Support requested for graphing FFT from ISB messages in DataFlash files

[peterbarker]

Issue details

We have a new feature coming in ArduPilot which allows fft-able data to be contained in many dataflash files, rather than just those with raw IMU sampling data enabled.

There is a new format field type, a. It is an array of 32 16-bit signed integers.

There are two new messages, ISBH and ISBD. One contains header information for a batch sample, the other the data for the batch sample. Many ISBDs are expected for a single ISBH message, and one is related to the other using the N field.

The ISBH message contains type and instance fields, with type specifying an IMU sensor type (gyro or accel) and instance the instance number of that sensor on the flight controller.

Version

These patches are as yet unmerged into master. https://github.com/peterbarker/ardupilot/tree/fft is the WIP for support. We do not expect the format of the messages to change before that's merged in.

Platform

[ * ] All
[ ] AntennaTracker
[ ] Copter
[ ] Plane
[ ] Rover
[ ] Sub

Logs

df-isb-fft-sample.bin.zip

pbarker@bluebottle:~$ mavlogdump.py df-isb-fft-sample.bin --t FMT | grep ISB
Skipped 443 bad bytes in log at offset 3827, type=(163, 149, 248)
2017-09-22 09:33:08.14: FMT {Type : 248, Length : 29, Name : ISBH, Format : QHBBHQf, Columns : TimeUS,N,type,instance,mul,SampleUS,smp_rate}
2017-09-22 09:33:08.14: FMT {Type : 249, Length : 207, Name : ISBD, Format : QHHaaa, Columns : TimeUS,N,seqno,x,y,z}

pbarker@bluebottle:~$ mavlogdump.py df-isb-fft-sample.bin --t ISBH,ISBD | head
Skipped 443 bad bytes in log at offset 3827, type=(163, 149, 248)
2017-09-22 09:33:08.14: ISBH {TimeUS : 98933484, N : 27, type : 1, instance : 0, mul : 938, SampleUS : 0, smp_rate : 1001.12475586}
2017-09-22 09:33:08.14: ISBD {TimeUS : 98933499, N : 27, seqno : 0, x : array('h', [3, 2, 1, 0, 0, 1, 0, 2, 2, 3, 3, 2, 1, 1, 1, 2, 2, 1, 0, 0, 1, 1, 2, 1, 1, 1, 0, 0, 0, 0, 1, 1]), y : array('h', [-1, 0, 0, -1, 0, 0, 1, 1, 2, 1, 0, 0, 0, -1, -1, 0, 1, 0, 0, 0, -1, -1, -1, -1, 0, 0, 0, 0, 0, 0, 2, 1]), z : array('h', [0, 1, 2, 1, 0, 0, -1, -1, -1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])}
2017-09-22 09:33:08.14: ISBD {TimeUS : 98933542, N : 27, seqno : 1, x : array('h', [0, 0, 0, 1, 2, 4, 5, 3, 0, 0, 0, 2, 2, 1, 2, 2, 1, 0, 0, 1, 1, 0, 0, 2, 3, 3, 2, 2, 2, 1, 1, 1]), y : array('h', [0, 0, 0, 0, 0, -2, -2, 0, 0, 0, 0, 1, 2, 2, 1, 1, 0, 0, 0, 0, 0, 2, 2, 2, 0, 0, 0, 0, 0, 1, 0, 0]), z : array('h', [0, 0, 0, -1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])}
2017-09-22 09:33:08.25: ISBD {TimeUS : 99044768, N : 27, seqno : 2, x : array('h', [1, 0, 1, 2, 2, 2, 2, 1, 0, 2, 2, 3, 2, 2, 3, 2, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 3, 4, 2, 0, 1, 3]), y : array('h', [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, -1, -2, -1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0]), z : array('h', [0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, -1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0])}
2017-09-22 09:33:08.25: ISBD {TimeUS : 99044832, N : 27, seqno : 3, x : array('h', [3, 0, 0, 0, 0, 1, 2, 1, 0, 0, 0, 1, 2, 0, 0, 0, 2, 2, 1, 2, 2, 2, -1, -2, 2, 5, 8, 9, 3, -2, -3, -1]), y : array('h', [0, 0, 0, 0, 0, 0, -1, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 1, 0, -2, 0, 2, 1, -3, -6]), z : array('h', [0, 0, 0, 0, -1, -1, -2, -2, -2, -2, -1, 0, 1, 1, 0, 0, 2, 1, 0, 0, 2, 1, -4, -4, 0, -6, -1, 8, 2, -2, 0, 0])}

mavfft_isb.py (https://github.com/ArduPilot/pymavlink/blob/master/tools/mavfft_isb.py) interprets these messages to produce pretty pictures:

Note that this tool takes the average of all of the per-sensor/instance FFTs to produce the graphs, however animating each sensor/instance, or producing overlays on e.g. a motor-output graph might be other good uses for this data.

[meee1]

added support here 7151df4

[tridge]

thanks Michael!