Sparse Trajectory Data Research Articles

A graph CNN-LSTM neural network for short and long-term traffic forecasting based on trajectory data

Traffic forecasting is an important research area in Intelligent Transportation Systems that is focused on anticipating traffic in order to mitigate congestion. In this work we propose a deep neural network that simultaneously extracts the spatial features of traffic, using graph convolution, and its temporal features by means of Long Short Term Memory (LSTM) cells to make both short-term and long-term predictions. The model is trained and tested using sparse trajectory (GPS) data coming from the ride-hailing service of DiDi in the cities of Xi'an and Chengdu in China. Besides, presenting the deep neural network, we also propose a data-reduction technique based on temporal correlation to select the most relevant road links to be used as input. Combining the suggested approaches, our model obtains better results compared to high-performance algorithms for traffic forecasting, such as LSTM or the algorithms presented in the TRANSFOR19 forecasting competition. The model is capable of maintaining its performance over different time-horizons from 5 min to up to 4 h with multi-step predictions.

Density-weighted ant colony algorithm for ship trajectory reconstruction

With the development of the international ocean shipping and the increase in the automatic identification system-receiving equipment, the availability of ship trajectory data has shown explosive growth. To reduce the uncertainty in the application of sparse ship trajectory data in the open sea, this paper proposes the use of an ant colony algorithm based on node weights and edge weights for ship trajectory reconstruction. Ship trajectory is reconstructed from independent trajectory points through prior knowledge base construction, solving candidate path sets and optimal path generation. In addition, in order to further experiment the proposed method for the reconstruction of trajectory in the large-scale, this paper also uses the reconstructed trajectory to extract the distribution of ship traffic flow in the study area. The experimental results show that, the proposed trajectory reconstruction method can reconstruct ship trajectories in open seas more accurately.

Identifying Foreign Tourists’ Nationality from Mobility Traces via LSTM Neural Network and Location Embeddings

The interest in human mobility analysis has increased with the rapid growth of positioning technology and motion tracking, leading to a variety of studies based on trajectory recordings. Mapping the routes that people commonly perform was revealed to be very useful for location-based service applications, where individual mobility behaviors can potentially disclose meaningful information about each customer and be fruitfully used for personalized recommendation systems. This paper tackles a novel trajectory labeling problem related to the context of user profiling in “smart” tourism, inferring the nationality of individual users on the basis of their motion trajectories. In particular, we use large-scale motion traces of short-term foreign visitors as a way of detecting the nationality of individuals. This task is not trivial, relying on the hypothesis that foreign tourists of different nationalities may not only visit different locations, but also move in a different way between the same locations. The problem is defined as a multinomial classification with a few tens of classes (nationalities) and sparse location-based trajectory data. We hereby propose a machine learning-based methodology, consisting of a long short-term memory (LSTM) neural network trained on vector representations of locations, in order to capture the underlying semantics of user mobility patterns. Experiments conducted on a real-world big dataset demonstrate that our method achieves considerably higher performances than baseline and traditional approaches.

Data Cleansing Method for Sparse Trajectory Data: A Case Study of Shared Electric Bicycles in Tengzhou

Abstract. Location based service (LBS) technologies provides a new perspective for the spatiotemporal dynamics analysis of urban systems. Previous studies have been performed by using data of mobile communications, public transport vehicles (taxis and buses), wireless hotspots and shared bicycles. However, the analysis based on shared electric bicycles (e-bike) has yet to be studied in the literature. Data cleansing and the extraction of origin-destination (O-D) are prerequisites for the study of urban systems spatiotemporal patterns. In this study, based on a dataset that contains a week of shared e-bike GPS data in Tengzhou City (Shandong Province), sparse characteristics of discontinuities and non-uniformities of trajectory GPS and a lack of riding status are captured. Based on the characteristics and combining with the actual road, we proposed a method for the extraction of O-D pairs for every trajectory segments from continuous and stateless trajectory GPS data. This method cleans the incomplete and invalid trajectory records, which is suitable for sparse trajectory data. Finally, a week-long shared e-bike GPS data in Tengzhou City is scrubbed, and by sampling method, the extraction accuracy of 91% is verified. In summary, we provide a preliminary cleansing rules for the sparse trajectory data of shared e-bikes at the first time, which is highly reliable, and is suitable for data mining from other forms of sparse GPS trajectory data.

Green Travel Mode: Trajectory Data Cleansing Method for Shared Electric Bicycles

Location-based service (LBS) technologies provide a new perspective for the analysis of the spatiotemporal dynamics of urban systems. Previous studies have been performed using data from mobile communications, public transport vehicles (taxis and buses), wireless hotspots and shared bicycles. However, corresponding analyses based on shared electric bicycle (e-bike) have not yet been reported in the literature. Data cleaning and extraction of the origin-destination (O-D) are prerequisites for the study of the spatiotemporal patterns of urban systems. In this study, based on a dataset of a week of shared e-bike GPS data in the city of Tengzhou (Shandong Province), sparse characteristics of discontinuities and nonuniformities of the GPS trajectory and a lack of riding status are observed. Based on the characteristics and the actual road, we proposed a method for the extraction of O-D pairs for every trajectory segment from continuous and stateless trajectory GPS data. This method cleans the incomplete and invalid trajectory records, which is suitable for sparse trajectory data. A week of shared e-bike GPS data in Tengzhou is scrubbed and, by the sampling method, the extraction accuracy of 91% is verified. We provide preliminary cleaning rules for sparse trajectory shared e-bike data for the first time, which are highly reliable and suitable for data mining from other forms of sparse GPS trajectory data.

A Personal Location Prediction Method Based on Individual Trajectory and Group Trajectory

With the rapid development of communication technology, a large amount of spatiotemporal trajectory data has been produced. One of the critical applications of trajectory data is location prediction that is important for urban traffic planning and location-based services. Although many methods for personal location prediction have been proposed, to the best of our knowledge, some users always have sparse historical trajectory data in practical applications, resulting in poor location prediction precision. Targeting on this challenge, we propose an Individual Trajectory-Group Trajectory (ITGT) location prediction model by utilizing the pattern of group travels. First, the model performs the stay point extraction and conducts the spatial clustering to construct the clustering link. Second, Fano's inequality and clustering link are used to evaluate the predictability of location. Third, two variable order Markov models, named prediction by partial match (PPM) and probabilistic suffix tree (PST), are adopted to predict the clustering link. Finally, our approach is evaluated by using 608 712 points from 5000 volunteers at Shenzhen, China. The results show that: 1) when using individual trajectory, the PPM individual model is superior to the conventional N-order Markov model and PST individual model; 2) when all group trajectories are used, the PPM group model is not as accurate as the PPM individual model; and 3) when it classifies the group trajectory into different traffic zones, the PPM zone model is better than the PPM group model and the PPM individual model. The prediction precision of 1-3 order PPM zone model is 83.21%, 86.75%, and 87.35%, respectively, which introduces approximately 3% performance gains by utilizing the characters of traffic zone groups.

BVis: urban traffic visual analysis based on bus sparse trajectories

Public urban transport network has the characteristic of wide coverage, while buses have the features of stable running routes and static parking places, which is helpful to study urban traffic and bus station congestion patterns. Unlike the common GPS trajectory data, our data includes the pertinent records of the buses arrival and departure from the relevant bus stations due to data compression. In this paper, a visual analysis system called BVis is presented to analyze the urban traffic applying the large-scale real sparse buses dataset. This system covers the four modules of bus data visualization, first, the sparse trajectory data cleaning and mapping, second, the global traffic states and section traffic patterns analysis of roads, third, the bus station congestion patterns analysis using the station parking time, finally, an importance analysis of bus stations in the complex public transport network. Furthermore, an enhanced node importance evaluation algorithm is presented, which combines the dynamic properties of the bus station, such as traffic volume of station and station parking time. Using the real bus GPS dataset, three cases are described to demonstrate the performance and effectiveness of the system.

Identification of individual coherent sets associated with flow trajectories using coherent structure coloring.

We present a method for identifying the coherent structures associated with individual Lagrangian flow trajectories even where only sparse particle trajectory data are available. The method, based on techniques in spectral graph theory, uses the Coherent Structure Coloring vector and associated eigenvectors to analyze the distance in higher-dimensional eigenspace between a selected reference trajectory and other tracer trajectories in the flow. By analyzing this distance metric in a hierarchical clustering, the coherent structure of which the reference particle is a member can be identified. This algorithm is proven successful in identifying coherent structures of varying complexities in canonical unsteady flows. Additionally, the method is able to assess the relative coherence of the associated structure in comparison to the surrounding flow. Although the method is demonstrated here in the context of fluid flow kinematics, the generality of the approach allows for its potential application to other unsupervised clustering problems in dynamical systems such as neuronal activity, gene expression, or social networks.

Understanding the geometry of transport: Diffusion maps for Lagrangian trajectory data unravel coherent sets.

Dynamical systems often exhibit the emergence of long-lived coherent sets, which are regions in state space that keep their geometric integrity to a high extent and thus play an important role in transport. In this article, we provide a method for extracting coherent sets from possibly sparse Lagrangian trajectory data. Our method can be seen as an extension of diffusion maps to trajectory space, and it allows us to construct "dynamical coordinates," which reveal the intrinsic low-dimensional organization of the data with respect to transport. The only a priori knowledge about the dynamics that we require is a locally valid notion of distance, which renders our method highly suitable for automated data analysis. We show convergence of our method to the analytic transfer operator framework of coherence in the infinite data limit and illustrate its potential on several two- and three-dimensional examples as well as real world data.

A hybrid HMM model for travel path inference with sparse GPS samples

In this study, we propose a novel method for a travel path inference problem from sparse GPS trajectory data. This problem involves localization of GPS samples on a road network and reconstruction of the path that a driver might have been following from a low rate of sampled GPS observations. Particularly, we model travel path inference as an optimization problem in both the spatial and temporal domains and propose a novel hybrid hidden Markov model (HMM) that uses a uniform cost search (UCS)-like novel combinational algorithm. We provide the following improvements over the previous studies that use HMM-based methods: (1) for travel path inference between matched GPS positions, the proposed hybrid HMM algorithm evaluates all candidate paths to find the most likely path for both the temporal and spatial domains. In contrast, previous studies either create interpolated trajectories or connect matched GPS positions using the shortest path assumption, which might not be true, especially in urban road networks (Goh et al. 2012; Lou et al. 2009). (2) The proposed algorithm uses legal speed limits for the evaluation of discrepancy in the temporal domain as in Goh et al. (2012), and Lou et al. (2009) only if there is not sufficient historical average speed data; otherwise, we use historical average speed computed from data. Our experiments with real datasets show that our algorithm performs better than the state of the art VTrack algorithm (Thiagarajan et al. 2009), especially for cases where GPS data is sampled infrequently.

A rough-and-ready cluster-based approach for extracting finite-time coherent sets from sparse and incomplete trajectory data.

We present a numerical method to identify regions of phase space that are approximately retained in a mobile compact neighbourhood over a finite time duration. Our approach is based on spatio-temporal clustering of trajectory data. The main advantages of the approach are the ability to produce useful results (i) when there are relatively few trajectories and (ii) when there are gaps in observation of the trajectories as can occur with real data. The method is easy to implement, works in any dimension, and is fast to run.

Visual Exploration of Sparse Traffic Trajectory Data.

In this paper, we present a visual analysis system to explore sparse traffic trajectory data recorded by transportation cells. Such data contains the movements of nearly all moving vehicles on the major roads of a city. Therefore it is very suitable for macro-traffic analysis. However, the vehicle movements are recorded only when they pass through the cells. The exact tracks between two consecutive cells are unknown. To deal with such uncertainties, we first design a local animation, showing the vehicle movements only in the vicinity of cells. Besides, we ignore the micro-behaviors of individual vehicles, and focus on the macro-traffic patterns. We apply existing trajectory aggregation techniques to the dataset, studying cell status pattern and inter-cell flow pattern. Beyond that, we propose to study the correlation between these two patterns with dynamic graph visualization techniques. It allows us to check how traffic congestion on one cell is correlated with traffic flows on neighbouring links, and with route selection in its neighbourhood. Case studies show the effectiveness of our system.

Hierarchical spatiotemporal modeling for dynamic video trajectory analysis

Normalcy decision in video security involves highly uncertain phenomena due to its inherited insufficient knowledge, intrinsic ambiguity in human cognition, measurement error, etc. This paper presents a hierarchical spatiotemporal trajectory modeling for dynamic trajectory analysis using sparse trajectory data. The trajectory data is assumed to be given by people detection and tracking methods, which are also challenging issues due to occlusion, noise, illumination changes, etc. The proposed method partitions the trajectory feature space into the attributes of trajectory position, direction, and speed. Inherent uncertainty of video trajectory is tackled by employing the uncertainty propagation model of a trajectory segment and Markov random field in analyzing the uncertainty attributes of object movement direction and speed. The proposed method can be used in online learning for incremental adaptation as well as offline learning for optimality guarantee. The method was evaluated using both synthetic trajectories and video streams of multiple people movements acquired from multiple video cameras developed by GE Global Research Center and KAL beta sites. Extensive experiments were performed using video sequences in real world and synthesized trajectories, which achieved very encouraging results.