Open Martygp95 opened 4 years ago
The algorithm is explicitly designed to properly propagate pose uncertainties. Since the turtlebot3 example has perfect pose information because it is a simulation, the pose has therefore zero covariance.
Chapter 3 of the corresponding PhD thesis excellently explains the details of the algorithm.
Hi I have a simple question; why in the turtlebot3 branch, in the tf_to_pose_publisher.py, we obtain a pose with zero covariance as an input of the algorithm? The algorithm should be able to handle with pose uncertainity of the robot, isn't it?