Navigation with Rtabmap and Stereo Camera
Posted by imran05 on
URL: http://official-rtab-map-forum.206.s1.nabble.com/Navigation-with-Rtabmap-and-Stereo-Camera-tp10644.html
Hello,
I am using the OAK-D-PRO-POE camera with a stereo sensor and successfully generated a 2D/3D map following the Stereo Handheld Mapping tutorial on the RTAB-Map ROS wiki. My goal is to navigate my robot autonomously using this map.
To achieve this, I am working with the move_base package and following the steps in the Stereo Outdoor Navigation tutorial. However, I am uncertain whether I need to convert the point cloud data to laser scan format for the navigation stack. Several sources suggest this conversion is necessary for proper operation with move_base. Currently, when I run my move_base launch file, I encounter the following error:
pocess[planner/move_base-1]: started with pid [51922]
[ WARN] [1730784550.229903768]: global_costmap: Pre-Hydro parameter "static_map" unused since "plugins" is provided
[ INFO] [1730784550.232415848]: global_costmap: Using plugin "static_layer"
[ INFO] [1730784550.245615792]: Requesting the map...
[ INFO] [1730784550.854107903]: Resizing costmap to 421 X 421 at 0.050000 m/pix
[ INFO] [1730784550.953536180]: Received a 421 X 421 map at 0.050000 m/pix
[ INFO] [1730784550.961638562]: global_costmap: Using plugin "obstacle_layer"
[ INFO] [1730784550.970935649]: Subscribed to Topics:
[ INFO] [1730784550.982091918]: global_costmap: Using plugin "inflation_layer"
[ WARN] [1730784551.058547847]: local_costmap: Pre-Hydro parameter "static_map" unused since "plugins" is provided
[ INFO] [1730784551.060172258]: local_costmap: Using plugin "obstacle_layer"
[ INFO] [1730784551.070438626]: Subscribed to Topics: point_cloud_sensor
[ INFO] [1730784551.099451642]: local_costmap: Using plugin "inflation_layer"
[ INFO] [1730784551.167750201]: Created local_planner base_local_planner/TrajectoryPlannerROS
[ INFO] [1730784551.179935695]: Sim period is set to 0.05
[ WARN] [1730784553.146037583]: The openni_points observation buffer has not been updated for 2.06 seconds, and it should be updated every 0.50 seconds.
[ INFO] [1730784553.161125806]: Recovery behavior will clear layer 'obstacles'
[ INFO] [1730784553.168741668]: Recovery behavior will clear layer 'obstacles'
[ WARN] [1730784553.646013966]: The openni_points observation buffer has not been updated for 2.56 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784554.146029932]: The openni_points observation buffer has not been updated for 3.06 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784554.646104295]: The openni_points observation buffer has not been updated for 3.56 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784555.150279887]: The openni_points observation buffer has not been updated for 4.06 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784555.646011850]: The openni_points observation buffer has not been updated for 4.56 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784556.146003108]: The openni_points observation buffer has not been updated for 5.06 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784556.646024576]: The openni_points observation buffer has not been updated for 5.56 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784557.146082286]: The openni_points observation buffer has not been updated for 6.06 seconds, and it should be updated every 0.50 seconds.
[ WARN] [1730784557.646060362]: The openni_points observation buffer has not been updated for 6.56 seconds, and it should be updated every 0.50 seconds.
Could you advise on whether point cloud data must be converted to laser scan format, and if so, recommend an approach to make this integration work with move_base?
Thank you!