This study proposes an offline reinforcement learning framework based on Critic Regularized Regression (CRR) to optimize speed guidance at signalized intersections under mixed traffic conditions. Using real-world trajectory data combined with signal phase and timing (SPaT) information, the framework learns safe and efficient driving policies without requiring online interactions. A structured data-processing pipeline converts raw vehicle trajectories into Markov Decision Process (MDP) format, effectively encoding vehicle motion states and signal timing information to facilitate realistic decision-making. SUMO simulation experiments demonstrate substantial improvements over rule-based baselines in safety, comfort, and efficiency: time-to-collision increases from 2.75 to 8.53 seconds, jerk is reduced by over 50%, and time headway is consistently maintained at approximately 1.68 seconds. Trajectory visualizations confirm smoother and more adaptive driving behavior. Comparisons with state-of-the-art approaches including Model Predictive Control (MPC), Behavior Cloning (BC), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Batch Constrained Q-learning (BCQ) highlight CRR's stable performance across multiple evaluation metrics. Ablation studies reveal the critical role of different components in ensuring robust policy behavior, while communication loss simulations demonstrate framework resilience. The method's computational efficiency, requiring considerably less time than MPC, and robustness to communication failures reinforce its practicality for real-time deployment. 

With increasing autonomous vehicle (AV) penetration, understanding the factors contributing to AV crashes is crucial for addressing ongoing road safety challenges. This study aims to reveal the effects of vehicle characteristics, road conditions, environmental factors, and precrash movements on the occurrence of head-on, rear-end, and side-impact crashes. In particular, factors associated with various types of Level 2 and Level 4 AV crashes were analyzed. Our data, obtained from the California Department of Motor Vehicles and the National Highway Traffic Safety Administration, spans from October 2014 to March 2024. Association rule mining techniques identify the significant patterns and interdependencies among the factors contributing to AV crashes. The findings demonstrate that the factors influencing crash types include weather, roadway surface, lighting, and vehicle precrash movements. For example, head-on crashes frequently occur at intersections under poor lighting conditions, whereas rear-end crashes occur more frequently on high-speed highways, particularly because of unexpected braking by the vehicle ahead. Side-impact crashes commonly result from merging maneuvers, especially under adverse weather and lighting conditions. These findings provide new insights into the causal mechanisms behind different types of AV crashes and underscore the need to strengthen traffic management, enhance AV sensing and decision-making capabilities, and implement targeted safety measures in high-risk areas.

Semantic scene completion (SSC) plays a pivotal role in achieving comprehensive perceptions of autonomous driving systems. However, existing methods often neglect the high deployment costs of SSC in real-world applications, and traditional architectures such as three-dimensional (3D) convolutional neural networks (3D CNNs) and self-attention mechanisms struggle to efficiently capture long-range dependencies within 3D voxel grids, limiting their effectiveness. To address these challenges, we propose MetaSSC, a novel meta-learning-based framework for SSC that leverages deformable convolution, large-kernel attention, and the Mamba (D-LKA-M) model. Our approach begins with a voxel-based semantic segmentation (SS) pretraining task, which is designed to explore the semantics and geometry of incomplete regions while acquiring transferable meta-knowledge. Using simulated cooperative perception datasets, we supervise the training of a single vehicle's perception via the aggregated sensor data from multiple nearby connected autonomous vehicles (CAVs), generating richer and more comprehensive labels. This meta-knowledge is then adapted to the target domain through a dual-phase training strategy—without adding extra model parameters—ensuring efficient deployment. To further enhance the model's ability to capture long-sequence relationships in 3D voxel grids, we integrate Mamba blocks with deformable convolution and large-kernel attention into the backbone network. Extensive experiments show that MetaSSC achieves state-of-the-art performance, surpassing competing models by a significant margin while also reducing deployment costs.