Efficient Motion Sickness Assessment: Recreation of On-Road Driving on a Compact Test Track

2024-12-19

arXiv.org

AI Summary

Public-road experiments for assessing motion sickness in autonomous vehicle passengers suffer from poor reproducibility, while dedicated test tracks are constrained by prototype vehicle availability and limited spatial scale. Method: This study proposes a novel method to faithfully reconstruct real-road acceleration profiles within a compact, 70 m × 175 m enclosed test track. It innovatively integrates model predictive control (MPC)-based trajectory reconstruction with a within-subject crossover experimental design—enabling, for the first time, accurate replication of open-road motion-sickness–inducing dynamics at reduced scale. Contribution/Results: In a study with 47 participants, no significant difference in motion sickness incidence was observed between the test track and real-road conditions (p > 0.05), and individual motion sickness susceptibility showed high consistency (Pearson’s r = 0.82). These results validate the method’s breakthrough contributions to safety, experimental reproducibility, and ecological validity in autonomous vehicle motion sickness assessment.

Technology Category

• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters

Application Category

• Security and Privacy: Large-scale security measurements
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

The ability to engage in other activities during the ride is considered by consumers as one of the key reasons for the adoption of automated vehicles. However, engagement in non-driving activities will provoke occupants'motion sickness, deteriorating their overall comfort and thereby risking acceptance of automated driving. Therefore, it is critical to extend our understanding of motion sickness and unravel the modulating factors that affect it through experiments with participants. Currently, most experiments are conducted on public roads (realistic but not reproducible) or test tracks (feasible with prototype automated vehicles). This research study develops a method to design an optimal path and speed reference to efficiently replicate on-road motion sickness exposure on a small test track. The method uses model predictive control to replicate the longitudinal and lateral accelerations collected from on-road drives on a test track of 70 m by 175 m. A within-subject experiment (47 participants) was conducted comparing the occupants'motion sickness occurrence in test-track and on-road conditions, with the conditions being cross-randomized. The results illustrate no difference and no effect of the condition on the occurrence of the average motion sickness across the participants. Meanwhile, there is an overall correspondence of individual sickness levels between on-road and test-track. This paves the path for the employment of our method for a simpler, safer and more replicable assessment of motion sickness.

Problem

• Replicating on-road motion sickness exposure on compact test tracks
• Assessing motion sickness factors for automated vehicle acceptance
• Developing reproducible testing methods using model predictive control

Innovation

• Model predictive control replicates accelerations
• Compact test track enables reproducible experiments
• Within-subject validation shows equivalent sickness levels