plot_flight_data.py¶

Source: tools/plot_flight_data.py

Plot AURORA flight data — recorded telemetry from the flight computer or simulated streams produced by sim_flight_kalman.py.

What it does¶

This is the dedicated plotting tool for the AURORA flight stack. It covers two use cases:

Standalone CLI: python3 plot_flight_data.py --flight DIR segments the log under DIR into individual flights on ARMED → BOOST transitions and produces one multi-panel plot per flight, with state-machine transitions drawn as labelled vertical markers on every panel. DIR is expected to contain flights.influx (line-protocol telemetry) and state_audit (state-machine transition log).
Imported module: sim_flight_kalman.py imports the module and calls plot_flight(...) to render the seven-panel plot of its simulated output.

Panels are created only for the streams a log actually contains, so the plot covers whatever the recording holds: barometer, Kalman altitude and velocity, state-machine acceleration inputs, body acceleration, rotation rate, magnetometer, orientation, battery voltage, and the replayed apogee votes.

The tool does not run the Kalman filter or attitude tracker itself. What it does carry, on top of pure plotting, is the log-loading and replay glue needed to turn raw telemetry into the curves that the plot panels expect — pressure → altitude conversion, NIS-gate replay against the logged Kalman altitude, the three-vote apogee replay, and the boost/landed-recovery heuristics for logs whose state machine never reached BOOST.

Note

The replay panels (NIS gate, apogee votes) approximate the firmware behaviour from logged streams alone. The on-board covariance is not telemetered, so the gate is evaluated under a steady-state assumption. Treat the markers as a sanity check, not as a perfect re-run.

Usage¶

python3 tools/plot_flight_data.py --flight DIR
                                  [--theme {light,dark,both}]
                                  [--show] [--disable-votes]
                                  [--pre-boost SECONDS]
                                  [--post-end SECONDS]
                                  [--title TITLE]
                                  [--trim [SECONDS]]
                                  [--r-meas R_MEAS]

Data selection¶

--flight DIR — required; flight directory containing flights.influx and state_audit.
--pre-boost SECONDS — seconds of pre-boost data to keep per flight (default 10).
--post-end SECONDS — seconds of data to keep after the flight close-out transition, trimming the long post-parachute tail (default 2).

Replay¶

--r-meas R_MEAS — assumed Kalman measurement variance (m²) used to approximate the NIS gate from logged baro vs sm_pose. Match CONFIG_FILTER_R_MILLISCALE / 1000 for the flight build (default 6.0).
--disable-votes — drop the apogee-vote panel (useful when the logged streams are known to be unreliable, e.g. a stuck pointer in the logger).

Output¶

--theme {light,dark,both} — render plots matching the Furo Sphinx docs theme (default both).
--show — open the plots in an interactive matplotlib window in addition to writing them to disk.
--title TITLE — override the plot title (applied to every plot the run produces).
--trim [SECONDS] — write a trimmed copy of flights.influx next to the original, covering each flight’s plot window expanded by SECONDS on each side (default 10 when --trim is given without a value). Useful for shrinking long pad-time logs before sharing them.

Plots are written to flight<N>.png in the current directory, with _light / _dark suffixes when --theme both is used.

Reuse from other scripts¶

The plotting and log-handling helpers are designed to be importable. The most useful entry points are:

parse_influx(path), parse_state_audit(path), segment_flights(events) — load telemetry and split a log into flights.
slice_real_flight(...) — trim raw streams to one flight window.
compute_log_votes(sliced), compute_log_gated(sliced, r_meas) — replay the apogee votes and NIS gate from logged streams.
altitude_to_pressure(h) / pressure_to_altitude(p, p_ref) — ISA conversions.
plot_flight(...) — seven-panel plot used by sim_flight_kalman.py.
plot_raw_flight(...) — multi-panel raw-telemetry plot used by the CLI above. Pass out_path=None to build the figure without writing it (this is how flight_log_gui.py embeds it in a Tk canvas), and panels=[...] to draw a subset.
available_panels(sliced) — the panel keys a sliced flight has data for, in stack order; PANEL_LABELS maps each to a readable name.

Requirements¶

Python 3.10+
numpy, matplotlib