4. Far-side occupant motion simulation
Far-side occupant motion is fundamentally a three-dimensional problem. When a belted driver is displaced toward the passenger side during a far-side impact, the restraint system no longer controls the torso through the diagonal shoulder belt in the way it was designed to. The occupant slides laterally out of the belt geometry, the head travels across the center console region, and the interaction with the far-side interior — or with a far-side occupant — becomes the dominant injury mechanism. Capturing this motion requires a simulation environment that can represent multi-body kinematics, belt routing, contact, and detailed human body deformation simultaneously. MADJMO's hybrid approach — rigid multi-body bodies coupled to finite element belts and FE human body models — is the standard tool chain for this class of problem.
Model architecture and the hybrid coupling strategy
The practical starting point is deciding what must be FE and what can remain multi-body. In a far-side simulation, the occupant's gross trajectory is driven by inertia and belt forces, which multi-body dynamics handles efficiently and robustly. However, the local deformation at the shoulder, the thorax deflection under the belt, and head-to-interior contact benefit from FE resolution. A typical far-side model therefore uses an FE human body model (THUMS, GHBMC, or the MADYMO Human Body Model) coupled to a multi-body seat and interior, with the belt modeled as FE seatbelt elements rather than multi-body belt segments.
The coupling itself is defined through a MADYMO coupling element that exchanges forces and displacements between the FE and MB subsystems at each time step. The critical setup detail is the coupling surface: the FE model's outer surface must be declared as a coupling surface, and the multi-body bodies that interact with it (seat cushion, console, door trim) must be assigned contact characteristics that are consistent with the FE contact definitions. Mismatched contact stiffness between the two subsystems is one of the most common sources of non-physical penetration or oscillation in far-side runs.
Positioning, belt routing, and the far-side initial state
Far-side simulation is sensitive to the initial occupant position in a way that frontal simulation is not. The lateral offset of the occupant relative to the seat centerline, the initial belt fit at the shoulder and lap, and the pre-impact posture all determine whether the occupant escapes the belt early or late. A defensible setup procedure is:
- Place the FE human body model in the seat using the H-point and pelvis angle from the target vehicle's seating reference, then apply a gravity settling phase of 200–300 ms so the cushion and seat back reach equilibrium before the impact pulse is applied.
- Route the FE belt through the D-ring, across the shoulder, through the buckle, and to the anchor points. Verify belt-to-body clearance at the clavicle and the lap belt position over the ASIS; a lap belt sitting too high will produce early submarining that masks the far-side motion you are trying to study.
- Apply a pre-tensioner and load-limiter profile consistent with the restraint configuration under study. For far-side, the load limiter level strongly influences how much the torso is allowed to translate laterally before the belt slips off the shoulder.
- Define the impact pulse and the far-side boundary — typically a moving deformable barrier or a prescribed door/console intrusion — and confirm the pulse is applied to the correct subsystem.
Coordinate conventions must be fixed before any of this: use the vehicle coordinate system with X forward, Y to the occupant's left (passenger side for a left-hand-drive vehicle), and Z up, and state explicitly whether the far-side direction is +Y or −Y for the specific occupant. Every output plot and every validation comparison should carry that convention.
Validation: what to check before trusting a far-side result
Far-side models are validated in two layers. The first layer is the human body model itself: the FE HBM must reproduce published frontal and lateral certification corridors — chest deflection, head acceleration, and pelvis response — before it is used in any vehicle-specific study. The second layer is the restraint and interior model: belt pull tests, seat cushion stiffness, and console force-deflection characteristics should be checked against component-level test data.
For the far-side scenario specifically, the most informative validation target is the lateral head trajectory. If a physical sled test with a far-side pulse is available, compare the head center-of-gravity Y-displacement time history against the simulation. A well-validated model typically tracks the head excursion within roughly 10–15 mm through the first 120 ms, with divergence growing after head contact. If the simulated head moves laterally too early, the usual causes are an over-stiff lap belt, a shoulder belt routed too far inboard, or a seat cushion that is too soft and lets the pelvis translate before the belt engages. If the head lags the test, check the pre-tensioner timing and the load limiter level.
Key outputs and how to read them
The output set for a far-side study should be chosen to answer the countermeasure question, not to fill a report. The four channels that matter most are:
| Output | What it tells you | Typical observation |
|---|---|---|
| Head CoG lateral excursion (Y vs. time) | Whether the head reaches the far-side interior or the far-side occupant | Excursion onset and peak timing; compare against console inner face position |
| Shoulder belt force and slip time | When the belt loses control of the torso | Force drop or plateau coincident with shoulder slip |
| Thorax deflection (FE, mm) | Local injury-relevant loading under the belt | Peak deflection and location; compare to AIS-relevant thresholds |
| Head contact force / HIC path | Severity of head-to-console or head-to-occupant contact | Contact time, peak force, and contact location |
Read these together rather than in isolation. A far-side countermeasure that reduces head excursion but increases belt force on the thorax may simply trade one injury mode for another. The useful engineering question is whether the countermeasure — an airbag, a modified belt geometry, a console energy absorber — changes the timing of shoulder slip relative to head contact. If shoulder slip occurs before the head reaches the console, the belt is no longer contributing to restraint at the moment of contact, and the countermeasure must address the contact path directly.
Practical setup checklist
- Confirm the coordinate convention and the far-side direction before building anything; document it in the model header.
- Settle the occupant under gravity for 200–300 ms and verify no residual penetration at the seat interface.
- Check belt fit at the clavicle and ASIS; a high lap belt will invalidate the far-side result.
- Match contact stiffness across the FE/MB coupling interface; run a short static penetration check.
- Validate the HBM against certification corridors, then validate the far-side trajectory against sled data if available.
- Extract head Y-excursion, belt force, thorax deflection, and head contact force as a single time-aligned set.
With the model built and validated in this way, the far-side simulation becomes a reliable instrument for comparing countermeasure concepts rather than a source of plausible-looking but unverified animation.