Converting raw Lidar data to something more useful

I have a small robot (turtlebot3) with a Lidar. For aa demonstration I would like to process the data coming from the lira in the following way:

  1. Pay attention only to a range of angles (e.g. -15 degrees to +15 degrees)
  2. Filter it into
    a) distance to obstacle
    b) rate of change of a)
    c) rate of change of b)

I am assuming that this is elementary and there are nodes/libraries in ROS that do just this. I am not asking for code but pointers to where to look and what to read. Thanks for your help!

Might get more responses if you post this on ROS answers

Yeah the mysteries of what folks accept here vs. ros answers. It’s a constant thing.

Please read the Support guidelines for what to post here versus answers.ros.org.

At the most simple a question should go on answers.ros.org. So this would be more appropriate on answers.ros.org as it’s a question.

ROS has a wrapper (http://wiki.ros.org/pcl) for Point Cloud Library (http://pointclouds.org) which has a lot of functionality to process lidar data. To cut out certain region from PointCloud you may want to take a look at PassThrough filter (http://pointclouds.org/documentation/tutorials/passthrough.php).

Interesting, I will dig into that. Thanks!

I tried it on answers and didn’t get any answers. Here I did get a couple of useful ones. So if “questions” are supposed to go onto answers.ros.org then discourse is only for announcements and declarations and actually not for any discourse.

I’m only being somewhat ironic. I thought of answers for geeky detailed questions about ROS (like stack overflow) and discourse for more conceptual, theoretical, and general questions. This is I guess a misunderstanding.

For example I had such a general question: “If I want to process the liar data so that I can find out a good current estimate of the closet obstacle plus its rate of change, should I be learning about Kaman filters or is that the wrong kind of processing?” I would have posted that question here. But I guess I will try somewhere else, right?

Actually I think that particular non-ROS-specific question would probably get answered quickest on the robotics stack exchange.

1 Like