To alleviate decision-making pressure on rail transit dispatchers in the emergency handling process, this work sorts out the scenario elements of rail transit emergency cases, establishes a scenario element system, and uses the information weight method to determine the weight of each scenario element. Based on the information of the key decision points, the complete process of emergencies is divided into various scenarios, and an emergency case representation model is constructed. The model establishes a database of historical emergency cases in rail transit, utilizes the scenarios as the search object to match the similarity of emergencies, and provides the decision-making support information to handle the current emergencies. Furthermore, the model constructed in this paper is subjected to an actual emergency case for analysis and calculation, which verify the validity and feasibility of the proposed model.

Semi-automated urban rail transit systems still rely on human intervention during safety-critical events, yet emergency takeover performance has received far less attention than in SAE Level-3 road automation. This study focuses on the reaction phase of emergency takeover, defined as the interval from anomaly onset to the train operator’s first control action. We propose a conditional two-stage evaluation framework that jointly assesses event recognition and control execution quality. A simulation-based experiment was conducted to replicate GoA2 operating conditions under controlled emergency scenarios. Three indicators were extracted: (i) event recognition accuracy derived from eye-tracking and retrospective recall, (ii) takeover reaction time, and (iii) initial action accuracy reflecting compliance with operational speed or braking limits. An attention-enhanced multilayer perceptron (MLP) was developed to dynamically weight input features and improve interpretability. The proposed model achieved stable subject-wise performance, with an average accuracy of 0.86 and a macro F1-score of 0.857. These results support the feasibility of interpretable learning-based evaluation for human-in-the-loop safety assessment and provide practical implications for improving operator readiness monitoring and operational safety management in semi-automated metro systems.

To investigate the regulatory effects of emotional states on the evolution of fatigue in metro drivers, this study conducts an experimental investigation based on an urban rail transit driving simulation platform. A total of 21 participants complete a 90 min simulated driving task, during which electroencephalogram (EEG) and electrocardiogram (ECG) signals are synchronously collected from drivers for fatigue assessment and emotion recognition, respectively. An emotion recognition model based on a multi-scale convolutional neural network (MSCNN) combined with an attention mechanism is constructed. The proposed model uses ECG signals to classify three emotional states—neutral, positive, and negative—where the neutral state is defined as an emotionally undefined baseline that is neither positive nor negative. The model achieves a classification accuracy of 86.96% on the DREAMER dataset. By temporally aligning the emotion recognition results with EEG frequency-domain fatigue indicators, the results show that fatigue exhibits the highest growth and largest fluctuation in amplitude under negative emotions, demonstrating a pronounced fatigue-accelerating effect. Under positive emotions, fatigue decreases considerably and has smaller fluctuations, indicating a certain buffering and restorative effect. In contrast, the neutral emotional state exhibits intermediate and transitional fatigue characteristics. This study innovatively integrates ECG-based emotion recognition with EEG-based fatigue assessment to reveal the mechanisms based on which emotions influence fatigue in metro driving tasks from a physiological perspective. This work provides a basis for emotion-aware fatigue monitoring and safety intervention strategies.

In the field of intelligent inspection, high-definition video data collected by quadruped robot dogs face severe transmission and storage constraints. Although existing advanced lossy video coding standards can significantly improve compression efficiency, they inevitably introduce severe compression artifacts in low-bit-rate scenarios. To address this issue, this paper proposes a video decoding quality enhancement network named Video Quality Restoration Network (VQRNet), based on a dual-stream architecture. Specifically, the Local Feature Extraction component incorporates a Progressive Feature Fusion Module (PFFM) with a four-stage progressive structure. By integrating reparameterized convolution and attention mechanisms, PFFM focuses on capturing high-frequency texture details to repair small-scale distortions. Simultaneously, the Multi-Scale Lightweight Spatial Attention Module (MLSA) performs spatial feature recalibration, leveraging multi-scale convolution to adaptively identify and enhance key spatial regions, specifically addressing multi-scale distortion. In the Global Feature Extraction component, the State-Space Attention Module (SSAM) combines State-Space Models (SSMs) with attention mechanisms to capture long-range dependencies and contextual information, for large-scale distortions caused by high-intensity compression. To verify the performance of the proposed algorithm, a dedicated dataset comprising 20 real-world video sequences captured by quadruped robot dogs (partitioned into 15 training and 5 testing sequences) was constructed, and the VTM 23.4 reference software was employed to simulate compression degradation using four quantization parameters (QP 30, 35, 40, and 45). Experimental results demonstrate that VQRNet outperforms state-of-the-art quality enhancement methods in terms of core metrics, including PSNR and SSIM, specifically including MIRNet, NAFNet, TRRHA, and CTNet. In the QP = 30 scenario, VQRNet achieves an average PSNR of 40.33 dB, a significant improvement of 3.32 dB over the VTM 23.4 baseline (37.01 dB), while demonstrating significant advantages in computational complexity and parameter efficiency—requiring only 5.27 G FLOPs and 1.40 M parameters, with an average inference latency of only 11.82 ms per 128 × 128 patch. This work provides robust technical support for the efficient video perception of quadruped robot dogs.