Search

Cooperative driving is crucial for improving traffic efficiency and safety for connected and automated vehicles (CAVs), especially in traffic bottlenecks. However, most of the state-of-the-art cooperative driving strategies neglect the issue of fairness. Fairness is essential to properly allocate road resources and improve the travel experience. In this paper, we focus on the fairness concerns in the on-ramp cooperative driving problem. First, we note that enhancing traffic efficiency usually leads to unfairness, but we propose solutions to balance both aspects. Using the fundamental relation in traffic flow theory, we illustrate the existence of the trade-off at congested on-ramps. We then make some modifications to the cooperative driving strategies to incorporate fairness considerations. Simulation results show that the modified strategies achieve trade-offs in agreement with the theoretical one, laying the foundation for implementing the trade-off in real-world scenarios. These findings are enlightening for the increasing research on fairness issues in cooperative driving, and contribute to the optimization of traffic management strategies.

Virtual simulation-based testing of autonomous vehicles (AVs) needs massive challenging corner cases to reach high testing accuracy. Current methods achieve this goal by finding testing scenarios with low sampling frequency in the empirical distribution. However, these methods neglect modeling heterogeneous driving behavior, which actually is crucial for finding corner cases. To fill this gap, we propose an interpretable and operable method for sampling corner cases. Firstly, we initialize a testing scenario and allocate testing tasks to AV. Then, to simulate the variability in driving behaviors, we design utility functions with several hyperparameters and generate aggressive, conservative, and normal driving strategies by adjusting hyperparameters. By changing the heterogeneous driving behavior of surrounding vehicles (SVs), we can sample the challenging corner cases in the scenario. Finally, we conduct a series of simulation experiments in a typical lane-changing scenario. The simulation results reveal that by adjusting the occurrence frequency of heterogeneous SVs in the testing scenario, more corner cases can be found in limited rounds of simulations.

There are generally two kinds of traffic control strategies to relieve traffic congestion in lane-drop bottlenecks: variable speed limits (VSL) control and lane-changing (LC) control. However, VSL has limited or even no effect due to many mandatory LC maneuvers near bottlenecks, while LC fails to reduce traffic congestion when traffic demand is high. Although a few control methods combine VSL and LC, they do not consider the interaction between VSL and LC, which rules out many potentially good alternatives. We instead propose an integrated VSL and LC control method under a connected and automated vehicle (CAV) environment, which can consider the interaction and simultaneously find the values of LC numbers and speed limits to maximize traffic efficiency. Our control is in the framework of the model predictive control (MPC), which consists of prediction, optimization, and implementation. We adopt an improved multi-class cell transmission model (CTM) for traffic state prediction, then use the genetic algorithm (GA) for optimization which optimizes traffic network performance, and implement our control method in the SUMO platform. Simulation results demonstrate that our control method greatly improves the capacity of the road and is robust to different traffic demands and scenarios. Our control outperforms no control and VSL-only control in travel time and exhaust emissions, which reduces total travel time by 23.86% to 44.62% and exhaust emissions by 10.29% to 48.19%.

Connected and automated vehicle (CAV) can obtain the precise vehicle location information via vehicle to infrastructure (V2I) communication to achieve the path tracking task under the intelligent vehicle-infrastructure cooperative system (i-VICS). However, in some special scenarios, the location information of vehicles might become inaccurate or lost, e.g., the vehicle loses contact with the road-side unit (RSU). To improve the reliability of vehicle location information, we concentrate on solving two kinds of tracking problems, i.e., vehicle location information is inaccurate and lost in this paper. First, a predicted method using vehicle dynamics is presented to recognize the scenarios in which the vehicle location information becomes inaccurate when the vehicle moves. Then, a modified Kalman filter is employed to adapt to a more complex scenario with variable position error. After that, a multilayer perceptron (MLP) model is designed to predict vehicle location information when vehicle location information is lost. At last, simulation results demonstrate that the proposed methods have a marked influence on handling the driving scenario and make our path tracking algorithm more robust.

Accurate and robust lane detection, especially the curve lane detection, is the premise of lane departure warning system and forward collision warning system. In this article, an algorithm based on improved river flow and random sample consensus is proposed to detect curve lane under challenging conditions including the dashed lane markings and vehicle occlusion. The curve lanes are modeled as hyperbola pair. To determine the coefficient of curvature, an improved river flow method is presented to search feature points in the far vision field guided by the results of detected straight lines in near vision field or the curved lines from the last frame, which can connect dashed lane markings or obscured lane markings. As a result, it is robust on dashed lane markings and vehicle occlusion conditions. Then, random sample consensus is utilized to calculate the curvature, which can eliminate noisy feature points obtained from improved river flow. The experimental results show that the proposed method can accurately detect lane under challenging conditions.

Traffic data are the information source for traffic control and management. With the development and integration of Intelligent Transportation Systems, many applications and their respective sensors and detectors are a rich source of data about transportation system characteristics and performance. However, because of the limitation of databases and devices, the huge amounts of traffic data can not be stored without reduction. In this paper, an approach for urban traffic volume compression based on artificial neural network is proposed. The lossy compression of data is realized by using a set of three-layer back-propagation neural networks to remove the correlation within traffic volumes. The model has both a small reproduction error and a relatively high compression ratio.