The standard test environment for SFR evaluations.
This document defines the reference test conditions used when evaluating a simulation system against the SFR framework. It establishes the reference vehicle concept, reference motion events, reference physics conditions, and reference measurement procedures. The methodology is architecture-neutral and does not reference specific manufacturers or proprietary systems.
A reference test methodology defines the conditions under which an evaluation is conducted. Without a reference methodology, evaluations conducted by different evaluators may reach different conclusions not because the system performed differently, but because the test conditions were different. The reference test methodology eliminates test-condition variance as a source of classification disagreement.
This methodology applies to the evaluation of any simulation system for which SFR classification is being determined. It is architecture-neutral: it does not assume any particular motion system design, and it applies equally to in-the-loop, surface-level, and out-of-the-loop systems under evaluation.
The reference test methodology exists to make evaluation outcomes a function of system performance, not a function of test conditions.
The reference vehicle is an abstract conceptual construct, not a specific production vehicle. It is defined by its dynamics properties, not its manufacturer or model designation. Any simulation system under evaluation must be capable of representing the reference vehicle's dynamics properties within its stated operating envelope.
The reference vehicle is defined by the following properties:
The reference vehicle is not intended to replicate any specific vehicle. It is intended to produce physically consistent and repeatable dynamics events that can be used to assess whether the simulation system under evaluation is correctly deriving its motion output from live physics state.
Four reference motion events are defined. Each event is selected to isolate specific aspects of the three fundamental criteria. Events are conducted in sequence, with sufficient rest time between events to allow system state reset.
| Event | Description | Primary Purpose | Criteria Addressed |
|---|---|---|---|
| Event 1 Steady-State Lateral |
Sustained cornering at a constant lateral acceleration level within the grip envelope. Maintained for a duration sufficient to allow system state to stabilize and for vestibular steady-state to develop. | Verifies that continuous lateral force is derived from the live physics state, not from a canned or scripted motion profile. | A — Causative Accuracy |
| Event 2 Transient Yaw Onset |
Turn-in from a straight-line condition with rapid yaw rate development. The event captures the onset of yaw rotation from rest and the rate of change of yaw in the initial entry phase. | Verifies that the yaw cue arrives within a valid temporal relationship to the visual yaw onset. Tests the timing structure of the motion system under a directionally pure axis event. | B — Temporal Coherence |
| Event 3 Combined Axis |
Simultaneous braking and turning, producing compound yaw and pitch events within a single driving action. This event creates simultaneous demand on multiple independent axes. | Verifies that the motion system can deliver independent timing for simultaneous yaw and pitch events without axis coupling degrading either signal's temporal accuracy. | B — Temporal Coherence |
| Event 4 Limit-State Threshold |
Progressive lateral acceleration to the grip threshold, then to the limit state. The event tests whether the participant's control corrections are driven by physical sensation or by visual information. | Verifies that the sensory cue is sufficient to alter natural control behavior at the threshold. If the participant relies primarily on visual input for threshold identification, the criterion is not met. | C — Human Response Relevance |
Events must be conducted in the specified order. Results from events conducted outside this sequence are not valid for SFR classification purposes.
The following physics conditions apply to all reference events. These conditions define the operational range within which reference events are conducted. They are specified qualitatively here, as numerical thresholds are subject to revision when instrumentation standards are finalized for SFR v1.0.
Must span from below the vestibular detection threshold to at or above the grip limit threshold of the reference vehicle. Events 1 and 4 operate within this range. The range must be sufficient to produce a clearly distinguishable onset and a clearly distinguishable limit-state event.
Events 2 and 3 must produce yaw rate values above the vestibular angular velocity detection threshold. The yaw onset in Event 2 must be sufficiently rapid to produce a measurable temporal relationship between motion cue delivery and visual yaw indication.
Events 3 must produce simultaneous longitudinal deceleration in addition to lateral acceleration. The braking component must be above vestibular detection threshold and must produce a measurable forward weight transfer in the physics model.
All events are conducted at a vehicle speed that produces physics events within the vestibularly relevant range. Static or near-static conditions are excluded. The speed range is sufficient to produce events in the range where driver control corrections are neurologically triggered.
Event 3 requires that yaw and pitch events occur simultaneously within the same driving action. The physics model must generate both axes concurrently, not sequentially. The test is invalid if the axes are experienced as separate sequential events.
Each reference event must be repeatable within the same session with consistent physics output. Variation in vehicle behavior across repeated runs of the same event must be within the measurement precision of the telemetry system being used.
Measurement procedures define the sequence of steps taken before, during, and after reference events. Adherence to these procedures is required for evaluation results to be valid.
Before any physical test runs, the evaluator must review the system's architecture documentation, physics model specification, and actuator specification. This review establishes the Tier 2 evidence baseline against which telemetry results (Tier 1) will be compared. Discrepancies between documentation and measured behavior are recorded.
All measurement instruments must be calibrated to a known reference before the test session begins. Calibration records must be retained as part of the evaluation evidence package. Uncalibrated instrumentation produces Tier 3 (observed behavior) evidence at best, not Tier 1 (measured telemetry).
Before conducting reference events, a baseline run is performed through the standard test sequence at reduced intensity to confirm that the telemetry system is capturing all required data channels and that the system under test is in its standard operating configuration.
Reference events are conducted in the sequence defined in Section 3. Each event is repeated a minimum of three times. The median result across repetitions is used for criterion assessment. Individual outlier runs may be noted but do not replace the median unless a documented technical anomaly explains the outlier.
Following reference event execution, synchronization between the physics telemetry, motion telemetry, and visual system timing is verified. This verification uses the Event 2 (Transient Yaw Onset) data, as it produces the clearest temporal relationship between physics event and motion cue delivery.
All telemetry, calibration records, documentation review notes, and synchronization verification results are assembled into the evaluation evidence package. This package is the primary input to the criterion assessment process defined in the Evaluation Process document.
This reference test methodology is one of three normative methodology documents that together define how SFR evaluations are conducted. The methodology defines what to test. The inputs document defines what data to collect. The process document defines how to assess the data against the criteria.
No single document is sufficient for a complete evaluation. All three must be applied together.
Evaluation results that cannot be reproduced under the same conditions are not evaluation results. They are observations. The reference test methodology converts observation into repeatable measurement by fixing the conditions under which the measurement is taken.
Architecture-neutrality is a requirement, not a preference. The methodology must apply equally to any system being evaluated, regardless of its design. A methodology that only applies to certain architectures cannot produce a general classification standard.