This study tested and further optimized an application-oriented parameter set for the 6DOF-SVC (six degrees-of-freedom subjective vertical conflict) model to estimate motion sickness incidence (MSI) in passengers engaged in a non-driving task with a lowered gaze. The authors had previously proposed these parameters; here they validated them using data from five real-world driving experiments and adjusted parameters to improve generalizability across those datasets.
The work focuses on translating established sensory-conflict modeling into an application-relevant predictor of passenger MSI that could inform vehicle behavior or route selection. The authors report that, after parameter optimization, the model can produce usable estimates of MSI across different test tracks, while also identifying remaining inaccuracies in predicted MSI trajectories.
Validation and optimization were performed using five previously conducted motion-sickness driving experiments. The manuscript reports that these datasets represent real-world driving conditions and different driving dynamic profiles, route characteristics, and driving styles. Specific experimental details (sample sizes, exact track layouts, vehicle types, or individual trial protocols) were not reported in the abstract and therefore are not reproduced here.
The approach aimed to examine whether a single, application-oriented parameter set for the 6DOF-SVC model can generalize across heterogeneous real-world driving data and whether model outputs align with experimentally measured MSI.
A principal finding reported is that the expected natural fluctuations in MSI—derived from the driving dynamic profile, route characteristics, or driving style—are approximately 10%. The authors note that this magnitude of fluctuation is similar to the variation seen in experimentally determined MSI across the studies.
This ~10% variability is described as an inherent consequence of varying driving dynamics and route features, implying a baseline level of fluctuation that predictive models must accommodate when estimating passenger MSI in operational contexts.
The authors observed that participants seem to possess an individual MSI threshold that correlates with their susceptibility to motion sickness. In other words, some individuals require higher cumulative or peak stimulus to reach the same MSI outcome compared with others.
However, the manuscript also reports that this individual threshold fluctuates substantially across experiments and thus lacks the stability required for reliable practical application (for example, as a personal calibration parameter for predictive systems). The abstract does not provide numerical metrics for inter- or intra-individual variability beyond the qualitative statement that fluctuation was strong.
Overall, the optimized 6DOF-SVC model was capable of estimating motion sickness risk across the different test tracks included in the five studies. Despite successful estimation at a risk level, the authors emphasize that inaccuracies remain in the predicted MSI trajectories. These inaccuracies suggest that while average or aggregate risk predictions may be acceptable, temporally resolved predictions (trajectory shapes or moment-to-moment MSI changes) are less precise.
The abstract does not enumerate specific sources of residual error (for example, sensor noise, unmodeled passenger behavior, or adaptive physiological responses) nor quantify the magnitude of trajectory mismatch beyond the reported natural fluctuation.
The work frames reliable MSI prediction as a potential enabler for intelligent-vehicle interventions intended to reduce motion-sickness–related drive abortion. Possible applications include adaptive driving strategies or route planning that minimize predicted MSI exposure for susceptible passengers. The authors position the 6DOF-SVC model, with the optimized parameter set, as a candidate tool for estimating passenger motion-sickness risk in such systems.
At the same time, the reported ~10% natural fluctuation and the instability of individual MSI thresholds indicate that further development is needed before the model can be deployed for individualized, operational decision making.
This manuscript is reported as a preprint and has not been certified by peer review. The abstract provides limited methodological detail; specific experiment-level parameters, sample sizes, model-fitting procedures, and quantitative performance metrics beyond the stated fluctuation were not reported in the abstract and therefore cannot be expanded upon here.
The underlying research projects received funding from the German Federal Ministry of Education and Research (BMBF) with additional project funding from ZF Friedrichshafen AG. The authors declare that the funding sources had no involvement in study design, data analysis, interpretation of results, manuscript writing, or the decision to submit the article for publication.
In summary, the optimized 6DOF-SVC model estimated MSI across five real-world driving studies and revealed an intrinsic approximate 10% fluctuation in predicted and observed MSI attributable to driving dynamics and route variability. Individual susceptibility appears to relate to an MSI threshold, but that threshold's instability limits immediate practical use for individualized prediction. Remaining inaccuracies in predicted MSI trajectories highlight the need for further refinement before operational deployment in intelligent-vehicle systems.