Jiangbo Yu

High manual annotation costs and poor generalizability often limit the robustness of deep learning models for pavement crack segmentation. This study addresses these challenges by developing and validating a diffusion-based generative artificial intelligence framework to support data-efficient crack segmentation models for infrastructure inspection. The proposed framework formalizes the generation and utilization of synthetic, pixel-level annotated crack data and is empirically evaluated across multiple learning paradigms. To implement and validate the framework, an annotated synthetic dataset (AI500) was generated using Google’s Imagen 4 Ultra and designed to capture diverse crack morphologies, lighting conditions, and surface textures. The utility of the synthetic data was systematically assessed through three experimental strategies: synthetic-only training, hybrid dataset training, and sim-to-real transfer. Model performance was evaluated using three deep learning architectures and five real-world datasets. Results show that synthetic-only models provide a robust, transferable performance baseline, achieving cross-domain generalization comparable to the target-only baseline (TOB) for models trained on real-world datasets. Augmenting AI500 with only 10% real-world data yielded performance comparable to TOB. Furthermore, sim-to-real transfer using the same 10% fraction achieved near parity with TOB. Cost analysis revealed the potential for double-digit savings without compromising model robustness. Additionally, the development of G3P-500 using Gemini 3 Pro demonstrated superior photorealism and diversity, requiring minimal human verification, with only 3.2% of pixels added to generated annotations. These findings demonstrate that high-quality synthetic data within a structured generative framework can substantially reduce annotation requirements while enabling accurate, generalizable pavement crack segmentation models. However, further validation remains necessary in more challenging environments.

Transport emissions remain substantial, while eco-driving studies are often constrained by incomplete context information, inconsistent evaluation methods, and ineffective driver guidance. This study proposes an agentic framework integrating emission prediction, decision evaluation, and driver-facing intervention. The emission prediction component combines large language model (LLM)-enabled road-context and vehicle-proxy agents with a supervised machine learning model to estimate directed-link CO$_2$ emissions. The evaluation component couples reinforcement-learning policy and evaluation agents in a policy--evaluation feedback loop for route--style optimization through consistent CO$_2$ and travel-time calculations. The intervention component employs an LLM-enabled advice agent to translate selected decisions into natural-language guidance. A Montreal case study shows that enriched covariates improve CO$_2$ prediction, that emissions and travel time should be evaluated jointly, and that optimized decisions can be communicated as practical guidance. This framework delivers a transparent, deployable workflow for eco-driving strategies, bridging the gap before electric and autonomous vehicles dominate the market.

Agentic systems are changing how complex operational tasks are coordinated by connecting heterogeneous data sources, introducing a new paradigm for simplifying tasks, connecting datasets, and automating processes. Electric bus fleets provide a relevant test case for this paradigm. Their operation requires continuous coordination between service reliability, battery state-of-charge, charger availability, electricity prices, route-energy uncertainty, and vehicle-to-grid (V2G) opportunities. This paper proposes an agentic aggregator framework that streamlines this decision environment by coupling an optimization-based electric bus scheduling model with supervisory agents for disturbance detection, tariff adaptation, and schedule evaluation. The optimization core enforces physical feasibility across routes, chargers, batteries, and V2G exchanges, while the agentic layer interprets changing operating conditions, triggers real-time re-optimization when needed, and defines how flexibility value is allocated between the aggregator and the public transport operator (PTO). A realistic depot case study evaluates day-ahead and real-time operations under profit-based and operation-based coordination modes, considering service delays, route-energy deviations, electricity price shocks, and combined disturbances. The results show that agentic aggregation can support adaptive fleet-grid coordination by maintaining feasible schedules, activating re-optimization selectively, and improving the use of charging and V2G flexibility. However, they also reveal a critical trade-off: the same agentic capability that reduces operational complexity can extract value from the PTO when configured around profit-oriented pricing. These findings suggest that agentic aggregators can become useful tools for managing electric bus V2G operations, but their deployment in public-fleet contexts requires transparent coordination modes, auditable tariff-setting, and explicit value-sharing rules.

Society is witnessing rapid advances and deployments of intelligent systems exhibiting varying degrees of agency, enabled in part by multimodal large language models. These AI agents demonstrate continuously improving capabilities such as contextual intention formation and interpretation, goal-directed planning and action, purposeful social engagement, external tool use, and adaptation through memory and learning. These agentic capabilities present significant opportunities as well as substantial risks. This paper synthesizes the concepts of agentic AI and AgTS and provides a systematic review of relevant transportation literature. Existing developments show that early-stage AI agents and agentic workflows have been developed across stakeholders and lifecycle phases, including (i) real-time traffic prediction and control, (ii) vehicle operation and fleet management, (iii) asset monitoring and maintenance, and (iv) systems data analysis, demand forecasting, and strategic planning. The review further indicates that current implementations understandably only exhibit low and a subset of agentic capabilities. A more critical concern is the predominantly siloed development, with nearly no attention to their strategic integration in environments where these agents will inevitably interact—directly or indirectly—with humans and with one another. If unaddressed proactively, this fragmentation may become too late to adjust, risking the reproduction and amplification of chronical challenges that arose in part from poor coordination and conflicts among historical human stakeholders that also had their agency. The paper concludes with policy and practice recommendations for the responsible and sustainable development for the continuous development of AgTS in different socioeconomic, regulatory, and environmental contexts.

The advent of large language models (LLMs) and multimodal foundation models holds significant potential to further enhance the depth, adaptivity, and semantic richness of Vehicle-to-Everything (V2X) connectivity. This paper reviews a rapidly growing body of work toward LLM-enabled Agentic Vehicle-Infrastructure Cooperation (AgVIC) and examines how an explicit reasoning-and-interaction layer reshapes cooperation loops among vehicles, infrastructure, and humans in next-generation Intelligent Transportation Systems (ITS). Across the reviewed literature, four recurring themes emerge: vehicle-assisted traffic control, infrastructure-assisted vehicle intelligence, vehicle-assisted infrastructure health monitoring and maintenance, and the operational enablers that sustain AgVIC under realistic deployment constraints. The latter theme focuses on information transmission, particularly communication efficiency, resilience (e.g., cybersecurity and robustness), and privacy preservation. We synthesize the key agentic capabilities each study emphasizes and the role these capabilities play in their target tasks. Our synthesis shows that current progress remains largely prototypical and that major gaps persist in maintaining shared semantic understanding, coordinating heterogeneous agents in open traffic, and translating semantic outputs into safe, real-time, closed-loop actions. We outline future directions toward more efficient, safeguarded, and deployable agentic cooperation.

As cities face increasing climate-induced disruptions, shared autonomous electric vehicles (SAEVs) emerge as a dual-purpose solution to sustain passenger mobility and provide emergency power. This study introduces a framework that combines GIS-based spatial analysis and multi-objective optimization to evaluate SAEV fleet performance during power outages. Using Montreal as a testbed, we simulate the operations of a mid-sized SAEV fleet under varying power disruption scenarios. Our analysis reveals a critical operational trade-off: the fleet can meet up to 2% of daily mobility demand or supply 28% of energy needs in affected zones, but not simultaneously. The sensitivity analyses indicate that improving charging infrastructure yields greater operational benefits than increasing battery capacity. Further, revenue from energy provision increases significantly with larger fleets and higher charger power. The findings underscore the importance of coordinated infrastructure planning and incentive design to enable SAEVs to effectively support transport continuity and urban energy resilience.

Drivers’ mental and physiological states critically influence traffic safety, making reliable estimation essential for timely warnings and interventions. However, these states evolve over time and are only indirectly observable through noisy, modality-dependent responses, which can induce biased labels and fragile detectors. To improve the accuracy and reliability of state estimation, we propose a unified nonlinear state-space framework that jointly models the dynamics of latent driver states, the influence of driving context, and their effects on multi-sensory responses. Within this framework, an offline multi-view clustering method fuses complementary indicators to establish robust benchmark states, while an online ResNet-Transformer maps multi-sensory data to states in real time. We evaluated the proposed framework in a driver-fatigue case study, where fatigue labels were inferred offline from self-reports, reaction-time variability, and eye-aspect-ratio variability; and were subsequently detected online using blood volume pulse, electrodermal activity, and facial images. Our results show that the offline-derived levels exhibit statistically significant separation across multiple fatigue indicators, validating the benchmark labels, while the online model achieved 82.5% mean accuracy with 10-second windows and up to 89.5% with 30-second windows, outperforming representative baselines without evidence of class-specific bias. These findings demonstrate that the proposed state-space representation effectively captures latent driver-state dynamics and that the online-offline estimation approach offers reliable and effective means of inferring these states from diverse observational sources, supporting proactive warnings and safer human–vehicle interaction.