Next Generation Simulation Research Articles

Two-Step Quadratic Programming for Physically Meaningful Smoothing of Longitudinal Vehicle Trajectories

Longitudinal vehicle trajectories suffer from errors and noise because of detection and extraction techniques, challenging their applications. Existing smoothing methods either lack physical meaning or cannot ensure solution existence and uniqueness. To address this, we propose a two-step quadratic programming method that aligns smoothed speeds and higher-order derivatives with physical laws, drivers’ behaviors, and vehicle characteristics. Unlike the well-known smoothing splines method, which minimizes a weighted sum of discrepancy and roughness in a single quadratic programming problem, our method incorporates prior knowledge of position errors into two sequential quadratic programming problems. Step 1 solves half-smoothed positions by minimizing the discrepancy between them and raw positions, subject to physically meaningful bounds on speeds and higher-order derivatives of half-smoothed positions. Step 2 solves smoothed positions by minimizing the roughness while maintaining physically meaningful bounds and allowing the deviations from raw data of smoothed positions by at most those of the half-smoothed positions and prior position errors. The second step’s coefficient matrix is not positive definite, necessitating the matching of the first few smoothed positions with corresponding half-smoothed ones, with equality constraints equaling the highest order of derivatives. We establish the solution existence and uniqueness for both problems, ensuring their well-defined nature. Numerical experiments using Next Generation Simulation (NGSIM) data demonstrate that a third-order derivative constraint yields an efficient method and produces smoothed trajectories comparable with manually re-extracted ones, consistent with the minimum jerk principle for human movements. Comparisons with an existing approach and application to the Highway Drone data set further validate our method’s efficacy. Notably, our method is a postprocessing smoothing technique based on trajectory data and is not intended for systematic errors. Future work will extend this method to lateral vehicle trajectories and trajectory prediction and planning for both human-driven and automated vehicles. This approach also holds potential for broader smoothing problems with known average error in raw data. History: This paper has been accepted for the Transportation Science Special Issue on ISTTT25 Conference. Funding: The authors extend their gratitude to the Pacific Southwest Region University Transportation Center and the University of California Institute of Transportation Studies (UC ITS) Statewide Transportation Research Program (STRP) for their valuable financial support.

A Game-Theoretic Framework for Generic Second-Order Traffic Flow Models Using Mean Field Games and Adversarial Inverse Reinforcement Learning

A traffic system can be interpreted as a multiagent system, wherein vehicles choose the most efficient driving approaches guided by interconnected goals or strategies. This paper aims to develop a family of mean field games (MFG) for generic second-order traffic flow models (GSOM), in which cars control individual velocity to optimize their objective functions. GSOMs do not generally assume that cars optimize self-interested objectives, so such a game-theoretic reinterpretation offers insights into the agents’ underlying behaviors. In general, an MFG allows one to model individuals on a microscopic level as rational utility-optimizing agents while translating rich microscopic behaviors to macroscopic models. Building on the MFG framework, we devise a new class of second-order traffic flow MFGs (i.e., GSOM-MFG), which control cars’ acceleration to ensure smooth velocity change. A fixed-point algorithm with fictitious play technique is developed to solve GSOM-MFG numerically. In numerical examples, different traffic patterns are presented under different cost functions. For real-world validation, we further use an inverse reinforcement learning approach (IRL) to uncover the underlying cost function on the next-generation simulation (NGSIM) data set. We formulate the problem of inferring cost functions as a min-max game and use an apprenticeship learning algorithm to solve for cost function coefficients. The results show that our proposed GSOM-MFG is a generic framework that can accommodate various cost functions. The Aw Rascle and Zhang (ARZ) and Light-Whitham-Richards (LWR) fundamental diagrams in traffic flow models belong to our GSOM-MFG when costs are specified. History: This paper has been accepted for the Transportation Science Special Issue on ISTTT25 Conference. Funding: X. Di is supported by the National Science Foundation [CAREER Award CMMI-1943998]. E. Iacomini is partially supported by the Italian Research Center on High-Performance Computing, Big Data and Quantum Computing (ICSC) funded by MUR Missione 4-Next Generation EU (NGEU) [Spoke 1 “FutureHPC & BigData”]. C. Segala and M. Herty thank the Deutsche Forschungsgemeinschaft (DFG) for financial support [Grants 320021702/GRK2326, 333849990/IRTG-2379, B04, B05, and B06 of 442047500/SFB1481, HE5386/18-1,19-2,22-1,23-1,25-1, ERS SFDdM035; Germany’s Excellence Strategy EXC-2023 Internet of Production 390621612; and Excellence Strategy of the Federal Government and the Länder]. Support through the EU DATAHYKING is also acknowledged. This work was also funded by the DFG [TRR 154, Mathematical Modelling, Simulation and Optimization Using the Example of Gas Networks, Projects C03 and C05, Project No. 239904186]. Moreover, E. Iacomini and C. Segala are members of the Indam GNCS (Italian National Group of Scientific Calculus).

An improved data-driven predictive optimal control approach for designing hybrid electric vehicle energy management strategies

This paper proposes an improved “prediction + optimal control” method for energy management in hybrid electric vehicles equipped with planetary gears. A differentiable predictor and a differentiable optimal controller are developed using supervised learning and reinforcement learning approaches, respectively. Three training steps are performed for the initial predictor, the optimal controller, and the final predictor. This method improves the traditional energy management predictive optimal control approach by incorporating an additional step of retraining the differentiable predictor. This adjustment ensures that the predictor does not blindly improve its performance based on evaluation criterion irrelevant to energy management, which was commonly used in previous studies. Instead, it focuses on enhancing the overall performance of energy management under the “prediction + optimal control” framework. The approach introduced in this paper is compared with the globally optimal dynamic programming results and traditional predictive optimal control methods on the Next Generation Simulation (NGSIM) data. Our method outperforms traditional approaches in energy management on both the training dataset and the test dataset. This further illustrates that the conventional practice of presumptuously optimizing predictors in “prediction + optimal control” methods can be improved using the proposed method.

Ego‐planning‐guided multi‐graph convolutional network for heterogeneous agent trajectory prediction

AbstractAccurate prediction of the future trajectories of traffic agents is a critical aspect of autonomous vehicle navigation. However, most existing approaches focus on predicting trajectories from a static roadside perspective, ignoring the influence of autonomous vehicles’ future plans on neighboring traffic agents. To address this challenge, this paper introduces EPG‐MGCN, an ego‐planning‐guided multi‐graph convolutional network. EPG‐MGCN leverages graph convolutional networks and ego‐planning guidance to predict the trajectories of heterogeneous traffic agents near the ego vehicle. The model captures interactions through multiple graph topologies from four distinct perspectives: distance, visibility, ego planning, and category. Additionally, it encodes the ego vehicle's planning information via the planning graph and a planning‐guided prediction module. The model is evaluated on three challenging trajectory datasets: ApolloScape, nuScenes, and next generation simulation (NGSIM). Comparative evaluations against mainstream methods demonstrate its superior predictive capabilities and inference speed.

An uncertainty cognition-based game model for lane-changing process in mixed driving environment

The non-instantaneous nature of lane-changing demands real-time adaptability for autonomous vehicles (AVs) to respond continuously changing traffic conditions. In the mixed environment where AVs coexist with human-driven vehicles (HVs), the lack of inter-vehicle information exchange necessitates the Nash Equilibrium as best response. In addition, the unpredictable intentions of HV introduce uncertainty, posing a challenge for the solution of equilibrium. This paper introduces an aggressiveness parameter reflecting human drivers' yielding tendencies to autonomous vehicles and enables human-like uncertainty cognition during lane changes. To meet the practical solution requirements of the uncertainty cognition-based game model, we propose Proactive Equilibrium Strategy Algorithm (PESA) based on two-stage Nash equilibrium and anticipation of the opponent's next-stage strategy. Utilising Next Generation Simulation (NGSIM) as environmental data, PESA shows safer and more efficient lane-changing behaviour and leads to more favourable post-lane-changing traffic conditions compared to actual data outcomes.

VSFL: Trajectory prediction framework based on validity-aware semi-asynchronous federated learning in internet of vehicles

Accurate prediction of vehicle trajectories on the Internet of Vehicles is crucial for improving traffic safety. Federated learning has been proposed to address the issue of islands in shared data. Traditional synchronous federated learning requires waiting for each node to complete training before model aggregation can occur, and the aggregation algorithm of the server cannot adjust the aggregation weight in real-time, which leads to increased communication overhead, time cost, and decreased accuracy. To balance these issues, this paper proposes a Validity-aware Semi-asynchronous Federated Learning-based framework for predicting vehicle trajectories. The framework makes the following improvements - firstly, based on the asynchronous cooperation algorithm of generalization research, the models of vehicle nodes within the same RoadSide Unit (RSU) range are aggregated into this RSU to reduce communication overhead and time costs. To further ensure the privacy and security of all participants, we added differential privacy algorithms during the model training process. Secondly, the server uses a weighted aggregation algorithm based on effective perception to aggregate all RSU models into a global model to address potential accuracy issues. This article uses a Social-LSTM like model to validate the effectiveness of this framework in the Next Generation Simulation (NGSIM) Vehicle Trajectories and Support Data dataset. Compared with traditional federated averaging algorithms and trust aware federated learning algorithms, this framework reduces distance errors by 27.80 % and 19.23 %, respectively, and improves accuracy by 22.56 % and 13.59 %.

An improved stochastic car-following model considering the complete state information of multiple preceding vehicles under connected vehicles environment

Studying the car-following model contributes to improving vehicle driving efficiency and traffic flow stability within the context of connected automated vehicles (CAVs). Therefore, an improved stochastic car-following model (SMLVM) is proposed in this paper, building upon the traditional FVD model and BLVD model. This new model comprehensively considers four deterministic factors: acceleration, velocity, headway, and optimal speed memory of multiple preceding vehicles, as well as the comprehensive influence of stochastic factors on car-following behavior during the traveling process. Meanwhile, weight functions are utilized to represent the varying strengths of influence from vehicles ahead with different positions and speeds on the following vehicle. The SMLVM model is categorized and discussed based on practical application scenarios in this paper. When the focus is primarily on fundamental traffic flow behavior, relative motion between vehicles, and other basic principles, the SMLVM model degenerates into the MLVM model. Through stability analysis and simulation experiments on MLVM model, critical stability conditions are derived. Comparisons with other classical models demonstrate that the MLVM model exhibits stronger stability and resistance to disturbances. When the focus is primarily on various realistic driving scenarios, through simulation experiments on SMLVM models in different scenarios, it is found that the influence of stochastic factors on SMLVM models mainly depends on the magnitude of the intensity of stochastic fluctuations, which is rarely affected by the specific functional expressions of the stochastic terms. Finally, empirical analysis is conducted by selecting five sets of processed data from the next-generation simulation (NGSIM) dataset. Based on parameter estimation using the SMLVM model, and a comparison is made with the FVD model and BLVD model. The results confirm that the model proposed in this paper fits vehicle speeds better and absorbs disturbances more effectively, and exhibiting greater stability in traffic flow.

LSTM-based graph attention network for vehicle trajectory prediction

Vehicle Trajectory Prediction (VTP) is one of the key technologies for autonomous driving, which can improve the safety and collaboration of the autonomous driving system. The interaction behavior among vehicles in reality has an impact on VTP. However, many methods ignore the interaction among vehicles, which results in limited accuracy of prediction results. Therefore, we propose a Long Short-Term Memory (LSTM)-based Graph Attention Network (GAT) method for VTP, which encodes vehicle trajectory information with LSTM networks and represents vehicle interactions with GAT. Firstly, in order to capture the temporal relationship between positions and consider their influence, we use LSTM model to encode the position data. Meanwhile, to comprehensively model vehicle motion and use multidimensional feature representation, we employ another LSTM model to encode the motion data, including position, velocity and acceleration. Secondly, to learn distinct feature representation, we use one GAT module to process the LSTM position encoding features for capturing spatial relationships of position information. Another GAT module is employed to process the LSTM motion encoding features for fully considering multidimensional motion dynamics and spatial–temporal dependencies. Finally, the LSTM decoder receives all features and predicts the vehicle trajectory. The experimental results show that the proposed method demonstrates superior predictive performance by using the Next Generation Simulation (NGSIM) dataset.

Traffic state estimation from vehicle trajectories with anisotropic Gaussian processes

Accurately monitoring road traffic state is crucial for various applications, including travel time prediction, traffic control, and traffic safety. However, the lack of sensors often results in incomplete traffic state data, making it challenging to obtain reliable information for decision-making. This paper proposes a novel method for imputing traffic state data using Gaussian processes (GP) to address this issue. We propose a kernel rotation re-parametrization scheme that transforms a standard isotropic GP kernel into an anisotropic kernel, which can better model the congestion propagation in traffic flow data. The model parameters can be estimated by statistical inference using data from sparse probe vehicles or loop detectors. Moreover, the rotated GP method provides statistical uncertainty quantification for the imputed traffic state, making it more reliable. We also extend our approach to a multi-output GP, which allows for simultaneously estimating the traffic state for multiple lanes. We evaluate our method using real-world traffic data from the Next Generation simulation (NGSIM) and HighD programs, along with simulated data representing a traffic bottleneck scenario. Considering current and future mixed traffic of connected vehicles (CVs) and human-driven vehicles (HVs), we experiment with the traffic state estimation (TSE) scheme from 5% to 50% available trajectories, mimicking different CV penetration rates in a mixed traffic environment. We also test the traffic state estimation when traffic flow information is obtained from loop detectors. The results demonstrate the adaptability of our TSE method across different CV penetration rates and types of detectors, achieving state-of-the-art accuracy in scenarios with sparse observations.

Investigating autonomous vehicle discretionary lane-changing execution behaviour: Similarities, differences, and insights from Waymo dataset

Recently released autonomous vehicle datasets like Waymo can provide rich information (and unprecedented opportunities) to investigate lane-changing behaviour of autonomous vehicles, requiring data from multiple drivers and lanes with different objectives. As such, the study investigates the discretionary lane-changing execution behaviour of autonomous vehicles and compares its behaviour with human-driven vehicles from Waymo and Next Generation Simulation (NGSIM) datasets. Several behavioural factors are statistically analysed and compared, whereas the discretionary lane-changing execution time (or duration) is modelled by a random parameters hazard-based duration modelling approach, which accounts for unobserved heterogeneity. Descriptive analyses suggest that autonomous vehicles maintain larger lead and lag gaps, longer discretionary lane-changing execution time, and lower acceleration variation than human-driven vehicles. The random parameters duration model reveals heterogeneity in discretionary lane-changing execution behaviour, which is higher in human-driven vehicles but decreases significantly for autonomous vehicles. Whilst contradictory to a general hypothesis in the literature that autonomous vehicles will eliminate heterogeneity, our finding indicates that heterogeneous behaviour also exists in autonomous vehicles (although to a lesser extent than in human-driven vehicles), which can be contextual to prevailing traffic conditions. Overall, autonomous vehicles show safer discretionary lane-changing behaviour compared to human-driven vehicles.

Vehicle lane-change intention recognition based on BiLSTM Attention model for the Internet of vehicles

In terms of lane-changing and other driver actions, precise identification of the intentions of nearby vehicles is crucial to autonomous vehicle performance safety. At present, research in this domain primarily focuses on ideal environments without considering data packet loss. Therefore, this paper considered the impact of packet loss in the Internet of Vehicles on the performance of the lane change intent recognition model. To achieve this, an enhanced BiLSTM Attention model, which combines the bidirectional long short-term memory network structure and attention mechanism, is proposed based on LSTM. The NGSIM (Next Generation Simulation) dataset was utilized to extract vehicle lane-change behaviors for model training and testing. A long short-term memory (LSTM) model was employed to conduct comparative experiments using various input frequencies and packet loss rates. The performance of the proposed BiLSTM Attention model was evaluated through ablation experiments. A comparison was made between the model’s performance in the absence of packet loss and its performance under a scenario with 30% packet loss. Additionally, the impact of continuous packet loss on the recognition of the lane-change intent model was analyzed. Experiments show that it outperforms basic LSTM and BiLSTM models, including the LSTM Attention method, with impressive improvements of 7.84%, 2.22%, and 4.89% (F1macro) and 2.83%, 1.03%, and 2.18% for the area under the receiver operating characteristic curve (AUC), respectively. Even under extreme (30%) packet loss conditions, the proposed model outperforms the same models by 8.23%, 2.68%, and 5.38% (F1macro) and 2.94%, 1.03%, and 2.29% (AUC), respectively. For 30% packet loss, the proposed model’s performance decreased by 0.108% (F1macro) and 0.102% (AUC); however, the LSTM, BiLSTM, and LSTM Attention model performances decreased by 0.468% and 0.209%, 0.554% and 0.103%, and 0.569% and 0.208% for F1macro and AUC, respectively. Thus, the proposed model is the least affected by packet loss.

BAT: Behavior-Aware Human-Like Trajectory Prediction for Autonomous Driving

The ability to accurately predict the trajectory of surrounding vehicles is a critical hurdle to overcome on the journey to fully autonomous vehicles. To address this challenge, we pioneer a novel behavior-aware trajectory prediction model (BAT) that incorporates insights and findings from traffic psychology, human behavior, and decision-making. Our model consists of behavior-aware, interaction-aware, priority-aware, and position-aware modules that perceive and understand the underlying interactions and account for uncertainty and variability in prediction, enabling higher-level learning and flexibility without rigid categorization of driving behavior. Importantly, this approach eliminates the need for manual labeling in the training process and addresses the challenges of non-continuous behavior labeling and the selection of appropriate time windows. We evaluate BAT's performance across the Next Generation Simulation (NGSIM), Highway Drone (HighD), Roundabout Drone (RounD), and Macao Connected Autonomous Driving (MoCAD) datasets, showcasing its superiority over prevailing state-of-the-art (SOTA) benchmarks in terms of prediction accuracy and efficiency. Remarkably, even when trained on reduced portions of the training data (25%), our model outperforms most of the baselines, demonstrating its robustness and efficiency in predicting vehicle trajectories, and the potential to reduce the amount of data required to train autonomous vehicles, especially in corner cases. In conclusion, the behavior-aware model represents a significant advancement in the development of autonomous vehicles capable of predicting trajectories with the same level of proficiency as human drivers. The project page is available on our GitHub.

A decision model based on gene expression programming for discretionary lane-changing move

ABSTRACT This study focuses on modeling Discretionary Lane-Changing (DLC), which accounts for the majority of lane-change moves in traffic flows. A binary decision model for lane-changing moves was improved with the method of Gene Expression Programming (GEP). The decision to prefer GEP is due to its high performance in a variety of engineering solutions in the literature. The GEP model was trained with Next Generation SIMulation (NGSIM) trajectory data gathered at the I-80 Freeway in Emeryville, California, and then tested with data gathered at the U.S. Highway 101 in LA, California. The test results indicate that the model made decisions of “change lane” with 92.98% accuracy, and “do not change lane” with 99.65% accuracy. A sensitivity analysis was also conducted to discover potential limits of the performance of the GEP model. The performance of this model was compared with other high-performance decision models developed with the NGSIM's DLC data in the literature and with TransModeler's gap acceptance model. This comparison indicates that the GEP model is the most successful decision model for discretionary lane-changing moves. The GEP model has a high potential to be applied in DLC decision support systems in (semi-) automated vehicles, as well as traffic simulation software.

Reliability Assessment of Vehicle-to-Vehicle Communication Networks through Headway Distribution and Information Propagation Delay

Several developed vehicle spaces and time headway distribution models in traffic flow theory have been widely used in the literature, reflecting the primary uncertainty in drivers’ car-following movements and explaining the traffic flow stochastic features. Moreover, effective vehicle-to-vehicle (V2V) communication is a key to decentralizing traffic information systems. Accurate vehicle headway distribution estimation will ensure reliable communication and benefit passengers’ safety and comfort. Consider several proposals for headway distribution in the literature; this paper studies the effect of space-headway distribution on information propagation delay in assessing the reliability of the V2V communication networks. We utilize the properties of reliability measurement for headway data by introducing the quasi-maximum likelihood estimator (QMLE) to measure the effects that different headway models have on estimating the parameters of headway distribution in a V2V communication network. The statistical analysis is then applied to the real Next-Generation Simulation (NGSIM) data. It validates the proposed methodology and formulations by measuring the effects of model selection on headway data and information propagation delay on the reliability of the V2V network. The results show that the effect of headway distribution when the vehicle’s transmission range is smaller than the road segmentation is not negligible, especially when cars are very distant. Based on our results, we recommend new metrics based on Kullback–Leibler divergence for model selection of headway data, thereby enhancing the reliability of the V2V network.

Deep spatial‐temporal embedding for vehicle trajectory validation and refinement

AbstractHigh‐angle cameras are commonly used for trajectory data collection in transportation research. However, without refinement and validation, trajectory data obtained through video processing software may be unreliable, inaccurate, or incomplete. This paper focuses on a critical issue in the field of trajectory data acquisition and analysis—there is still no reliable and fully vetted trajectory dataset in the research community. The current practice for validating video‐based trajectory can be classified as indirect methods and direct methods. Indirect methods of trajectory validation use algorithms to efficiently correct data anomalies without human intervention but may overlook detailed driving behaviors, whereas direct methods involve meticulous manual verification to preserve data fidelity but are labor‐intensive and less scalable. The spatial‐temporal maps (STMaps) method offers an additional layer of verification to affirm the accuracy and reliability of trajectory data. To enhance the performance, the deep spatial‐temporal embedding model is proposed for trajectory instance segmentation on STMaps using the contrastive learning framework. The parity constraints at both pixel and instance levels guide the deep neural network to learn the embedding spaces that can be built on different backbone networks. The reconstructed Next Generation Simulation (NGSIM) highway dataset trajectory dataset is thoroughly validated against manually processed ground truth, and the error‐free NGSIM data are refined to be a reliable resource for transportation research based on car‐following behaviors, lane‐change frequency, consistency, and jerk value measurements.

Generative adversarial network for car following trajectory generation and anomaly detection

Car-following trajectory generation and anomaly detection are critical functions in the sensing module of an automated vehicle. However, developing models that capture realistic trajectory data distribution and detect anomalous driving behaviors could be challenging. This paper proposes ‘TrajGAN’, an unsupervised learning approach based on the Generative Adversarial Network (GAN) to exploit vehicle car following trajectory data for generation and anomaly detection. The proposed TrajGAN consists of two modules, an encoder-decoder Long Short-Term Memory (LSTM)-based generator and an LSTM-multilayer perceptron (MLP) based discriminator, whose former component is used to generate vehicular car following trajectories and the latter one is for trajectory anomaly detection. By letting these two modules game with each other in training, we can simultaneously achieve robust trajectory generators and anomaly detectors. Trained with the Next Generation Simulation (NGSIM) dataset, TrajGAN can generate realistic trajectories with a similar distribution of training data and identify a manifold of anomalous trajectories based on an anomaly scoring scheme. Simulation results indicate that the proposed approach is efficient in reproducing artificial trajectories and identifying anomalous driving behaviors.

GIVA: Interaction-aware trajectory prediction based on GRU-Improved VGG-Attention Mechanism model for autonomous vehicles

Predicting future trajectories is crucial for autonomous vehicles, as accurate predictions enhance safety and inform subsequent decision-making and planning modules. This is however a challenging task due to the complex interactions between surrounding vehicles. Existing methods struggled to extract deep representations and often overlook spatial dependence. To address this problem, this paper introduces GIVA, an interaction-aware trajectory prediction method based on the Gated Recurrent Unit (GRU)-Improved Visual Geometry Group (VGG)-Attention Mechanism model. GIVA first encodes the historical trajectories of the target vehicle and its surrounding vehicles using a GRU Encoder. Next, an Interaction Module, which combines the Improved VGG Pooling Module and the Attention Mechanism Pooling Module, effectively captures spatial interaction features between vehicles. The Improved VGG Pooling Module extracts more detailed and effective interaction information, while the Attention Mechanism Pooling Module emphasizes the importance of surrounding vehicles for the target vehicle’s future trajectory. Lastly, the dynamic encoding feature of the target vehicle and the fused interaction feature are concatenated and input into a GRU Decoder to generate the future trajectory. Experiments on the public Next Generation Simulation (NGSIM) dataset showcase the effectiveness of GIVA compared to existing prediction approaches, demonstrating its potential for improving autonomous vehicle performance.

Application of machine learning models to predict driver left turn destination lane choice behavior at urban intersections

When there are multiple lanes to choose from downstream of a turning movement, drivers should choose the innermost lane so that drivers at other approaches of the intersection may make concurrent turning movements in the outermost lane(s). However, human drivers do not always choose the innermost lane, which could lead to crashes with other vehicles. Therefore, predicting human driver behaviors is vital in reducing crashes, as the need to share the roadways with automated vehicles (AVs) continues to grow. In this research, various machine learning models have been used to predict the left turn destination lane choice of human-driven vehicles (HDVs) at urban intersections based on several quantifiable parameters. A total of 174 subject vehicles were extracted and analyzed in Los Angeles, California, and Atlanta, Georgia, using HDV trajectory data from the Next Generation SIMulation (NGSIM) database. Five machine learning techniques, namely binary logistic regression, k nearest neighbors, support vector machines, random forest, and adaptive neuro-fuzzy inference system, were applied to the extracted data to predict the lane choice behavior of drivers. The k nearest neighbors model showed the most promising results for the evaluated data with a correct decision score of over 93% for the unseen test data. This model may be programmed into: (i) AVs, in conjunction with sensors, to predict if an HDV is about to turn into the incorrect destination lane; and (ii) microscopic traffic simulation tools so that modelers can identify potential conflicts when HDVs do not select the appropriate destination lane.

TCN-SA: A Social Attention Network Based on Temporal Convolutional Network for Vehicle Trajectory Prediction

Vehicle trajectory prediction can provide important support for intelligent transportation systems in areas such as autonomous driving, traffic control, and traffic flow optimization. Predicting vehicle trajectories is an extremely challenging task that not only depends on the vehicle’s historical trajectory but also on the dynamic and complex social-temporal relationships of the surrounding traffic network. The trajectory of the target vehicle is influenced by surrounding vehicles. However, existing methods have shortcomings in considering both time dependency and interactive dependency between vehicles or insufficient consideration of the impact of surrounding vehicles. To address this issue, we propose a hybrid deep learning model based on a temporal convolutional network (TCN) that considers local and global interactions between vehicles. Specifically, we use a social convolutional pooling layer to capture local interaction features between vehicles and a multihead self-attention layer to capture global interaction features between vehicles. Finally, we combine these two features using an encoder-decoder structure to predict vehicle trajectories. Through experiments on the Next-Generation Simulation (NGSIM) public dataset and ablation experiments, we validate the effectiveness of our model.

Modeling and Optimization of Connected and Automated Vehicle Platooning Cooperative Control with Measurement Errors.

This paper presents a cooperative control method for connected and automated vehicle (CAV) platooning, thus specifically addressing the challenge of sensor measurement errors that can disrupt the stability of the CAV platoon. Initially, the state-space equation of the CAV platooning system was formulated, thereby taking into account the measurement error of onboard sensors. The superposition effect of the sensor measurement errors was statistically analyzed, thereby elucidating its impact on cooperative control in CAV platooning. Subsequently, the application of a Kalman filter was proposed as a means to mitigate the adverse effects of measurement errors. Additionally, the CAV formation control problem was transformed into an optimal control decision problem by introducing an optimal control decision strategy that does not impose pure state variable inequality constraints. The proposed method was evaluated through simulation experiments utilizing real vehicle trajectory data from the Next Generation Simulation (NGSIM). The results demonstrate that the method presented in this study effectively mitigates the influence of measurement errors, thereby enabling coordinated vehicle-following behavior, achieving smooth acceleration and deceleration throughout the platoon, and eliminating traffic oscillations. Overall, the proposed method ensures the stability and comfort of the CAV platooning formation.