Robotics Stack Exchange is a question and answer site for professional robotic engineers, hobbyists, researchers and students. It only takes a minute to sign up.
Sign up to join this community
http://docs.ros.org/api/nav_msgs/html/msg/MapMetaData.html
The description says it is 'The time at which the map was loaded'- should this be the time the map was initialized? The time it was last updated (making it redundant with the header timestamp)?
PS: I am asking the same question at https://answers.ros.org/question/114726/what-is-map_load_time-in-nav_msgsmapmetadata/ that nobody has answered so far.
I have a lot of ros packages checked out into a big catkin workspace, looks like it is redundant with the header stamp:
$ git grep --no-index -n -C 2 --heading map_load_time
other/jsk_common/jsk_ros_patch/multi_map_server/src/main.cpp
155- ROS_INFO("Loading map from image \"%s\"", mapfname.c_str());
156- map_server::loadMapFromFile(&map_resp_,mapfname.c_str(),res,negate,occ_th,free_th, origin);
157: map_resp_.map.info.map_load_time = ros::Time::now();
158- map_resp_.map.header.frame_id = frame_id;
159- map_resp_.map.header.stamp = ros::Time::now();
--
ros/grid_map/grid_map_ros/src/GridMapRosConverter.cpp
275- occupancyGrid.header.frame_id = gridMap.getFrameId();
276- occupancyGrid.header.stamp.fromNSec(gridMap.getTimestamp());
277: occupancyGrid.info.map_load_time = occupancyGrid.header.stamp; // Same as header stamp as we do not load the map.
278- occupancyGrid.info.resolution = gridMap.getResolution();
279- occupancyGrid.info.width = gridMap.getSize()(0);
--
ros/navigation/map_server/src/main.cpp
154- // To make sure get a consistent time in simulation
155- ros::Time::waitForValid();
156: map_resp_.map.info.map_load_time = ros::Time::now();
157- map_resp_.map.header.frame_id = frame_id_;
158- map_resp_.map.header.stamp = ros::Time::now();
--
ros/rviz/src/test/map_test.py
17- grid.header.frame_id = "map"
18- grid.header.stamp = rospy.Time.now()
19: grid.info.map_load_time = rospy.Time.now()
20- grid.info.resolution = 1.0
21- grid.info.width = 3
GridMapRosConverter.cpp deserves credit over the others for for re-using the header stamp and putting in a comment instead of having sequential ros::Time::now() calls for time stamps in the same message which will result in very slightly different timestamps which will confuse the issue of whether there is a meaningful difference between the header stamp and map_load_time.
But still I don't fully understand 'Same as header stamp as we do not load the map', probably looking elsewhere in grid_map_ros would reveal what those authors believed map_load_time to mean.
