Connected vehicles enabled by communication technologies have the potential to improve traffic mobility and enhance roadway safety such that traffic information can be shared among vehicles and ...infrastructure. Fruitful speed advisory strategies have been proposed to smooth connected vehicle trajectories for better system performance with the help of different car-following models. Yet, there has been no such comparison about the impacts of various car-following models on the advisory strategies. Further, most of the existing studies consider a deterministic vehicle arriving pattern. The resulting model is easy to approach yet not realistic in representing realistic traffic patterns. This study proposes an Individual Variable Speed Limit (IVSL) trajectory planning problem at a signalized intersection and investigates the impacts of three popular car-following models on the IVSL. Both deterministic and stochastic IVSL models are formulated, and their performance is tested with numerical experiments. The results show that, compared to the benchmark (i.e., without speed control), the proposed IVSL strategy with a deterministic arriving pattern achieves significant improvements in both mobility and fuel efficiency across different traffic levels with all three car-following models. The improvement of the IVSL with the Gipps’ model is the most remarkable. When the vehicle arriving patterns are stochastic, the IVSL improves travel time, fuel consumption, and system cost by 8.95%, 19.11%, and 11.37%, respectively, compared to the benchmark without speed control.
High-granularity vehicle trajectory data can help researchers develop traffic simulation models, understand traffic flow characteristics, and thus propose insightful strategies for road traffic ...management. This paper proposes a novel vehicle trajectory extraction method that can extract high-granularity vehicle trajectories from aerial videos. The proposed method includes video calibration, vehicle detection and tracking, lane marking identification, and vehicle motion characteristics calculation. In particular, the authors propose a Monte-Carlo-based lane marking identification approach to identify each vehicle's lane. This is a challenging problem for vehicle trajectory extraction, especially when the aerial videos are taken from a high altitude. The authors applied the proposed method to extract vehicle trajectories from several high-resolution aerial videos recorded from helicopters. The extracted dataset is named by the High-Granularity Highway Simulation (HIGH-SIM) vehicle trajectory dataset. To demonstrate the effectiveness of the proposed method and understand the quality of the HIGH-SIM dataset, we compared the HIGH-SIM dataset with a well-known dataset, the NGSIM US-101 dataset, regarding the accuracy and consistency aspects. The comparison results showed that the HIGH-SIM dataset has more reasonable speed and acceleration distributions than the NGSIM US-101 dataset. Also, the internal and platoon consistencies of the HIGH-SIM dataset give lower errors compared to the NGSIM US-101 dataset. To benefit future research, the authors have published the HIGH-SIM dataset online for public use.
•Collect high-resolution traffic flow videos from a helicopter.•Propose a deep learning-based vehicle trajectory extraction method to extract vehicle trajectories from aerial videos.•Propose a Monte-Carlo-based lane marking identification approach to identify each vehicle's lane.•The extracted long-coverage trajectory dataset has been published online for public use.
•Design a mixed traffic framework at signalized intersections with multi-lane roads.•Propose a CAV trajectory smoothing concept with an HV lane-change-aware mechanism.•Find induced HV lane changes ...cause reduction of half or more expected benefits.•Find the proposed model yields significant benefits when the CAV MPR is not high.
Trajectory smoothing is an effective concept to control connected automated vehicles (CAVs) in mixed traffic to reduce traffic oscillations and improve overall traffic performance. However, smoother trajectories often lead to greater gaps between vehicles, which may incentivize human driven vehicles (HVs) from adjacent lanes to make cut-in lane changes. Such cut-in lane changes may compromise the expected performance from CAV trajectory smoothing. To figure out the reasons behind the issue, this paper designs a mixed traffic framework at a signalized intersection with multi-lane roads considering detailed trajectory control, car following and lane changing maneuvers all together. Based on the framework, this paper proposes a decentralized lane-change-aware CAV trajectory optimization model including discretionary lane change restraining and mandatory lane change yielding strategies. Riding comfort and traffic mobility are considered as a joint objective. And the complex non-linear lane-change-aware constraints are linearized to convert the proposed problem to a quadratic optimization problem. The linearization allows the investigated problem to be easily fed into a commercial solver. Numerical experiments are conducted to study the performance of the proposed model and to compare it with other models (e.g., a cooperative lane change model and a trajectory optimization model without the lane-change-aware mechanism) in different scenarios. First, results show that the HV lane changes cause reduction of half or more expected benefits of trajectory smoothing along a multi-lane segment adjacent to a signalized intersection. Then, we find that the proposed model outperforms the other models. Especially, the proposed model yields extra benefits in the system joint objective (10–25%), riding comfort (10–25%), travel time (1–8%), fuel consumption (3–15%) and safety (5–25%) compared with the trajectory optimization model without the lane-change-aware mechanism when CAV market penetration rate is not high. Sensitivity analyses on road segment lengths, signal cycle lengths, traffic saturation rates and through-vehicle rates show that the proposed model yields better system performance under most scenarios, e.g., 20% extra benefit at a short road segment length, 30% extra benefit at a long signal cycle length, 25% extra benefit at a high traffic saturation rate, and 25% extra benefit at a high through-vehicle rate.
Amid escalating energy crises and environmental pressures, electric vehicles (EVs) have emerged as an effective measure to reduce reliance on fossil fuels, combat climate change, uphold sustainable ...energy and environmental development, and strive towards carbon peaking and neutrality goals. This study introduces a nonlinear integer programming model for the deployment of dynamic wireless charging lanes (DWCLs) and EV charging strategy joint optimization in highway networks. Taking into account established charging resources in highway service areas (HSAs), the nonlinear charging characteristics of EV batteries, and the traffic capacity constraints of DWCLs. The model identifies the deployment of charging facilities and the EV charging strategy as the decision-making variables and aims to minimize both the DWCL construction and user charging costs. By ensuring that EVs maintain an acceptable state of charge (SoC), the model combines highway EV charging demand and highway EV charging strategy to optimize the DWCL deployment, thus reducing the construction cost of wireless charging facilities and user charging expenses. The efficacy and universality of the model are demonstrated using the classical Nguyen–Dupius network as a numerical example and a real-world highway network in Guangdong Province, China. Finally, a sensitivity analysis is conducted to corroborate the stability of the model. The results show that the operating speed of EVs on DWCLs has the largest impact on total cost, while battery capacity has the smallest. This comprehensive study offers vital insights into the strategic deployment of DWCLs, promoting the sustainable and efficient use of EVs in highway networks.
•Propose a decentralized connected automated vehicle trajectory optimization model.•Consider travel time, fuel consumption and safety risks simultaneously.•Consider mixed traffic containing connected ...automated and human driven vehicles.•Compare the decentralized optimization model with a centralized optimization model.
It is concerned that system-level benefits of connected automated vehicle control might only prevail in a far-future centralized control environment, whereas the benefits could be much offset in a near-future decentralized control system. To address this concern, this paper proposes a decentralized control model for connected automated vehicle trajectory optimization at an isolated signalized intersection with a single-lane road where each connected automated vehicle aims to minimize its own travel time, fuel consumption and safety risks. To improve the computational tractability, the original complex decentralized control model is reformulated into a discrete model. A benchmark centralized control model is also formulated to compare with the decentralized control model. The DIRECT algorithm is adopted to solve the above models. Numerical results show that the decentralized control model has better computational efficiency (with an average solution time of 10 s) than the centralized control model (with an average solution time of 60 s) without significant loss of the system optimality (with an average of 3.91%). Finally, analysis on connected automated vehicle market penetration shows that the extra benefit of the centralized control model is not obvious either in under-saturated traffic (less than 1%) or at a low connected automated vehicle market penetration rate in critically-saturated and over-saturated traffic (less than 3% when the market penetration rate is lower than 20%). The results suggest that, as apposed to the earlier concern, the near-future decentralized control scheme that requires less technology maturity and infrastructure investment can achieve benefits similar to the far-future centralized control scheme with much simpler operations in under-saturated traffic, or in critically-saturated traffic and over-saturated traffic with a low connected automated vehicle market penetration rate.
The physics of shockwaves is a fundamental traffic characteristic that is useful for microscopic traffic flow modeling. Thanks to connected vehicle technologies, it is possible to capture shockwaves ...by collecting the information of multiple downstream vehicles. Numerous classical physics-based models have utilized the physics of shockwaves to predict vehicle trajectory dynamics, yet their predictability is often limited due to the volatile and complex nature of highway traffic composed of human-driven vehicles. Recent learning-based trajectory prediction models utilize historical trajectories of surrounding vehicles to improve predictability. However, those learning-based models are purely data-driven, thus lacking interpretability and physical insights, or even missing opportunities for further improving model predictability. To leverage the advantages of both learning-based and physics-based models, this study proposes a physics-aware learning-based model for a trajectory prediction of congested traffic in a connected vehicle environment. A newly collected highway trajectory dataset is adopted for training and validation. Experiment results show that the proposed hybrid model yields better predictability, compared with the learning-based models (e.g., long short-term memory neural networks and convolutional neural networks), with an 8.7% predictability reduction of position errors and a 6.5% reduction of speed errors, which further verify the positive impacts of adopting physics of shockwaves in hybrid learning models. Moreover, result analysis shows that predictability improves as the market penetration rate increases, and the proposed hybrid model is better than the learning-based models with 3-10% improvements in predictability.
•Combine physics insights with data-driven features to predict vehicle safety.•Propose a customized loss function to give more attention to more risky events.•Predict a time series of safety ...indicators to give drivers enough reaction time.•Produce continuous safety indicators to reflect risk levels.
Real-time vehicle safety prediction is critical in roadway safety management as drivers or vehicles can be altered beforehand to take corresponding evasive actions and avoid possible collisions. This study proposes a physics-informed multi-step real-time conflict-based vehicle safety prediction model to enhance roadway safety. Physics insights (i.e., traffic shockwave properties) are combined with data-driven features extracted from deep-learning techniques to improve prediction accuracy. A time series of future vehicle safety indicators are predicted such that vehicles/drivers have enough time to take precautions. The safety indicator at each time stamp is a continuous value that the sign reflects the presence of conflict risks, and the absolute value indicates the conflict risk level to advise different magnitudes of evasive actions. A customized loss function is developed for the proposed prediction model to give more attention to risky events, which are the focus of safety management. The prediction superiority of the proposed model is proven through numerical experiments by comparing it with three benchmarks constructed based on the literature. Further, sensitivity analysis on key model parameters is carried out to advise parameter selections in developing real-world conflict-based vehicle safety prediction applications.
•Propose interpretable dimensionality reduction techniques in trajectory prediction.•Piecewise Taylor series approximation and piecewise Fourier series approximation.•Reduce device cost and computing ...energy consumption by simplifying computation.•Verify the techniques’ effectiveness by comparing them with benchmarks.•Prove the techniques’ robustness against data noises.
To facilitate low-cost connected automated vehicle (CAV) system development, this study proposes two interpretable dimensionality reduction techniques in vehicle trajectory prediction, i.e., the piecewise Taylor series approximation (PTA) and the piecewise Fourier series approximation (PFA), to lower computation complexity, reduce device investment, and decrease computation energy consumption. Two benchmarks are developed, the long short-term memory (LSTM)-based model without dimensionality reduction and the LSTM-based model with encoder-decoder (a widely used dimensionality reduction technique). Results show that the four predictions have similar accuracy, and the training time (proportional to computation energy consumption) of models with dimensionality reduction techniques is greatly reduced. The reduction is even more significant when PTA/PFA is used. Sensitivity analysis advises PFA/PTA parameter selections to reduce computation complexity without significant loss of prediction accuracy. Further, the robustness of the LSTM PTA/PFA is proven by the investigation of data noises.
Increasing commercial vehicles are equipped with automated driving features. Adaptive cruise control, a critical longitudinal control system of commercial automated vehicles (AVs), may have ...significant impacts on fuel consumption. To investigate the impacts, this paper collected high-resolution trajectory data of commercial AVs with different operating scenarios, speed ranges, and headway settings on the highway system. The AVs’ fuel consumption was calculated by several state-of-the-art or classical vehicle fuel consumption models. From empirical analyses, we found that as the AV headway setting increases, the corresponding fuel consumption decreases. Also, we found that as the speed of AV traffic increases, the impacts of AV headway settings on fuel consumption decrease. Moreover, we compared the fuel consumption of AVs and human-driven vehicles (HVs). We found that for the same settings, the AVs always require less fuel consumption than the HVs. Following these findings, a set of managerial insights were provided into relevant stakeholders for future AV traffic.
Car-following safety is related to both observed driving characteristics (e.g. car-following behaviour) and unobserved driver heterogeneity (e.g. drivers' psychological features). Two major issues ...remain in the existing literature, i.e. limiting to longitudinal characteristics and not addressing the confounding effects of unobserved driver heterogeneity. This study takes a matched case-control approach to model car-following safety with both longitudinal and lateral driving characteristics. Unobserved driver heterogeneity is controlled by matching preceding and following vehicle IDs. Results show that unstable lateral movements of preceding vehicles and following vehicles contribute to higher crash risks. Comparison results on two datasets with different congestion levels reveal that it is safer in more congested traffic when the following vehicle maintains more stable longitudinal and lateral behaviours, and greater speed difference, headway, and spacing regarding its preceding vehicle. This study provides insights in enhancing roadway safety management and benefiting the automated vehicle development by warnings on associated risks.