Abstract A connected autonomous vehicle (CAV) network can be defined as a set of vehicles including CAVs that operate on specific spatial scope may road network, corridor, or segment. The constitutes an environment where traffic information is shared and instructions are issued for controlling the movements. Within such scope, high‐level cooperation among fostered by joint planning control their movements greatly enhance safety mobility performance operations. Unfortunately, highly...

Abstract Autonomous vehicle (AV) stakeholders continue to seek assurance of the safety performance this new technology through AV testing on in‐service roads, AV‐dedicated road networks, and test tracks. However, recent AV‐related fatalities roads have exacerbated public skepticism eroded some trust in operations. Further, tracks are unable characterize adequately real‐world driving environment. For reason, simulators serve as an attractive means testing. most simulators, operation is based...

Despite significant progress in autonomous vehicles (AVs), the development of driving policies that ensure both safety AVs and traffic flow efficiency has not yet been fully explored. In this paper, we propose an enhanced human-in-the-loop reinforcement learning method, termed Human as AI mentor-based deep (HAIM-DRL) framework, which facilitates safe efficient mixed platoon. Drawing inspiration from human process, first introduce innovative paradigm effectively injects intelligence into AI,...

Highway agencies continue to show interest in measuring pavement condition effects on safety. This paper estimates univariate negative binomial (UNB) and random-parameters seemingly-unrelated (RPSUNB) regression models. The latter account for unobserved heterogeneity correlation crash frequencies across the severity levels. analysis was carried out two-lane multi-lane highways, results suggest that at latter, generally has a far more significant safety impact compared former. could be due...

To standardize definitions and guide the design, regulation, policy related to automated transportation, Society of Automotive Engineers (SAE) has established a taxonomy consisting six levels vehicle automation. The SAE defines each level based on capabilities system. It does not fully consider infrastructure support required for level. This can be considered critical gap in practice because existing account fact that operational design domain (ODD) any system must describe specific...

The peculiar nature of bridge infrastructure condition data persistently poses challenges in predicting component deterioration that necessitate the continued investigation probabilistic modeling techniques. These include uncertainty characterizes due to inherent random factors and existence other variables are not typically measured (unobserved responsible for deterioration), panel its consequent observation-specific correlation heterogeneity bias, lack knowledge type past interventions. To...

The evolution of scientific advances has often been characterized by the amalgamation two or more technologies. With respect to vehicle connectivity and automation, recent literature suggests that these emerging transportation technologies can will jointly profoundly shape future transportation. However, it is not certain how individual synergistic benefits be earned from related their prevailing levels development. As such, may considered useful revisit primary concepts automation...

Automation and connectivity based platforms have great potential for managing highway traffic congestion including bottlenecks. Speed harmonisation (SH), one of such platforms, is an Active Traffic Management (ATM) strategy that addresses flow breakdown in real-time by adjusting upstream speeds. However, SH has limitations the need supporting roadway infrastructure immovable limited coverage; inability to enact control beyond its range; dependence on human driver compliance. These issues...

Trajectory prediction for heterogeneous traffic agents plays a crucial role in ensuring the safety and efficiency of automated driving highly interactive environments. Numerous studies this area have focused on physics-based approaches because they can clearly interpret dynamic evolution trajectories. However, methods often suffer from limited accuracy. Recent learning-based demonstrated better performance, but cannot be fully trusted due to insufficient incorporation physical constraints....

Connected Autonomous Vehicle (CAV) Network can be defined as a collection of CAVs operating at different locations on multilane corridor, which provides platform to facilitate the dissemination operational information well control instructions. Cooperation is crucial in CAV systems since it greatly enhance operation terms safety and mobility, high-level cooperation between expected by jointly plan within network. However, due highly dynamic combinatory nature such number agents (CAVs)...

Abstract Construction projects are often associated with partial or full road closures, which result in user costs and community disruptions terms of reduced business productivity. A number studies have addressed the problem scheduling construction based on a variety stakeholder objectives. Yet still, there seems to exist few gaps regarding (1) possible tradeoffs between cost reduction optimal scheduling, (2) role project type (rehabilitation capacity expansion) solution methodology, (3)...

Purpose Perception has been identified as the main cause underlying most autonomous vehicle related accidents. As key technology in perception, deep learning (DL) based computer vision models are generally considered to be black boxes due poor interpretability. These have exacerbated user distrust and further forestalled their widespread deployment practical usage. This paper aims develop explainable DL for driving by jointly predicting potential actions with corresponding explanations. The...

Abstract Dynamic rerouting has been touted as a solution for urban traffic congestion. However, its implementation is stymied by the complexity of traffic. To address this, recent studies suggest efficacy novel technologies like fog computing and deep reinforcement learning. there exist significant challenges in this regard: (1) sorting massive amounts data associated with large networks, (2) action space that hinders learning efficiency, (3) impairment due to overreliance on regional/local...