ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

localization in crowd environment

asked 2011-07-25 21:08:03 -0600

Hi everybody, I'm trying to figure out how to achieve robust navigation in crowded environment. I'm using a 2-d laser scanner and amcl algorithm. Is somebody facing the same problem? Are there any other stacks/tools in ROS that address this situation?

My idea is to filter people and other moving objects from the scan, setting the occluded rays to max range and then to pass this filtered scan to the amcl. Has anybody already tried this solution? Can anybody give me a hint about it?

Thanks in advance!

edit retag flag offensive close merge delete

2 Answers

Sort by ยป oldest newest most voted

answered 2011-09-02 12:12:04 -0600

Mac gravatar image

There's substantial scientific literature on this; the short answer is that it's tricky, but good techniques exist. There's no up-to-date ROS package for doing this.

edit flag offensive delete link more


could you give me some references?
luca gravatar image luca  ( 2011-09-05 00:54:14 -0600 )edit

answered 2012-09-13 23:24:18 -0600

Ibraim gravatar image

Although this is an old post maybe someone new comes looking for the same information. I have found references of doing this with the name of SLAM and DATMO. This means SLAM with detection and tracking of moving objects.

In this method the people are treated in no different way as any other moving object in the scene and probably this is a better approach as there are significant different objects in indoor and outdoor environments.

You can take a look at the following references:

Hope this help anyone who arrives here.

edit flag offensive delete link more

Question Tools

1 follower


Asked: 2011-07-25 21:08:03 -0600

Seen: 1,192 times

Last updated: Sep 13 '12