| Commit message (Collapse) | Author | Age |
|
|
|
|
|
| |
This should avoid mis-programming devices with incorrect firmware.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
This way, updating satellite information doesn't drop all of the
regular GPS data on the floor.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
The conversion was losing the replicated top four bits by shifting by
the wrong amount.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
| |
These look much like TeleMega v2.0 log files, except that the mag
sensor data now comes from the mpu9250 instead of an external
hmc5883. The gyro and 3-axis accel data from the mpu9250 are the same
as the mpu6000.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
| |
* Catch and report CRC errors
* Deal with corrupted flight records
* Add ability to immediately graph new data
* Check before overwriting existing files
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
This avoids a crash dealing with corrupted flight data
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
| |
This makes the API more consistent, and means that the listener is
responsible for mangaing the temp gps state. In particular, the
AltosDataListener set_gps API now calls the cal_data function.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
|
|
| |
When using the remote link, there are two separate configuration data
blocks, that for the local device and for remote. Make the link report
both versions, depending on whether it is in remote mode or not.
Request config data in remote mode when running idle monitoring so
that the presented data is for the remote device, not the local one.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
The ground accel is the basis of the accel adjustment, so it needs to
be delivered in un-adjusted form.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Small floating point rounding errors could lead to NaNs.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Just check for null before writing as a precaution.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
|
| |
When generating a KML file, we want to position markers near the start
of the flight section. This is done by looking for a GPS coordinate
'before' the starting point of the flight, which doesn't work well
when the first GPS coordinate is later than that. Pick the first point
after the chosen time if there isn't an earlier one.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
| |
Provide two paths, one using GPS data the other baro. Replace separate
path segments for each state with markers so that the path is a single
unit, able to be displayed in the elevation profile widget.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
| |
Also switch smoothing window to Kaiser and change default accel filter
width to 1 second instead of 4 seconds.
Now users can play with the filter and see what it does.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
Not terribly useful, but did help validate firmware handling of tick
wrapping, so we'll keep it.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
Check the CRC status in the packet before creating a new telemetry
object.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
| |
When the device being analyzed has no flight state, we want to use the
'stateless' state so that the UI can display reasonable
information. This bit was lost in the recent AltosState shuffle and
this patch brings it back.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
It's now version 6. Also removed duplicate time values and made radio
values conditional on having radio data.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
This lets AltosDroid record which frequency each station was heard on.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
An empty AltosState is allocated when parsing JSON data; it's not a
mistake.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
Was trying to match 'IMU call along' instead of 'IMU cal along',
causing the line to not match and losing the IMU accel cal values.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Provides a GUI for re-calibrating accelerometers
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
Allocate throw-away freq array to get the class pointer.
Add null-ary AltosFrequency constructor for JSON code.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Useful to have drogue/main deployment speeds
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
Attempting to fetch sat data from wrong byte led to array bounds
exception.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
If there's no product, assume we've got some ancient log file.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Use new computed state_time value.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Push landed time out if we computed silliness.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Combine boost/fast/coast sections when computing ascent stats.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Especially the nested comment which borked the compile
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
| |
Export euler to quaternion (instead of half_euler).
Provide angles to rotate rather than rates and time.
Add comments to quaternion and rotation code.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
This way we can create it as needed.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
The API and ABI have changed a bit since 1.7
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
No longer in use
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
Need to take the computed tick time and tell the listener what time
that cooresponds to.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
AltosState no longer has imu and mag objects, instead it just holds
the values now.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
| |
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
This makes the igniter fire without a multi-second delay.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
The API for idle monitoring changed, but these modules hadn't been
fixed.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
| |
IMU bits arrive with X/Y/Z labels and are converted to the correct
axes as they leave this object.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
This makes using the graph way easier.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Different from EasyMini v1.0
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
| |
Also fix AltosConfigData.java
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
|
| |
The HMC5883 sensor data is ordered x, z, y. Relabel everything to
match that to preserve compatibility with existing firmware. With the
data correctly ordered, fix the labling of the axes along, across and
through.
Signed-off-by: Keith Packard <keithp@keithp.com>
|
|
|
|
|
|
|
|
| |
Config values that didn't apply to the current product used to get set
to -1, but now they're set to AltosLib.MISSING to be consistent with
the rest of the code. Deal with this in the UI bits.
Signed-off-by: Keith Packard <keithp@keithp.com>
|