Urban Traffic Dynamics Prediction—A Continuous Spatial-temporal Meta-learning Approach

Urban traffic status (e.g., traffic speed and volume) is highly dynamic in nature, namely, varying across space and evolving over time. Thus, predicting such traffic dynamics is of great importance to urban development and transportation management. However, it is very challenging to solve this problem due to spatial-temporal dependencies and traffic uncertainties. In this article, we solve the traffic dynamics prediction problem from Bayesian meta-learning perspective and propose a novel continuous spatial-temporal meta-learner (cST-ML), which is trained on a distribution of traffic prediction tasks segmented by historical traffic data with the goal of learning a strategy that can be quickly adapted to related but unseen traffic prediction tasks. cST-ML tackles the traffic dynamics prediction challenges by advancing the Bayesian black-box meta-learning framework through the following new points: (1) cST-ML captures the dynamics of traffic prediction tasks using variational inference, and to better capture the temporal uncertainties within tasks, cST-ML performs as a rolling window within each task; (2) cST-ML has novel designs in architecture, where CNN and LSTM are embedded to capture the spatial-temporal dependencies between traffic status and traffic-related features; (3) novel training and testing algorithms for cST-ML are designed. We also conduct experiments on two real-world traffic datasets (taxi inflow and traffic speed) to evaluate our proposed cST-ML. The experimental results verify that cST-ML can significantly improve the urban traffic prediction performance and outperform all baseline models especially when obvious traffic dynamics and temporal uncertainties are presented.

INVESTIGATION AND MODELING OF TRANSPORT NOISE DEPENDENCE ON TRAFFIC SPEED AND IMPACT ON POPULATION ANNOYANCE

Transport noise is a serious problem in cities and has a negative impact on both health and economics. In addition to the aforementioned unnoticed health effects, traffic noise has also been identified as one of the leading causes of sleep disorders, annoyance and negative cardiovascular effects. This research consists of three parts: part one involves onsite measurements of traffic noise in Trakai town; part two simulates traffic noise at different average vehicle speeds; part three assesses the number of people affected by traffic noise. The carried-out simulation has demonstrated that the noise level changes very slightly at different average vehicle speeds. It should be noticed that more noise is generated at average vehicle speed of 30 km/h rather than at 50 km/h. The assessment of the annoyance level has disclosed that an average vehicle speed of 30 km/h should cause the highest level of annoyance (highest – 26.8%).

Evaluating the citywide Edinburgh 20mph speed limit intervention effects on traffic speed and volume: A pre-post observational evaluation

Objectives Traffic speed is important to public health as it is a major contributory factor to collision risk and casualty severity. 20mph (32km/h) speed limit interventions are an increasingly common approach to address this transport and health challenge, but a more developed evidence base is needed to understand their effects. This study describes the changes in traffic speed and traffic volume in the City of Edinburgh, pre- and 12 months post-implementation of phased city-wide 20mph speed limits from 2016–2018. Methods The City of Edinburgh Council collected speed and volume data across one full week (24 hours a day) pre- and post-20mph speed limits for 66 streets. The pre- and post-speed limit intervention data were compared using measures of central tendency, dispersion, and basic t-tests. The changes were assessed at different aggregations and evaluated for statistical significance (alpha = 0.05). A mixed effects model was used to model speed reduction, in the presence of key variables such as baseline traffic speed and time of day. Results City-wide, a statistically significant reduction in mean speed of 1.34mph (95% CI 0.95 to 1.72) was observed at 12 months post-implementation, representing a 5.7% reduction. Reductions in speed were observed throughout the day and across the week, and larger reductions in speed were observed on roads with higher initial speeds. Mean 7-day volume of traffic was found to be lower by 86 vehicles (95% CI: -112 to 286) representing a reduction of 2.4% across the city of Edinburgh (p = 0.39) but with the direction of effect uncertain. Conclusions The implementation of the city-wide 20mph speed limit intervention was associated with meaningful reductions in traffic speeds but not volume. The reduction observed in road traffic speed may act as a mechanism to lessen the frequency and severity of collisions and casualties, increase road safety, and improve liveability.