SIL · Software-in-the-Loop
Exercise production control logic against simulated sensors, actuators, and vehicle physics.
DEVELOPMENT · TEAM-WIDE CAPABILITY
SIMLAB
Simulation, Integration, Verification, and Engineering Data Laboratory
Explore how physics, software, hardware, and operators interact before the real test. SIMLAB is DRT’s planned integrated environment for discovering system faults, understanding uncertainty, and making better engineering decisions.
Initial program charter · Incremental development roadmap
PURPOSE & MISSION
Understand the complete system.
SIMLAB aims to bring propulsion, avionics, controls, ground systems, vehicle dynamics, and mission operations into controlled, repeatable tests under nominal and failure conditions.
Detailed CFD, FEA, and fluid-network analyses supply reduced-order models; physical tests supply the evidence to validate and improve them. Start with one useful model, validate it, then connect the next subsystem.
PLANNED CAPABILITIES
From control logic to integrated missions
PIL · Processor-in-the-Loop
Run software on representative embedded processors to evaluate timing, scheduling, and resource use.
HIL · Hardware-in-the-Loop
Connect real avionics and DAQ hardware to a simulated vehicle through realistic inputs and outputs.
HITL · Human-in-the-Loop
Give Mission Control a repeatable environment for countdowns, fault response, aborts, and safing rehearsals.
Fault injection & regression
Introduce sensor drift, valve delays, packet loss, and combined failures; check expected responses after software changes.
Monte Carlo & uncertainty
Explore parameter distributions and sensitivity to understand performance margins and the drivers of system risk.
Engineering data & calibration
Clean and synchronize test telemetry, compare predictions with measurements, and calibrate models within validated ranges.
Computing performance
Profile CPU, memory, network load, and timing; scale parameter sweeps and simulation campaigns across computing workers.
Machine learning
Explore anomaly detection, validated surrogate models, and scenario generation as supplements to physics and deterministic verification.
INTEGRATION / CONCEPTUAL SIGNAL FLOW
Close the loop.
Real controllers command a simulated vehicle. Its response returns as sensor measurements, completing the feedback loop.
Hardware-in-the-loopILLUSTRATIVE SEQUENCE
Simulated values, not live telemetry. A simplified sequence: open valve → pressure rises → close valve → pressure falls.
REPRESENTATIVE FUTURE SCENARIO
Rehearse a static fire. Challenge the response.
Begin with an Ambition One digital test stand, then extend validated models toward Dragon One. Real control logic would command simulated loading, arming, and ignition while Mission Control monitors realistic telemetry.
01 · Run
Model valve timing, fluid flow, chamber-pressure rise, and thrust buildup.
02 · Inject
At T+2.5 seconds, introduce downward drift in a simulated fuel-pressure sensor.
03 · Verify
Check fault detection, abort initiation, valve closure, shutdown timing, and the final safe state against requirements.
CROSS-TEAM INTERFACES
Built with every discipline
Propulsion
Engine models, operating limits, and test data inform transient simulation and uncertainty studies.
Avionics, Controls & GNC
Flight software, state machines, and communications connect to virtual sensors, actuators, and timing checks.
Stage 0 & Operations
Ground-system interfaces and procedures support virtual servicing, countdowns, and operator rehearsals.
Turbomachinery
Turbopump models, performance maps, operational data informing transient & steady-state conditions.
Future interoperability with Synnax-based telemetry and Mission Control would let operators use familiar interfaces across simulation, HIL, and physical testing. Test data feeds back into model calibration and verification.
PROPOSED ROADMAP
One validated step at a time
A development sequence from the initial charter; milestones will evolve with team needs and validation evidence.
- Phase 0
Architecture & requirements
Define model interfaces, scenario standards, data schemas, and the computing foundation.
- Phase 1
Ambition One digital test stand
Validate simplified tank, valve, sensor, and engine models against static-fire data and control sequences.
- Phase 2
Automated scenarios
Build repeatable fault schedules, expected responses, pass/fail checks, and result logging.
- Phase 3
Monte Carlo framework
Add distributed execution, statistical summaries, sensitivity studies, and reporting.
- Phase 4
Dragon One system models
Integrate feed systems, pumps, GSE, and reduced-order startup and shutdown models.
- Phase 5
Hardware-in-the-Loop
Introduce control computers, DAQ, communications, and eventually representative avionics racks.
- Phase 6
Vehicle & mission simulation
Expand into flight dynamics, GNC, staging, recovery, and complete mission rehearsals.
LONG-TERM VISION
Thousands of virtual missions before flight.
Grow a persistent digital engineering environment spanning ground servicing, ignition, ascent, staging, and recovery. Success means finding faults earlier, quantifying risk, reproducing test data, and giving the team stronger evidence before operating the real system.