I want to construct 2D occupancy grids for mapping of an unknown environment. Could you guide me how can I achieve it?

What I have researched is that robot can construct its map given sensor data using recursive Bayesian estimation algorithm. Is it correct?


1 Answer 1


Yes you do need some sort of sensor data such as LIDAR data.

If you are planning to perform 2D SLAM then you could start playing around with the ROS package gmapping which will help you construct a 2D occupancy grid using LIDAR data.

Here is a simple tutorial that shows you how to use gmapping for this use case. Here is another tutorial that shows you how to do this in simulation.

This youtube video is also a good guide.

  • $\begingroup$ Could you provide me any useful info or link of how to implement gmapping using vrep software? $\endgroup$
    – Saad
    Commented Jun 28, 2019 at 7:02
  • $\begingroup$ Hi Saad, I have never used V-REP but from the looks of it it has some level of ROS support as well. Have a look at this forum post . Someone seems to have figured out how to perform gmapping on V-REP! Hope this helps! $\endgroup$
    – JJerome
    Commented Jul 3, 2019 at 0:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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