Combining visual and icp odometry

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

Combining visual and icp odometry

Hendrik C
Hello everyone

I was looking through the paper "RTAB-Map as an Open-Source Lidar and Visual SLAM Library for Large-Scale and Long-Term Online Operation". It states that RTAB-map has two ways to generate odometry from exteroceptive sensors, namely visual from camera and icp from LIDAR.

I was wondering if it is possible to generate both the visual (in my case stereo) odometry and icp odometry, and to fuse them together either internally in RTAB-map (if it has such a function) or by using for instance the robot_localization package (as shown in the sensor_fusion.launch example).

Thanks in advance
Hendrik
Reply | Threaded
Open this post in threaded view
|

Re: Combining visual and icp odometry

matlabbe
Administrator

Hi,

stereo_odometry and icp_odometry can be started independently in parallel, their outputs could be fed to robot_localization.

cheers,
Mathieu
Reply | Threaded
Open this post in threaded view
|

Re: Combining visual and icp odometry

Hendrik C
Thanks Matthieu!
Reply | Threaded
Open this post in threaded view
|

Re: Combining visual and icp odometry

Masoumeh
In reply to this post by matlabbe
hi,

what are the output topics related to stereo_odometry and icp_odometry?

thanks
Reply | Threaded
Open this post in threaded view
|

Re: Combining visual and icp odometry

matlabbe
Administrator
They both provide odometry topics, one should make sure to remap their output topic name to avoid issues where 2 nodes are publishing on the same topic.