Predicting occupant head displacements in evasive maneuvers; tuning and comparison of a rotational based and a translational based neck muscle controller

Front Bioeng Biotechnol. 2024 Jan 12:11:1313543. doi: 10.3389/fbioe.2023.1313543. eCollection 2023.

Abstract

Objective: Real-life car crashes are often preceded by an evasive maneuver, which can alter the occupant posture and muscle state. To simulate the occupant response in such maneuvers, human body models (HBMs) with active muscles have been developed. The aim of this study was to implement an omni-directional rotational head-neck muscle controller in the SAFER HBM and compare the bio-fidelity of the HBM with a rotational controller to the HBM with a translational controller, in simulations of evasive maneuvers. Methods: The rotational controller was developed using an axis-angle representation of head rotations, with x, y, and z components in the axis. Muscle load sharing was based on rotational direction in the simulation and muscle activity recorded in three volunteer experiments in these directions. The gains of the rotational and translational controller were tuned to minimize differences between translational and rotational head displacements of the HBM and volunteers in braking and lane change maneuvers using multi-objective optimizations. Bio-fidelity of the model with tuned controllers was evaluated objectively using CORrelation and Analysis (CORA). Results: The results indicated comparable performance for both controllers after tuning, with somewhat higher bio-fidelity for rotational kinematics with the translational controller. After tuning, good or excellent bio-fidelity was indicated for both controllers in the loading direction (forward in braking, and lateral in lane change), with CORA scores of 0.86-0.99 and 0.93-0.98 for the rotational and translational controllers, respectively. For rotational displacements, and translational displacements in the other directions, bio-fidelity ranged from poor to excellent, with slightly higher average CORA scores for the HBM with the translational controller in both braking and lane changing. Time-averaged muscle activity was within one standard deviation of time-averaged muscle activity from volunteers. Conclusion: Overall, the results show that when tuned, both the translational and rotational controllers can be used to predict the occupant response to an evasive maneuver, allowing for the inclusion of evasive maneuvers prior to a crash in evaluation of vehicle safety. The rotational controller shows potential in controlling omni-directional head displacements, but the translational controller outperformed the rotational controller. Thus, for now, the recommendation is to use the translational controller with tuned gains.

Keywords: SAFER HBM; active human body model; controller tuning; omni-directional control; pre-crash.

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. The work was conducted at SAFER—Vehicle and Traffic Safety Centre at Chalmers, Gothenburg, Sweden and carried out by Chalmers University of Technology, Autoliv Research, Dynamore Nordic and Volvo Cars in the project Active human body models for virtual occupant response, step 5. The work was funded by FFI (Strategic Vehicle Research and Innovation), by VINNOVA, the Swedish Transport Administration, the Swedish Energy Agency and the industrial partners. Simulations were performed on resources at Chalmers Centre for Computational Science and Engineering (C3SE) provided by the Swedish National Infrastructure for Computing (SNIC), partially funded by the Swedish Research Council through grant agreement no. 2018-05973. PAF received funding through the Dutch Research Council (NWO) Talent Program Vidi (VI.Vidi.203.066).