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

Revision history [back]

click to hide/show revision 1
initial version

Very good solution! Saved me a lot of research as I need to use hector slam with move_base. However, I think there is no need for the duration to be so high. In my case it works with 0.09-0.1 secs. Looking on how gmapping works with move_base I saw the following lines:

ros::Time tf_expiration = ros::Time::now() + ros::Duration(tf_delay_);
 tfB_->sendTransform( tf::StampedTransform (map_to_odom_, tf_expiration, map_frame_, odom_frame_));

with tf_delay to be:

 if(!private_nh_.getParam("tf_delay", tf_delay_))
    tf_delay_ = transform_publish_period;

In the case of hector_slam I think the tranform period is the published /scan period and it depends on the laser. In my case with the hokuyo urg-04lx is 0.1 sec (10hz).

Works fine so far with move_base. The only thing i have noticed is more cpu load by hector_slam.