Maximilian Wulf

Hi @blablebli-robots, > the returns corresponding to the laser beams going through the hole are outside the local map (because of height they go too far), so there is no...

Hi @blablebli-robots, have fun. Please also make sure to check out the traversability estimation package from RSL. There might be useful information for you. https://github.com/leggedrobotics/traversability_estimation

Hi @blablebli-robots, the only difference considering other depth sensors with respect to elevation mapping is the noise parameters of the camera. In general, you should be good to go.

HI @blablebli-robots, could you share more insights into the observed noise of the ZED stereo cameras?

@blablebli-robots thank you for your reply. I will close this issue for now. Feel free to re-open.

Hi @rokusottervanger, at the moment we don't support nav_msgs/Odometry. There is no reason to not use it. If anyone wants to write a feature to support odometry messages feel free...

It seems you are that your sensor is so far that you run into the egg carton effect. I think it should be solved by moving the camera closer. Also...

Hey @Bibbidi-Babbidi-Boo I can not confirm the issue on my setup. Are you using ubuntu 18.04 and ROS melodic?

Hey @Bibbidi-Babbidi-Boo, no not on my machine. Can you build it in debug and show the backtrace, that is the only possibility I see to debug it.

Attach a gdb instance to your launch file: http://wiki.ros.org/roslaunch/Tutorials/Roslaunch%20Nodes%20in%20Valgrind%20or%20GDB And then as soon as you run into the segmentation fault enter `bt`. https://sourceware.org/gdb/onlinedocs/gdb/Backtrace.html