Step-by-step evaluation of a hypothetical hexapod platform where motion is transformed by a washout algorithm before reaching the actuators
Educational demonstration only. No real products, manufacturers, or organizations are referenced. This hypothetical system was constructed to illustrate what a Surface-Level classification looks like from the inside — and specifically why the washout layer and hexapod coupling produce a different result than Reference System A.
Educational reference only. This evaluation is a hypothetical demonstration. The system described does not exist and is not based on any real product. The classification shown here does not apply to any real simulation system. For the evaluation methodology this demonstration is based on, see Evaluation Process.
Reference System B is a hypothetical hexapod motion platform that uses a motion cueing algorithm (MCA) to translate vehicle physics state information into platform commands. This architecture represents the class of systems where motion is present and physically substantial, but where the motion delivered to the participant is a transformed approximation of the physics-derived motion — not the physics-derived motion itself.
The two architectural features that distinguish this system from Reference System A are the motion cueing algorithm and the hexapod mechanical configuration. The MCA is not merely a scaling factor — it applies washout filtering that modifies the timing and magnitude relationships between the physics state and the motion output. The hexapod configuration means that all six degrees of freedom are resolved simultaneously through six actuator legs: there is no independent axis at the mechanical level.
The same five required inputs were collected as in Reference Evaluation A. The evidence is complete. The issue is not missing evidence — it is what the evidence shows. Two inputs directly document non-compliance with the structural criteria.
| Input | Description | Tier | Finding |
|---|---|---|---|
| R1 Motion Telemetry |
Time-stamped actuator command logs cross-referenced with physics model force outputs during reference motion events | Tier 1 | Actuator commands do not correspond directly to physics outputs. Comparison shows washout-modified signal: onset timing is delayed, magnitude is attenuated, and high-frequency content is filtered. The physics signal is the input to the MCA, not the source of the actuator command. |
| R2 Physics Architecture |
System architecture documentation showing signal path from physics engine to motion controller | Tier 2 | Documentation confirms MCA layer between physics engine output and actuator command. Signal path: physics engine → MCA input → washout processing → platform inverse kinematics → actuator commands. The MCA is not optional and is not bypassable in standard operation. |
| R3 Actuator Specification |
Engineering specification for the hexapod leg actuators and kinematic configuration | Tier 2 | Hexapod configuration confirmed. All six degrees of freedom are resolved through the kinematic solution for six actuator legs. There is no independent actuator per axis: a command to produce yaw rotation requires coordinated changes across all six legs simultaneously. Mechanical coupling is inherent to the hexapod geometry. |
| R4 Synchronization Data |
Timing logs for motion command issuance relative to visual frame render during reference motion events | Tier 1 | Motion command issuance occurs within the visual frame cycle. Gross synchronization is present. Note: the washout filter introduces additional latency between physics onset and motion onset that is not captured by frame-level synchronization data alone. |
| R5 Control Telemetry |
Logs tracing participant control inputs through physics state changes to resulting motion commands | Tier 1 | Control inputs affect physics state. Physics state changes affect MCA input. However, the MCA's washout processing means the relationship between physics state and actuator command is not direct. The motion output is an approximation of the physics-derived motion, not the physics-derived motion itself. |
The same three criteria are assessed as in Reference Evaluation A. The evidence for this system produces a different outcome: two failures and one Insufficient Data result.
R1 (motion telemetry) shows that actuator commands reflect a washout-modified version of the physics signal, not the physics signal itself. R2 (architecture documentation) confirms that the MCA with washout filtering is the mandatory processing stage between physics output and actuator command.
Criterion A fails. The actuator command does not originate from the physics model output at time of generation. The physics model output is the input to a motion cueing algorithm. The MCA applies washout filtering that modifies the timing, magnitude, and frequency content of the signal before it becomes an actuator command. The motion delivered to the participant is a transformed approximation of the physics-derived motion. The causal relationship between vehicle physics state and participant motion input is indirect and mediated by the MCA.
Primary evidence: R1 (Tier 1), R2 (Tier 2)
R3 (actuator specification) documents hexapod geometry with six actuator legs resolving all six degrees of freedom through coordinated kinematic solution. No independent axis control exists at the mechanical level. Reference point confirmed as platform geometric center, not vehicle center of mass.
Criterion B fails on two grounds. First, the hexapod configuration means that no axis of motion can be commanded independently: every motion command requires a coordinated change across all six legs through the inverse kinematics solution. Axis independence does not exist at the mechanical level. Second, the reference point for all calculations is the platform geometric center, not the vehicle center of mass. Both conditions required for Criterion B to pass are absent.
Primary evidence: R3 (Tier 2)
R1 (motion telemetry) and R4 (synchronization data) provide motion magnitude and timing data. However, given that Criteria A and B have failed, the relevance of this data to the criterion question requires different framing: the question is not whether the motion corresponds to the physics state (it does not, per Criterion A), but whether the modified motion characteristics still fall within physiological detection parameters for the depicted events.
Criterion C returns Insufficient Data. The washout filter modifies timing and magnitude in ways that depend on MCA configuration parameters not provided in the evidence package. Without knowing the specific washout filter settings and their effect on onset timing, it is not possible to determine whether the inner ear receives a signal consistent with the depicted vehicle events or a signal that misrepresents them. This is not a question of inadequate evidence quantity — it is a gap in the evidence content that the available inputs cannot resolve.
Primary evidence: R1 (Tier 1), R4 (Tier 1)
Motion is present and physically delivered to the participant, but it is not causatively derived from the vehicle physics model. The MCA washout layer breaks the direct causal chain (Criterion A fail). The hexapod configuration prevents independent axis operation (Criterion B fail). Criterion C returns Insufficient Data because the evidence package does not include the MCA configuration parameters needed to assess physiological signal validity. Two failures are sufficient for a Surface-Level classification; the third criterion does not need to pass or fail for the classification to be determined.
The two reference systems are deliberately constructed to be similar in surface appearance — both have six axes of motion, both are driven by vehicle physics state information, both have synchronized motion and visual output. The difference is entirely in the architecture. This comparison shows precisely where the two systems diverge.
| Reference System A (In-the-Loop) | Reference System B (Surface-Level) | |
|---|---|---|
| Motion source | Direct physics model output | Physics model output processed through MCA washout filter |
| Signal path | Physics engine → motion controller → actuators | Physics engine → MCA → washout → inverse kinematics → actuators |
| Axis independence | Independent closed-loop actuator per axis | All six DOF coupled through hexapod inverse kinematics |
| Reference point | Vehicle center of mass | Platform geometric center |
| Criterion A | Pass — direct causal chain intact | Fail — washout layer breaks direct causal relationship |
| Criterion B | Pass — independent axes, CoM reference | Fail — hexapod coupling, geometric center reference |
| Criterion C | Pass — physics-derived motion within physiological parameters | Insufficient Data — MCA configuration not provided |
| Classification | In-the-Loop | Surface-Level |