-3
$\begingroup$

Please give me guidance how should I proceed to know about SLAM algorithm? I am following some youtube videos but those are not so much helpful for me.....

$\endgroup$
  • 1
    $\begingroup$ Read books, articles, robotics sites, Coursera/EdX courses? $\endgroup$ – Eugene Sh. Nov 20 '14 at 14:44
  • 1
    $\begingroup$ SLAM is a very large topic that requires a the mastery of a lot of background material, including probability theory, bayesian inference, kalman filters, etc. It is really hard to give advice without knowing your background and what you want to accomplish from SLAM. $\endgroup$ – Paul Nov 21 '14 at 3:13
2
$\begingroup$

You should approach it as you would any subject in which you'd like to learn more.

Start by clarifying for yourself:

  • why do you want to know more about SLAM algorithms
  • how much time do you want to put into the subject
  • how do you learn best -- listening to a speaker, reading a text, writing some code and observing how it works, having a discussion with others

Based on your objectives then it is possible to come up with a learning plan to progress through the subject.

Understand the basics, read the wikipedia article, look at a presentation... SLAM stands for Simultaneous Localization and Mapping. Understand what localization and mapping are independently, then why you'd want to do that simultaneously and why it is difficult.

  • read "simultaneous localization and mapping" on wikipedia
  • google "introduction to robotics slam" and follow some links, such as the pdf from ai robotics school

Build up the pre-requisite knowledge if you don't have it. If you were confused watching some youtube videos such as this Intro Course maybe you need to back up and watch the first 12 lectures first. https://www.youtube.com/watch?v=xoULHsN3bjQ

If you feel like you already have all the basics, then I guess trying to apply it in practice would be a good way to learn more about the different algorithms. You can probably shortcut a bit by using existing robot platforms like LEGO Mindstorms or a TurtleBot running ROS or whatever you have available to you.

  • google "slam lego"
  • google "slam ros" or "intro to navigation ros" and read more about gmapping

When you get stuck, back up and dig into part of the subject in more depth and repeat.

Good luck.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.