ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Fusing odom, imu, magnetometer with robot localization

asked 2020-04-07 08:18:28 -0600

elgarbe gravatar image

Hi, I'm looking for information about fusing odometry with imu and magnetometer. Just in 2D. My sensors are connected to an embbeded systems and I'm publishing raw data to ROS using rosserial. Right now I'm publishing odometry, IMU (raw Gyro and accel data) and raw (after calibration) magnetometer data. I'm not providing orientation data, just raw data. What is better, fusing all raw data in robot_localization? make some complimentary filter between gyro and magnetometer and publish orientation in IMU msg? has someone a config file for robot_localization with this sensors? I'm not sure what magneticfield data could be added to RL config file. I think I need to, at least, calculate heading with magnetometer an publish it...


edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2020-05-05 02:56:07 -0600

Tom Moore gravatar image

If it were me, I'd feed the IMU and magnetometer data to imu_filter_madwick. Then I'd fuse the output of that node with the wheel odometry data in an r_l EKF instance. That's a pretty straightforward setup, and the EKF config example would be a good place to start.

edit flag offensive delete link more


Thank you very much Tom for you time. I will do that.

elgarbe gravatar image elgarbe  ( 2020-05-05 11:37:36 -0600 )edit

Hi Tom, Could you kindly let me know what an r_l EKF instance is? Thank you

p_user2001 gravatar image p_user2001  ( 2021-08-31 20:43:42 -0600 )edit

r_l = robot_localization EKF instance = Extended Kalman Filter node instance. As in, an instance of the ekf_localization_node.

Tom Moore gravatar image Tom Moore  ( 2021-09-06 07:01:57 -0600 )edit

Question Tools



Asked: 2020-04-07 08:18:28 -0600

Seen: 1,333 times

Last updated: May 05 '20