Hi, when I use the slam_gmapping_replay node, I get messages like this (a few of these per second):
[ WARN] [1472656696.029278429]: Dropping old scan: Lookup would require extrapolation into the future. Requested time 1467913048.867926076 but the latest data is at time 1467913048.867113403, when looking up transform from frame [pioneer/odom] to frame [pioneer/laser]
This doesn't happen when playing back with rosbag and running slam_gmapping. Does that suggest that slam_gmapping_replay isn't perfectly emulating running gmapping in real time? Or running gmapping in realtime drops scans in the same way, just doesn't emit that message? In that case, perhaps a similar message should be added to the realtime mode.
The code which does that was introduced in 06b5d606b and 052ad1011 by @vrabaud, do you perhaps know what's going on?
Hi, when I use the slam_gmapping_replay node, I get messages like this (a few of these per second):
[ WARN] [1472656696.029278429]: Dropping old scan: Lookup would require extrapolation into the future. Requested time 1467913048.867926076 but the latest data is at time 1467913048.867113403, when looking up transform from frame [pioneer/odom] to frame [pioneer/laser]
This doesn't happen when playing back with rosbag and running slam_gmapping. Does that suggest that slam_gmapping_replay isn't perfectly emulating running gmapping in real time? Or running gmapping in realtime drops scans in the same way, just doesn't emit that message? In that case, perhaps a similar message should be added to the realtime mode.
The code which does that was introduced in 06b5d606b and 052ad1011 by @vrabaud, do you perhaps know what's going on?