Peter Bochmann
474e3cac8a
Fix what the thermal camera's first run on tower 111 showed
...
- The thermal frame was numbered camera 0 - label RGB, and the colour file of
the same time overwritten - because Vimba counts no camera before open().
The composite now takes each source's configured camera count, and two
enabled cameras may no longer share a label.
- The SDK switches LC_NUMERIC to de_DE when loaded, which made "hotx":0,4336.
Its calls now run under a guard that restores the locale, and the JSON's
numbers no longer go through printf.
- The camera's frame counter is always 0: new frames are told by their time,
so the stream no longer stops after the first one.
- The LUT is empty for the first frame: it is asked again on later ones.
- The SDK is libirbgrab_x86_64-linux.so in the ld cache.
- The null channel traces the CamEvent payload it would have sent.
Tested on tower 111 with the TarisIR mini, the motors mocked: cap_net_raw is
enough, RGB/TIR pairs with one timestamp, the TIR file decodes to the counts
its CamEvent describes, profile switches 0-1-0. 536/536 in the 22.04 container.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-14 21:01:33 +02:00
Peter Bochmann
3d90af8490
Take a thermal picture beside the colour one at every waypoint
...
A TarisIR mini (driver = taris) is a second camera source behind a
CompositeCameraSource: the colour camera is asked first, then the thermal one
for its newest frame, and both frames carry the colour frame's timestamp.
Counts are saved 16-bit lossless under TIR/, and the CamEvent gains cal,
missed, tmax, tmed, hotx and hoty (docs/mqtt-api.md). The SDK is declared in
IrbgrabApi.h and loaded with dlopen, so no SDK file is in the repository; a
missing or unreachable camera stays offline (Health cams_offline) while the
scan goes on. scripts/fgc-setcap gives the binary cap_net_raw after a build.
Built and tested in an Ubuntu 22.04 container (mock, MQTT, Vimba+MQTT,
534/534), plus a mock run with RGB/TIR pairs. Not yet run against the camera.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-14 20:37:25 +02:00
pgdalmeida
6fa43f1e08
Slow down, then pause, the scan when the tower computer runs hot or cannot keep up
2026-10-08 17:21:58 +02:00
pgdalmeida
ce6ef3f79a
Keep each tower's config, scan grid and camera profiles in ~/.config/fire_gimbal_control, out of a deploy's reach
2026-10-05 22:02:49 +02:00
pgdalmeida
70a30132bf
Merge counting the picture interval from the last picture and waiting for a slow camera's next frame
2026-10-05 13:02:13 +02:00
pgdalmeida
1fc87bf1ee
Rewrite the 12-bit camera test report as a numbered paper with an abstract, figure and table captions, and units
2026-10-05 11:28:52 +02:00
pgdalmeida
cfd8ba9978
Save raw 12-bit camera frames as 12-bit colour images, keep a waypoint's picture once the image has settled, and hold the scan when saving falls behind
2026-10-05 10:14:53 +02:00
pgdalmeida
62131d8083
Wait for a slow camera's next frame instead of restarting the stream, and read integer camera features for the exposure check
2026-10-04 16:12:44 +02:00
pgdalmeida
fb0f5117df
Count the picture interval from the last picture, and let a settle wait of 0 wait for the exposure to settle
2026-10-02 18:28:40 +02:00
pgdalmeida
4488a2181d
Set in config.ini whether the tower homes and which mode it starts in, instead of the --init and --start flags
2026-09-30 16:42:48 +02:00
pgdalmeida
f2c61e030e
Let a yaw-only tower aim with gimbal move, and stop it publishing a made-up pitch
2026-09-29 19:50:39 +02:00
pgdalmeida
895c2f22fd
Add a command to keep taking pictures where the gimbal points, drop unused camera commands, and fit the help on one page
2026-09-29 15:48:07 +02:00
pgdalmeida
a21be0d4e8
Take the scan waypoints only from the CSV file, never generate them
2026-09-29 15:02:19 +02:00
pgdalmeida
ab9fa3dbb0
Let a tower set its encoder scale, and send no speed after a failed homing
2026-09-28 18:21:37 +02:00
pgdalmeida
8d81e03a3f
Support the legacy towers' DHT11 and fan, and fix the installer's reflash and test steps
2026-09-28 14:43:05 +02:00
pgdalmeida
79225bc3cd
Rework the dashboard keys: a/p/m for modes, n for connectivity, page keys to scroll, heading setup in the activity strip
2026-09-25 18:57:25 +02:00
pgdalmeida
f4e67666d8
Give the tower three modes, auto, pan and manual, switched by command
2026-09-25 17:50:48 +02:00
pgdalmeida
4a69d65afb
Take any console command over MQTT, and answer it
2026-09-25 16:43:45 +02:00
pgdalmeida
9dc9b90f6d
Call the dashboard's messaging panel COMMS, to match its m key
2026-09-25 15:33:39 +02:00
pgdalmeida
d8d2fee8af
Test the camera with forced dark and bright settings, and recover a camera that stops sending frames
2026-09-24 17:35:37 +02:00
pgdalmeida
5a0515ab13
Test that the camera delivers frames and follows its settings to black and white
2026-09-24 15:47:58 +02:00
pgdalmeida
86f930a671
Add a one-shot test command and a tool that edits the config files safely
2026-09-23 18:52:44 +02:00
pgdalmeida
4049a85b9b
Check the firmware was built with the pins hardware.ini lists
2026-09-23 18:26:14 +02:00
pgdalmeida
b6308c1eda
Check the tower is running its own settings, not the firmware's fallback
2026-09-23 16:07:07 +02:00
pgdalmeida
00c998301a
Bound the report files too, and only prune what this program generated
2026-09-22 18:27:43 +02:00
pgdalmeida
582047d3a2
Keep a log file, so a tower's record outlives the tmux session
2026-09-22 18:15:09 +02:00
pgdalmeida
57a166c9a4
Warn when the homed travel falls below zero, and say why the centre matters
2026-09-21 23:06:37 +02:00
pgdalmeida
575e3c590b
Say when a paused dashboard was frozen, so it cannot be mistaken for a hung one
2026-09-21 21:53:04 +02:00
pgdalmeida
e2bdc8debf
Trust the firmware's reset-cause name over a decode that cannot be right for both boards
2026-09-21 19:03:55 +02:00
pgdalmeida
816242725a
Add the handshake bring-up runbook (plus the mock/test changes)
2026-09-18 17:36:19 +02:00
pgdalmeida
041ad2a18f
Understand the NEEDCFG state instead of reading it as a dead link
2026-09-18 17:15:16 +02:00
pgdalmeida
842d95eeca
Check every declared value, not just the current
2026-09-18 16:44:08 +02:00
pgdalmeida
522ba072ec
Answer the firmware's NEED lines from hardware.ini
2026-09-18 16:16:04 +02:00
pgdalmeida
579c4ec166
Let a declared sensor enable itself, and retire the duplicate flag
2026-09-18 15:57:10 +02:00
pgdalmeida
b5d1fac4cb
Declare axis speed in degrees per second, not VMAX
2026-09-18 15:44:19 +02:00
pgdalmeida
4f57651c9b
Refuse to start when the rig is missing hardware it declares
2026-09-18 14:50:50 +02:00
pgdalmeida
31da8012d7
Stop addressing an axis this tower does not have, and carry the live calibration
2026-09-17 18:14:13 +02:00
pgdalmeida
9a61167aa6
Parse the motor current the firmware echoes back in DUMP
2026-09-17 17:58:46 +02:00
pgdalmeida
ffaa41ac2f
Read the drivetrain per axis from hardware.ini, and compare INI values numerically
2026-09-17 17:58:45 +02:00
pgdalmeida
3476e5adc7
Convert motor current from milliamps to TMC register values
2026-09-17 17:58:45 +02:00
pgdalmeida
76e777f5af
Add hardware.ini assembly config, build identity in the header, and ;prev calibration history
...
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018GsFaaLUTCYy3ZjaNTJzMv
2026-09-11 21:28:18 +02:00
pgdalmeida
1a572cb074
Changed heading logic; added snapshot feature
2026-09-11 17:42:01 +02:00
pgdalmeida
cd8491dcc0
removed Env MQTT topic
2026-09-09 13:43:29 +02:00
pgdalmeida
8eda0b4a36
Added IMU pitch to CamEvent; Changed IMU reading to before jxl encoding to avoid lag
2026-09-09 13:04:25 +02:00
pgdalmeida
5fce32aee7
IMU readings fixed; IMU mounting is part of config.ini; device config verified at startup
2026-09-09 12:19:53 +02:00
pgdalmeida
a5ad814260
Camera settings now come from VimbaX profile XMLs; explicit per-camera config
2026-09-04 14:12:05 +02:00
pgdalmeida
7db8d2092a
Quitting the TUI now asks whether to shut down or detach
2026-09-02 13:36:50 +02:00
pgdalmeida
bc4d2ec002
Added an expanded connectivity view with MQTT message history
2026-09-02 12:39:21 +02:00
pgdalmeida
6c1941cc80
SHT41 bug resolved. It now logs reading failures.
2026-09-02 12:08:31 +02:00
pgdalmeida
ef96513b58
MQTT messages working
2026-08-26 17:56:56 +02:00
Peter Bochmann
pgdalmeida