Use the idea from #28 to fix the unstable timestamp, see #25 for details.
Add a parameter time_alpha to compute final timestamp as
t1 = a dt_ros + (1-a) dt_dev + t0
time_alpha defaults to 0 which is the fix from #28.
User can check /imu/dt for the difference between 2 consecutive time stamps and /imu/dnow for the deviation from ros time.
Various cleanup.
Use binary group 1 only.
Remove ypr publisher
Normally I would just merge this, but since there are other people using this driver, it would be nice if someone could review this PR.
@IanTheEngineer @clalancette @Triocrossing
Thanks.
Use the idea from #28 to fix the unstable timestamp, see #25 for details. Add a parameter
time_alpha
to compute final timestamp as t1 = a dt_ros + (1-a) dt_dev + t0 time_alpha defaults to 0 which is the fix from #28. User can check/imu/dt
for the difference between 2 consecutive time stamps and/imu/dnow
for the deviation from ros time.Various cleanup.
Use binary group 1 only.
Remove ypr publisher
Normally I would just merge this, but since there are other people using this driver, it would be nice if someone could review this PR. @IanTheEngineer @clalancette @Triocrossing Thanks.