Easy SLAM with ROS using slam_toolbox

UPDATE: If you're on humble or newer, please note that "params_file" has changed to "slam_params_file".

SLAM is an important process, underpinning the autonomous behaviour of many mobile robots. The amazing slam_toolbox package for ROS makes getting started with SLAM incredibly simple!

📃 Companion blog post coming soon

00:00 - Intro
00:48 - SLAM Overview
03:22 - ROS and SLAM
07:28 - Setting up for slam_toolbox
10:50 - SLAM with slam_toolbox
15:13 - Localisation with slam_toolbox
17:20 - Localisation with amcl
22:43 - slam_toolbox on our real robot
24:48 - Outro

[jombrs.]

For those having problems loading map. The "params_file" parameter has been changed to "slam_params_file". Hope this help :)
Edit: I'm using ros2 humble

[ZoGi]

Did not watch this video in particular but all your videos are structured always fine. Highly appreciated that you provide this content for us. There is no other content creator that explains the usage of ROS2 better than your channel. Keep up the good work! we need this.

[alfonsopanadero]

Hello Josh, At the beginning of July I found your videos, a little late right jejeje.
The truth is that you explain it very well and detailed. I've been learning something about ROS1 for a long time. Today I know more with your perfect videos and explanations. Today I have everything working until this video. For which I am very grateful for your effort. Thank you.
Note: If someone wants to use more powerful motors, they can use the BTS7960 motor controller that can be directly replaced by the L298.

[carldraper]

Another great video, had to watch it a few times to digest and implement but wonderful content.

I also would love to see how you might go about IMU implementation for fused ODOM data. I have totally rewritten by ROS stack based on the Articubot code you provided us. Super well structured and things run so much better so far!

[usimian]

Great video! I've built pretty much the same thing but with a Jetson Nano. I used the same lidar and differential wheel odometry plus an IMU thru the robot_localization package.
What I'd like to learn is how to integrate a pointcloud device (I have a sparkfun VL53L5CX) or use two cameras to do visual slam. I agree, slam_toolbox is quite nice. Thanks!

[benoitlagassedelocht]

Finally an updated tutorial series about ROS 2 that is simple, yet very detailed! This will be very helpfull for my next project. keep up the amazing work🦾

[MartinBgelund]

Thanks for a great video! I'd like to see more on SLAM, especially using IMU's and maybe even distance sensors (ultrasound and time-of-flight).

[alexanderyang]

Thank you for sharing this video. I just begin learning SLAM and your video is friendly to new user. Thank you! :)

[rukey]

Can’t wait for next video..!! Love to see a video of sensor fusion.

[messedmushroom]

It would be great to see a more deep dive into SLAM, other sensors like IMUs, depth cameras, etc!

[MrPrinzovdarkness]

I would like more videos on SLAM with other sensors as well as how to fuse these sensors and algorithms together

[ricardocasimiro]

great job! thanks for the video.
it could be interesting to compare with cartographer and also add imu and gps
also it could be very interesting to show the base math behind these algorithms

[yourileenman]

Awesome video, looking forward to autonomous navigation

[ctrlaltdel]

The missile knows where it is at all times. It knows this because it knows where it isn't.

[noahHAyoumentionedme]

Hey I love your videos, you are so clear and precise. It's great!
Would you be willing to do a video on vSLAM? I'm particularly interested in orb-slam monocular and similar single camera SLAM method.
Thank you so much again!

[solutionhacker]

Thanks alot! Very insightful video! Looking forward to the next one too.

[ថាន្នីសុគុណ]

I can't understand from Localization with slam toolbox and Localization with amcl . Could you make a video to explain more about these two points? Or show me how to load the saved map into the rviz.

[volrin]

Hello. I have a problem when adding the Map in Rviz. I get an error with the message "No map received" and at the transform it says "Could not transform from [] to [odom]"
I can't figure out what I did wrong.

[xdboys]

in ros humble, the name of the argument that points to slam params file, should be 'slam_params_file'

[sharma]

I want to learn if I can just change the map without any parameter or node difference?