Ask Your Question

Transform GPS pose to map frame in robot_localization

asked 2021-03-18 01:53:20 -0500

James Wang gravatar image

I tried to using navsat_transform_node in robot_localization to transform gps/fix to the map frame and further fuse with pose estimate provided by SLAM or point cloud matching. I input three topics GPS measurement (gps/fix), SlAM measurement (odometry/filtered) and IMU measurement (imu/data) to navsat_transform_node. However, the output of navsat_transform_node, odometry/gps, did not align well with the SLAM measurement (odometry/filtered). It looks like that odometry/gps was not well transformed to the map frame. Did I miss something in using this node? How to do this transformation correctly?

edit retag flag offensive close merge delete


Have you checked your tf tree? What is the frame for your slam measurements? Who is publishing map->odom?

xaru8145 gravatar image xaru8145  ( 2021-03-23 06:50:12 -0500 )edit

1 Answer

Sort by ยป oldest newest most voted

answered 2021-04-27 04:13:40 -0500

Tom Moore gravatar image

I'm going to need a lot more information.

  • What SLAM package are you using?
  • Please provide your config for navsat_transform_node
  • Please provide a sample message for every input

However, I can tell you now that if you are actually performing live SLAM and trying to integrate GPS data, I don't think it's going to work well. Depending on your SLAM package, I would assume that once you have a loop closure, your map origin could change. Whatever transform navsat_transform_node computes at the start would therefore be invalid.

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools



Asked: 2021-03-18 01:53:20 -0500

Seen: 94 times

Last updated: Apr 27