To identify and quantify the factors that influence the risky riding behaviors of electric bike riders, we designed an e-bike rider behavior questionnaire (ERBQ) and obtained 573 valid samples through tracking surveys and random surveys. An exploratory factor analysis was then conducted to extract four scales: riding confidence, safety attitude, risk perception, and risky riding behavior. Based on the exploratory factor analysis, a structural equation model (SEM) of electric bike riding behaviors was constructed to explore the intrinsic causal relationships among the variables that affect the risky e-bike riding behavior. The results show that the relationship between riding confidence and risky riding behavior is mediated by risk perception and safety attitudes. Safety attitude was found to be significantly associated with risky riding behaviors. Specifically, herd mentality is most closely related to safety attitudes, which means that those engaged in e-bike traffic management and safety education should pay special attention to riders’ psychological management and education. Risk perception has a direct path to risky riding behaviors. Specifically, stochastic evaluation and concern degree are significantly related to e-bike riders’ risk perception. The findings of this study provide an empirical basis for the creation of safety interventions for e-bike riders in China.

The effective assessment of drivers’ driving capability under the condition of an advanced driver assistance system is of great significance to the precise switching of driving rights between human and machine and the promotion of the development of man-computer collaboration. In this study, real-time collected warning data from bus driver state monitoring system (DSMS) and advanced driver assistance system (ADAS) were utilized to determine the drivers’ comprehensive driving capability indicators. The information utility and interaction of the indicators were considered, and an integrated weight method based on standard deviations was proposed. This method was used to combine the entropy weight method and improved analytic network process (ANP), to evaluate the drivers’ comprehensive driving capability under man-computer cooperative driving conditions in real time. The results show that the entropy weight method and improved ANP algorithm have good consistency and are significantly correlated and that the integrated weight method is effective and dependable. The top four indicators in the integrated weighting results were eye closure (0.241), yawn (0.210), rapid deceleration (0.186), and lane departure (0.159). Drivers' comprehensive driving capability scores were concentrated in the score range of 1 to 6, with the lowest scores in zones A and B for stages 2, 11 and 21. Therefore, it is necessary to further explore the relationship between driver behavior, vehicle status and road traffic environment within the score range of 1 to 6 so that the man-computer interaction can be optimized and the driver's comprehensive driving capability can be improved.

Multi-sensor fault diagnosis, especially when using heterogeneous sensors, substantially enhances the accuracy of fault detection in asynchronous motors operating under high-interference conditions. A critical challenge in multi-sensor fault diagnosis lies in effectively fusing data from different sensors. Deep learning offers a promising solution by transforming multi-sensor data into a unified representation, thereby facilitating robust data fusion. However, existing approaches often fail to fully exploit inter-sensor correlations and inherent prior physical knowledge. To address this limitation, we propose a novel graph neural network-based model that emphasizes graph structure construction for heterogeneous multi-sensor information fusion. Our framework includes (1) a multi-task enhanced autoencoder for node feature extraction, enabling discriminative representation learning, particularly with heterogeneous sensor data; (2) an adjacency matrix builder integrated with physical prior constraints to improve the generalization and robustness of the model; and (3) a graph isomorphism network to derive graph-level representations for fault classification. Our experimental results demonstrate the model’s effectiveness in diagnosing faults, as it achieves superior performance compared to conventional methods on two heterogeneous asynchronous motor datasets.

This study evaluates the risk level of human-machine collaborative driving takeover in a highway environment under the interaction of non-driving related tasks and takeover request prompt scenarios. Using a driving simulator, a 5x5 factor analysis examined non-driving tasks and takeover prompts. Takeover impact and risk indicator features were extracted, with risk indicators weighted using an improved ANP method. On this basis, the XGBoost algorithm was employed to identify crucial variables that reflect the level of takeover risk and to construct a driver takeover risk assessment model. The evaluation results indicate that the risk indicator feature with the greatest influence on the takeover risk level was the minimum TTC, which had the highest correlation (r=-0.81); the scene factor in the takeover influence feature had the highest correlation with the takeover risk level (r=-0.78), which had the greatest influence on driver takeover safety; Although non-driving related tasks exhibited a weak correlation with takeover reaction time, steering reaction time, and minimum TTC, the effect was minor. The XGBoost algorithm-based risk assessment model demonstrated superior performance over LightGBM and SVM, with 87.1% accuracy. Overall, this study highlights the significant influence of takeover scenes and minimum TTC on collaborative driving risk, enabling accurate risk modeling.

With the development of smart cities and smart transportation, cities can gradually provide people with more information to facilitate their life and travel, and parking is also inseparable from both of them. Accurate on-street parking demand prediction can improve parking resource utilization and parking management efficiency, as well as potentially improve urban traffic conditions. Previous parking demand prediction methods seldom consider the correlation between the parking demand of a road section and its surroundings. Therefore, in order to capture the correlation of parking demand in the temporal and spatial dimensions as carefully as possible and enrich the relevant features in the prediction model so as to achieve more accurate prediction results, we designed a parking demand prediction structure that considers different features from two perspectives: overall and internal. We used gated recurrent units (GRU) to extract demand influences in the temporal dimension. The GRU is used in combination with a graph convolutional neural network (GCN) to extract demand influencing factors in the spatial dimension. Additionally, a more detailed representation is designed to express spatial dimensional features. Then, based on the historical parking demand features extracted using encoder–decoder, we fuse the extracted spatio-temporal features with them to finally obtain an on-street parking demand prediction model combining the overall and the internal information. By combining them, we can integrate more correlation factors to achieve a more accurate parking demand prediction. The performance of the model is evaluated by real parking data in Xiufeng District of Guilin. The results show that the proposed model achieves good prediction performance compared with other baselines. In addition, we also design feature ablation experiments. Through the comparison of the results, we find that each feature considered in the proposed model is important in parking demand prediction.

In order to improve the adaptation of driver to the advanced driver assistance system (ADAS) and optimize the active safety control technology of vehicle under man-computer cooperative driving, this paper investigated the correlation between driver’s improper driving behaviors and abnormal vehicle states under the ADAS. Based on the warning data collected from the driver’s assistance warning system equipped on buses, the interaction between improper behaviors, between abnormal vehicle states, and between improper behaviors and abnormal vehicle states were quantitatively analyzed through the hierarchical clustering method and improved Apriori algorithm. The results showed that eye closure and yawn were high in concurrency (probability: 0.888) and interaction (average probability: 0.946); the interaction among lane departure, rapid acceleration, and rapid deceleration are frequent (average probability: 0.7224); eye closure (average probability: 0.452) and yawn (average probability: 0.444) are likely to induce abnormal vehicle states such as rapid acceleration and rapid deceleration. Some suggestions proposed based on the results are as follows. First, it is suggested that the ADAS should combine the warning modes of eye closure and yawn; second, when the driver closes eyes or yawns, the control of the ADAS over the lateral and longitudinal performance of vehicle should be enhanced; third, the extent of control by the ADAS should be determined according to the relationship probability; finally, the lateral control over the vehicle by the ADAS should be strengthened when there is a forward collision warning.