Flight Dynamics
6-DoF longitudinal rigid-body dynamics with second-order actuator modeling, SU2-anchored aerodynamic surrogates, and event-driven CFD verification.
The Verified Bridge
Flight dynamics connects CFD to real-world motion. An aerodynamic surrogate built from SU2 RANS data feeds a Newton-Euler 6-DoF integrator, which drives an event-driven verification hook that fires SU2 snapshots at critical flight conditions - closing the loop between fast surrogate prediction and high-fidelity CFD.
SU2 CFD RANS
Compressible RANS with SA turbulence at 5 AoA points
0° to 16°
32K nodes
AeroSurrogate
Linear interpolation over AoA with flap delta effects
scipy interp1d
JSON serialized
6-DoF Dynamics
Newton-Euler integration at 200 Hz with second-order actuator
[u,w,q,θ,x,z]
200 Hz
Verification Hook
Event-driven SU2 snapshot at peak gust and cruise recovery
2 events
Active
Interactive Simulation Replay
Watch the aircraft respond to a 5 m/s “1-cos” vertical gust at 50 m/s cruise. The simulation runs at 200 Hz with second-order actuator dynamics. Nose-up pitch indicates positive AoA excursion.
Steady vs Gust Response
A 5 m/s vertical gust (duration 0.5s, onset at t=1.0s) excites the longitudinal dynamics. The aircraft experiences a transient AoA increase of 0.6° and a peak load factor of 2.48g before recovering to trim.
Steady Cruise
Peak Gust Response
Verification Event Log
The verification hook monitors the simulation state stream for key flight conditions. On trigger, it serializes the flight state and requests a high-fidelity SU2 RANS snapshot for comparison against the aerodynamic surrogate prediction.
| Event | Time | AoA | V | Surrogate CL | Surrogate CD | SU2 Status |
|---|---|---|---|---|---|---|
| Peak Gust | 1.26s | 1.73° | 50.6 m/s | 0.194 | 0.083 | Snapshot Requested |
| Cruise Recovery | 1.70s | 0.80° | 51.0 m/s | 0.090 | 0.080 | Snapshot Requested |
Physics Model
The longitudinal dynamics integrate Newton-Euler rigid-body equations in body-axis coordinates. Aerodynamic forces are computed from the surrogate model and rotated from wind axes. Static stability is enforced via a negative static margin ($C_{m_\alpha} = C_L \cdot \overline{x}_{cg}$).
Newton-Euler Equations
$$ \dot{u} = -q w - g\sin\theta + \frac{F_x}{m} $$
$$ \dot{w} = q u + g\cos\theta + \frac{F_z}{m} $$
$$ \dot{q} = \frac{M}{I_{yy}} $$
$$ \dot{\theta} = q $$
Wind-to-Body Force Rotation
$$ F_x = L\sin\alpha - D\cos\alpha + T $$
$$ F_z = -L\cos\alpha - D\sin\alpha $$
$$ M = \bar{q} S c\,(C_m + C_{L}\overline{x}_{cg} + C_{m_q}\hat{q}) $$
Second-Order Actuator
$$ \ddot{\delta} = \omega_n^2(\delta_{cmd} - \delta) - 2\zeta\omega_n\dot{\delta} $$
$\zeta = 0.7$, $\omega_n = 10$ rad/s, $f_s = 200$ Hz
Gust Profile
$$ V_g(t) = \frac{A}{2}\left[1 - \cos\frac{2\pi(t-t_0)}{T}\right] $$
$A = 5$ m/s, $t_0 = 1.0$ s, $T = 0.5$ s
Full Telemetry
The dual-panel dashboard shows the flight trajectory with AoA overlay (top) and the actuator commanded vs actual deflection with load factor (bottom). The gust event at t=1.0s is visible as a transient in both panels.
Module Architecture
The flight dynamics pipeline is implemented across six modules in src/ and one in physics/:
| Module | Class / Function | Purpose |
|---|---|---|
physics/aero_surrogate.py | AeroSurrogate | CL/CD/CM interpolation from SU2 data |
src/dynamics.py | LongitudinalDynamics | 6-DoF Newton-Euler integration |
src/actuator.py | Actuator | Second-order control surface model |
src/gust_profile.py | gust_1minuscos() | Smooth vertical wind gust generator |
src/verification_hook.py | VerificationHook | Event-driven SU2 snapshot trigger |
src/plot_telemetry.py | plot_telemetry() | Dracula-themed telemetry dashboard |
run_flight_dynamics.py | main() | Orchestrator and entry point |