Gradient-Based Distributed Cruise Control Under Intermittent V2V Communication for Smoothing Traffic Flow

Cooperative formation and control of autonomous vehicles (AVs) promise\nincreased efficiency and safety on public roads. In mixed traffic flow\nconsisting of AVs and human-driven vehicles (HDVs), the prevailing platooning\nof multiple AVs is not the only choice for cooperative formation. In this\npaper, we investigate how different formations of AVs impact traffic\nperformance from a set-function optimization perspective. We first reveal a\nstability invariance property and a diminishing improvement property of\nnoncooperative formation when AVs adopt typical Adaptive Cruise Control (ACC)\nstrategies. Then, we focus on the case of cooperative formation where the AV\ncontrollers are cooperatively designed %redesign the control strategies of AVs\nin different formations and investigate the optimal formation of multiple AVs\nusing set-function optimization. Two predominant optimal formations, i.e.,\nuniform distribution and platoon formation, emerge from extensive numerical\nexperiments. Interestingly, platooning might have the least potential to\nimprove traffic performance when HDVs have poor string stability behavior.\nThese results suggest more opportunities for cooperative formation of AVs,\nbeyond platooning, in practical mixed traffic flow.\n

Human-driven Vehicles (HVs) and Autonomous Vehicles (AVs) will inevitably coexist in the future, resulting in mixed traffic conditions. Modeling car-following behavior in such mixed traffic is crucial for better understanding the differences in driving behavior between AVs and HVs, thereby facilitating the development of AV driving strategies to enhance traffic safety. In such mixed traffic, the car-following modes typically include AVs following HVs, HVs following HVs and HVs following AVs. Previous studies have compared the driving behavior of HVs following HVs and HVs following AVs, but few have considered the driving behavior of AVs following HVs in the comparison. To bridge this gap, this paper first decomposes car-following events in mixed traffic extracted from a Lyft level-5 dataset into driving primitives using the Hidden Markov Model with Gaussian Mixture Model (GMM-HMM). Then, the Latent Dirichlet Allocation with GMM (GMM-LDA) is introduced to cluster the driving primitives into different car-following behavior patterns. Based on the identified patterns, the differences across the three car-following modes are further explored. The results show that the driving behavior under three car-following modes differs from each other. Specifically, compared to the HVs following HVs mode, in the AVs following HVs mode, AVs tend to maintain car-following behavior patterns with a longer distance gap and adopt a lower speed to ensure safety. This study offers a comprehensive understanding of car-following behavior in mixed traffic, which will facilitate the development of human-like driving strategies for AVs and enhance the safety of mixed traffic.

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 development of intelligent connected vehicle and artificial intelligence technologies, mixed traffic scenarios where autonomous and human-driven vehicles coexist are becoming increasingly common. Autonomous vehicles need to accurately predict the lane change behavior of preceding vehicles to ensure safety. However, lane change behavior of human-driven vehicles is influenced by both environmental factors and driver preferences, which increases its uncertainty and makes prediction more difficult. To address this challenge, this paper focuses on the mining of driving preferences and the prediction of lane change behavior. We clarify the definition of driving preference and its relationship with driving style and construct a representation of driving operations based on vehicle dynamics parameters and statistical features. A preference feature extractor based on the SimCLR contrastive learning framework is designed to capture high-dimensional driving preference features through unsupervised learning, effectively distinguishing between aggressive, normal, and conservative driving styles. Furthermore, a dual-branch lane change prediction model is proposed, which fuses explicit temporal features of vehicle states with implicit driving preference features, enabling efficient integration of multi-source information. Experimental results on the HighD dataset show that the proposed model significantly outperforms traditional models such as Transformer and LSTM in lane change prediction accuracy, providing technical support for improving the safety and human-likeness of autonomous driving decision-making.

Mixed traffic capacity estimation of autonomous vehicles impact based on empirical data

The authors present the real-time three-dimensional (3D) traffic simulator RASY (Real-time autonomous vehicle traffic simulator developed by Yonsei University), for the modeling and analysis of various congestion scenarios, including interactions among human-driven and autonomous vehicles. Vehicles with different control models can be generated to investigate the effects of vehicle models and traffic environment on the traffic flow during simulations of congestion. A data log of the simulation can be exported for post-processing analysis. Following an analysis of simulation data under congested conditions, the authors found that a faster reaction time of an individual vehicle can raise the overall flow rate of a road, raising the critical vehicle density for congestion.

In this paper, a control program architecture is proposed for autonomous vehicles. This set of programs is structured in a hierarchy and it is inspired by the human driver behavior. It contains three levels connected by specific links: the strategic, the tactical, and the operational level. Autonomous vehicles and human-driven vehicles will share the roads in future transportation systems. The human-like driving behavior will allow other participants to predict the autonomous vehicle's behavior.

Transitioning to a driverless city: Evaluating a hybrid system for autonomous and non-autonomous vehicles

To explore the impact of autonomous vehicles (AVs) on human-driven vehicles (HDVs), a solution for AV to coexist harmoniously with HDV during the car following period when AVs are in low market penetration rate (MPR) was provided. An extension car following framework with two possible soft optimization targets was proposed in this article to improve the experience of HDV followers with different following strategies by deep deterministic policy gradient (DDPG) algorithm. The pretreated Next Generation Simulation (NGSIM) dataset was used for the experiments. 1027 car following events with being redefined were extracted from it, in which 600 of the events were used for training and 427 of the events were used for testing. The different driving strategies obtained from the classical car following models were embedded into virtual environment built by OpenAI gym. The reward function combined safety, efficiency, jerk, and stability was used to encourage the agent with DDPG algorithm to maximize it. The final result reveals that disturbance of HDV followers decreases by 2.362% (strategy a), 8.184% (strategy b), and 13.904% (strategy c), respectively. The disturbance of HDV follower decreases by 14.961% (strategy a), 12.020% (strategy b), and 13.425% (strategy c), respectively. HDV followers with different strategies get less jerk in both soft optimizations. AV passengers get a loss on jerk and efficiency, but safety is enhanced. Also, AV car following performs better than HDV car following in both soft and brutal optimizations. Moreover, two possible solutions for harmonious coexistence of HDVs and AVs when AVs are in low MPR are proposed.

To address the travel delay caused by start-up lag in queued traffic at signalized inter-sections and fill the research gap of unseparated start-up loss characteristics between human-driven vehicles (HDVs) and autonomous vehicles (AVs), this study proposes an analytical start-up loss delay model applicable to both unsaturated and saturated traffic states. The model explicitly quantifies queue-length-dependent cascading propagation effects of start-up loss, and integrates vehicle type-specific parameters based on traffic queue theory and mixed traffic flow field observation data. Conventional models are limited by underestimated delay from neglecting platoon-level start-up loss propagation and failure to account for intrinsic HDV-AV start-up mechanism differences; to resolve these, we first distinguished the two vehicle types’ start-up behaviors (reaction time, acceleration, platoon coordination), then decoupled their start-up loss mechanisms and quantified their delay contributions via theoretical derivation, with validation against field test data and comparison with classical Webster and Clayton models. Field results revealed an order-of-magnitude difference in start-up response: HDVs had an average 2.05 s reaction time with large individual variability, while AVs maintained a stable 0.3–0.5 s response; HDV platoons reached saturated flow at the sixth vehicle, versus the third for AV platoons due to consistent acceleration and shorter headways. Model validation showed that under unsaturated conditions, red light duration significantly affects HDV delay, and the AV mix ratio is exponentially negatively correlated with additional delay. Under saturated conditions, green light duration increases start-up loss delay for both vehicle types, yet the growth rate of AVs (3.1–12.3%) is far lower than that of HDVs (18.2–67.5%), and arranging AVs in the leading position of mixed platoons can further reduce delay. The proposed model improves the accuracy of delay estimation in mixed HDV-AV traffic environments, and provides a theoretical basis for the optimized design of signal control strategies and the efficient management of intersection travel delay.

With the development of autonomous vehicles, a mixed traffic flow scenario of Connected Autonomous Vehicles (CAV) and Human-Driven Vehicles (HVs) would be popular in the near future. The traditional traffic light control systems (TLCSs) for HVs do not make full use of traffic information collected via VANET; meanwhile emerging traffic light systems for CAV assume complicated and quick reactions, which human drivers may not be able to handle. Therefore, they are not suitable for the mixed scenario. This paper proposed a novel TLCS, named PALM, for tackling the challenge for handling mixed traffic scenarios. PALM considers the traffic flow at each intersection and adjacent ones and adjusts the traffic lights schedule for the next few phases accordingly. It also optimizes the signal timing and phases to better serve the platoons formed by CAV. The simulation results show that our approach achieves up to 75.34% and 33.02% drop in the average waiting time compared to the static and actuated TLCS, respectively.

This study aims to quantify the impact of discretionary lane-changing (DLC) on following vehicles (FVs) in the target lane using real-world dataset. The Waymo Open Dataset is used to identify the differences between autonomous vehicles (AVs) DLC and human-driven vehicles (HDVs) DLC maneuvers and compare their impacts on the driving volatility. Then, a block maxima (BM) model is applied to estimate crash risks. Finally, multivariate adaptive regression splines (MARS) is adopted to model gap acceptance behaviors of AV and HDV. Compared to HDV DLC, AV DLC leads to lower speed and yaw rate volatility and smaller acceleration rates of FVs. Further, the BM model reveals that the crash risk in AV DLC events is half of that in HDV DLC events. Additionally, MARS show that AV and HDV accept different lead gap. These findings highlight the benefits of mixing AVs in traffic and guide the improvement of AV controllers.

Liability Design for Autonomous Vehicles and Human-Driven Vehicles: A Hierarchical Game-Theoretic Approach

A stochastic minimax optimal time-delay state feedback control strategy for uncertain quasi-integrable Hamiltonian systems is proposed. First, a stochastic optimal state feedback control problem of uncertain quasi-integrable Hamiltonian system with time delay in feedback control subjected to Gaussian white noise is formulated. Then, the time-delayed state feedback control forces are approximated by the control forces without time delay and the original problem is converted into a stochastic optimal state feedback control problem of an uncertain system without time delay. After that, by following a procedure based on the stochastic averaging method and stochastic differential game, the worst-case disturbances and the optimal controls are obtained from solving a Hamiltonian–Jacobi–Isaacs (HJI) equation. As an example, the stochastic minimax optimal state feedback control of a Duffing oscillator with parametric disturbances and time-delayed feedback control is worked out in detail to illustrate the procedure and effectiveness of the proposed control strategy.

Autonomous driving has attracted significant research interests in the past two decades as it offers many potential benefits, including releasing drivers from exhausting driving and mitigating traffic congestion, among others. Despite promising progress, lane-changing remains a great challenge for autonomous vehicles (AV), especially in mixed and dynamic traffic scenarios. Recently, reinforcement learning (RL) has been widely explored for lane-changing decision makings in AVs with encouraging results demonstrated. However, the majority of those studies are focused on a single-vehicle setting, and lane-changing in the context of multiple AVs coexisting with human-driven vehicles (HDVs) have received scarce attention. In this paper, we formulate the lane-changing decision-making of multiple AVs in a mixed-traffic highway environment as a multi-agent reinforcement learning (MARL) problem, where each AV makes lane-changing decisions based on the motions of both neighboring AVs and HDVs. Specifically, a multi-agent advantage actor-critic (MA2C) method is proposed with a novel local reward design and a parameter sharing scheme. In particular, a multi-objective reward function is designed to incorporate fuel efficiency, driving comfort, and the safety of autonomous driving. A comprehensive experimental study is made that our proposed MARL framework consistently outperforms several state-of-the-art benchmarks in terms of efficiency, safety, and driver comfort.