Re: /zed/odom deviates from rtabmap pose but the 3D map is good

Posted by matlabbe on Sep 27, 2020; 7:39pm
URL: http://official-rtab-map-forum.206.s1.nabble.com/zed-odom-deviates-from-rtabmap-pose-but-the-3D-map-is-good-tp6950p6961.html

Do you feed an imu to rtabmap? It is like the odometry path is optimized using IMU on rtabmap side. For the big gap, it is mostly odometry drift, which is corrected after the loop closure is detected.