| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
Hello
Im using Hector SLAM and it works fine. My sensor package consist of IMU, laser and Kinect camera. Would like to know how to use IMU for better pose estimation. I know that pose estimation is based in laser scan matching but Im not using the sources if IMU. So i think fusing this IMU data would gives me better orientation. ANY help on this???Thanks
Hi,
You should take a look into data fusion if you want to get better estimation of your pose. By using data fusion algorithm, such as Kalman Filters, you can obtain better measurement by combining the precision of each sensors. Theoretically, the more sensors you have, the more precision you can get. This family of algorithm is really robust and can typically run asynchronously.
In ROS, I have found 2 EKF (Extend Kalman Filter) implementation : robot_pose_ekf and hector_localization, but I have not been able to make either work. There is also this open source code http://kalman.sourceforge.net/
Hope it helps!
wish stefan Kohlbrecher answer this question :(
Wish I had the time for an extensive answer :) Will forward this, maybe someone else can chime in.
Asked: 2012-12-21 12:42:21 -0500
Seen: 2,325 times
Last updated: Jun 07 '13
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
