27 June 2024 in IEEE Transactions on Intelligent Vehicles
This study examines the impact of conditionally automated driving vehicles (CADVs) handing control over to human drivers on the string stability of mixed traffic. It specifically considers various penetration rates of CADVs and different time-to-collision (TTC) thresholds set at 5, 7, and 9 s. To gather relevant vehicle trajectory data, driving simulator tests were performed, capturing both the normal car-following and the post-takeover processes. Then, the intelligent driver model parameters were calibrated for these two processes, respectively. Additionally, a car-following model was developed to incorporate the dynamic takeover process of CADVs within the heterogeneous traffic. Theoretical analysis was conducted to determine the string stability boundary of heterogeneous traffic, with numerical simulation tests performed to validate the results. The results indicate that the transition of control from CADVs to human drivers significantly contributed to the instability of mixed traffic. Specifically, a higher TTC correlated with more CADVs transferring control to human drivers. The increase in the takeover rate from CADVs to human drivers resulted in pronounced instability within heterogeneous traffic flows. Additionally, the simultaneous control transition of multiple CADVs was associated with reduced minimum platoon speed, exacerbating traffic congestion. Conversely, enhancing the permeability of CADVs led to smoother traffic conditions and suppressed the formation of unstable traffic waves. The findings underscore the significant impact that the transition of control from CADVs to human drivers has on the stability of heterogeneous traffic. This lays a crucial foundation for implementing cooperative safety control measures within such traffic systems. Moreover, implementing a 5-second takeover request lead time in automated driving systems can effectively reduce the takeover rate of CADVs, thereby enhancing overall traffic stability.
6 May 2025 in IEEE Transactions on Intelligent Transportation Systems
Accurately predicting the occurrence, amplitude, and propagation speed of traffic flow shockwaves are essential for dynamic traffic control to mitigate traffic congestion. However, traditional approaches primarily rely on macroscopic traffic data from loop detectors, which often fail to capture the fine-grained interactions among vehicles and roadway features, leading to limited prediction accuracy. To address this gap, we propose a novel vehicle trajectory prediction method based on geometric deep learning (GNN-T), integrating a graph neural network with an attention mechanism. The graph neural network captures vehicle and lane features, and the attention mechanism models interactions between vehicles and between vehicles and lanes, significantly enhancing prediction performance. The evaluation results show that GNN-T outperforms conventional car-following models and other baseline methods, achieving improvements in trajectory prediction accuracy by 7.1% to 9.5%. Furthermore, a traffic flow shockwave prediction model, based on wavelet transform and predicted trajectories, is developed to predict shockwave occurrence, timing, propagation speed, and amplitude. Compared to the Intelligent Driver Model and traditional traffic flow detection methods, our approach demonstrates superior performance, with accuracy rates in shockwave prediction up to 93.5% in validation datasets. These results indicate the potential of the proposed model for practical applications in traffic flow management.