRacer Camera ROS package creates the camera_node that is responsible for reading the raw data from the one/two cameras connected to the USB slots at the front of the AWS DeepRacer device and publishing them as CameraMsg.
The DeepRacer Control ROS package creates the ctrl_node that manages the different mode of operation of the device [manual, autonomous, calibration].
The DeepRacer Device Info ROS package creates the device_info_node that is responsible for providing the hardware version of the AWS DeepRacer device and software version of aws-deepracer-core debian package.
Learn to deploy an object detection model that enables the AWS DeepRacer device to identify and follow an object.
The DeepRacer I2C ROS package creates the battery_node that is responsible for providing the vehicle battery level information.
The DeepRacer Inference ROS package creates the inference_node that is responsible for running the inference on the machine learning model that is selected using the Intel OpenVino Inference Engine APIs.
The DeepRacer Interfaces ROS package is a foundational package that creates the custom service and message types that are used in the core AWS DeepRacer application.
Build robotics applications with AWS DeepRacer device software and hardware: https://www.amazon.com/dp/B07JMHRKQG
In this project, use the AWS DeepRacer car to draw a map with SLAM (Simultaneous Localization and Mapping), a technique for creating a map of an environment by estimating a device’s current location as it moves through a space.
The DeepRacer Model Optimizer ROS package creates the model_optimizer_node that is responsible for running the Intel OpenVino Model Optimizer script on the DeepRacer reinforcement learning models to obtain the optimized artifacts.
The DeepRacer Navigation ROS package creates the deepracer_navigation_node that is responsible for collecting the RL model inference results and mapping it to the servo message based on the action space for the particular model selected.
Provides a jailbreak experience of AWS DeepRacer, giving us more control over the training/simulation process and RL algorithm tuning
Use a series of QR codes as waypoints to navigate the AWS DeepRacer around a custom path. Create your own custom path by placing the basic waypoint codes as a sequence along the track or encode additional instructions in them to invent new applications.
The DeepRacer Sensor Fusion ROS package creates the sensor_fusion_node that is responsible for collecting the messages from all the sensors (Camera+LiDAR) and publish the combined sensor message.
The DeepRacer Servo ROS package creates the servo_node that is responsible to map the input servo throttle and steering angle messages to raw PWM values to move the car.
The DeepRacer Status LED ROS package creates the status_led_node responsible for the blink and solid light effects on Wi-Fi and Power status led lights by the side of the DeepRacer device.
The DeepRacer Systems ROS package creates the software_update_node, model_loader_node, otg_control_node, network_monitor_node which are responsible for managing system level functionalities required in the DeepRacer application.
The DeepRacer USB Monitor ROS package creates the usb_monitor_node which monitors the connection/disconnection of the USB drive and provides functionality to watch for specific files/folders in it.
The DeepRacer Web Server ROS package creates the web_publisher_node which launches a Flask application as a background thread and creates service clients and subscribers for all the services and topics that are required by the APIs called from the DeepRacer vehicle console.
A Python library to interface "RL Lab - DeepRacer" environment.
A python library to manipulate "RL Lab - DeepRacer" environment configurations over the UDE side-channel.
The AWS DeepRacer Track Geometry package provides a simple interface to access "RL Lab - DeepRacer" track information.
An open-source reinforcement learning environment build toolkit for ROS and Gazebo.
ROS1 kinesis video streamer
ROS1 Video Recorder to save MP4 files from camera topic and upload to s3 bucket
An open source Python library for virtualizing reinforcement learning environments.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent