Attention: Answers.ros.org is deprecated as of August the 11th, 2023. Please visit robotics.stackexchange.com to ask a new question. This site will remain online in read-only mode during the transition and into the foreseeable future. Selected questions and answers have been migrated, and redirects have been put in place to direct users to the corresponding questions on Robotics Stack Exchange. Additional details are available here.

Can octomap be used on sparse pointclouds?

octomap_mapping

asked 2015-11-11 15:43:15 -0600

Lolita390
31 ●4 ●5 ●7

updated 2015-11-11 17:39:13 -0600

Hello,

I am receiving a sparse point cloud from a monocular slam algorithm. My question is if octomap_mapping is suitable for obstacle detection if my input pointcloud is sparse and I do not use a laser as sensor? If so, what changes would have to be made to octomap?

EDIT: Thanks for the reply. I configured the launch file of octomap and remapped it to the right point cloud topic, but nothing occurs. I checked the publisher topics of octomap, but I get the message that the topics have not been published yet. What could be the cause of this? Could you give me any hint about this please.Also I was not sure about the configuration of the following line in the launch file: param name="sensor_model/max_range" value="5.0"

edit retag flag offensive close merge delete

add a comment

1 Answer

Sort by » oldest newest most voted

answered 2015-11-11 15:59:13 -0600

Stefan Kohlbrecher

24361 ●173 ●299 ●444 https://www.energy-rob...

Octomap per default will use a pointcloud to update it's map representation. For the updating, it should not matter how sparse it is. Of course, depending on what application you want to use the octomap for, you could run into problems due to the sparse data, for instance if obstacle data is missing and your robot runs into things.

edit flag offensive delete link

Comments

I cannot launch octomap. The following warning comes up : "Nothing to publish, octree is empty". The frame I write in the launch file is the same as the fixed frame in rviz, which is "world". No transform is required, since the slam algorithm outputs the point cloud already in the world frame