ABSTRACT This study introduces an innovative approach, the Predictive Approach, employing the Temporal Convolutional Network (TCN) algorithm to estimate traffic density. We used naturalistic vehicle trajectories captured by drones at a three-way signalized intersection in Athens, Greece, as part of the pNEUMA initiative. This method calculates the densities of input approaches at intersections with non-uniform MPRs, using these predictions to estimate the target approach density. With accuracy ranged from 92% to 95%, using the Predictive Approach showed that improving traffic density predictions can be achieved through factors such as accounting for MPR variations over time and between different intersection approaches while considering practical scenarios. Results also highlighted that excluding Signal Phase and Timing (SPaT) data in certain cases can enhance model performance. It offers practical applications in optimizing traffic flow and reducing congestion in smart cities and traffic control centers, particularly when rapid and real-time computations are required.

ABSTRACT This study compares pedestrian behaviours in station areas of Hong Kong and San Francisco to explore how culture and urban design shape human-environment interactions. With field observation, multisensory and video data from 283 street intersections, the study evaluates pedestrian utilization of dedicated space and signal timing, walking speeds, and rule violations. It finds that despite the observed differences in infrastructure design and policy contexts, pedestrians exhibit similar behaviours under comparable conditions of perceived risk and right-of-way. The study contributes to a global discussion on context-sensitive street design and sustainable transportation planning across cultures.

Mitigating Grade Crossing Blockage Queues by Modifying Signal Timing Plans: A Network-Level Microsimulation Approach Using Train Detection and Probe-Based Traffic Data

Chance-constrained eco-driving control of connected autonomous vehicles in mixed traffic environment at signalized intersections with uncertain signal timings

This study develops a drone-based multi-class traffic counting system using fine-tuned YOLO11 integrated with the SORT tracking algorithm on a one-way urban corridor. Aerial data was captured at an altitude of 25 meters using a DJI Mavic 3 drone from a nadir perspective to minimize intermodal occlusion. System performance was validated against manual ground truth using Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error (RMSE) metrics. The system achieved an overall MAPE of 18% and an RMSE of 45.07. Notably, the car class demonstrated perfect accuracy (0% error), confirming that these automated counts are suitable for direct application in fundamental traffic engineering metrics. Conversely, significant overcounting occurred in the motorcycle class (+34.9%), primarily attributed to Non-Maximum Suppression (NMS) failures under conditions of dense spatial proximity. In civil engineering practice, utilizing uncalibrated automated counts risks overestimating the Volume-to-Capacity (V/C) ratio, leading to a false degradation of the reported Level of Service (LOS). Consequently, a specific calibration factor of -26% for motorcycle counts is essential to ensure data validity for high-precision infrastructure design and signal timing optimization. Keywords: Drone, Traffic Counting, NMS, SORT, YOLO

Timing and Characteristics of Carotid Microembolic Signals during Pentaspline Pulsed Field Ablation.

The past couple of decades have witnessed an increasing amount of diversity in alternative intersection (AI) designs to handle an ever-increasing traffic demand. Therefore, the need for reliable analysis tools to assess current intersection operations and to predict future performance is of crucial importance. This paper assesses the accuracy of macroscopic and microscopic simulation tools to describe the operational performance of two AI designs (continuous flow intersections and offset T-intersections). Drones were employed to simultaneously capture the input and output variables to ensure that the observed outputs and performance metrics were produced by the observed inputs. Saturation flow rate, queues, and traffic signal timing data were considered as the primary model calibration data, and time-in-system, which consists of control delay at each intersection and extra travel time because of rerouting, was employed for model validation. Data collection results show that field-measured saturation flow rates at the two study sites were lower than their defaults in the Highway Capacity Manual guidance for signalized intersections. This will probably generate lower movement capacities than are currently being assumed in simulation modeling tools. Model validation results show that for the two AI designs, the tested analytical models tended to overstate the field travel times, especially for turning movements affected by upstream platooning, and particularly under high-demand conditions. This was mainly because of: (a) not accounting for initial and final queues; and (b) ignoring varying platoon flow rates at a downstream signal. By comparison, microsimulation performed better at the study sites.

Urban intersections are critical nodes for roadway safety, congestion management, and autonomous vehicle coordination. Traditional traffic control systems based on fixed-time signals and static sensors lack adaptability to real-time risks such as red-light violations, near-miss incidents, and multimodal conflicts. This study presents a grid-enabled framework integrating computer vision and machine learning to enhance real-time intersection intelligence and road safety. The system overlays a computational grid on the roadway, processes live video feeds, and extracts dynamic parameters including vehicle trajectories, deceleration patterns, and queue evolution. A novel active learning module improves detection accuracy under low visibility and occlusion, reducing false alarms in collision and violation detection. Designed for edge-computing environments, the framework interfaces with signal controllers to enable adaptive signal timing, proactive collision avoidance, and emergency vehicle prioritization. Case studies from multiple intersections typical of US cities show improved phase utilization, reduced intersection conflicts, and enhanced throughput. A grid-based heatmap visualization highlights spatial risk zones, supporting data-driven decision-making. The proposed framework bridges static infrastructure and intelligent mobility systems, advancing safer, smarter, and more connected roadway operations.

To meet the requirements of non-mechanical beam scanning and acquisition in space laser communication, this study proposes a two-dimensional scanning and acquisition method based on a silicon-based optical phased array (OPA). The OPA utilizes thermo-optic phase modulation to achieve horizontal beam pointing, while vertical beam pointing is controlled by wavelength tuning. By combining the OPA with a rectangular spiral scanning strategy, non-mechanical scanning is realized and beam acquisition experiments are carried out. Experimental results demonstrate that for an 8° step signal, the horizontal and vertical rise times are 156.8 μs and 214.76 ms, respectively. A full scan of 440 points covering a ±4° field of view is completed in 8.119 s. Acquisition experiments were conducted assuming a Gaussian-distributed uncertainty region (standard deviation σ=1°). Out of 106 independent trials, a success rate of 97.17% was achieved with an average acquisition time of 0.41 s. This work experimentally applies a rectangular spiral scanning strategy to an OPA-based acquisition system, addressing a capability that has been largely missing in previous studies. These results verify that the OPA technology has good scanning efficiency and acquisition robustness in space laser communication applications.

Abstract Experiments on entire ecosystems have contributed knowledge on effects of atmospheric CO 2 and climate change, environmental pollutants, trophic cascades, response of fisheries to management, consumer interactions, ecosystem resilience and stability, and early warning indicators of critical transitions in ecosystem state. Rate of change in external drivers of an ecosystem such as climate warming, inflow of water or nutrients, or harvest of apex predators may affect signals of critical transitions but rates of change of drivers are rarely considered in whole-ecosystem studies. We studied effects of drivers’ rates of change on indicators of critical transitions using models for whole lake manipulations of nutrient enrichment, light-absorbing substances, and apex predators. Results show that times of signals from indicators relative to times of critical transitions can vary depending on the rate of external drivers, including the rate of manipulation in whole-ecosystem experiments.

ObjectiveSignaling was introduced to the otolaryngology match in 2021, with 5 signals allotted to applicants in 2021, 4 in 2022, 7 in 2023, and 25 in 2024. This study investigated the modifying effect of signaling volume on the relationship between away rotations and matching in otolaryngology from 2018 to 2024.Study DesignCross‐sectional.SettingNational survey of US medical students.MethodsWe used the Texas Seeking Transparency in Application to Residency (STAR) survey responses of otolaryngology applicants from 2018 to 2024. Using multivariate logistic regression, we determined the odds of matching where away rotations were performed and how these odds varied across the pre‐volume (2018‐2020), low‐volume (2021‐2023), and high‐volume (2024) signaling eras.ResultsIn total, 28.3% (n = 855) of otolaryngology applicants from 2018 to 2024 completed the Texas STAR survey. Using multivariate logistic regression, adjusting for applicant characteristics, and including an interaction term between performing away rotations and signaling time period, applicants in the high‐volume signaling era were found to be significantly less likely to match at programs where away rotations were performed (odds ratio [OR]: 0.56, 95% CI: 0.33‐0.95; P < .05) compared to the pre‐signaling era. The same trend was seen in the low‐volume signaling era, though not statistically significant (OR: 0.76, 95% CI: 0.47‐1.22, P = .24). The most impactful factor on matching across all study years was performing an away rotation (OR: 12.1, 95% CI: 9.0‐16.5, P < .001).ConclusionThe introduction of signaling and the recent increase in signal number are associated with decreased likelihood of matching at a program where an away rotation was performed compared to the pre‐signaling era.Level of EvidenceV.

Traffic congestion poses significant challenges in historic cities striving to balance modern mobility needs and her- itage preservation. This paper proposes a self-adaptive fuzzy logic control system for traffic signals optimized by a recurrent neural network (RNN) for vehicular density prediction. The fuzzy controller dynamically adjusts sig- nal timing based on real-time traffic density data at in- tersections in the colonial cities. The RNN component forecasts traffic density to tune the fuzzy membership functions, enabling adaptive signal control. Simulation experiments demonstrate noticeable reductions in queue length using the proposed neuro-fuzzy method compared to uncontrolled and fuzzy logic only techniques. Improve- ments are positively correlated to street length, although less significant in very short streets. The system demon- strates promising capabilities to reduce congestion and emissions through adaptive optimization in complex ur- ban environments. Keywords: Fuzzy logic control, neural networks, intelli- gent transportation systems, traffic signal timing, conges- tion mitigation

Clock Asynchronous Traffic Signal Timing for Multi-Intersections Based on a Joint Traffic Prediction and Control Method

Seasonal instability in the determinants of vulnerable road user crashes: a partially temporally constrained modeling approach.

As the semiconductor industry advances toward lower technology nodes, the adoption of multi-voltage designs within channel-based SoC architecture presents both significant opportunities and complex challenges, particularly in the context of stringent power optimization requirements. These designs inherently introduce complex issues related to voltage domain transitions, which are critical to managing signal integrity and timing closure. Transition phenomena between voltage domains represent a major concern, directly impacting the chip’s power, performance, and area (PPA) metrics. This paper provides an in-depth analysis of transition mechanisms in multi-voltage channel-based SoC designs, identifying root causes and quantifying their effects on timing and signal integrity. It further proposes robust, practical methodologies and design techniques to mitigate transition-related issues, ensuring these solutions integrate seamlessly without compromising design integrity or chip specifications. By systematically addressing the intricacies of multi-voltage transitions in channel-based SoCs, we deliver comprehensive and efficient strategies validated through real-world implementations.

The imbalance between vehicle growth and road capacity increases traffic density at intersections, leading to congestion. This issue can be addressed through various strategies, including infrastructure development, traffic flow management, and traffic signal control. This study aims to optimize traffic signal timing by minimizing the total vehicle waiting time as a strategy to alleviate congestion. The congestion reduction strategy is implemented by grouping compatible traffic flows into the same signal phase and calculating their optimal durations. The research was conducted at the Air Putih Intersection in Samarinda City on Monday, May 5, 2025, during peak traffic hours. Graph theory particularly the concept of compatible graphs is applied by identifying traffic flows that can proceed simultaneously without conflict. These compatible flows are then modeled into a compatible graph structure to determine the optimal vehicle waiting time durations. These groupings were modeled using compatible graphs to determine the optimal vehicle waiting durations. The results show that total vehicle waiting time can be reduced from 522 seconds to 120 seconds during the morning peak, from 507 seconds to 120 seconds at midday, and from 552 seconds to 120 seconds in the evening. This approach has proven effective in designing a more efficient and adaptive traffic signal control system that responds to actual traffic conditions.

Traffic congestion and delayed emergency response remain major challenges in urban transportation due to the limitations of conventional fixed-time traffic signal systems. These systems operate with predefined signal duration and failed to adapt to real-time traffic density or provide automatic priority to emergency vehicles. This paper presents “Green Lane – Smart Signal with Emergency Priority”, an intelligent traffic signal system that dynamically controls signal timing based on real-time vehicle density while enabling priority passage for emergency vehicles. The proposed system integrates camera-based vehicle detection using Python and Open CV with an ESP32 microcontroller for adaptive traffic signal control. Emergency vehicle detection triggers an immediate signal override to create a dedicated green lane, ensuring faster clearance. Experimental results obtain from a prototype implementation show reduced waiting time improve traffic flow efficiency and quicker emergency vehicle movement compared to traditional fixed time systems. The system is low-cost, scalable, and suitable for smart city traffic management applications.

Adaptive traffic signal control is a critical component of intelligent transportation systems, and multi-agent deep reinforcement learning (MARL) has attracted increasing interest due to its scalability and control efficiency. However, existing methods have two major drawbacks: (i) they are largely driven by current and historical traffic states, without explicit forecasting of upcoming traffic conditions, and (ii) their coordination mechanisms are often weak, making it difficult to model complex spatial dependencies in large-scale road networks and thereby limiting the benefits of coordinated control. To address these issues, we propose TG-MADDPG, which integrates short-term traffic prediction with a graph attention network (GAT) for regional signal control. A WT-GWO-CNN-LSTM traffic forecasting module predicts near-future states and injects them into the MARL framework to support anticipatory decision-making. Meanwhile, the GAT dynamically encodes road-network topology and adaptively captures inter-intersection spatial correlations. In addition, we design a reward based on normalized pressure difference to guide cooperative optimization of signal timing. Experiments on the SUMO simulator across synthetic and real-world networks under both off-peak and peak demands show that TG-MADDPG consistently achieves lower average waiting times, shorter queue lengths, and higher cumulative rewards than IQL, MADDPG, and GMADDPG, demonstrating strong effectiveness and generalization.

Traffic congestion is a growing concern in rapidly urbanizing cities, particularly in Dhaka, where mixed traffic conditions and poor infrastructure exacerbate delays and reduce mobility. This study aims to evaluate traffic flow characteristics and volume along the corridor from Sony Square to Mirpur 10 and identify the key contributors to congestion. The research follows a systematic methodology involving manual traffic volume counts using tally counters and video recordings at strategic locations across four daily time slots over ten consecutive days. The collected data were categorized by vehicle type and analyzed through conversion to Passenger Car Units (PCU) to standardize heterogeneous traffic. Microsoft Excel was employed for data tabulation, graphical analysis, and service flow rate calculations. The Highway Capacity Manual (HCM) framework was used to evaluate the Level of Service (LOS) at different times of day. The findings reveal that evening peak hours exhibit the most critical congestion levels (LOS F), while morning peaks also approach capacity (LOS E). Midday and afternoon periods show relatively smoother flow (LOS B and LOS C/D, respectively). Contributing factors to congestion include excessive vehicular traffic, the absence of dedicated lanes, inefficient signal timings, unauthorized street parking, and a lack of traffic discipline among both pedestrians and drivers. Based on these insights, the study proposes practical recommendations, including the implementation of dedicated lanes, optimized signal coordination, strict enforcement of parking regulations, and enhanced public transportation operations. These strategies aim to alleviate congestion and support efficient traffic management in urban corridors, such as Mirpur 10

A growing problem for cities around the world is urban traffic congestion, which lowers quality of life, increases pollution, and causes economic losses. Traditional traffic signal systems operate on pre-timed schedules that are often inadequate in responding to real-time traffic fluctuations. Artificial Intelligence (AI) offers a dynamic alternative through adaptive traffic signal control systems that respond to real-time traffic data, optimize flow, and reduce congestion. The core technologies of AI-based traffic signal control are examined in this paper, with an emphasis on computer vision, reinforcement learning, and machine learning algorithms. Use cases in multimodal traffic management, emergency vehicle prioritisation, and dynamic signal timing are presented. Case studies from smart cities globally demonstrate the effectiveness of AI in reducing congestion and travel time. Ethical considerations, including privacy, accessibility, and algorithmic fairness, are discussed alongside technical challenges such as sensor reliability, system scalability, and integration with legacy infrastructure. Future directions include the integration of connected vehicle data, edge computing, and decentralized traffic control systems. AI-driven adaptive signal control systems are key components of intelligent transportation networks, enabling more efficient, responsive, and sustainable urban mobility.