0
$\begingroup$

Rosanswers logo

From the hector slam tutorial, it is possible to replace gmapping with hector_slam, with or without the presence of the odom frame. What is the difference?

For gmapping, the odometry is used when there are no feature matches. What about hector_slam? What happens in an open space with sparse features, or if we are using short range laser that cannot "see" anything within range?

On the other hand, if odometry is not well calibrated, gmapping will have problems. How does this affect hector_slam if I use the odom frame?

Originally posted by weiin on ROS Answers with karma: 2268 on 2012-06-12

Post score: 5

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

Rosanswers logo

The odom frame does not have any effect on mapping performance. hector_slam relies on features in laser scans only which can be a blessing if no odometry data is available, but also a curse if you have very wide open spaces and odometry available. We haven't implemented odometry support so far, as it is not needed for most indoor/search and rescue scenarios we look at (and gmapping covers the "with odometry" case nicely).

Originally posted by Stefan Kohlbrecher with karma: 24361 on 2012-06-12

This answer was ACCEPTED on the original site

Post score: 2

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.