A data-driven, kinematic feature-based, near real-time algorithm for injury severity prediction of vehicle occupants.

Accurate real-time prediction of occupant injury severity in unavoidable collision scenarios is a prerequisite for enhancing road traffic safety with the development of highly automated vehicles. Specifically, a safety prediction model provides a decision reference for the trajectory planning system in the pre-crash phase and the adaptive restraint system in the in-crash phase. The main goal of the current study is to construct a data-driven, vehicle kinematic feature-based model to realize accurate and near real-time prediction of in-vehicle occupant injury severity. A large-scale numerical database was established focusing on occupant kinetics. A first-step deep-learning model was established to predict occupant kinetics and injury severity using a convolutional neural network (CNN). To reduce the computational time for real-time application, the second step was to extract simplified kinematic features from vehicle crash pulses via a feature extraction method, which was inspired by a visualization approach applied to the CNN-based model. The features were incorporated with a low-complexity machine-learning algorithm and achieved satisfactory accuracy (85.4 % on the numerical database, 78.7 % on a 192-case real-world dataset) and decreased computational time (1.2 ± 0.4 ms) on the prediction tasks. This study demonstrated the feasibility of using data-driven and feature-based approaches to achieve accurate injury risk estimation prior to collision. The proposed model is expected to provide a decision reference for integrated safety systems in the next generation of automated vehicles.

Safety envelope of pedestrians upon motor vehicle conflicts identified via active avoidance behaviour.

Human reaction plays a key role in improved protection upon emergent traffic situations with motor vehicles. Understanding the underlying behaviour mechanisms can combine active sensing system on feature caption and passive devices on injury mitigation for automated vehicles. The study aims to identify the distance-based safety boundary ("safety envelope") of vehicle-pedestrian conflicts via pedestrian active avoidance behaviour recorded in well-controlled, immersive virtual reality-based emergent traffic scenarios. Via physiological signal measurement and kinematics reconstruction of the complete sequence, we discovered the general perception-decision-action mechanisms under given external stimulus, and the resultant certain level of natural harm-avoidance action. Using vision as the main information source, 70% pedestrians managed to avoid the collision by adapting walking speeds and directions, consuming overall less "decision" time (0.17-0.24 s vs. 0.41 s) than the collision cases, after that, pedestrians need enough "execution" time (1.52-1.84 s) to take avoidance action. Safety envelopes were generated by combining the simultaneous interactions between the pedestrian and the vehicle. The present investigation on emergent reaction dynamics clears a way for realistic modelling of biomechanical behaviour, and preliminarily demonstrates the feasibility of incorporating in vivo pedestrian behaviour into engineering design which can facilitate improved, interactive on-board devices towards global optimal safety.

Seating preferences in highly automated vehicles and occupant safety awareness: A national survey of Chinese perceptions.

Objective: The potential challenge for providing occupant protection accompanying seating preferences is an essential safety prerequisite for highly automated vehicle (HAV) popularization. This research is aimed toward identifying Asia-specific individualized seating preferences in HAVs and occupant safety awareness via a national survey in China.Methods: An online questionnaire survey was performed to investigate seating preferences (i.e., sitting posture, seating orientation, and position) and occupant safety awareness (i.e., seat belt usage and receptiveness to extended or additional restraints beyond the conventional three-point seat belt). We assessed whether perceptions were modulated by individual characteristics via bivariate and correlation analyses. The possibility of wearing seat belts was estimated by binary logistic regression.Results: The final survey data set includes 1,018 respondents after a rigorous validity check (response rate: 59.2%). The results show that preferred sitting postures and seating orientation were significantly associated with sociodemographic characteristics (e.g., gender, age, city tier) (p < 0.05). The rear seat was preferred in both the conventional (65.6%) and "face-to-face mode" seating configurations (77.6%), largely due to the fact that customers subjectively viewed it as being safer than sitting in a front seat in case of collisions. Despite the current trend of an increasing usage rate of seat belts, 48.5% of respondents preferred to be unrestrained in rear seats, especially for the subgroups who were from less developed cities and with a higher usage rate of public transport (p < 0.01). Low receptiveness to extended restraint and high comfort requirements were confirmed for the young, high-frequency road users, and for those who were from developed areas (p < 0.05).Conclusions: Diversified and specific seating preferences of Chinese occupants were identified facing emerging use of HAVs. Next generation occupant protection systems shall be adapted to account for the individualized expectations and needs on seating designs from certain population groups. Balanced restraint design between safety and comfort was required to exceed the existing strong dependence on exogenous causes of restraint use (e.g., legal restrictions) in Asia.

Influence of population variability in ligament material properties on the mechanical behavior of ankle: a computational investigation.

Biomechanical behavior of ankle ligaments varies among individuals, with the underlying mechanism at multiple scales remaining unquantified. The present probabilistic study investigated how population variability in ligament material properties would influence the joint mechanics. A previously developed finite element ankle model with parametric ligament properties was used. Taking the typical external rotation as example loading scenario, joint stability of the investigated population was consistently shared by specific ligaments within a narrow tolerance range, i.e. 62.8 ± 8.2 Nm under 36.1 ± 5.7° foot rotation. In parallel, the inherent material variability significantly alters the consequent injury patterns. Three most vulnerable ligaments and the consequent rupture sequences were identified, with the structural weak spot and the following progressive stability loss dominated by the relative stiffness among ligaments. This study demonstrated the feasibility of biofidelic models in investigating individual difference at the material level, and emphasized the importance of probabilistic description of individual difference when identifying the injury mechanism of a broad spectrum.