package
Vehicles.Rdd2.Test
RDD2 model-level test missions
TruthWaypointMission is the truth-feedback baseline. Compare it with WaypointMission for optical-flow-aided ESKF navigation and GlobalWaypointMission for GPS-aided geodetic waypoints. The corresponding UKF missions use the same controller and plant assumptions.
GPS-to-Mocap handoff
Paired surveyed and ideal missions share noise and coverage timing. The surveyed rig offset is 5 cm east and 3 cm north. Compare estimate minus truth at matching ticks: errors must agree before coverage and recover that vector during both the first 0.5 seconds and remaining coverage. The residual budget is 3 sqrt(2) times the per-axis 1 cm Mocap noise. It does not shrink with sample count because errors are correlated. Missing windows, mismatched ticks and non-finite errors fail.
These are deterministic simulation checks, not independent-sample confidence intervals or a substitute for flight qualification.
Members
| Name | Kind | Description |
|---|---|---|
| WaypointMission | model | Optical-flow-navigated takeoff, box, and landing mission in the local East-North-Up frame |
| ReducedControllerWaypointMission | model | Optical-flow box mission flown by the reduced-authority controller variant |
| OpticalFlowGpsDeniedMission | model | GPS-aided box mission with a mid-flight GPS outage carried by optical flow |
| TruthWaypointMission | model | RDD2 waypoint mission using plant truth as the controller baseline |
| GlobalWaypointMission | model | GPS-navigated RDD2 waypoint mission authored in global latitude/longitude/altitude |
| MocapWaypointMission | model | Motion-capture-navigated takeoff, box, and landing inside a rig volume |
| MocapHandoffMission | model | Fly GPS, cross into rig coverage, manoeuvre, and cross back out |
| MocapHandoffIdealMission | model | The same two-way handoff with a perfect rig survey |
| LongRangeFlowMission | model | Two hundred metres of optical-flow-aided flight, to characterize drift |
| LongRangeGpsMission | model | The same two hundred metres on GPS aiding, as the drift control |
| ManualFlightMission | model | GPS-navigated flight handed from the mission to the pilot and back |
| VibrationGlobalWaypointMission | model | GPS-navigated RDD2 waypoint mission flown with rotor vibration on the IMU |
| UnfilteredVibrationGlobalWaypointMission | model | Vibration mission flown without the sensor anti-alias low-pass |
| SevereVibrationGlobalWaypointMission | model | Vibration mission flown with a worn or badly unbalanced rotor set |
| FohGlobalWaypointMission | model | GPS-navigated waypoint mission with first-order-hold IMU composition |
| UkfWaypointMission | model | Optical-flow waypoint mission using the StrapdownINS manifold UKF |
| UkfGlobalWaypointMission | model | GPS waypoint mission using the StrapdownINS manifold UKF |
| ConsistencyEnsemble | package | Seed-varied mission wrappers for the NEES/NIS consistency study |
Modelica declarations and implementation
Show parameters, interfaces and source
Parameters, inputs, outputs, defaults and equations from the source file.
package Test "RDD2 model-level test missions"
annotation(Documentation(info = "<html>
<p><a href=\"modelica://Vehicles.Rdd2.Test.TruthWaypointMission\">TruthWaypointMission</a>
is the truth-feedback baseline. Compare it with
<a href=\"modelica://Vehicles.Rdd2.Test.WaypointMission\">WaypointMission</a>
for optical-flow-aided ESKF navigation and
<a href=\"modelica://Vehicles.Rdd2.Test.GlobalWaypointMission\">GlobalWaypointMission</a>
for GPS-aided geodetic waypoints. The corresponding UKF missions use the
same controller and plant assumptions.</p>
<h4>GPS-to-Mocap handoff</h4>
<p>Paired surveyed and ideal missions share noise and coverage timing.
The surveyed rig offset is 5 cm east and 3 cm north. Compare estimate
minus truth at matching ticks: errors must agree before coverage and
recover that vector during both the first 0.5 seconds and remaining
coverage. The residual budget is 3 sqrt(2) times the per-axis 1 cm Mocap
noise. It does not shrink with sample count because errors are correlated.
Missing windows, mismatched ticks and non-finite errors fail.</p>
<p>These are deterministic simulation checks, not independent-sample
confidence intervals or a substitute for flight qualification.</p>
</html>"));
end Test;