Traffic simulation is essential for validating the safety and reliability of autonomous driving systems, yet data-driven simulation methods often struggle with distribution shifts, limiting their generalizability across diverse datasets (domains). To address this, we present Causal Driving Pattern Transfer (CDPT), a novel two-stage knowledge distillation framework built upon diffusion model to enhance cross-domain generalizability. In Phase I, we implement hybrid self-distillation within the source domain by integrating feature-, response-, and contrastive-level distillation, which enables the model to decompose complex driving behaviors into their core causal components, including scene-conditioned driven patterns, multi-agent interaction dynamics and casual saliency. In Phase II, we introduce a continual distillation strategy: few-shot samples from the target domain are used to initiate generation of diverse synthetic scenarios, allowing the student model to continually adapt to novel environments without retraining on large-scale data. Extensive experiments demonstrate that CDPT achieves strong generalization in both open-loop and closed-loop simulations, effectively generating realistic, interaction-aware behaviors that are critical for scalable and reliable autonomous driving testing.

Human drivers’ control quality in the first seconds after a handover is critical to shared-driving safety; potentially unsafe steering or pedal inputs therefore require detection and correction by the automated vehicle’s safety-fallback system. Yet performance in this window is vulnerable because cognitive states fluctuate rapidly, causing purely rationality-driven, cognition-unaware models to miss early control dynamics. We present an interpretable driver model grounded in bounded rationality with online adaptation that predicts early-stage control quality. We encode boundedness by embedding cognitive constraints in reinforcement learning and adapt latent cognitive parameters in real time via particle filtering from observations of driver actions. In a vehicle-in-the-loop study (n=41), we evaluated predictive performance and physiological validity. The adaptive model not only anticipated hazardous takeovers with higher coverage and longer lead times than non-adaptive baselines but also demonstrated strong alignment between inferred cognitive parameters and real-time eye-tracking metrics. These results confirm that the model captures genuine fluctuations in driver risk perception, enabling timely and cognitively grounded assistance.

During the ongoing development and proliferation of autonomous driving, human-driven vehicles (HDVs) and connected automated vehicles (CAVs) will coexist in mixed traffic environments for the foreseeable future. However, current autonomous driving systems often face challenges in ensuring optimal safety and efficiency, particularly in complex conflict scenarios. To address these shortcomings and improve cooperation in mixed traffic environments, this paper presents a game theoretic decision-making method. The proposed framework accounts for both CAV-CAV cooperation and CAV-HDV interaction in mixed traffic at un-signalized intersections. It introduces a parameter updating mechanism based on twin games to dynamically adjust HDVs’ parameters to better predict and respond to variable human driving behaviors. To validate the effectiveness of the proposed cooperative driving framework, the comparative analysis of its safety and efficiency with other established methods is conducted. The results demonstrate that our method successfully ensures both safety and efficiency in mixed traffic environments. Compared with reinforcement learning approaches such as IPPO, it achieves a 35–55% improvement in success rate while maintaining decision stability and traffic efficiency. In contrast to methods that enforce strict safety guarantees, our approach improves the average vehicle speed by 0.1–0.6 m/s and the average CAV speed by 0.7–1.7 m/s, without compromising safety. Additionally, several validation experiments are conducted using a hardware-in-the-loop and human-in-the-loop experimental platform, confirming the practical applicability of the method.

Autonomous Vehicles (AVs) have entered the stage of commercialization, yet their performance in interactive scenarios remains unsatisfactory due to challenges such as decision interpretability, human driver (HV) heterogeneity, and scenario diversity. Recent advances in Large Language Models (LLMs) provide a promising avenue to enhance AV interaction capabilities, but their high computational demand hinders practical deployment. To address these challenges, this paper introduces a parallel Actor–Reasoner framework designed to enable explicit and real-time bidirectional AV-HV interactions. First, the Reasoner employs a localized LLM with CoT reasoning and human instructions to progressively infer HV intent, style, AV action, and eHMI displays during the training stage. During testing, it continues to infer he above information except AV action. The Actor, in turn, is constructed as an interaction memory through the Reasoner’s interactions with heterogeneous simulated HVs across diverse scenarios, where the memory partition and two-layer retrieval modules are employed in the construction process. During testing, the Actor is used to retrieve feasible actions for the AV. Ablation studies across multiple scenarios demonstrate that the proposed modules improve interaction success rates by an average of 15% and 12%, respectively. Moreover, comparison studies in multi-vehicle scenarios further show that the proposed Actor–Reasoner framework achieves superior safety while simultaneously improving efficiency. Finally, with the integration of external Human–Machine Interface (eHMI) information derived from the Reasoner’s reasoning and feasible actions retrieved from the Actor, the framework is validated in real-world field interactions. Our code is available athttps://github.com/FanGShiYuu/Actor-Reasoner

Microscopic traffic simulation models are promising for modeling detailed traffic features from the individual vehicle level, but they present challenges in optimization due to their computationally time-consuming, and non-analytic nature. This study addresses the difficulty of incorporating microscopic simulations into the optimization framework for multiple types of Active Traffic Management (ATM) strategies design by adopting the Simulation-Based Optimization (SBO) paradigm. A novel discrete SBO algorithm for highway ATM strategies designs, embedded within a Cell Transmission Model (CTM)-based meta model and a problem-specific Adaptive Hyperbox Algorithm (AHA) is proposed. The effectiveness of the proposed SBO method is validated through two cases: a toy highway and a real-world highway. The method is benchmarked versus the commonly utilized general-purpose model based-SBO algorithm. The experiments indicate that the proposed problem specific SBO algorithm outperforms the general-purpose model-based algorithms to obtain promising solutions in ATM strategies design problems, especially when the simulation resources are tight. Under ATM strategies derived by the proposed algorithm, the observed total delay on the highway shows a substantial 70% decrease. The ATM strategy obtained through the microscopic model-based optimization is proven to be superior to the macroscopic model-based approach. The introduced SBO algorithm in this work advances the current SBO theoretical framework by enhancing its applicability to highway traffic issues and overcoming obstacles related to the optimization of discrete variables.

The application of external human–machine interface (EHMI) on autonomous vehicles (AVs) facilitates information exchange. Existing research fails to consider the impact of the sequence of actions, as well as the effects of EHMI applications and deception, raising the question of whether benevolent, well-intentioned deception should be permitted (i.e., misleading statements that are intended to benefit both parties). We established a game theory based EHMI information disclosure framework for AVs in this study. In considering benevolent deception, this framework divided the decision-making process into three stages, respectively encompassing three key questions: whether to disclose, when to disclose, and what type of intention information to disclose. The results show that theoretical advantages of deception exist in certain cases when AV expects to maximize the safety of the interaction. In 40 out of 484 cases (8.3%), safety can be enhanced through successful deception. Those successful deceptions fall into two categories: 1) In 28 of these cases, the straight-going AV expected the left-turning human-driven vehicle (HV) to yield, while HV exhibited lower speed and higher acceleration; 2) In 12 of these cases, AV expected HV to proceed first, while HV exhibited higher speed and lower acceleration. We also conducted a VR-based driving simulation experiment, and the results confirmed our conclusion. Additionally, we found that when participants had low trust in the EHMI, its use negatively impacted interaction efficiency instead. This study serves as an exploratory behavioral mechanism study based on specific hypotheses for future EHMI design and ethical decision-making of autonomous driving system.

202602 bcjz

Autonomous driving has entered the testing phase, but due to the limited decision-making capabilities of individual vehicle algorithms, safety and efficiency issues have become more apparent in complex scenarios. With the advancement of connected communication technologies, autonomous vehicles equipped with connectivity can leverage vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications, offering a potential solution to the decision-making challenges from individual vehicle's perspective. We propose a multi-level vehicle-infrastructure cooperative decision-making framework for complex conflict scenarios at unsignalized intersections. First, based on vehicle states, we define a method for quantifying vehicle impacts and their propagation relationships, using accumulated impact to group vehicles through motif-based graph clustering. Next, within and between vehicle groups, a pass order negotiation process based on Large Language Models (LLM) is employed to determine the vehicle passage order, resulting in planned vehicle actions. Simulation results from ablation experiments show that our approach reduces negotiation complexity and ensures safer, more efficient vehicle passage at intersections, aligning with natural decision-making logic.

The safe deployment of autonomous vehicles (AVs) into real-world traffic requires robust interaction with human drivers exhibiting heterogeneous behavioral tendencies, spanning from rational cooperation to adversarial aggression. Existing simulation frameworks often lack the capacity to systematically model such behavioral diversity, limiting their applicability for rigorous A V evaluation. To address this challenge, we propose a multi-agent reinforcement learning framework that generates dynamically controllable traffic through Tendency-Guided Policy Optimization (TGPO). Central to TGPO is the Adversary-Rationality-Tendency (ART), a continuous hyperparameter that enables fine-grained control over the spectrum of driving behaviors by fusing separately learned adversarial and rational value functions. Furthermore, we design an ART -guided policy network incorporating multi-head mechanisms to resolve high-dimensional multi-agent observations, adaptively prioritizing context features aligned with assigned driving tendencies. Extensive experiments across urban and highway scenarios demonstrate that TGPO generates traffic with enhanced behavioral controllability and diversity, which provides a scalable solution for simulating realistic interactions with various driving tendencies, thereby facilitating the development of AV systems capable of handling complex real-world corner cases.

A lack of understanding of interactions and the inability to effectively resolve conflicts continue to impede the progress of Connected Autonomous Vehicles (CAVs) in their interactions with Human-Driven Vehicles (HDVs). To address this challenge, we propose the Recognize then Resolve (RtR) framework. First, a Bilateral Intention Progression Graph (BIPG) is constructed based on CAV-HDV interaction data to model the evolution of interactions and identify potential HDV intentions. Three typical interaction breakdown scenarios are then categorized, and key moments are defined for triggering cooperative conflict resolution. On this basis, a constrained Monte Carlo Tree Search (MCTS) algorithm is introduced to determine the optimal passage order while accommodating HDV intentions. Experimental results demonstrate that the proposed RtR framework outperforms other cooperative approaches in terms of safety and efficiency across various penetration rates, achieving results close to consistent cooperation while significantly reducing computational resources. Our code and data are available at: https://github.com/FanGShiYuu/RtR-Recognize-then-Resolve/.

Advancements in Autonomous Vehicles (AVs) technology necessitate Active Traffic Management (ATM) strategies that incorporate fine-grained microscopic traffic features, accounting for AVs' distinct driving behaviors, decision-making patterns, and the specific road geometry design required for increasing AV penetration. While microscopic traffic simulation models are well-suited for capturing these microscopic features, their high computational demands and non-analytic nature have confined their use to ATM strategies optimization. To bridge this gap, this study employs a Simulation-Based Optimization (SBO) paradigm to integrate microscopic models into the design of ATM strategies for highways. We introduce an SBO framework that combines a Cell Transmission Model (CTM)-based meta model with the Adaptive Hyperbox Algorithm (AHA), where the CTM-based meta model approximates microscopic traffic conditions, and the AHA explores high-dimensional, discrete decision spaces to determine optimal ATM strategies. Our experiments demonstrate that the proposed SBO frame-work substantially outperforms other general-purpose SBO methods. This work advances ATM optimization by facilitating the integration of microscopic simulation models into practical optimization frameworks, ensuring the consideration of evolving traffic characteristics with the increasing presence of AVs.

Networked urban systems facilitate the flow of people, resources, and services, and are essential for economic and social interactions. These systems often involve complex processes with unknown governing rules, observed by sensor-based time series. To aid decision-making in industrial and engineering contexts, data-driven predictive models are used to forecast spatiotemporal dynamics of urban systems. Current models, such as graph neural networks, have shown promise but face a tradeoff between efficacy and efficiency due to computational demands. Hence, their applications in large-scale networks still require further efforts. This article addresses this tradeoff challenge by drawing inspiration from physical laws to inform essential model designs that align with fundamental principles and avoid architectural redundancy. By understanding both micro- and macro-processes, we present a principled interpretable neural diffusion scheme based on transformer-like structures, whose attention layers are induced by low-dimensional embeddings. The proposed scalable spatiotemporal transformer (ScaleSTF), with linear complexity, is validated on large-scale urban systems including traffic flow, solar power, and smart meters, showing state-of-the-art performance and remarkable scalability. Our results constitute a fresh perspective on the dynamics prediction in large-scale urban networks.

Given the complex nature of interaction under ambiguous right-of-way scenarios, the interactions between Autonomous Vehicles (AVs) and Human-driven Vehicles (HVs) present considerable challenges to the safety and efficiency of the traffic system. Existing AVs struggle to comprehend and apply common HV social norms, especially the proactive behavior exhibited by adept human drivers in ambiguous right-of-way scenarios. In this study, we propose a novel framework to leverage expert priors for proactive-aware decision-making in ambiguous right-of-way, merging Reinforcement Learning (RL) with parameterized modeling. Building upon unprotected-turning interactions from real-world driving datasets, we select typical cases under ambiguous right-of-way as human-expert priors, which are utilized to guide the learning of the RL agent. Then, a Hidden Markov Model (HMM), which is governed by interpretable parameters derived from expert priors, introduces human decision updating mechanism into AV strategy. Experimenting with typical driving tasks, our approach achieves balanced safety and efficiency in tackling ambiguities of right-of-way, with superior decision-making performance via the guidance of expert priors when compared with established baselines. Furthermore, the results indicate that the proposed method enables AVs to accelerate the convergence during the interaction by consistent probing and decision updates.

Scenario-based testing has become a crucial method for certifying the safety of autonomous vehicles amidst the rapid advancement of autonomous driving technology. The exploration of worst-case scenarios, which plays a vital role in safety assessments and accelerated improvements of autonomous driving systems, has posed challenges due to the high dimensionality of parameters and the uncertainty of the decisions of autonomous driving systems. To address these challenges, a Bayesian optimization framework was proposed herein as a solution. Surrogate models, including two Gaussian process-based and five non-Gaussian process-based models, are employed to replace part of the costly simulation models, to overcome the challenge of high dimensionality. The Jackknife and Bootstrap methods are used to estimate variance and calculate the acquisition function. Additionally, a differential evolutionary algorithm is employed to achieve efficient solutions while balancing exploration and exploitation, to overcome the challenge of output uncertainty. The proposed framework is validated through experimental results. It successfully identifies the worst-case scenarios in all three cases and the average number of tests for the optimal model in the most complex 9-dimensional scenario is less than 400. Notably, the approach using the Random Forest (RF) surrogate model achieves the highest search success rate in all three cases.

The development of autonomous vehicles has shown great potential to enhance the efficiency and safety of transportation systems. However, the decision-making issue in complex human-machine mixed traffic scenarios, such as unsignalized intersections, remains a challenge for autonomous vehicles. While reinforcement learning (RL) has been used to solve complex decision-making problems, existing RL methods still have limitations in dealing with cooperative decision-making of multiple connected autonomous vehicles (CAVs), ensuring safety during exploration, and simulating realistic human driver behaviors. In this paper, a novel and efficient algorithm, Multi-Agent Game-prior Attention Deep Deterministic Policy Gradient (MA-GA-DDPG), is proposed to address these limitations. Our proposed algorithm formulates the decision-making problem of CAVs at unsignalized intersections as a decentralized multi-agent reinforcement learning problem and incorporates an attention mechanism to capture interaction dependencies between ego CAV and other agents. The attention weights between the ego vehicle and other agents are then used to screen interaction objects and obtain prior hierarchical game relations, based on which a safety inspector module is designed to improve the traffic safety. Furthermore, both simulation and hardware-in-the-loop experiments were conducted, demonstrating that our method outperforms other baseline approaches in terms of driving safety, efficiency, and comfort.

Ensuring and improving the safety of autonomous driving systems (ADS) is crucial for deployment of highly automated vehicles, especially in safety-critical events. To address the rarity issue, adversarial scenario generation methods are developed, in which behaviors of traffic participants are manipulated to induce safety-critical events. However, existing methods still face two limitations. First, identification of the adversarial participant directly impacts the effectiveness of the generation. However, complexity of real-world scenarios, with numerous participants and diverse behaviors, makes identification challenging. Second, potential of generated safety-critical scenarios to continuously improve ADS performance remains underexplored. To address these issues, we propose LLM-attacker: a closed-loop adversarial scenario generation framework leveraging large language models (LLMs). Specifically, multiple LLM agents are designed and coordinated to identify optimal attackers. Then, the trajectories of attackers are optimized to generate adversarial scenarios. These scenarios are iteratively refined based on the performance of ADS, forming a feedback loop to improve ADS. Experimental results show that LLM-attacker can create more dangerous scenarios than other methods, and the ADS trained with it achieves a collision rate half that of training with normal scenarios. This indicates the ability of LLM-attacker to test and enhance the safety and robustness of ADS. The framework’s closed-loop design enables continuous scenario evolution compliant with regulatory standards, supporting both safety assurance and policy verification for ADS. Video demonstrations are provided at: https://drive.google.com/file/d/15rROV_ 8LUcc2jXSuSNBHVOHMCFKSn__B/view

Spatiotemporal traffic data (STTD) displays complex correlational structures. Extensive advanced techniques have been designed to capture these structures for effective forecasting. However, because STTD is often massive in scale, practitioners need to strike a balance between effectiveness and efficiency using computationally efficient models. An alternative paradigm based on multilayer perceptron (MLP) called MLP-Mixer has the potential for both simplicity and effectiveness. Taking inspiration from its success in other domains, we propose an adapted version, named NexuSQN, for STTD forecast at scale. We first identify the challenges faced when directly applying MLP-Mixers as series- and window-wise multivaluedness. To distinguish between spatial and temporal patterns, the concept of ST-contextualization is then proposed. Our results surprisingly show that this simple-yet-effective solution can rival SOTA baselines when tested on several traffic benchmarks. Furthermore, NexuSQN has demonstrated its versatility across different domains, including energy and environment data, and has been deployed in a collaborative project with Baidu to predict congestion in megacities like Beijing and Shanghai. Our findings contribute to the exploration of simple-yet-effective models for real-world STTD forecasting.

This paper revisits the Reactive Dynamic User-Equilibrium (RDUE) model for dynamic traffic assignment (DTA) of macroscopic traffic flow in two-dimensional continuum space, focusing on the Eikonal equation—a crucial partial differential equation (PDE) with specific boundary conditions. Traditionally, solving Eikonal equations has relied on iterative numerical methods through the discretization of the continuum space. However, this discretization compromises the precision of numerical solutions and could lead to non-convergence issues during iterative processes. This study refers to Physics-Informed Neural Networks (PINNs) and develops the Distance-Informed Neural Eikonal Solver (NES-DI) for solving Reactive Dynamic User-Equilibrium models. While the previously proposed Neural Eikonal Solver (NES) performs badly in a strong heterogeneous cost field with large cost differences, NES-DI explicitly considers the influence of solid boundaries during the factorization process by incorporating accurate distance information. Numerical examples of RDUE at both the static and dynamic levels are presented to illustrate the performance and applications of the NES-DI framework. The results demonstrate that NES-DI greatly outperforms both NES and the fast sweeping method. Moreover, NES-DI overcomes the limitations of discretization, enabling predictions of solutions at arbitrary locations within the computational domain. At the dynamic level, transfer learning is employed to leverage historical solutions to solve RDUE problems more efficiently. Overall, NES-DI shows the potential of solving reactive dynamic problems with strong heterogeneity, which offers a promising alternative to discretization-reliant numerical methods.

The balance between model capacity and generalization has been a key focus of recent discussions in long-term time series forecasting. Two representative channel strategies are closely associated with model expressivity and robustness, including channel independence (CI) and channel dependence (CD). The former adopts individual channel treatment and has been shown to be more robust to distribution shifts, but lacks sufficient capacity to model meaningful channel interactions. The latter is more expressive for representing complex cross-channel dependencies, but is prone to overfitting. To balance the two strategies, we present a channel-aware low-rank adaptation method to condition CD models on identity-aware individual components. As a plug-in solution, it is adaptable for a wide range of backbone architectures. Extensive experiments show that it can consistently and significantly improve the performance of both CI and CD models with demonstrated efficiency and flexibility. The code is available at https://github.com/tongnie/C-LoRA.

With the rise of autonomous driving technology, Autonomous Vehicles (AVs) are becoming increasingly prevalent on public roads, influencing traffic conditions and the operation of Manual Vehicles (MVs). It is anticipated that a mixed traffic flow comprising both AVs and MVs will persist for an extended period. Current research concerning the impact of AVs often lacks a comprehensive focus on network-level effects and presents varying perspectives on network performance. Some studies have shown that AVs have a smaller headway and are therefore able to improve road capacity and reduce energy consumption and emissions. Some other scholars have found that existing commercial AVs may generate and spread oscillations, which can aggravate congestion in bottleneck areas if no coordination strategy is implemented. Such discrepancies may stem from differing application scenarios, datasets, and assumptions utilized in each study. Furthermore, these investigations have predominantly examined individual vehicles or fleets within localized road networks, such as single bottlenecks or small-scale networks, thus offering limited guidance at a macroscopic level. This paper seeks to elucidate the impact of AVs on macroscopic traffic patterns, particularly within mixed traffic scenarios, through simulation experiments conducted on both a grid network and a real-world network in Beijing.

Missing data is a pervasive issue in both scientific and engineering tasks, especially for the modeling of spatiotemporal data. Existing imputation solutions mainly include low-rank models and deep learning models. The former assumes general structural priors but has limited model capacity. The latter possesses salient expressivity, but lacks prior knowledge of the underlying spatiotemporal structures. Leveraging the strengths of both two paradigms, we demonstrate a low rankness-induced Transformer to achieve a balance between strong inductive bias and high expressivity. The exploitation of the inherent structures of spatiotemporal data enables our model to learn balanced signal-noise representations, making it generalizable for a variety of imputation tasks. We demonstrate its superiority in terms of accuracy, efficiency, and versatility in heterogeneous datasets, including traffic flow, solar energy, smart meters, and air quality. Promising empirical results provide strong conviction that incorporating time series primitives, such as low-rankness, can substantially facilitate the development of a generalizable model to approach a wide range of spatiotemporal imputation problems.

Autonomous vehicles (AV) are consistently criticized for their inadequacies in harmoniously interacting with human-driven vehicles (HV), primarily attributed to the lack of sociality, a key human trait that balances individual and group rewards. Understanding sociality is essential for smooth AV navigation but remains challenging. To address this, we propose a Game-Theoretic Inverse Reinforcement Learning (GT-IRL) approach to quantify individualized sociality in driving interaction. Our approach identifies the sociality-preference parameters of a pre-designed reward function that integrates the ego agent’s rewards and the group rewards shared by all the interacting agents. Instead of presuming an agent is propelled by the maximization of its own rewards, the game-theoretical mechanism is utilized within the IRL structure to capture the fact that human drivers take into account others’ interests. We validated our method using human driving data from an unprotected left-turn scenario. The results demonstrate that the proposed GT-IRL outperforms state-of-the-art methods in better reproducing the evolution of the left-turn interaction at both semantic and trajectory levels. Additionally, cross-dataset analysis reveals variations in sociality due to geographical differences (China vs. the U.S.) and the nature of the interacting entities (AV vs. HV or HV vs. HV).

Existing studies on Autonomous Intersection Management (AIM) primarily focus on regular vehicles (e.g., cars), while ignoring bus priority demands. This paper aims to optimize the bus operations at autonomous intersection with trajectory planning and priority control. First, an intersection trajectory planning approach is proposed for turning movements within the intersection considering significant passenger volume and large size of buses. A two-stage trajectory planning method is adopted that employs a combination of transition and circular curves to ensure the smooth turning movements for vehicles within the intersection. Next, a bus priority control model for autonomous intersection (AIM-BP) is developed to minimize weighted combinations of total bus delay and car delay. In particular, the model incorporates the introduction of the dynamic bus lane designed to clear the vehicles in front of buses, thereby creating a relatively exclusive space for the buses. The proposed model simultaneously optimizes the lane choice on the road section, the route choice within the intersection and the time to enter the intersection for each vehicle, while determining whether to deploy the dynamic bus lane and which lane serves as the dynamic bus lane in the approach. The model is formulated as a Mixed Integer Linear Programming (MILP) problem, compiled in AMPL (A Mathematical Programming Language) and solved by CPLEX. Results demonstrate significant reductions in average bus delay and passenger delay under the proposed model, and sensitivity analysis further examines its effectiveness.

To validate whether Highly Automated Vehicles (HAVs) can live up to human expectations, it is essential to estimate their risk rate within the context of the naturalistic driving environment. Due to the low probability of exposure to risky events, the testing process is exceedingly time-consuming. To tackle this issue, we proposed a Surrogate-based Monte Carlo Method to accelerate the risk assessment for HAVs in scenario-based simulation. Surrogate Models (SMs) were utilized to approximate the outcomes of untested scenarios, hence facilitating the identification of risky scenarios. Therefore, the large number of samples required by Monte Carlo did not need to be tested entirely. Naturalistic distributions fitted from HighD data was used to generate samples. The Car-following and Cut-in scenarios were selected for the case study, as they represented two distinct testing spaces. As such, the capabilities of different SMs can be further analyzed. We proved that the performances of six mainstream SMs were greatly distinguished from each other. Inverse Distance Weighted (IDW), the proven most capable SM, was combined with Monte Carlo. Compared with Monte Carlo, up to 95% of tested samples can be saved. Compared with the Importance Sampling method, another popular improvement in Monte Carlo simulation, the proposed method can further save 58.1% of CPU execution time.

Interacting with human road users is one of the most challenging tasks for autonomous vehicles. For congruent driving behaviors, it is essential to recognize and comprehend sociality, encompassing both implicit social norms and individualized social preferences of human drivers. To understand and quantify the complex sociality in driving interactions, we propose a Virtual-Game-based Interaction Model (VGIM) that is parameterized by a social preference measurement, Interaction Preference Value (IPV). The IPV is designed to capture the driver’s relative inclination towards individual rewards over group rewards. A method for identifying IPV from observed driving trajectory is also developed, with which we assessed human drivers’ IPV using driving data recorded in a typical interactive driving scenario, the unprotected left turn. Our findings reveal that (1) human drivers exhibit particular social preference patterns while undertaking specific tasks, such as turning left or proceeding straight; (2) competitive actions could be strategically conducted by human drivers in order to coordinate with others. Finally, we discuss the potential of learning sociality-aware navigation from human demonstrations by incorporating a rule-based humanlike IPV expressing strategy into VGIM and optimization-based motion planners. Simulation experiments demonstrate that (1) IPV identification improves the motion prediction performance in interactive driving scenarios and (2) the dynamic IPV expressing strategy extracted from human driving data makes it possible to reproduce humanlike coordination patterns in the driving interaction.

Interacting with other vehicles while ensuring safety is a routine task for human drivers, but it can pose a challenge for autonomous vehicles. To address this challenge, we derived a generic utility term of interaction activeness (UTIA) from the driving interaction formulation which considers the rationality of interacting counterparts. Our research shows that incorporating UTIA as a supplementary utility term can improve the active interaction capability of both sampling-based and game-theoretic baseline motion planners without compromising safety. Through simulation experiments, we observed that on average, incorporating the weighted UTIA into the utility function of baseline planners can result in an 8.8% increase in the success rate of exiting a highway within a set distance.

Previous empirical behavior analysis on Autonomous Vehicles (AV) mainly focused on vehicles with Adaptive Cruise Control (ACC) system due to the lack of high-level AV dataset. Recently released SAE Level-4 AV datasets such as the Waymo Open Dataset provide great opportunities to evaluate their behavioral impact on traffic flow. In this study, we aim to characterize the empirical Car Following (CF) behaviors of the Waymo autonomous vehicle and compare its feature with human-driven Vehicles (HV), and capture such behavioral differences using the IDM CF model. Our main findings include: (a) AV is much safer than HV, based on our analysis using surrogate safety measures, as time headways and jam spacings of the AV are significantly larger than HV; (b) the response time of AV is also significantly larger than that of HV in response to various types of stimuli; (c) despite the short length of trajectories in the Waymo Open Dataset, we have confirmed that these trajectories are suitable for calibrating some of the IDM parameters; and the calibration results of IDM are consistent with our empirical analysis. Moreover, the modelling results, reveal that the proportion of string unstable behavior of AV is less than that of HV; and (d) for HV, there is generally no significant difference between following AV and following HV except a smaller jam spacing when following AV. Overall, we conclude that currently AV behaves in a conservative way to ensure its safety at the cost of traffic efficiency.

With the rapid development of autonomous driving, Autonomous Vehicles (AVs) have started to appear on public roads, which has inevitably affected current traffic conditions and the operations of Manual Vehicles (MVs). Current research on AVs’ influence has mainly been conducted at individual level of driving behaviors, while few studies have focused on the overall network level to consider the traffic flow pattern due to mixed traffic. In this work, considering varying signal control schemes and demand loading patterns, we conducted simulation experiments based on a grid network and a real-world network in Beijing using SUMO. Traffic flow with the mixture of MVs, low-level AVs (LAVs), and high-level AVs (HAVs) were emulated so to investigate how the network performs at various levels of mixed traffic. Driving behaviors between the three types of vehicles were calibrated using driving data drawn from OpenACC dataset, and Waymo Open Dataset. The capacity and critical accumulation of the Macroscopic Fundamental Diagram (MFD) were chosen as the key indicators of network performance. We found that AVs positively boost network capacity (up to 19.0% increase) but a negative influence on critical accumulation was also observed (up to 9.0% decrease). However, the positive impact of OpenACC and Waymo’s AVs on macroscopic traffic is still far from ideal since they may be too conservative. AVs can boost flow when traffic is in unsaturated or saturated states. However, when traffic flow is oversaturated, AVs can instead cause flow and average speed to drop faster than that in the MV-only scenario.

Path dispersion (the spatial distribution of vehicular paths) is an important feature of traffic flow inside intersections and differs from traffic flow running along traffic lanes at road segment, especially under conflicting movements. The path dispersion reflects the operational features of traffic flow and is related to driving behaviour, arrival flow patterns, layout design, and the traffic control and management scheme. This study aims to improve the understanding of the overall path dispersion of unprotected left-turns and the opposing through movement. A behavioural simulation model was established to represent the overall path dispersion. Human behaviours regarding vehicle trajectory planning with and without conflicting vehicles were modelled based on optimal control and integrated into the proposed discrete event simulation framework. The descriptive power and accuracy of the proposed simulation model were validated using empirical data. The effects of the spatial size of the intersection, crossing angle, and traffic volume on the path dispersion of the left-turn and through movement were explored based on numerical experiments. The results show that the proposed simulation model can represent the path dispersion of left-turn and opposing through movement well for both the calibrated intersections and newly added intersections without model parameter recalibration with an average error of 8.92%.

One essential goal for Highly Automated Vehicles (HAVs) safety test is to assess their risk rate in naturalistic driving environment, and to compare their performance with human drivers. The probability of exposure to risk events is generally low, making the test process extremely time-consuming. To address this, we proposed a surrogate-based method in scenario-based simulation test to expediate the assessment of the risk rate of HAVs. HighD data were used to fit the naturalistic distribution and to estimate the probability of each concrete scenario. Machine learning model-based surrogates were proposed to quickly approximate the test result of each concrete scenario. Considering the different capabilities and domains of various surrogate models, we applied six surrogate models to search for two types of targeted scenarios with different risk levels and rarity levels. We proved that the performances of different surrogate models greatly distinguish from each other when the target scenarios are extremely rare. Inverse Distance Weighted (IDW) was the most efficient surrogate model, which could achieve risk rate assessment with only 2.5% test resources. The required CPU runtime of IDW was 2% of that required by Kriging. The proposed method has great potential in accelerating the risk assessment of HAVs.

Vehicle trajectories can provide a clear picture of traffic flow, which facilitates traffic state estimation and signal control optimization at intersections. Connected and Automated Vehicles (CAVs) can not only report their own trajectories, but also continuously collect surrounding vehicles’ trajectories using onboard sensors, which creates an opportunity to reconstruct fully-sampled vehicle trajectories. However, this data source brings challenges such as low penetration rate of CAVs and complex detection environment at intersections. To address these problems, this study proposes a novel framework under micro-perspective, in which trajectory estimation and fusion algorithms are integrated. The spatiotemporal correlations of detected trajectories are analyzed and classified into four regions, and four corresponding trajectory estimation algorithms based on extended car-following model are established to estimate the undetected part of each trajectory. Furthermore, a trajectory fusion algorithm based on Particle Filter is developed to fuse the estimated trajectories separately derived from upstream and downstream with minimized errors. The proposed method was comprehensively evaluated at field and simulated signalized intersections. The results show that compared with Variational Theory method, queue location error, time error and cumulative distance error of the proposed method were 76.3%, 44.4% and 54.5% lower, respectively, and the proposed trajectory fusion algorithm improved the accuracy and smoothness with the above three indices decreased by 17.3%, 47.7% and 6.2%, respectively. It was also found that the proposed method can adapt to different traffic conditions and penetration rates of CAVs.

Automated Vehicles (AVs) need to be thoroughly evaluated in order to ensure their driving capabilities. However, comprehensive evaluations are intractable due to both time and monetary costs. To address this problem, we propose an Adaptive Design of Experiments (ADOE) method to evaluate the safety of AVs. Using this method, a Surrogate Model (SM) is established and updated iteratively. SM in ADOE is used to approximate the results of AV testing and help to select the next concrete scenario to be tested in each iteration. Two different ADOE approaches are proposed in this study for different testing purposes. Since the choice of the surrogate model has a profound impact on the performance of the ADOE method, 6 surrogate models were compared with two logical scenarios at different scales – a car following logical scenario and a cut-in logical scenario. Results show that Extreme Gradient Boosting (XGB) is suitable for both ADOE approaches. And both proposed ADOE approaches achieved desired performance. Scenario-oriented ADOE made full use of each concrete scenario, capturing one collision case for every 1.12 test runs in the car following logical scenario, while SM-oriented ADOE successfully depicted the boundary between safety and danger. Using 0.46% test resources compared to enumeration, the SM-oriented ADOE found 93.9% dangerous scenarios with 90.9% precision. ADOE approaches have great potential in accelerating the evaluation of AV safety.

Highly Automated Vehicles (HAVs) must undergo strict safety testing before being released to the public. Mileage-based on-road testing suffers from unaffordable time costs and high safety risks. Simulated scenario-based testing has been found to be a trustworthy alternative for testing HAVs’ built-in algorithms and functionalities. Test automation is typically used to generate target scenarios. This approach facilitates customized testing and avoids wasting time on simple and redundant scenarios. This study aims to review test automation methods and discuss how to accentuate their strengths rather than be trapped in their weaknesses under certain applicable conditions. According to their main purposes, we classify test automation methods into coverage-oriented, unsafe-scenario-oriented, and naturalistic-assessment-oriented categories. To further demonstrate the differences of these methods, we then design numerical experiment to compare the capabilities of seven test automation methods. Finally, we compile our observations to form a comprehensive guide for selecting test automation methods with different test requirements in mind.

Deep learning models for short-term travel speed prediction on urban expressways, such as the convolutional neural network (CNN), still present several limitations in multiscale spatiotemporal feature extraction. Hence, in this paper, three hybrid CNN models are proposed to improve the basic CNN model with regard to three target aspects for short-term (i.e., 5 min) travel speed prediction on urban expressways. More specifically, long short-term memory (LSTM), AutoEncoder (AE), and Inception module are incorporated into the basic CNN model to capture multiscale spatiotemporal features of travel speed data effectively and improve the accuracy and robustness of the basic CNN model. Based on loop detector data collected on the Yan’an expressway in Shanghai, the proposed hybrid CNN models are trained and tuned. To validate the improvements on the target aspects, a comprehensive comparison is conducted using a classical statistical model (i.e., autoregressive integrated moving average), a typical shallow neural network model (i.e., artificial neural network), and two basic deep learning models (i.e., recurrent neural network and CNN). Results show that the prediction accuracies of all the proposed hybrid CNN models exceed 96% and the mean absolute errors are less than 2.5 km/h, which are superior to other models. In terms of target improving aspects, two new metrics were introduced, and the proposed models, especially the AE–CNN model, showed better robustness under various input data structures and traffic states. The LSTM–CNN model outperformed the other models in learning time-series features, and the Inception–CNN model is superior in reproducing the dynamics of traffic congestion patterns on urban expressways.

Highly automated vehicles (HAVs) have been introduced to the transportation system for the purpose of providing safer mobility. Considering the expected long co-existence period of HAVs and human-driven vehicles (HDVs), the safety operation of HAVs interacting with HDVs needs to be verified. To achieve this, HAVs’ Operational Design Domain (ODD) needs to be identified under the scenario-based testing framework. In this study, a novel testing framework aiming at identifying the Safety performance boundary (SPB) is proposed, which assures the coverage of safety-critical scenarios and compatible with the black-box feature of HAV control algorithm. A surrogate model was utilized to approximate the safety performance of HAV, and a gradient descent searching algorithm was employed to accelerate the search for SPB. For empirical analyses, a three-vehicle following scenario was adopted and the Intelligent Driver Model (IDM) was tested as a case study. The results show that only 4% of the total scenarios are required to establish a reliable surrogate model. And the gradient descent algorithm was able to establish the SPB by identifying 97.42% of collision scenarios and only false alarming 0.29% of non-collision scenarios. Furthermore, the concept of safety tolerance was proposed to measure the possibilities of boundary scenarios dropping in safety performance. The applications of helping to construct ODD and compare different control algorithms were discussed. It shows that the IDM performs better than the Wiedemann 99 (W99) model with larger ODD.

A freeway stretch with even one pair of unmeasured on/off-ramps is not fully observable in traffic states. Flow observability is essential for freeway traffic modeling, surveillance, and control. It is a longstanding and tricky issue to estimate flows for unmeasured ramp pairs. This problem seems to be hardly tractable in conventional approaches, and this paper intends to handle it based on machine learning. The work was partially inspired by transfer learning. Consider that no measurements are available for a target ramp pair, and the knowledge about ramp flow estimation may be drawn from other (measured) ramp pairs, provided that measured and unmeasured ramp pairs would share similarities in some key traffic flow patterns. Two simple machine learning algorithms, random forest (RF) and gradient boosting machine (GBM), were employed to this end. RF and GBM were driven by real measurement data to establish models that relate ramp flows to adjacent mainstream traffic conditions. The models were then applied for our task. The estimation performance was evaluated using real measurement data from the Shanghai Urban Expressway and the Intercity Highway in California, with satisfactory results obtained.

Intersection queue length estimation using high-resolution probe vehicle trajectory data has received increasing attentions in recent years. Existing methods for cycle-based queue length estimation still face the challenge of low and/or unstable estimation accuracies under the condition of sparse vehicle trajectory data, i.e., there is no greater than one vehicle trajectory per cycle on average. To address this challenge, this study proposed a novel approach for cycle-based queue length estimation by fusing real-time and historical probe vehicle trajectory data, through a statistical parameter estimation method, i.e., maximum likelihood estimation (MLE). With known signal timing information, firstly, the historical probe trajectory data are used to acquire the arrival flow rate distribution over the entire study period. Then, a likelihood function of queue length is derived by fully exploiting real-time traffic flow information provided by the queued and non-queued probe vehicles. Finally, the MLE method is adopted to estimate the cycle-based queue lengths with the maximum probability. The proposed approach is verified using both simulation and empirical data. Results indicate that precise estimation for cycle-based queue lengths can be realized based on sparse vehicle trajectory data, while showing superiority to a representative existing method. The proposed method is basically an offline method, but it can also work in an online manner if provided a priori arrival distribution either acquired from historical probe vehicle trajectory data or a theoretical assumption.

In this paper, a novel algorithm called bi-objective elite differential evolution (BOEDE) is proposed to optimize multivalued logic (MVL) networks. It is a multiobjective algorithm completely different from all previous single-objective optimization ones. The two objective functions, error and optimality, are put into evaluating the fitness of individuals in evolution simultaneously. BOEDE innovatively uses an archive population with different ranks to store elite individuals and offsprings. Moreover, a characteristic updating method based on this archive structure is designed to produce the parent population. Because of the particularity of MVL network problems, the performance of BOEDE to solve them is further improved by strictly distinguishing elite solutions and Pareto optimal solutions, and by modifying the method of dealing with illegal variables. The simulations show that BOEDE can collect a great number of solutions to provide decision support for a variety of applications. The comparison results also indicate that BOEDE is significantly better than the existing algorithms.

Performing a left turn in a non-protected phase at mixed-flow intersections is one of the most challenging driving maneuvers. In general, there are three typical behavioral features during this turning process: multiple conflicting objects, multi-layer interaction between vehicles, non-motorized vehicles, and pedestrians, and long-term interaction event chains. In current studies, few models consider the impact of these three typical features simultaneously. This paper proposes a Conv-LSTM model to predict the positions of turning vehicles at each moment during the turning process in these complicated environments. The model uses convolution, which is an essential part of a convolutional neural network (CNN), to extract higher-level features across different time segments. Afterward, a long short-term memory (LSTM) network is employed to obtain the feature sequence with long-term dependence on the features of historical periods. Finally, the initial prediction of the Conv-LSTM is modified by the non-holonomic constraints of vehicles. Left-turning trajectories extracted from a simulation environment are used for model training and testing. As a result, the Conv-LSTM model performs better compared with the CNN and LSTM models. The average offset of every point on the trajectories is reduced by 50.4% and 37.1%, respectively, relative to the CNN and LSTM, and the interactive behaviors during the left-turn process are well reproduced by the proposed model. Another round of testing shows that the generalization ability of proposed model to real environments was initially proved to be feasible with the extracted ground-truth trajectories in the field environment.

The cycle-based volume is critical for traffic state estimation and signal control optimization at signalized intersections. Traditional volume estimation mainly depends on fixed detectors represented by loop detectors, but limited spatial coverage and detection failure are also prominent. With the development of vehicle positioning, smartphone-based navigation, and connected-vehicle technologies, massive high-resolution trajectory data have recently become available, which can provide rich and timely information on the traffic arrival and departure processes at signalized intersections. Hence, the studies utilizing trajectory data for estimating the queue length and traffic volume at intersections has received increasing attention in the past few years. However, the most existing studies have demanded a comparatively high penetration rate and adopted site-specific assumptions for unsteady arrival patterns. In contrast, this paper solely used trajectory data for cycle-based flow estimation through a generic hybrid method that combined a probabilistic model and shockwave theory to maximize the utilization of limited captured trajectories, especially under a low penetration rate. In this method, within each cycle, the volume of stopped vehicles is estimated based on the shockwave theory, while the volume of non-stopped vehicles is modeled as a parameter estimation problem of a time-dependent constrained Poisson distribution, where the time headway correspondingly obeys an M3 distribution. The cycle-based volume is solved by a maximum likelihood estimation using an expectation-maximization procedure. An empirical case study was conducted with various signal timing schemes and the results showed satisfactory robustness with an accuracy of more than 90% under a penetration rate of 7.6%.

This paper is concerned with the target tracking problem over a filtering network with dynamic cluster and data fusion. A novel distributed consensus-based adaptive Kalman estimation is developed to track a linear moving target. Both optimal filtering gain and average disagreement of the estimates are considered in the filter design. In order to estimate the states of the target more precisely, an optimal Kalman gain is obtained by minimizing the mean-squared estimation error. An adaptive consensus factor is employed to adjust the optimal gain as well as to acquire a better filtering performance. In the filter's information exchange, dynamic cluster selection and two-stage hierarchical fusion structure are employed to get more accurate estimation. At the first stage, every sensor collects information from its neighbors and runs the Kalman estimation algorithm to obtain a local estimate of system states. At the second stage, each local sensor sends its estimate to the cluster head to get a fused estimation. Finally, an illustrative example is presented to validate the effectiveness of the proposed scheme.

The cross-river tunnel is an important facility for modern cities. It is usually a natural bottleneck due to the changes in gradient, which is known as sag. Once a queue propagates from the sag to upstream bottleneck, the induced multi-bottleneck congestion shows more complicated pattern than an isolated one. The lack of understanding on such congestion pattern hampers the application of appropriate control strategies. In this paper, the congestion pattern of the long-queued tunnel sag is investigated at the Xiangyin Tunnel in Shanghai. Three distinctive congestion stages are observed from day-to-day traffic flow data. The mechanisms of congestion are discussed. Based on the understanding of the congestion pattern, a novel control strategy is proposed which cooperatively controls the tunnel sag mainline and the on-ramp. The control objective is to maximize the throughput and minimize the breakdown probability using quantitative risk analysis method. The control strategy is validated in the field experiment. The traffic features (volume, speed, and productivity) in the tunnel sag are found to be significantly enhanced by the control.

In this paper, a new coupled map (CM) car-following model is proposed to describe the dynamic motion of vehicles moving along a single-lane road under inter-vehicle communication (IVC). In the model, the motion of a vehicle is affected by the information of the preceding vehicles' movements. Moreover, the mechanism of the information effect is considered, which depends on the communication topology of vehicles. The theoretical analysis shows that our model could keep the traffic flow stable under some conditions. The corresponding numerical simulations confirm the correctness of the theoretical analysis. Compared with previous works on CM model, our model is more reasonable and effective in suppressing the formation of traffic congestion.