When Automated Vehicles (AVs) navigate dynamic, interactive driving scenarios, they must consider the spatio-temporal layout of surrounding traffic, including social interactions among agents, to accurately predict their trajectories. Existing approaches often fail to handle complex inter-agent interactions with the necessary adaptive attention to dynamic contexts. This paper introduces a multi-agent trajectory prediction algorithm that leverages attentive spatio-temporal modelling to capture interactions among agents. Our approach integrates weighted Distance Graph Attention Networks (wDGAT) with dynamic attention assignment and Multi-Head Attention (MHA)-based Transformers to learn multi-headed social interaction patterns, preserving this critical information throughout the learning pipeline. This enables efficient aggregation of information from any number of neighbouring agents, allowing for robust processing of complex, time-dependent data and consistent retrieval of spatio-temporal knowledge across extended prediction horizons. We validate our model through extensive experiments on the NGSIM (US-101 and I-80) highway datasets. The results demonstrate that our approach consistently achieves the lowest prediction error over a 5-second horizon, producing diverse outcomes for different agents and outperforming state-of-the-art methods. Numerical results demonstrate that, compared with state-of-the-art models, the proposed model reduces the average prediction root-mean-square error over a five-second time horizon by 40% and achieves a median performance gain of 20% on large-scale public datasets. Ablation studies further confirm the effectiveness of our algorithm.

Traffic estimation using probe vehicle data is a crucial aspect of traffic management as it provides real-time information about traffic conditions. This study introduced a novel framework for traffic density estimation using Temporal Convolutional Network (TCN) for time series data. The study used two datasets collected from a three-leg intersection in Greece and a four-leg intersection in Germany. The model was built to predict the density in an approach of the signalized intersection using features extracted from the other approaches. The results showed that the highest accuracy was achieved when only probe vehicle data was used. This implies that relying solely on probe vehicle data from two approaches can effectively predict traffic density in the third approach, even when the Market Penetration Rate (MPR) is low. The results also indicated that having Signal Phase and Timing (SPaT) information may not be necessary for high accuracy in traffic estimation and that as the MPR increases, the model becomes more predictable.

Automated Vehicles (AVs) have been receiving increasing attention as a potential highly mechanised, intelligent, self-regulating futuristic mode of transport. AVs are predicted to address limitations and human factors associated with traditional modes of transportation. Beyond the typical operations of AVs which can perform rudimentary tasks, the intelligent embedded program fit in to process challenging scenarios and deep multi-dimensional/ agent intents and interaction of the roadway is the grey area yet to be explored to design an exclusive encoding of social functionality and operation in order to address human factors causing road crashes. The aim of this study is to design a data-driven prediction framework for AVs that utilises multiple inputs to prove a multimodal, probabilistic estimate of the future intentions and trajectories of surrounding vehicles in freeway operation. Our proposed framework is a deep multi-agent learning-based system designed to effectively capture social interactions between vehicles without relying on map information. Our approach excels in capturing the high-level behaviours of multiple vehicles and generating a multi-modal trajectory forecast. It employs a multi-headed neural architecture to learn from social interactions between vehicle pairs and generates diverse trajectories proportional to predicted target intents, thus enabling feature fusion. Additionally, a multi-head self-attention mechanism is incorporated for prediction refinement. We achieved a good prediction performance with a lower prediction error in real traffic data at highways. Evaluation of the proposed framework using the NGSIM (US-101 and I-80) and HighD datasets shows satisfactory prediction performance for long-term trajectory prediction of multiple surrounding vehicles. Additionally, the proposed framework has higher prediction accuracy and generalisability than state-of-the-art approaches.

This paper presents an early prediction framework to classify drivers’ intended intersection movements in a connected vehicle environment. Intersections are considered accident blackspots with major traffic violations that cause property damage, injuries and fatalities. An accurate perception of drivers’ intended movements at intersections is required for advanced red-light (ARLW) or turning warnings for vulnerable road users (TWVR). Early prediction of intersection movement and adequate warning assistance will ensure road users’ safety at the intersection. In this study, we adopted recurrent neural networks (RNN): long short-term memory (LSTM) and gated recurrent units (GRU) networks to predict driver intended movements at intersections using the vehicle kinematics extracted from the Cooperative Awareness Messages (CAMs). We used naturalistic driving data of the Ipswich Connected Vehicle Pilot (ICVP) project, Queensland, which was collected from 351 participants who drove their connected vehicles during the pilot period. The pilot study installed roadside equipment at 29 signalised intersections to enable the Cooperative Intelligent Transportation System (C-ITS) use cases. Vehicle speed, speed limit, longitudinal acceleration, lateral acceleration, and yaw rate are used as predictors and monitored in 100-millisecond intervals for 1s to 4s at different warning distances from the stop line. Separate prediction models are trained based on different monitoring windows. Furthermore, drivers’ intended intersection movements are predicted at two individual intersections to evaluate intersection-specific prediction performance and are found with improved prediction accuracy than overall prediction models trained with all 29 intersections data. Overall prediction models are useful for some intersections which lack available data for individual intersection-based prediction.

Bike sharing systems (BSSs) are becoming an important part of urban mobility in many cities given that they are sustainable and environmentally friendly. BSS operators spend great efforts to ensure bike and dock availability at each station. Measuring the quality of service (QoS) of each station and/or the entire system is critical for efficient system operations. The traditionally-known QoS measure reported in the literature is based on the proportion of problematic stations, which are defined as those with no bikes or docks available to users. This measure neither exposes the spatial dependencies between stations nor does it discriminate between stations in the BSS. Hence, we propose a novel QoS measure, namely the Optimal Occupancy, in which: 1) the temporal variations in arrival and pick up rates at individual stations are considered; 2) the discriminative property of the Optimal Occupancy is demonstrated using Analysis of Variance (ANOVA) procedures; and 3) geo-statistics, which have not been used before, are applied to explore the spatial Optimal Occupancy variations and model variograms for spatial prediction. This study uses an anonymized bike trip dataset from 34 stations in downtown San Francisco to compare the traditionally-known QoS measure and the proposed Optimal Occupancy measure. Results reveal that the Optimal Occupancy is beneficial, outperforms the traditionally-known QoS measure, and produces a better prediction of the QoS at nearby locations. In addition, the Optimal Occupancy can be used to predict candidate locations for the introduction of new stations in an existing BSS.

Human errors contribute to 94%(±2.2%) of road crashes resulting in fatal/non-fatal causalities, vehicle damages and a predicament in the pathway to safer road systems. Automated Vehicles (AVs) have been a potential attempt in lowering the crash rate by replacing human drivers with an advanced computer-aided decision-making approach. However, AVs are yet to progress in handling the unprecedented situations involving interactions with other road users. This raises a need for a sophisticated and robust methodological framework to predict human driver interaction and intention. It is of prime importance to develop a constructive knowledge on the existing literature for a proficient forward leap in the field. To address this, we aim to conduct a comprehensive review on motion prediction methods in automated driving context with a special emphasis on model-based and data-driven approaches. Over a hundred studies related to the motion prediction for AVs have been extensively reviewed. This study recommends that the field requires more intricate classification of motion prediction methods, as the conventional three-level categorisation scheme should be upgraded to a profound and present-day context. Therefore, we attempt to provide a clear categorisation of existing motion prediction solutions by adopting four principal strategies: 1. Prediction methods, 2. Classes, 3. Algorithms and 4. Datasets. An all-inclusive summary of the reviewed studies with their respective pros and cons are also presented. Furthermore, we summarise the standard evaluation metrics applied for road users’ intention estimation and trajectory prediction tasks. It is found that the recent studies are built upon multi-agent learning systems with interaction among multiple road users in the same road environment. These methods can provide reliable prediction performance in highly interactive situations over long periods of time. However, the limitation could be at the cost of higher computational complexity in comparison to conventional methods, which are simpler to design and computationally effective. It is also observed that the conventional methods can only operate over a narrow prediction horizon and seldom consider the interactions among the road users. This review contributes to knowledge in validation, addresses the discrepancies, to explicate the ambiguities and to streamline current research for a futuristic perspective beneficiary in motion prediction field.

Automated driving raises new human factors challenges. There is a paradox that allows drivers to perform non-driving related tasks (NDRTs), while benefiting from a driver who regularly attends to the driving task. Systems that aim to better manage a driver's attention, encouraging task switching and interleaving, may help address this paradox. However, a better understanding of how drivers self-interrupt while engaging in NDRTs is required to inform such systems. This paper presents a counterbalanced within-subject simulator study with N=42 participants experiencing automated driving in a familiar driving environment. Participants chose a TV show to watch on a HUD and mobile display during two 15min drives on the same route. Eye and head tracking data revealed more self-interruptions in the HUD condition, suggesting a higher likelihood of a higher situation awareness. Our results may benefit the design of future attention management systems by informing the visual and temporal integration of the driving and non-driving related task.

This paper proposes a novel supervised clustering algorithm to analyze large datasets. The proposed clustering algorithm models the problem as a matching problem between two disjoint sets of agents, namely, centroids and data points. This novel view of the clustering problem allows the proposed algorithm to be multi-objective, where each agent may have its own objective function. The proposed algorithm is used to maximize the purity and similarity in each cluster simultaneously. Our algorithm shows promising performance when tested using two different transportation datasets. The first dataset includes speed measurements along a section of Interstate 64 in the state of Virginia, while the second dataset includes the bike station status of a bike sharing system (BSS) in the San Francisco Bay Area. We clustered each dataset separately to examine how traffic and bike patterns change within clusters and then determined when and where the system would be congested or imbalanced, respectively. Using a spatial analysis of these congestion states or imbalance points, we propose potential solutions for decision makers and agencies to improve the operations of I-64 and the BSS. We demonstrate that the proposed algorithm produces better results than classical K -means clustering algorithms when applied to our datasets with respect to a time event. The contributions of our paper are: 1) we developed a multi-objective clustering algorithm; 2) the algorithm is scalable (polynomial order), fast, and simple; and 3) the algorithm simultaneously identifies a stable number of clusters and clusters the data.

This paper develops a novel two-layer hierarchical classifier that increases the accuracy of traditional transportation mode classification algorithms. This paper also enhances classification accuracy by extracting new frequency domain features. Many researchers have obtained these features from global positioning system data; however, this data was excluded in this paper, as the system use might deplete the smartphone's battery and signals may be lost in some areas. Our proposed two-layer framework differs from previous classification attempts in three distinct ways: 1) the outputs of the two layers are combined using Bayes' rule to choose the transportation mode with the largest posterior probability; 2) the proposed framework combines the new extracted features with traditionally used time domain features to create a pool of features; and 3) a different subset of extracted features is used in each layer based on the classified modes. Several machine learning techniques were used, including k-nearest neighbor, classification and regression tree, support vector machine, random forest, and a heterogeneous framework of random forest and support vector machine. Results show that the classification accuracy of the proposed framework outperforms traditional approaches. Transforming the time domain features to the frequency domain also adds new features in a new space and provides more control on the loss of information. Consequently, combining the time domain and the frequency domain features in a large pool and then choosing the best subset results in higher accuracy than using either domain alone. The proposed two-layer classifier obtained a maximum classification accuracy of 97.02%.

Real-time, accurate travel time prediction algorithms are needed for individual travelers, business sectors, and government agencies. They help commuters make better travel decisions, avert traffic congestion, help the environment by reducing carbon emissions, and improve traffic efficiency. Travel time prediction has begun to attract more attention with the rapid development of intelligent transportation systems (ITSs), and is considered one of the more important elements required for successful ITS subsystems deployment. However, the stochastic nature of travel time makes accurate prediction a difficult task. This paper proposes travel time modeling using a mixture of linear regressions. The proposed model consists of two normal components. The first component models the congested regime while the other models the free-flow regime. The means of the two components are modeled by two linear regression equations. The predictors used in the linear regression equation are selected out of the spatiotemporal speed matrix using a random forest machine-learning algorithm. The proposed model is tested using archived data from a 74.4-mile freeway stretch of I-66 eastbound connecting I-81 and Washington, D.C. The experimental results show the ability of the model to capture the stochastic nature of travel time and to predict travel time accurately.

Real-time automatic congestion identification is one of the important routines of intelligent transportation systems (ITS). Previous efforts usually use traffic state measurements (speed, flow, occupancy) to develop congestion identification algorithms. However, the impacts of weather conditions to identify congestion have not been investigated in the existing studies. In this paper, we proposed an algorithm that uses the speed probe data and the corresponding weather and visibility to build a transferable model. This model can be used on any road stretch. Our algorithm assumes traffic states can be classified into three regimes: congestion, speed at capacity and free-flow. Moreover, the speed distribution follows a mixture of three components whose means are functions in weather and visibility. The mean of each component is defined using a linear regression using different weather conditions and visibility levels as predictors. We used three data sets from VA, CA and TX to estimate the model parameters. The fitted model is used to calculate the speed cut-off between congestion and speed at capacity which minimize either the Bayesian classification error or the false positive (congestion) rate. The test results demonstrate the proposed method produces promising congestion identification output by considering weather condition and visibility.

The paper addresses the problem of stretch wide short-term prediction of traffic stream state. The problem is a multivariate problem where the responses are the speeds or flows on different road segments at different time horizons. Recognizing that short-term traffic state prediction is a multivariate problem, there is a need to maintain the spatiotemporal traffic state correlations. Two cutting-edge machine learning algorithms are used to predict the stretch-wide traffic stream traffic state up to 120 minutes in the future. Furthermore, the divide and conquer approach was used to divide the large prediction problem into a set of smaller overlapping problems. These smaller problems are solved using a medium configuration PC in a reasonable time (less than a minute), which makes the proposed technique suitable for practical applications.