Giter VIP home page Giter VIP logo

knut0815 / aws-deepracer-navigation-pkg Goto Github PK

View Code? Open in Web Editor NEW

This project forked from aws-deepracer/aws-deepracer-navigation-pkg

0.0 0.0 0.0 23 KB

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.

License: Apache License 2.0

Python 100.00%

aws-deepracer-navigation-pkg's Introduction

DeepRacer Navigation Package

Overview

The DeepRacer Navigation ROS package creates the deepracer_navigation_node which is part of the core AWS DeepRacer application and will be launched from the deepracer_launcher. More details about the application and the components can be found here.

This node is responsible for collecting the reinforcement learning model inference results and mapping it to the servo message with throttle and steering angle values based on the action space for the particular model selected.

License

The source code is released under Apache 2.0 (https://aws.amazon.com/apache-2-0/).

Installation

Prerequisites

The DeepRacer device comes with all the pre-requisite packages and libraries installed to run the deepracer_navigation_pkg. More details about pre installed set of packages and libraries on the DeepRacer, and installing required build systems can be found in the Getting Started section of the AWS DeepRacer Opensource page.

The deepracer_navigation_pkg specifically depends on the following ROS2 packages as build and execute dependencies:

  • deepracer_interfaces_pkg - This packages contains the custom message and service type definitions used across the AWS DeepRacer core application.

Downloading and Building

Open up a terminal on the DeepRacer device and run the following commands as root user.

  1. Switch to root user before you source the ROS2 installation:

     sudo su
    
  2. Source the ROS2 Foxy setup bash script:

     source /opt/ros/foxy/setup.bash 
    
  3. Create a workspace directory for the package:

     mkdir -p ~/deepracer_ws
     cd ~/deepracer_ws
    
  4. Clone the deepracer_navigation_pkg on the DeepRacer device:

     git clone https://github.com/aws-deepracer/aws-deepracer-navigation-pkg.git
    
  5. Fetch unreleased dependencies:

     cd ~/deepracer_ws/aws-deepracer-navigation-pkg
     rosws update
    
  6. Resolve the deepracer_navigation_pkg dependencies:

     cd ~/deepracer_ws/aws-deepracer-navigation-pkg && rosdep install -i --from-path . --rosdistro foxy -y
    
  7. Build the deepracer_navigation_pkg and deepracer_interfaces_pkg:

     cd ~/deepracer_ws/aws-deepracer-navigation-pkg && colcon build --packages-select deepracer_navigation_pkg deepracer_interfaces_pkg
    

Usage

Although the deepracer_navigation_node is built to work with the AWS DeepRacer application, it can be run independently for development/testing/debugging purposes.

Run the node

To launch the built deepracer_navigation_node as root user on the DeepRacer device open up another terminal on the DeepRacer device and run the following commands as root user:

  1. Switch to root user before you source the ROS2 installation:

     sudo su
    
  2. Source the ROS2 Foxy setup bash script:

     source /opt/ros/foxy/setup.bash 
    
  3. Source the setup script for the installed packages:

     source ~/deepracer_ws/aws-deepracer-navigation-pkg/install/setup.bash 
    
  4. Launch the deepracer_navigation_node using the launch script:

     ros2 launch deepracer_navigation_pkg deepracer_navigation_pkg_launch.py
    

Launch Files

The deepracer_navigation_pkg_launch.py is also included in this package that gives an example of how to launch the deepracer_navigation_node.

from launch import LaunchDescription
from launch_ros.actions import Node

def generate_launch_description():
    return LaunchDescription([
        Node(
            package='deepracer_navigation_pkg',
            namespace='deepracer_navigation_pkg',
            executable='deepracer_navigation_node',
            name='deepracer_navigation_node'
        )
    ])

Node Details

deepracer_navigation_node

Subscribed Topics

Topic Name Message Type Description
/inference_pkg/rl_results InferResultsArray This message holds the reinforcement learning inference results for the state input passed through the current model that is selected in the device console.

Published Topics

Topic Name Message Type Description
/deepracer_navigation_pkg/auto_drive ServoCtrlMsg Publish a message with steering angle and throttle data sent to the servo package to move the car.

Services

Service Name Service Type Description
action_space_service LoadModelSrv A service that is called when a new model is loaded and helps set the action space to be considered while mapping the inference results.
throttle_service NavThrottleSrv A service that is called to dynamically set the scale value to multiply to the throttle in autonomous mode for each action.

Resources

aws-deepracer-navigation-pkg's People

Contributors

amazon-auto avatar pratik-nichat avatar siddalingesha-ds avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.