This is a wave soaring sailplane simulator for planetary bodies with wind and mountains. Read these concepts for context.
- Mars time, gravity, and atmosphere
- First-order, 2-dimensional flight physics (altitude, attitude)
- First-order atmospheric gravity wave generator (aka. mountain lee waves)
- Static angle-of-attack
- Background wind direction and flight are aligned, mountain ridges are perpendicular
- 2-dimensional dynamics
- Linear atmospheric gravity wave models that work for small background winds, nominal ridge heights
- No wave breaking turbulence
- Constant scale height, which does not account for temperature-dependent density variations
- No turbulence modeling (e.g., wave breaking, lee rotors, etc.), assumes laminar flow fields
- Constant lift curve slope, no accounting for Mach (assuming M < 0.6) or Reynolds number effects on aerodynamics
Building and running is handled via docker, or can be achieved directly via cargo (Rust) or nix (Nix).
Build it: docker build -t wavepool:dev . (or call it something other than wavepool:dev if you like)
Run it: docker run -u $(id -u):$(id -g) -v "$(pwd)/tlm:/app/tlm" -t wavepool:dev (make sure wavepool:dev matches your name and tag of the image; also, this make sure that telemetry is written to the tlm/ directory locally with your user's permissions)
An example configuration file is located in config/, and the output telemetry channels are written to tlm/YYYY-MM-DDTHHMM (for example, tlm/2025-08-19T1028/) in Parquet format.
Inspect mars.toml for an example of a configuration for Mars. The configuration file is broken down into the aircraft and the planet. The planet has an atmosphere, and later will have a wind system.
- Dynamic angle-of-attack control system (currently static input)
- Autonomous soaring controller (wave detection, energy optimization, trajectory planning)
- Waypoint navigation system (follow pre-planned routes)
- Path planning algorithms (find optimal routes through wave fields)
- Control surface modeling (elevator, aileron, rudder deflections as they impact forces/moments)
- Control authority limits (max deflection rates, control saturation)
- Full 6-DOF dynamics (roll, pitch, yaw rates)
- Bank angle and coordinated turns (slip/skid modeling)
- Stall aerodynamics (post-stall lift/drag, spin dynamics)
- Enforce stall speed (minimum airspeed for controlled flight)
- Never-exceed speed (structural limits)
- Coordinate transformations
- Latitude-dependent Coriolis effect
- Great-circle navigation for long-distance flight planning
- Mars curvature effects (for flights > 1000 km)
- Spherical coordinate system (proper lat/lon tracking as position changes)
- Mach-dependent aerodynamics (compressibility effects at M > 0.5)
- Reynolds-dependent aerodynamics (boundary layer transition, drag variations)
- Variable stall characteristics (stall varies with Mach, Re, load factor)
- Wing loading effects (span-wise lift distribution, induced drag refinement)
- Ground effect modeling (for launch/landing phases)
- Turbulence modeling (stochastic velocity perturbations from wave breaking, lee rotors)
- Non-linear wave dynamics (hydraulic jumps, flow separation, wave-wave interactions)
- Thermal updrafts (convective, not just orographic waves)
- Time-varying atmospheric state (diurnal wind cycles, boundary layer evolution)
- Seasonal atmospheric variations (dust loading, temp profiles beyond simple L_s)
- Integration with Mars atmospheric databases (EMARS, MCD for validated profiles)
- Dust storm effects (visibility, heating/cooling, turbulence)
- Atmospheric forecast uncertainty (wind prediction errors)
- Realistic Mars terrain from elevation data (MOLA, HiRISE)
- Arbitrary terrain feature orientations (ridges not perpendicular to wind)
- Multiple interacting terrain features (wave interference patterns)
- Surface property variations (roughness, albedo affecting surface heating/thermals)
- Terrain collision detection and avoidance
- Operational altitude ceiling (atmosphere too thin above ~50 km)
- Energy accounting (KE + PE balance, dissipation tracking, wave energy extraction)
- Performance metrics (L/D ratio instantaneous & average, sink rate, glide ratio)
- Soaring efficiency metrics (altitude gain rate, distance covered, energy efficiency)
- Energy state constraints (can't violate energy conservation)
- Validation test suite (compare to analytical solutions, Earth glider data scaled to Mars)
- Trajectory visualization tools (3D path plots, energy-height diagrams)
- Launch phase modeling (release during EDL, i.e., EDF)
- Landing approach and touchdown (glide slope, flare, ground roll)
- Communication link simulation (downlink telemetry bandwidth, latency, windows)
- Adaptive time-stepping (smaller dt in high-gradient regions like wave breaking)
- Higher-order integration (RK45 with error control for stiff problems)
- Numerical stability analysis (ensure no spurious oscillations)
- Performance profiling and optimization (faster-than-realtime simulation target)
- Parallel simulation capability (Monte Carlo uncertainty analysis)
- Sensor noise modeling (GPS, IMU, airspeed, altimeter errors)
- State estimation (Kalman filter for noisy measurements)
- Navigation errors and corrections
- Monte Carlo analysis tools (sensitivity to parameters, atmospheric uncertainty)