Using NVIDIA Jetson Nano with ROS Python

Learn Deep Learning using NVIDIA Jetson Nano with IgnisBot

Course Overview


This course will take you from the basics of the NVIDIA Jetbot API, to a Deep Learning-based collision-avoidance system, and ending in a people-following ROS system that uses Deep Learning. And you will learn all that while building your own NVIDIA Deep Learning Robot.

Learning Objectives

  • The basics of NVIDIA jetson NANO setup.
  • Move a Jetbot based robot
  • Train a robot to do obstacle avoidance through deep learning
  • Track people and follow them.
  • Execute code designed for GPU-CUDA enabled hardware in only CPU systems.
  • Build your own IgnisBot, a robot designed for DeepLearning with JetsonNano Hardware.

Who is this course for?

  • If you are interested in Artificial Intelligence and Deep Learning, but you don’t know where to start, this course is for you.
  • If you want to have an affordable physical robot platform that is CUDA capable and has Deep Learning capabilities, this course is for you.
  • If you want a step-by-step guide to having a Jetbot fully set up for ROS, Deep Learning, and expandable for all your AI experiments, this is the course for you.

Simulation robots used in this course

IgnisBot, Ignisbot_PRO







What projects will you be doing?

[ROS Q&A] 168 - What are the differences between global and local costmap

Move the IgnisBot

Use the Jetbot API to move a two-wheeled robot

How to use the basics of the Jetbot NVIDIA API

Collision Avoidance with Deep Learning

Train Ignisbot to be able to navigate in a known environment, avoiding obstacles

Create a people follower that will allow your robot to detect people and follow them using a deep learning model

Create the People Follow ROS Script

ROSify the people tracker and create a people follower script

Ignisbot Mini Project

Combine everything you learned in this project

What you will learn

Course Syllabus

  • IgnisBot: Create your own NVIDIA Jetson Nano Robot
  • Hands-on Practice: PeopleFollower robot using DeepLearning

4 hrs.

Unit 1: Basics - Move Ignisbot

Understand how to use the Jetbot API to move a two-wheeled robot in simulation and the physical robot that uses ROS.

  • Setting Up your catkin_ws for Python3 environment
  • IgnisBot Move scripts
  • Move physical IgnisBot
  • How to connect and execute code in the Physical
  • How to start and execute the

5 hrs.

Unit 2 : Basics - Collision Avoidance with Deep Learning

In this unit, you will be able to understand:

  • How to gather the training images needed for the
    collision-avoidance training.
  • How to train a pre-trained model to avoid collisions in
    the current environment.
  • How to use the trained model.

7 hrs.

Unit 3: Create the people-follow ROS script

In this unit, you will learn how to:

  • Setup & ROSify the people-tracker so that any ROS
    system can access the detections.
  • Understand how the people-tracking data is published
    into ROS
  • Create a people-follower script that uses that data to
    create a behavior for Ignisbot.

4 hrs. 

Unit 4: IgnisBot Challenge

Combine everything you learned in this course to have Ignisbot (simulated and physical) navigate around an environment, searching for people, and reacting to them.

3 hrs. 

Ready to get started?

Start learning ROS & Robotics online quickly and easily

What’s next

People interested in this course also viewed

Learning Path

Machine Learning for Robots


Deep Learning with Domain Randomization

Top universities choose The Construct for Campus to teach ROS & Robotics.

Pin It on Pinterest