Occupant Safety in Vehicles Equipped with Automated Driving Systems, Part 3: Biofidelity Evaluation of GHBMC M50-OS Against Laboratory Sled Tests

Poulard, David; Lin, Hongnan; Panzer, Matthew B. · 2020 · ROSA P / United States. Department of Transportation. National Highway Traffic Safety Administration

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This report evaluates the biofidelity of computational human body models against experimental data to assess occupant safety in vehicles equipped with Automated Driving Systems (ADS). As ADS allows for occupant postures unconstrained by active driving, traditional safety evaluation tools based on standard seating positions may be insufficient. The study specifically validates the Global Human Body Models Consortium (GHBMC) male 50th percentile occupant (M50-O), the simplified M50-OS, and the Test Device for Human Occupant Restraint – Finite Element (THOR FE) models. These simulations were compared against the Gold Standard 3 (GS3) laboratory sled tests, which replicate a 30 km/h, 30° near-side oblique frontal impact using a 3 kN force-limited shoulder belt. Experimental data included three postmortem human subjects (PMHS) and four THOR anthropomorphic test dummy tests, instrumented with high-speed 3D motion capture and standard sensors. The researchers adapted existing finite element models to replicate the GS3 environment, including specific seat modifications and restraint anchor locations. Occupant positioning in the simulations was matched to averaged experimental parameters using computational frameworks and the LS-Prepost seat deformer tool. Belt modeling utilized a mixed technique of 1D and 2D elements to ensure proper interaction, with material properties and force-limiting mechanisms derived from experimental specifications. Model responses were quantitatively assessed using the CORA software, which evaluates correlation based on corridor, phase, magnitude, and slope metrics. Injury assessment for the M50-O model employed deterministic methods for clavicle, sternum, and cervical fractures, and probabilistic strain-based functions for rib fractures. Results indicated that the GHBMC M50-OS and M50-O models demonstrated stable simulations, with the M50-O requiring partial remeshing and timestep reduction. The M50-O model predicted a right clavicle fracture consistent with one PMHS case but failed to predict sternum or cervical fractures observed in other tests. Rib fracture risk was predicted at 85% for at least one fracture, aligning with the 66% incidence in PMHS tests, though the model did not predict the high number of fractures (≥6) seen in two subjects. The THOR FE model exhibited instability in the shoulder pads, resolved by increasing stiffness, and showed significant discrepancies in head acceleration due to artificial head-clavicle impacts not present in experiments. CORA ratings generally indicated fair correlation for most kinematic and force signals, though head acceleration and certain displacement metrics showed lower fidelity. The study concludes that while the GHBMC M50-O and M50-OS models provide reasonable biofidelity for kinematic and restraint load responses in oblique frontal impacts, injury prediction capabilities require refinement, particularly for rib fracture severity. The THOR FE model showed limitations in replicating specific kinematic interactions, such as head motion. These findings support the continued development and validation of computational models for assessing occupant safety in ADS scenarios, where non-standard postures necessitate robust simulation tools that accurately reflect human biomechanical responses.

Key finding

The GHBMC M50-O finite element model demonstrated the highest biofidelity with a CORA model score of 0.685, outperforming the simplified M50-OS (0.616) and the THOR FE (0.520) in replicating PMHS sled test responses.

Methodology

modeling

Sample size: 7

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• adas effectiveness

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Applied Guidance: standards test procedures
• Methodological Resource: validation psychometrics