Sudden traffic hazards trigger collision-avoidance behaviors in drivers that can significantly impact vehicle dynamics, potentially conflicting with the existing advanced driver assistance systems (ADAS), such as autonomous emergency braking and steering. This behavior can lead to unexpected vehicle movements, further complicating the situation and elevating the risk of accidents. Understanding and tailoring the inference of drivers’ perception-response time (PRT) is essential for optimizing ADAS activation in intelligent vehicles. This approach allows customization for individual drivers, improving safety and ensuring that interventions are personalized and minimally disruptive to normal driving patterns. To achieve this objective, this study performs high-fidelity simulation experiments to gather a comprehensive multidimensional dataset on drivers’ responses in safety-critical scenarios, primarily focusing on PRT and its influencing factors. Using the collision-avoidance behavior data, a driver evidence accumulation model is created to explain PRT distribution and facilitate real-time personalized inferences. We also analyze the relationship between model parameters and real-world physical significance, demonstrating that driver decisions rely on visual evidence accumulation influenced by dynamic interactions in different scenarios. Our proposed model, by offering a detailed understanding of drivers’ perceptual and decision-making processes, aids in developing personalized driver assistance system activation recommendations. This approach seeks to create personalized and adaptive systems within intelligent vehicles, thereby reducing human-machine conflicts and improving the overall safety of intelligent transportation systems.

Pedestrian kinematics is one of the important influence factors in pedestrian-vehicle collision accidents. The degree of pedestrian injury after collision is related to the initial pre-impact posture. The current research on pedestrian injury is based on the simulation collision of numerical pedestrian model, whose pre-impact posture comes from fixed gait sequence or manual construction. However, the former cannot represent the pedestrian posture of real accidents, and the latter takes a lot of time. This paper proposes an end-to-end framework to reconstruct pedestrian pre-impact posture from real-world accident images or videos for collision damage assessment. In order to obtain the most realistic pedestrian pre-impact posture, we first construct a video dataset of pedestrian-vehicle collision accidents. Then we use a human body 3D pose reconstruction method based on deep learning, SPIN (SMPL oPtimization IN the loop), which extracts the pose and body shape parameters of pedestrians in the image to generate SMPL (Skinned Multi-Person Linear) model. We exploit its similarity to the multi-body model to reconstruct the pre-impact pose of pedestrians. Experimental results show that our method can shorten the reconstruction time from minutes (327.25s) to milliseconds (0.260s), and the average error of joint angle is less than 3%. Through the computational simulation collision test, the collision damage results of the dummy constructed by our method is consistent with that of the real posture and the manually constructed posture. The proposed method is simple and easy to implement, which is helpful to enhance the performance of active and passive protection for pedestrians.

Trajectory prediction is significant for intelligent vehicles to achieve high-level autonomous driving, and a lot of relevant research achievements have been made recently. Despite the rapid development, most existing studies solely focus on normal and safe scenarios while largely neglecting safety-critical scenarios, particularly those involving imminent collisions. This oversight may result in autonomous vehicles lacking the essential predictive ability in such situations, posing a significant threat to safety. To tackle these, this paper proposes a risk-aware trajectory prediction framework tailored to safety-critical scenarios. Leveraging distinctive hazardous features, we develop three core risk-aware components. First, we introduce a risk-incorporated scene encoder, which augments conventional encoders with quantitative risk information to achieve risk-aware encoding of hazardous scene contexts. Next, we incorporate endpoint-risk-combined intention queries as prediction priors in the decoder to ensure that the predicted multimodal trajectories cover both various spatial intentions and risk levels. Lastly, an auxiliary risk prediction task is implemented for the ultimate risk-aware prediction. Furthermore, to support model training and performance evaluation, we introduce a safety-critical trajectory prediction dataset and tailored evaluation metrics. We conduct comprehensive evaluations and compare our model with several SOTA models. Results demonstrate the superior performance of our model, with a significant improvement in most metrics. This prediction advancement enables autonomous vehicles to execute correct collision avoidance maneuvers under safety-critical scenarios, eventually enhancing road traffic safety.

Abstract Scene text in natural images carries additional semantic information to aid in image classification. Existing methods lack full consideration of the deep understanding of the text and the visual text relationship, which results in the difficult to judge the semantic accuracy and the relevance of the visual text. This paper proposes image classification based on Cross modal Knowledge Learning of Scene Text (CKLST) method. CKLST consists of three stages: cross‐modal scene text recognition, text semantic enhancement, and visual‐text feature alignment. In the first stage, multi‐attention is used to extract features layer by layer, and a self‐mask‐based iterative correction strategy is utilized to improve the scene text recognition accuracy. In the second stage, knowledge features are extracted using external knowledge and are fused with text features to enhance text semantic information. In the third stage, CKLST realizes visual‐text feature alignment across attention mechanisms with a similarity matrix, thus the correlation between images and text can be captured to improve the accuracy of the image classification tasks. On Con‐Text dataset, Crowd Activity dataset, Drink Bottle dataset, and Synth Text dataset, CKLST can perform significantly better than other baselines on fine‐grained image classification, with improvements of 3.54%, 5.37%, 3.28%, and 2.81% over the best baseline in mAP, respectively.

There will be a time when automated vehicles coexist with human-driven ones. Understanding how drivers assess driving risks and modeling their differences is crucial for developing human-like and personalized behaviors in automated vehicles, gaining people’s trust and acceptance. However, existing driving risk models are usually developed at a statistical level, and no single model can accurately describe and explain the variations in risk perception among drivers. We propose a concise yet effective model known as the Potential Damage Risk (PODAR) model, which provides a universal and physically meaningful structure for estimating driving risk and explaining the reasons for differences in risk perception. Leveraging an open-access dataset collected from an obstacle avoidance experiment, this paper establishes individual risk perception models for drivers with high fitness performances. We conclude that the variations in risk perception among drivers stem from their assessments of potential damage, accounting for the uncertainty in both temporal and spatial dimensions. Our findings offer an explanation for human risk perceptions and present a promising risk model for autonomous vehicles to develop human-like behaviors and personalized services.

Driver attention cues contribute to the following intended maneuver prediction and provide a risk indicator for the advanced driver-assistance systems in complex driving scenarios. The diverse traffic scenes result in a challenging task to predict human visual attention with high generalization capability. The data heterogeneity caused by multiple sources with different data characteristics, such as video sources, was investigated in the current research and mitigated using the domain adaption modules (i.e., domain-specific batch normalization, Gaussian priors, and smoothing filter) and the domain-specific focal loss. Inspired by human attention mechanism, generic coders and task-driven attention modules were incorporated into a lightweight network to replicate human-like perceptual patterns, such as the perception of latent risk. Integrating these adaptive modules, we proposed the adaptive driver attention (ADA) model to predict salient regions in different traffic scenes. Consequently, the ADA model trained jointly on four driver attention datasets achieves the best performance against the state-of-the-art methods across seven metrics. Retrospective visualizations of the network and cross-validation results further explain the merit of the proposed method. Since these approaches are generic, adaptive modules are universally applicable for standard deep neural network architectures to alleviate the heterogeneity across datasets and easily extended for arbitrary traffic scenes in the real world.

Driver’s under-arousal occurred in automated driving systems (ADS) impairs takeover safety. This study aims to determine electrodermal activity (EDA) features’ importance for driver’s arousal quantification. A car-following simulator study was conducted with participants concurrently executing four levels of cognitive tasks, triggering four levels of arousal. Participants’ skin conductance (SC) data were collected and decomposed into tonic (skin conductance level, SCL) and phasic (skin conductance response, SCR) components. Seventeen features extracted from SC, SCL and SCR were compared. As a result, SCR-relevant features showed higher significance and larger effect size than SC and SCL features in response to cognitive load, which suggests the phasic component dominates changes in EDA under varying cognitive load. Moreover, the SCR rate TTP.nSCRs, identified by$0.03 ~\mu \text{S}$thresholds, attained the largest effect size among all features for driver’s arousal measurement. A varying time windows (TW) analysis showed that TTP.nSCRs was the most suggested arousal metric when TW was over 20 s, whereas the sum of SCRs amplitudes TTP.AmpSum was preferred when TW was less than 20 s. For driver’s arousal quantification with multi-features, the top five suggested features were TTP.nSCRs, SC_Rate5, CDA.SCR (or CDA.ISCR), CDA.AmpSum, and TTP.AmpSum. Although male drivers showed higher values of EDA features than female drivers, the sensitivity of the proposed EDA features stands across gender and individuals. This study promotes an improved understanding of EDA changes in human cognitive process. The sensitive EDA features proposed could be used from uni- or multi-modalities in driver state management and takeover-safety prediction for ADS.