Millimeter-wave (MMW) radar is essential in roadside traffic perception scenarios and traffic safety control. For traffic risk assessment and early warning systems, MMW radar provides real-time position and velocity measurements as a crucial source of dynamic risk information. However, due to MMW radar’s measuring principle and hardware limitations, vehicle positioning errors are unavoidable, potentially causing misperception of the vehicle motion and interaction behavior. This paper analyzes the factors influencing the MMW radar positioning accuracy that are of major concern in the application of transportation systems. An analysis of the radar measuring principle and the distributions of the radar point cloud on the vehicle body under different scenarios are provided to determine the causes of the positioning error. Qualitative analyses of the radar positioning accuracy regarding radar installation height, radar sampling frequency, vehicle location, posture, and size are performed. The analyses are verified through simulated experiments. Based on the results, a general guideline for radar data processing in traffic risk assessment and early warning systems is proposed.

In this paper, a gated recurrent unit–deep neural network (GRU-DNN) model integrated with multimodal feature embedding (MFE) is developed to evaluate the real-time risk of hazmat road transportation based on various types of data for contributing factors. MFE was incorporated into the framework of a deep learning model in which discrete variables, continuous variables, and images were uniformly embedded. GRU is a pre-trained sub-model, and the DNN is able to directly use the relative structure and weights of the GRU, improving the poor classification and recognition results due to insufficient samples. Additionally, the model is trained and validated based on hazmat road transportation database consisting of 2100 samples with 20 real-time contributing factors and four risk levels in China. The accuracy (ACC), precision (PR), recall (RE), F1-score (F1), and areas under receiver-operating-characteristic curves (AUC) of the proposed model and other commonly used models are compared as performance measurements in numerical examples. Finally, Carlini & Wagner attack and three defenses of adversarial training, dimensionality reduction and prediction similarity are proposed in the training to improve the robustness of the model, alleviating the impact of noise and error on small-sized samples. The results demonstrate that the average ACC of the model reaches 93.51% and 87.6% on the training and validation sets, respectively. The prediction of accidents resulting in injury is the most accurate, followed by fatal accidents. Combined with the RE of 89.0%, the model exhibits excellent performance. In addition, the proposed model outperforms other widely used models based on the overall comparisons of ACC, AUC, F1 and PR-RE curve. Finally, prediction similarity can be used as an effective approach for robustness improvement, with the launched adversarial attacks being detected at a high success rate.

Due to the harsh environment of highway tunnels and frequent breakdowns of various detection sensors and surveillance devices, the operational management of highway tunnels lacks effective data support. This paper analyzes the characteristics of operational surveillance data in highway tunnels. It proposes a multimodal information fusion method based on CNN–LSTM–attention and designs and develops a digital twin for highway tunnel operations. The system addresses issues such as insufficient development and coordination of the technical architecture of operation control systems, weak information service capabilities, and insufficient data application capabilities. The system also lacks intelligent decision-making and control capabilities. The developed system achieves closed-loop management of “accurate perception–risk assessment–decision warning–emergency management” for highway tunnel operations based on data-driven approaches. The engineering demonstration application underscores the system’s capacity to enhance tunnel traffic safety, diminish tunnel management costs, and elevate tunnel driving comfort.