Traffic Trajectory Research Articles

Markov modeling of traffic flow in Smart Cities

Modeling and simulating the traffic flow in large urban road networks are important tasks. A mathematically rigorous stochastic model proposed in [8] is based on the synthesis of the graph and Markov chain theories. In this model, the transition probability matrix describes the traffic dynamics and its unique stationary distribution approximates the proportion of the vehicles at the segments of the road network. In this paper various Markov models are studied and a simulation method is presented for generating random traffic trajectories on a road network based on the two-dimensional stationary distribution of the models. In a case study we apply our method to the central region of the city of Debrecen by using the road network data from the OpenStreetMap project which is available publicly.

Analysis of driver’s choice behavior based on the combined model of utility-regret

In the context of increasing urban traffic trajectory data, based on large-scale urban traffic trajectory data, in the field of intelligent transportation, it has become an urgent need to analyze travelers’ routing behaviors and establish effective routing models to provide travelers with efficient and reasonable driving route recommendations. It is often the case that urban taxi groups has better route choice behavior than other travel group under the cost-benefit constraints. Therefore, providing route recommendations for other travelers according to the route choice habits of urban taxi group is a feasible solution to the above need. This paper starts with the utility maximization model and the stochastic regret minimization model, for the problem that the traveler cannot satisfy the premise of the complete rationality of the utility maximization model when choosing routes and the random regret minimization model ignores the traveler’s sensitivity to the original value of the alternative attribute, and improve the calculation formula of the regret value in the random regret minimization model. A route choice model under the utility-regret combination rule is proposed based on the combination of the utility model and the improved regret model, and then, based on the historical trajectory data of taxis in Xi’an, the RP data required by the model is extracted, and the parameters of the model are estimated and verified. The experimental results show that the route choice probability calculated by the model is closer to the route choice probability in the real scene than the simple utility maximization model or the simple random regret minimization model.

Traffic clustering algorithm of urban data brain based on a hybrid-augmented architecture of quantum annealing and brain-inspired cognitive computing

In recent years, the urbanization process has brought modernity while also causing key issues, such as traffic congestion and parking conflicts. Therefore, cities need a more intelligent brain to form more intelligent and efficient transportation systems. At present, as a type of machine learning, the traditional clustering algorithm still has limitations. K-means algorithm is widely used to solve traffic clustering problems, but it has limitations, such as sensitivity to initial points and poor robustness. Therefore, based on the hybrid architecture of Quantum Annealing (QA) and brain-inspired cognitive computing, this study proposes QA and Brain-Inspired Clustering Algorithm (QABICA) to solve the problem of urban taxi-stand locations. Based on the traffic trajectory data of Xi'an and Chengdu provided by Didi Chuxing, the clustering results of our algorithm and K-means algorithm are compared. We find that the average taxi-stand location bias of the final result based on QABICA is smaller than that based on K-means, and the bias of our algorithm can effectively reduce the tradition K-means bias by approximately 42%, up to approximately 83%, with higher robustness. QA algorithm is able to jump out of the local suboptimal solutions and approach the global optimum, and brain-inspired cognitive computing provides search feedback and direction. Thus, we will further consider applying our algorithm to analyze urban traffic flow, and solve traffic congestion and other key problems in intelligent transportation.

A Real-Time Data Collection Mechanism With Trajectory Privacy in Mobile Crowd-Sensing

As a new paradigm to serve and sense the intelligent city, mobile crowd-sensing (MCS) usually requires participants’ real-time locations. However, uploading participants’ true locations to servers or third parties raises privacy concerns. In this letter, we propose a real-time data collection mechanism with trajectory privacy (RDCTP) in MCS, which achieves $w$ -event $\varepsilon $ -differential privacy for the crowd-sensing participants. Different from existing works, we focus on protecting the privacy of trajectories instead of individual locations. Specifically, RDCTP provides $\varepsilon $ -differential privacy for each sub-trajectory which consists of successive $w$ locations. To achieve this, a participant first allocates the trajectory privacy budget to each location. Then, he perturbs his true location and gets candidate location set which satisfies $\varepsilon $ -differential privacy. Last, he submits a location from the set by solving an optimization problem that aims to tradeoff between the privacy and utility. We utilize real world traffic trajectories of Shanghai taxis to evaluate the RDCTP, and the results show that it not only protects participants’ privacy, but also preserves the server’s utility.

Delivering Video-on-Demand services with IEEE 802.11p to major non-urban roads: A stochastic performance analysis

In this paper, we introduce a relatively simple and versatile analytical model to compute the performance of an IEEE 802.11-based Video-on-Demand (VoD) service for connected vehicles traveling along a highway or an expressway. We use a probabilistic approach to account for the intrinsic dynamics (mobility) of vehicles, the density of subscriber vehicles, the distance between Road Side Units (RSUs), as well as the video codec bit rate. Our solution delivers estimates for the attained throughput of each subscriber vehicle as well as for their interruption times in the playback of their video. We studied several scenarios to assess the accuracy of our approximate solution comparing its results with those of a discrete-event simulator with realistic vehicular traffic trajectories, video traffic, and networking protocols. Overall, our solution is found to be accurate delivering estimates in good agreement with the simulation results. Through several applications, we illustrate how our analytical solution can help in the design of a VoD service for vehicles for capacity planning purposes, or at improving its performance with the use of a blocking policy.

Functional clustering and missing value imputation of traffic flow trajectories

Patterns of traffic flow trajectories play an essential role in analysing traffic monitoring data in transportation studies. This research presents a data-adaptive clustering approach to explore traffic flow patterns and a unified algorithm to impute missing values for incomplete traffic flow trajectories. We recommend using subspace-projected functional data clustering with the assumption that each observed daily traffic flow trajectory is a realization of a random function sampled from a mixture of stochastic processes, and each subprocess represents a cluster subspace spanned by the mean function and eigenfunctions of the covariance kernel of the random trajectories. The unified algorithm combines probabilistic functional clustering with functional principal component analysis to propose a mixture prediction for missing value imputation. The proposed methods effectively unravel distinctive daily traffic flow patterns and improve the accuracy of missing value imputation. The advantage of the proposed approaches is demonstrated through numerical studies of a real traffic flow data application.

Traffic Congestion Identification and Reduction

Historical data is increasingly becoming more purposeful in the field of intelligent traffic systems. Traffic congestion is one of the major challenges in most cities around the world. It increases fuel wastage, monetary losses, and life-endangering. Historical trajectory data may be a suitable solution to reduce congestion on the road in VANET. This approach is a combination of data mining historical trajectory data to detect and predict traffic congestion and VANET in reducing the detected congestion events. The trajectories are first preprocessed before they are clustered using traj clustering algorithm. Congestion detection is then done for each cluster based on a certain speed threshold and also time duration of that particular event. The experiments are done on a data set which recorded traffic trajectories for one day and the same experiment is iterated on data sets for the day two and three respectively day and the detected congestion incidents are recorded in terms of their coordinates and time period in which they occurred. After detecting the congestion, we have also proposed a traffic congestion reduction solution. Results and simulations show that our proposed mechanism is suitable in VANET.

Artificial intelligence based commuter behaviour profiling framework using Internet of things for real-time decision-making

Road traffic environments are highly dynamic and volatile with a multitude of roadside and external environmental factors contributing to its dynamicity. Apart from infrastructure-related means such as traffic lights, planned and unplanned road events and different road networks, a core component which contributes towards the traffic environment is the human factor which is heavily overlooked in the current studies. Due to diverse travel patterns of day-to-day activities, the commuter behaviour is directly depicted in traffic patterns providing an opportunity to further explore human behaviours using road traffic. Conducting such analysis would reveal different commuter behavioural patterns that can be used for optimization and timely management of operations. However, to conduct such real-time behaviour analysis, large volumes of high-frequency data are required with high granularity, as well as, a suitable technology to manage such data. Addressing these needs, we propose an environment-driven commuter behavioural model that can be used to elucidate diverse behaviours in road traffic environments. We conceptualized, designed and developed an artificial intelligence based commuter behaviour profiling framework to detect diverse commuter behavioural profiles, fluctuating and routine patterns among commuters using traffic flow profiling and travel trajectory analysis. We evaluated the framework using 190 million data points captured from the Bluetooth sensor network of the Melbourne arterial road network, in the state of Victoria in Australia. The results demonstrate that traffic flow profiling of the proposed framework can provide insights on recurrent commuter behaviours that are distinct to a selected area with a high granularity. Moreover, traffic trajectory analysis provides insights on non-recurrent behaviours such as accidents with regard to how such incidents impact the dynamics of the network and how the impact is propagated through the network. Besides road traffic management, the proposed framework will enable real-time decision-making when planning road infrastructure and support decision-making of government and business entities to optimize operations.

Spatio‐temporal expand‐and‐squeeze networks for crowd flow prediction in metropolis

The use of deep learning methods to predict traffic flow in transportation systems has become a hot research project. The existing predictive model method faces problems such as long calculation time and difficult data pre-processing, especially for the prediction effect of high traffic area. In this study, the authors propose a novel framework ST-ESNet, spatio-temporal expand-and-squeeze networks, that designs several effective strategies for considering the complexity, non-linearity and uncertainty of traffic flow, and better captures traffic flow characteristics to adapt to the dynamic characteristics of traffic trajectory, traffic duration and traffic flow. Specially, we use extend-and-squeeze process rather than squeeze-and-extend process during the normal residual unit to capture farther spatial dependence among regions. Specifically, inverted residual and deformed convolution structures are utilised in the expanding process, and the convolution with stride 2 is utilised in the squeeze process. Furthermore, image feature scaling is used in each residual unit to obtain more fine-grained surface information, which improves the ability of the model to capture dynamic spatial dependence features. Finally, they use stochastic weight averaging to obtain an integration model. In summary, they propose a new predictive model ST-ESNet. The experimental results show that the authors' proposed network model has better prediction performance compared with the state-of-the-art model.

Safety and Operational Performance of Signalized Roundabouts: A Case Study in Doha

Abstract Traffic safety is one of the important challenges that urban transportation systems are facing. The common practice of assessing the safety of transportation systems starts with the collection of readily available crash data and crash reports from different resources, and processing these collecting data. This approach, however, requires having an operational transportation system and historical crash data that are extended for long periods that can, which are usually years before proper assessment and proposed countermeasures can be implemented. Therefore, a new trend in analyzing and assessing the traffic safety of different transportation systems has been developed and used lately. This trend relies on analyzing traffic conflicts that are generated through traffic simulation environment. Those simulated traffic conflicts are then associated with different traffic safety assessment measures. The simulated traffic conflicts can be obtained by analyzing observed vehicle trajectories, or simulated ones. Surrogate Safety Assessment Model (SSAM) is an analysis tool that processes simulated traffic trajectories. In this paper, the safety and operational performance of a signalized roundabout in the City of Doha ate assessed through the simulation approach and applying the surrogate safety assessment model. The results indicated that the signal timings and phasing schemes are associated with the level of safety at the signalized roundabout by reducing or increasing the potential traffic conflicts. Moreover, it was found that high traffic demand at signalized roundabouts is not necessarily related to the high traffic conflicts or high conflict severity.

TraLFM: Latent Factor Modeling of Traffic Trajectory Data

The widespread use of positioning devices (e.g., GPS) has given rise to a vast body of human movement data, often in the form of trajectories. Understanding human mobility patterns could benefit many location-based applications. In this paper, we propose a novel generative model called TraLFM via latent factor modeling to mine human mobility patterns underlying traffic trajectories. TraLFM is based on three key observations: (1) human mobility patterns are reflected by the sequences of locations in the trajectories; (2) human mobility patterns vary with people; and (3) human mobility patterns tend to be cyclical and change over time. Thus, TraLFM models the joint action of sequential, personal and temporal factors in a unified way, and brings a new perspective to many applications such as latent factor analysis and next location prediction. We perform thorough empirical studies on two real datasets, and the experimental results confirm that TraLFM outperforms the state-of-the-art methods significantly in these applications.

Visual analytics on big data of urban traffic trajectories

With the rapid development of Smart city and Big data concepts and technologies, how to quickly and accurately organize and reflect traffic data information has become the main topic of current research for urban traffic trajectory big data with increasing geometric quantity. Traffic trajectory visualization is favored for its ability to visually and vividly reflect traffic trajectory information. In this paper, the main research results of visual analysis of urban traffic trajectories in recent years are summarized from three main methods: direct visualization, aggregate visualization and feature visualization. The problems and future prospects of urban traffic trajectory data visual analysis research are summarized.

Research on visual analysis technology and application of urban road data

Many cities use the current wealth of multi-source data to create intelligent transportation systems, which visually process road data so that traffic consumers and traffic managers can better understand urban traffic conditions. Due to the heterogeneity, complexity and mass of these data, it is not easy to conduct effective analysis on them, and it is often necessary to integrate human perception in the analysis process, so as to cause extensive application of visualization. In this paper, we first briefly introduce the pre-processing of urban trajectory data, then systematically analyze the visualization of two data types of road traffic flow and traffic events, and finally briefly summarize the research trend of traffic visualization in recent years and put forward corresponding challenges. The results show that the early research focused on the visualization of road flow, including the main arrow diagram, Mosaic diagram, track wall, etc. With the deepening of visual analysis, the analysis of traffic flow increases to the analysis of traffic events. In recent years, there has been a new trend of mining and utilizing the attributes or semantics of traffic trajectories or traffic events. The research on traffic data visualization has been greatly expanded in breadth and depth.

Calibration of Vehicle-Following Model Parameters Using Mixed Traffic Trajectory Data

A number of models for car following have been proposed for homogeneous traffic and some of these have been modified or adapted to represent mixed traffic conditions in India. Vehicle-following behavior under mixed traffic is both complex and challenging and cannot be adequately captured by conventional lane-based following models and their variants. For example, the behavior of a subject vehicle in a mixed traffic condition depends on the behavior of lead vehicle as well as the influence of neighboring vehicles. Most existing models are based on the longitudinal spacing and the relative speed of the lead and the subject vehicles. However, the vehicular interactions also depend on the lateral movements such as lateral spacing and lateral speed. Furthermore, the response of the subject vehicle also depends on the type of vehicles involved and their maneuvers in the surrounding space. This study aims to address some of these gaps in the existing vehicle-following models for mixed traffic. Mixed traffic trajectory data collected from the mid-block section of a six-lane divided urban arterial road in Chennai city were used for this study. From the data set, leader–follower vehicle pairs identified based on three different methods: influence area method, headway method, and video data tracking are compared, and the most suitable method is chosen for further analysis. Variation of driver behavior due to different factors such as follower’s speed, relative spacing, lateral position, vehicle types, and following behaviors were examined. Calibration and validation of the models were done for different leader–follower vehicle pairs. The results show that the model parameters vary with not only by subject vehicle type, but also by leader–follower pairs. In addition, there is a significant effect of factors such as lateral position of vehicles and types of following behaviors. This study will find application in developing more realistic mixed traffic simulation models by including these factors.

Simulation-Based Framework for Estimating Crash Modification Factors (CMFs): A Case Study for ITS Countermeasures

The efficient movement of users and goods is the primary purpose of the surface transportation system. Roadway traffic crashes have devastating impacts on quality of life of the users as well as health of the system. While researchers are utilizing advanced computing and communication tools to reduce number of crashes on the roadways, there is still an absence of appropriate method to evaluate the safety performances of these advanced technologies in the planning stage. Development of crash modification factors (CMFs) is a standard method to evaluate the safety effect of proposed countermeasures. Though, the current practices of developing CMFs are not efficient and cost-effective in case of addressing impacts of Intelligent Transportation System (ITS) countermeasures. This study demonstrated a proof of concept of simulation-based framework for determining CMFs for ITS countermeasures. The proposed framework includes the application of traffic microsimulation model and Surrogate Safety Assessment Model (SSAM) developed by Federal Highway Administration (FHWA). The integration of these two models is suggested to estimate CMFs efficiently. However, the calibration of traffic microsimulation model and SSAM model is essential to portrait the real-world scenarios. A case study for estimating CMFs of ITS countermeasures was conducted to validate the proposed simulation-based approach. Four ITS countermeasures were considered: ramp metering, variable speed limit, junction control, and dynamic lane assignment. They were coded in traffic microsimulation environment and vehicle trajectory files were generated to import into SSAM model. After analyzing these trajectory files in SSAM tool, it was found that all proposed ITS countermeasures, except variable speed limit assignment, could reduce the number of crashes at crash prone locations.

AGV autonomous transport units in technological transport

The article discusses a number of aspects related to the application of surface autonomous technological transport vehicles. The attention of the authors is mainly focused on issues related to the location of the technological means of transport in the working space, planning the traffic trajectory for the implementation of the transport task, technical possibilities in mapping the working space and intelligent method of control of AGV vehicles focused on efficiency. The article synthetically collects key information on key aspects related to the efficient management of both individual AGVs and their entire teams.

Automatic Vehicle-Pedestrian Conflict Identification With Trajectories of Road Users Extracted From Roadside LiDAR Sensors Using a Rule-Based Method

Vehicle-pedestrian conflicts have been the major concern for traffic safety. Surrogate safety measures are widely applied for pedestrian safety evaluation. However, how to quickly identify the vehicle-pedestrian surrogate safety measures at the individual site is challenging due to the difficulty of obtaining the high-resolution trajectories of road users. This paper presented an effective method to generate the high-resolution traffic trajectories from the roadside deployed Light Detection and Ranging (LiDAR) sensor. The vehicle-pedestrian conflicts can then be identified from the trajectories simply using the speed-distance profile (SDP) of the vehicles. The SDP can be used to develop a rule-based method for vehicle-pedestrian identification. The events can be divided into different risk levels based on the spatial distribution of the SDP. The case study shows that the rule-based method can detect vehicle-pedestrian near-crash events effectively. The other indicators, such as widely used time-to-collision (TTC) or deceleration rate to avoid a crash (DRAC), can be also obtained from the SDP. The engineers can also adjust the thresholds in the rule-based method to meet the specific requirements at different sites. The proposed method can be extended to identify vehicle-vehicle conflicts or vehicle-bicycle conflicts in future studies.

Towards an integrated longitudinal and lateral movement data-driven model for mixed traffic

Abstract In recent years, there has been an increasing interest in modeling driving behavior in developing countries, where conditions, such as non–lane discipline and heterogeneity in vehicle types, prevail. Traffic flow in such conditions is very complex in nature and safety issues arise. Traffic simulation models have been formulated for lane–based conditions. The existing models do not consider the wider range of situations that drivers in mixed traffic may face compared to drivers in homogeneous lane–based traffic, such as multiple–leader following, and passing and lateral shifts. In this research, we define the concept of virtual lanes for modeling mixed traffic conditions. A methodology based on temporary virtual lanes and data–driven approaches is developed to simulate driving behavior in developing countries. The role of vehicle type in model efficiency is also explored. The proposed methodology is validated on mixed traffic trajectory data from India.

Trajectory clustering of air traffic flows around airports

We present a new approach to separate air traffic trajectories in an area constrained by operational procedures. This technique is applied on a set of real trajectories in Toulouse terminal manoeuvring area (TMA). The resulting clusters foster good understanding of the structure of traffic and of how controllers schedule landings at Toulouse–Blagnac airport; on the other hand, a group of peculiar trajectories emerge with useful information calling for further analysis and paving the way for a probabilistic approach to risk assessment in air traffic safety.

Dictionary-based Fidelity Measure for Virtual Traffic.

Aiming at objectively measuring the realism of virtual traffic flows and evaluating the effectiveness of different traffic simulation techniques, this paper introduces a general, dictionary-based learning method to evaluate the fidelity of any traffic trajectory data. First, a traffic pattern dictionary that characterizes common patterns of real-world traffic behavior is built offline from pre-collected ground truth traffic data. The corresponding learning error is set as the benchmark of the dictionary-based traffic representation. With the aid of the constructed dictionary, the realism of input simulated traffic flow data can be evaluated by comparing its dictionary-based reconstruction error with the dictionary error benchmark. This evaluation metric can be robustly applied to any simulated traffic flow data; in other words, it is independent of how the traffic data are generated. We demonstrated the effectiveness and robustness of this metric through many experiments on real-world traffic data and various simulated traffic data, comparisons with the state-of-the-art entropy-based similarity metric for aggregate crowd motions, and perceptual evaluation studies.