Infrastructure Engineering - Research Publications
Permanent URI for this collection
Search Results
1 - 5 of 5
-
ItemNo Preview AvailableTraffic Anomaly Detection: Exploiting Temporal Positioning of Flow-Density Samples(Institute of Electrical and Electronics Engineers (IEEE), 2023)It is of paramount importance to detect traffic data anomalies in a real-time manner as it helps efficient traffic control and management. Several unsupervised anomaly detection algorithms are proposed previously in the literature; however, lack of proper ground truth labels for traffic data has been always a substantial barrier to deploy and evaluate them. In this paper, we introduce a concept named Temporal Positioning of Flow-Density Samples (TP-FDS) that can be used by domain experts for fast and reliable traffic data labeling. We mathematically show that deviations in two-dimensional TP-FDS completely reflect point and subsequence anomalies previously defined in the literature of time series data. Furthermore, benefiting from this concept, we propose a novel anomaly detection framework with the help of Fast Angle Based Outlier Detection (Fast-ABOD) to be used for traffic data. Extensive data labeling experiments are conducted with the opinions of 20 different experts. Implementation of several machine learning algorithms, like KNN, OC-SVM, iForest, and LOF, is also adapted with two different setups of hyper-parameters to be used in the proposed framework. Results indicate that our framework integrated with Fast-ABOD is able to detect anomalies in traffic data better than other machine learning and state-of-the-art deep learning algorithms with more than 72% and 96% of F1 score and AUC.
-
ItemTransferable supervised learning model for public transport service load estimation(Springer Science and Business Media LLC, 2023-07-28)Detailed knowledge of service utilisation and passenger load profiles is the basis for the design, operation, and adjustment of a public transport service. The advancement in sensing technologies enable transit operators to monitor the variabilities in passenger flows continuously and consistently. There is a growing body of literature on using supervised learning models with direct passenger counts from historical observations. However, the incomplete, inaccurate, and biased data from automatic sensors pose challenges in this process. This paper proposes novel supervised learning models to estimate the onboard load profile of public transport services based on two main data sources: (1) limited data collected on a subset of service vehicles by automatic passenger counting (APC) systems, and (2) fare data collected by automatic fare collection (AFC) systems. The specific consideration is given to the fact that the developed models can be transferred across different routes. This is motivated by the commonly “limited coverage” of automated passenger counter devices on service vehicles. We introduce an array of new models, including a superior segment-based model, which demonstrates remarkable improvement in model transferability and accuracy. The proposed methodology utilises separate methods in different segments of a transit line. The proposed models were applied to three tram lines in Melbourne, Australia, where various types of shortcomings exist in the automated data. The test results demonstrate that the proposed models can be transferred and applied to other transit route without relying on historical observations. This would enable transit operators to reduce the number of required devices and monitor service utilisation in a more cost-efficiently manner, particularly in public transport networks where AFC coverage is usually incomplete and negatively skewed. The information on service utilisation will not only help operators to accommodate the variability in passenger demand but also assist passengers in journey planning to avoid overcrowding on services.
-
ItemTemporal pattern mining of urban traffic volume data: a pairwise hybrid clustering method(Taylor and Francis Group, 2023)Multiple pattern analyses of traffic data have been conducted previously; however, it has yet to be explored with an awareness of temporal factors in big real-world traffic data. In this paper, we introduce a hybrid method to measure the intensity of differences among various temporal factors’ data. The proposed method can efficiently process the historical data given temporal factors and provide insightful information about the intensity of variations. After data denoising with basis splines, we reshape the time series into a 2-D latent space using Principal Component Analysis (PCA) according to the type of analysis. Pairwise K-means clustering is then applied after anomaly elimination with DBSCAN to derive Adjusted Rand Index (ARI) matrices. Finally, these matrices are then systematically used to find similar patterns of different temporal perspectives. Multiple analyses are carried out with real data from Melbourne, Australia. Dissimilarities with intensities of up to 80% are detected that are not detectable with general clustering approaches.
-
ItemNo Preview AvailableIntelligent vehicle pedestrian light (IVPL): A deep reinforcement learning approach for traffic signal control(PERGAMON-ELSEVIER SCIENCE LTD, 2023-04)Deep reinforcement learning (RL) has been widely studied in traffic signal control. Despite the promising results that indicate the superiority of deep RL in terms of the quality of solution and optimality over fixed time signal control, the real-world multi-modal traffic flows, especially pedestrians, are not properly considered nor sufficiently investigated. This study presents a novel deep RL-based adaptive traffic signal model to control the vehicles and pedestrian flows by allocating an equitable green time to each, aiming at minimizing “total user delays” as opposed to “total vehicle delays” dominantly being used in the literature. Our proposed intelligent vehicle pedestrian light (IVPL) method can perform in the absence or presence of pedestrians, especially when there is jaywalking at the intersection, interrupting vehicle flows. To this end, an extended reward function is designed to capture delays due to vehicle-to-vehicle, vehicle-to-pedestrian, and pedestrian-to-pedestrian interactions, as well as red-light delays for vehicles and pedestrians. To evaluate the performance of IVPL, a microsimulation model of an intersection in city of Melbourne is used as a case-study. The real traffic signal parameters of an existing operation system (SCATS) are employed, and the simulation is calibrated using video-based camera data and loop detectors data collected at intersection. The experimental results demonstrate the superiority of the proposed model over fully actuated traffic signal, not only in terms of the quality of optimal solution, but also considering the fact that the proposed model can minimize the “total user delays”.
-
ItemNo Preview AvailableA Graph and Attentive Multi-Path Convolutional Network for Traffic Prediction(Institute of Electrical and Electronics Engineers, 2022-06-02)Traffic prediction is an important and yet highly challenging problem due to the complexity and constantly changing nature of traffic systems. To address the challenges, we propose a graph and attentive multi-path convolutional network (GAMCN) model to predict traffic conditions such as traffic speed across a given road network into the future. Our model focuses on the spatial and temporal factors that impact traffic conditions. To model the spatial factors, we propose a variant of the graph convolutional network (GCN) named LPGCN to embed road network graph vertices into a latent space, where vertices with correlated traffic conditions are close to each other. To model the temporal factors, we use a multi-path convolutional neural network (CNN) to learn the joint impact of different combinations of past traffic conditions on the future traffic conditions. Such a joint impact is further modulated by an attention generated from an embedding of the prediction time, which encodes the periodic patterns of traffic conditions. We evaluate our model on real-world road networks and traffic data. The experimental results show that our model outperforms state-of-art traffic prediction models by up to 18.9% in terms of prediction errors and 23.4% in terms of prediction efficiency.