ROS for Beginners II: Localization, Navigation and SLAM

Name: ROS for Beginners II: Localization, Navigation and SLAM
Rating: 4.4 (1607 reviews)

A practical approach to learn the foundation of mobile robots SLAM and Navigation with ROS

Created byAnis Koubaa

Last updated 4/2023

English

English [Auto],Korean [Auto],

What you'll learn

Theoretical foundations of 2D and 3D localization
Transformation between frames in 2D and 3D Spaces
The powerful feature of the tf package to represent frames and perform transformation and localization
Theoretical foundation of localization and mapping (SLAM)
Background on navigation concepts (global path planning, local path planning, collision avoidance)
Difference between Map-Based Navigation and Reactive Navigation
The navigation stack of ROS (move_base, amcl, gmapping)

Course content

9 sections • 77 lectures • 4h 20m total length

What is this course about?3:08
About the instructor1:17
This lecture provides a short biography on the instructor.
You can read about him in these links
http://www.riotu-lab.org/akoubaa/
https://scholar.google.com/citations?user=aEoY-0IAAAAJ&hl=en
[IMPORTANT] What is you face errors and need to debug and find solutions?0:59
[NEW] Important: Install ROS Navigation and SLAM0:17
[NEW] Important: Installation Instructions of TB30:56
[ROS Noetic] Install Turtlebot3 Simulator6:19
Turtlebot3 Simulation Enviornments5:19
Simple Turtlebot3 Demo3:35
Get the code from GitHub Repository1:40
Note about path setting for map-based navigation0:52
Link to the pre-installed virtual machine0:05

Navigation Demo Intro1:19
Explore how robotics navigation blends linear algebra, geometry, trigonometry, physics, and mechanics with practical ROS implementations through an interactive demo that links theory to real-world localization, navigation, and SLAM concepts.
Important note about the demo0:36
This folder contains the two map files needed to redo the experiments.
Put them into your home directory in ROS and use the correct location to the yaml file when you load the navigation stack
Start the Turtlebot3 Simulator1:02
If you have a problem starting Turtlebot3 simulator0:36
Setting the initial location of the robot1:24
Set the robot’s initial location and orientation to align the ARV navigation view with the real map, using the two depots estimate to place the robot and align laser scans.
Frames2:58
Explore localizing the robot using a grid map with an origin, and visualize frames and a fixed frame to relate location and orientation.
Location of the robot in different frames2:27
Explore how the robot’s location varies with reference frames, comparing the odom frame to the map frame and highlighting how frame shifts affect global map localization readings.
How orientation is represented in 3D space?1:09
Navigation Example1:43
General Question on the Navigation Demo

Section Overview and Learning Outcomes0:41
This lecture presents an overview of the section on frames.
Pose of a robot in a frame3:25
This video introduces the concept of frames and the pose of an object in a frame.
Pose of a robot
Dealing with Multiple Frames3:10
Explore how multiple coordinate frames affect localization, and learn to transform positions between world and robot frames to determine an object's location.
Transformations Overview3:01
2D Translation2:15
2D Rotation3:08
The General Transformation Matrix: Translation + Rotation2:53
In the previous lecture, we have presented the transformation matrix for a pure translation and the transformation of a pure rotation. However, in real world, transformation can be composed of translation plus rotation at the same time. In this lecture, we will present how to convert the coordinates between two frames with a general transformation.
Illustrative Example on 2D Transformation3:25
In the previous lecture, you have learned how to transform location coordinates between two frames that are shifted by a translation and a rotation. But, you may ask, why this is useful and in which context we need it?
In this lecture, I will present a simple example to show you why transformations are very important indeed.
Let us get started.
2D Transformations

The tf package overview0:54
What is TF in ROS?2:05
URDF: Language for the Description of Frames and Transformation in a Robot Model1:54
Why TF is important?2:25
Overview of tf package utilities4:15
Convert Orientation between Quaternion and Roll-Pitch-Yaw using TF5:56
Reading the Yaw of a Robot from its Orientation in Quaternion7:09
The tf package command line and utilities3:10
Use the tf package command line utilities to view and print frames and their transformations in a Gazebo ros setup, including translations and quaternion rotations for frames like base_footprint.
[PRE-ROS-NOETIC] view_frames command line2:24
[NEW in ROS Noetic] Note about view_frames1:06
more about view_frames2:01
tf_echo command line2:33
tf_monitor command line0:45
Use the tf_monitor command line to monitor transforms in ROS, view delays and broadcaster statistics for specific frames like odom to base_footprint, or display all transformations.
static transform publisher3:37
Broadcast a transformation in a ROS Node3:29
Learn to broadcast a transformation in a ROS node using Python or C++, creating a transform broadcaster and publishing translated and quaternion-rotated frames at a set rate.
Listen to a transformation in a ROS Node2:41

Mobile Robot Navigation Overview1:31
Map-based Navigation Overview5:16
SLAM Demonstration and Discussions7:56
Demonstrates building a 2D occupancy grid map via slam on a Turtlebot3 in gazebo using laser and odometry data, marking free, occupied, unknown cells, and saving the map for navigation.
Understand the structure of Maps in ROS3:33
Learn how to save a ROS map to file with the map server and saver, generating an image and a YAML with image path, resolution, origin, and thresholds.
Understand ROS Nodes and Launch File used for SLAM6:44
Map-based Navigation Demo6:19
The Recovery Behavior2:15
Understand the navigation launch file5:10
Writing a ROS Node for Robot Navigation (Python)12:00
Understand the Recovery Behaviors of the Navigation Stack4:51
Robot Setup to Support the ROS Navigation Stack7:34
Configure the robot to support the ROS navigation stack by implementing per-robot blue modules, publishing tf transforms, odometry, and sensor data for global and local planning.

Requirements

Prior knowledge of the basic concepts of the Robot Operating System (ROS) (required)
Knowledge in C++ and/or Python Programming language
Background on the concepts of Linear Algebra, Trigonometry and Geometry
Want to learn ROS
Eager to learn robotics navigation

Description

UPDATE

OCT 9, 2020: I added the installation instruction of Turtlebot3 on ROS Noetic

Overview

Localization, mapping, and navigation are fundamental topics in the Robot Operating System (ROS) and mobile robots. However, it is very complex to learn. Usually, beginners find it difficult to even know where to start. The typical tutorials in ROS give high-level information about how to run ROS nodes to performs mapping and navigation, but they do not give any details about the technical aspects. Some other courses focus more on the technical aspects, which is mathematically complex but does not give a clear link to how these concepts are tied with the ROS navigation stack.

This course addresses this gap and follows a practical approach to introduce new learners to mobile robot navigation foundations and how it is implemented in ROS. The course is designed to introduce you to the world of mobile robot navigation in a quick and effective manner.

In this course, I presented detailed coverage of the most important package in ROS for navigation: the tf package! Without understanding this package, it will be difficult to deeply understand how navigation works in ROS. Although there are tf tutorials, the tf package heavily relies on important theoretical concepts not presented in ROS tutorials. This course provides a systematic introduction to the necessary theoretical background and complement with demonstration and programming activities of the tf package utilities and API.

This course assumes that you have some background on the main concepts of Robot Operating System (ROS), such as ROS nodes, ROS topics, ROS services, and an understanding of the basic notion of motion with ROS. If you do not have these skills, I would recommend first enroll in my course ROS for Beginners: Basics, Motion, and OpenCV to get the necessary background.

My experience with ROS

I have been programming with ROS for many years both in academic and industrial projects. I am very passionate to develop programs with ROS. I have also been teaching ROS at the University and providing training programs. I am the R&D Director of Gaitech Robotics, and I have developed many ROS packages for robots and drones. I have been leading international scientific activities around ROS, and in particular, I am the editor of five volumes of books with Springer entitled Robot Operating System, The Complete Reference. I gained a lot of experience in what difficulties new users encounter to learn ROS and this contributed to pin right to the point addressing these problems through the different lectures of the course.

Who this course is for:

Beginner ROS developers and users
Students at Universities learning ROS
Anyone interest to know about the navigation concepts of ROS and mobile robots
Curious about robotics
Whoever wants to learn ROS navigation without wasting time

ROS for Beginners II: Localization, Navigation and SLAM

What you'll learn

Explore related topics

Course content

Course Overview11 lectures • 24min

ROS Navigation Introductive Demo9 lectures • 13min

2D Frames, Transformations and Localization8 lectures • 22min

3D Frames, Transformations and Localization4 lectures • 13min

Orientation in 3D Space4 lectures • 11min

The TF Package: Frames, Transformations and Localization in ROS Explained16 lectures • 46min

Map-based navigation11 lectures • 1hr 3min

Configuration and Tuning of the Navigation Stack Parameters8 lectures • 30min

Reactive Navigation6 lectures • 38min

Requirements

Description

Who this course is for: