ROS Deep Learning with TensorFlow 101 Python

The first step with ROS, Deep Learning, Tensor Flow, and Image Recognition

Course Overview


DeepLearning… A topic that we hear a lot about and that promises a lot. Deep learning is the technology behind intelligent drone flight, self-driving cars, robots recognizing a huge number of objects, people tracking video feeds, etc. But how can it be used?

In this course, you will focus on learning the essentials for doing image recognition with Deep Learning. You won’t learn each and every nook and cranny of deep learning, but the few elements you learn you will be able to put into use and apply in real life, integrated with ROS.

Learning Objectives

  • Use the Google TensorFlow Image Recognition DB to recognize hundreds of different objects with ROS
  • Generate your own TensorFlow Inference graph to make it learn a custom object.
  • Use TensorBoard Web visualizer to monitor how the learning process is going.

Simulation robots used in this course

Mira Robot, GarbageCollector Robot.





9h 10m

This course is part of this learning path:

What projects will you be doing?

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

Integrate ROS + TensorFlow

Create your own ROS package that recognizes images with TensorFlow

ROS Mini Challenge #2 - RViz

Train your model

Train your own TensorFlow Image Recognition Model


Image Detection

Detect with your new retrained model in ROS

Garbage Collecting Robot Challenge

Image recognition/ learning with TensorFlow

What you will learn

Course Syllabus

Unit 1: Introduction
  • TensorFlow Image: Introduction to the Course
  • Practical demo

1 hr.

Unit 2: Use existing TensorFlow Model

TensorFlow Image Create your own ROS Package that recognizes images with TensorFlow.

1 hr. 30 min.

Unit 3: Inspect a TensorFlow Model

TensorFlow Image Launch TensorBoard and inspect a TensorFlow Model.

1 hr. 30 min.

Unit 4: Part1: Train your own TensorFlow

TensorFlow Image Train your own TensorFlow Image Recognition model Part1.

2 hrs. 

Unit 5: Part2: Use your trained TensorFlow

TensorFlow Image Train your own TensorFlow Image Recognition model Part2.

1 hr. 

Unit 6: Garbage Collecting Robot

TensorFlow Image Garbage Collecting Robot.

3 hrs. 

Unit 7: Final Recommendations

TensorFlow Image FinalRecommendations.

10 min. 

Ready to get started?

Start learning ROS & Robotics online quickly and easily

What’s next

People interested in this course also viewed


Using OpenAI with ROS


Deep Learning with Domain Randomization

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

Pin It on Pinterest