I am trying to log some topics with sdlogger on a pixhawk with px4 v1.6.5 stable. I have several dropouts and strange sensor timestamps.
I have this /etc/logging/logger_topics.txt
vehicle_gps_position
actuator_outputs
sensor_accel
sensor_gyro
sensor_mag
sensor_baro
vehicle_local_position
vehicle_global_position
vehicle_attitude
With ulog2csv I get 10_53_20_sensor_accel_0.csv and 10_53_20_sensor_accel_1.csv. (Two IMUs are there on pixhawk, right?). If I look at the message timestamps (first column), both files start and end approximately at the same time, whereas the second has substantially less samples.
Here is a plot of the delta T between samples for the two files for a selected window of 10s. Why there is such apparent difference in the sampling rate for the two? One seem to operate at ~250Hz, whereas the other at 225. Is it correct, or there is something fundamental that I’ve misunderstood?
Second, many dropouts are visible. Is this normal? I have a SanDisk Extreme U3 32GB as suggested here. Is it possible to get no dropouts for the aforesaid set of topics? Are they too many? Somewhere I’ve read that I should increase the logging buffer, but I can’t find out how.
Thanks for the help, the full log it’s here: