GPS Trajectory Data Research Articles

The proposal of a 15-minute city composite index through integrating GPS trajectory data-inferred urban function attraction based on the Bayesian framework

The chrono-urbanism framework suggests that urban life quality decreases with increased time spent in transportation, particularly motorized modes. The 15-min city concept, aligned with chrono-urbanism, emphasizes the importance of reaching essential urban functions within a short active travel time and has gained global attention. This study aims to develop a 15-min city composite index score (CIS) using the chrono-urbanism framework. The CIS integrates spatial accessibility to six key urban functions, providing a holistic assessment of the 15-min city status in Wuhan, China. Urban function accessibility is computed based on visitor volume inferred from GPS trajectory data using the Bayesian framework. Results reveal CIS hotspots in Hongshan, Wuchang, Jiangan, and Qingshan districts, with Hongshan having the highest concentration. However, Wuhan faces challenges in achieving comprehensive 15-min city status, as hotspots are mainly concentrated in specific areas like university towns and traditional city centers. The proposed assessment approach is applicable to accurately evaluate the 15-min status in other urban contexts using GPS trajectory data. The study's findings can assist policymakers in understanding CIS hotspot distribution and developing future planning policies to enhance the overall 15-min city status.

Towards effective, robust and utility-preserving watermarking of GPS trajectories

Personal GPS trajectory is essential for businesses and emerging data markets due to its relevance in various data-driven methods, including traffic forecasting, accident prediction, and profiling driving behavior. Watermarking is a method that facilitates verification of data ownership and authenticity by embedding provenance information into the data. Whereas watermarking is commonly adopted in the image and audio domains, only a few initial watermarking methods exist for GPS trajectory data. GPS trajectory watermarking is particularly challenging due to the spatio-temporal data properties and easiness of data modification. As a result, existing watermarking methods often embed only minimal provenance information, lack robustness, and can fail to preserve data utility for downstream applications. In this work, we propose W-Trace - a novel, effective, robust, and utility-preserving GPS trajectory watermarking method. W-Trace transforms a GPS trajectory into a complex domain and applies the Fourier transformation to decompose the trajectory into the frequency representation. W-Trace embeds watermarks in the frequency representation and verifies them in a spatiotemporally-aware procedure. We demonstrate the effectiveness, robustness, and utility of the proposed W-Trace approach in realistic settings using real-world GPS trajectory datasets. In contrast to the baselines, the proposed W-Trace approach is robust to a wide range of trajectory modifications while preserving the GPS trajectory characteristics required for the downstream applications.

Space–Time Analysis of Refueling Patterns of Alternative Fuel Vehicles Using GPS Trajectory Data and Machine Learning

ABSTRACTThe urgency to combat climate change has led countries worldwide to embrace clean energy solutions across various sectors, including transportation, to reduce greenhouse gas emissions. This shift is evident in the growing popularity and adoption of alternative fuel vehicles (AFVs), such as natural gas vehicles and electric vehicles. AFVs have significantly lower carbon footprints compared to conventional petrol‐powered vehicles with internal combustion engines. Consequently, there arises a need to gain a deeper understanding of AFV refueling demand to optimize the distribution of refueling stations. To address this, our research proposes an innovative space–time method that integrates GPS trajectory data with the support vector machine technique to accurately identify and analyze patterns in AFV refueling behavior. The results highlight distinct space–time patterns, notably the clustering of refueling activities in areas like Shuiguohu, Shouyilu, Houhu, Hanshuiqiao, Zhoutou, and Yongfeng around noon, influenced by taxi drivers' breaks. This underscores the importance of increasing staff levels at refueling stations in these areas during peak refueling periods, forming alliances with local eateries, and coordinating taxi shift hours to evenly distribute refueling demand throughout the day, ultimately reducing congestion during peak refueling periods in these areas. The proposed method by this research is applicable to urban contexts worldwide and equips policymakers and planners with a powerful tool for effective planning of future AFV refueling stations.

Spatial‐Temporal Distribution Characteristics of Linear Heritage Hiking Tourism Based on GPS Data Analysis: A Case Study of the Great Wall in Beijing, China

ABSTRACTHiking plays a significant role in experiencing linear heritage, and gaining insights into the spatiotemporal distribution of hikers' behavior is imperative for effectively utilizing heritage resources. This study employed GPS trajectory data to examine the spatiotemporal patterns of hiking behavior on the Great Wall in Beijing and investigate the utilization patterns of heritage resources in hiking tourism. The results revealed that: there were notable variations in trajectory quantity among different months with extended activity durations, early finishing times, and an absence of nighttime activities. Spatially, the trajectories exhibited dispersion and uneven distribution, leading to the underutilization of numerous heritage resources. The distribution of starting and ending points of trajectories demonstrated a substantial correlation with neighboring natural villages. Consequently, this study offers valuable insights for informing decision‐making processes related to the development, construction, and optimization of the scenic area along the Great Wall and the planning and design of tourist routes.

Map matching on low sampling rate trajectories through deep inverse reinforcement learning and multi-intention modeling

Analyzing freight vehicle movements using GPS trajectory data presents challenges due to environmental conditions and hardware limitations impacting data accuracy. Map matching, the process of aligning GPS signals with road networks, facilitates accurate route reconstruction. However, existing methods have limitations, particularly with low and ultra-low sampling rates. They often assume the shortest path between points, overlook historical data insights and neglect diverse driving behaviors, which may not align with real-world scenarios where shortest paths are not always optimal and different drivers exhibit varied behaviors. These limitations affect existing methods’ reliability, especially when we face low sampling rate trajectories. In this study, we propose multi-intention deep inverse reinforcement learning for map matching (MIDIRL) to address these challenges. MIDIRL integrates deep neural networks and multi-intention capturing mechanisms with inverse reinforcement learning to model complex driving preferences from historical trajectories, improving map matching accuracy, especially in ultra-low-frequency trajectories. Our experiments on real-world datasets demonstrate MIDIRL’s improved accuracy and efficiency of map matching compared to previous methods, even with limited training data.

Uncovering driving factors and spatiotemporal patterns of urban passenger car CO2 emissions: A case study in Hangzhou, China

While passenger cars offer the convenience of door-to-door travel, they are responsible for a substantial share of urban traffic-related carbon emissions, accounting for 44% of the global transport sector's emissions in 2022. Existing studies have not adequately identified the driving factors and spatiotemporal patterns of Passenger car CO2 Emissions (PCEs). To fill this gap, a comprehensive analytical framework for analysing PCE, encompassing several key components, has been developed. First, the PCE is estimated at a fine-grained level by utilising million-scale GPS trajectory data. Secondly, road network and urban functional features are extracted based on multi-source data, examining their nonlinear impacts on PCE. The driving factors of PCE during peak and off-peak periods are identified based on interpretable factor analysis. Finally, latent space clustering techniques are applied to reveal the spatiotemporal patterns of PCE, verified by significance tests. The effectiveness of this framework is demonstrated via a case study in Hangzhou, China. Notably, during peak hours, the driving factors of PCE are the road network's closeness centrality, service-related Points of Interest (POI) density, and trunk road density. In addition, five distinctive PCE patterns are identified. These findings facilitate the development of targeted and differentiated passenger car traffic management strategies, promoting carbon emission reduction in the transport sector.

Travel-mode inference based on GPS-trajectory data through multi-scale mixed attention mechanism

Identifying travel modes is essential for modern urban transportation planning and management. Recent advancements in data collection, especially those involving Global Positioning System (GPS) technology, offer promising opportunities for rapidly and accurately inferring users' travel modes. This study presents an innovative method for inferring travel modes from GPS trajectory data. The method utilizes multi-scale convolutional techniques to capture and analyze both temporal and spatial information of the data, thereby revealing the underlying spatiotemporal relationships inherent in user movement and behavior patterns. In addition, an attention mechanism is integrated into the model to enable autonomous learning. This mechanism enhances the model's capacity to identify and emphasize key information across different time periods and spatial locations, thus improving the accuracy of travel mode inference. Evaluation on the open-source GPS trajectory dataset, GeoLife, demonstrates that the proposed method attained an accuracy of 83.3%. This result highlights the effectiveness of the method, demonstrating that the model can more accurately understand and predict user travel modes through the integration of multi-scale convolutional technologies and attention mechanisms.

Design, Analysis and Detection of Urban Mobility Patterns in Guayaquil through Supervised Learning Algorithms

This work explores the use of GPS trajectory data (Git Hub) and supervised learning algorithms to analyze mobility patterns in the city of Guayaquil-Ecuador. The analysis reveals that identifying mobility patterns is crucial to improve urban planning and optimize public transport in terms of its innovations from 2024. The methodology has included data collection from various open access sources, data preprocessing and the use of a Random Forest Classifier to detect mobility patterns. The results indicate a model accuracy of 50%, opening the possibilities for future specialized research to optimize the model and collect more data to improve its performance and propose new StartUPS for the city. The data compiled supports the implementation of the Single Card for Public Transport (TUT), which modernizes and unifies the payment system, facilitating the collection of essential data for analysis. The research concludes that the integration of data from applications such as Waze and Moovit is crucial for efficient and sustainable urban planning and that supervised learning algorithms are effective tools for urban mobility analysis and leaving open the possibility of exploring additional data and other algorithms to improve the accuracy of models and the generation of new mobility routes.

AnimalEnvNet: A Deep Reinforcement Learning Method for Constructing Animal Agents Using Multimodal Data Fusion

Simulating animal movement has long been a central focus of study in the area of wildlife behaviour studies. Conventional modelling methods have difficulties in accurately representing changes over time and space in the data, and they generally do not effectively use telemetry data. Thus, this paper introduces a new and innovative deep reinforcement learning technique known as AnimalEnvNet. This approach combines historical trajectory data and remote sensing images to create an animal agent using deep reinforcement learning techniques. It overcomes the constraints of conventional modelling approaches. We selected pandas as the subject of our research and carried out research using GPS trajectory data, Google Earth images, and Sentinel-2A remote sensing images. The experimental findings indicate that AnimalEnvNet reaches convergence during supervised learning training, attaining a minimal mean absolute error (MAE) of 28.4 m in single-step prediction when compared to actual trajectories. During reinforcement learning training, the agent has the capability to replicate animal locomotion for a maximum of 12 iterations, while maintaining an error margin of 1000 m. This offers a novel approach and viewpoint for mimicking animal behaviour.

Association between outdoor jogging behavior and PM2.5 exposure: Evidence from massive GPS trajectory data in Beijing

Outdoor jogging is one of the most popular practised exercises worldwide, providing various benefits for health and wellbeing. However, PM2.5 exposure risks of jogging behaviors were rarely explored. This study aims to investigate the association between jogging behavior and PM2.5 exposure with big data. PM2.5 exposure concentration and dose inhalation of individuals were calculated by integrating hourly PM2.5 concentration data and jogging GPS trajectory recorded by a sports app during 2015 in Beijing, after which relationships between jogging behaviors and PM2.5 exposure were unpacked using statistics analysis and structural equation modelling. Experimental results on massive jogging trajectories show that: (1) the average jogging PM2.5 exposure concentration is 60.43 μg/m3, and female joggers inhaled significantly less air pollution dose (19.70 μg) than men (24.91 μg). (2) There exist significant spatiotemporal disparities in jogging exposure to PM2.5. Joggings in the city center, in the morning, on weekdays and in autumn and winter seasons were exposed to higher pollution concentrations. (3) Jogging behavior characteristics, especially distance, activity space size, duration and rotation, were systematically associated with PM2.5 exposure across space and time. (4) The role of gender directly shaped joggers' dose inhalation of PM2.5 pollution and indirectly via duration, timing choice and distance. (5) The effects of weather conditions on joggers' exposure to PM2.5 are mainly via direct effects, whereas the direct impacts of precipitation and wind speed are mitigated by indirect effects stemming from jogging behavior patterns. Our findings provide insights for personal guidance and policy intervention for the sake of promoting physical activity and reducing PM2.5 exposure.

How Social Media Data Mirror Spatio-Temporal Behavioral Patterns of Tourists in Urban Forests: A Case Study of Kushan Scenic Area in Fuzhou, China

Exploring the spatial distribution of tourist attractions and comprehending the spatio-temporal behaviors of tourists within tourist attractions can provide local planning agencies, destination marketing organizations, and government departments with essential evidence for decision-making processes. This study examines the spatio-temporal behavior patterns of tourists in the Kushan Scenic Area by analyzing GPS trajectory data acquired from social media platforms. The investigation primarily utilizes three research methodologies: grid analysis, Markov chain, and K-means clustering. The grid analysis results reveal three spatial distribution patterns within the scenic area, while the outcomes from the Markov chain and K-means clustering delineate six tourist movement patterns, along with three choices regarding travel time. This finding holds significant practical implications for enhancing the attractiveness of scenic areas, optimizing spatial layout, and improving tourists’ experiences.

Non-Uniform Spatial Partitions and Optimized Trajectory Segments for Storage and Indexing of Massive GPS Trajectory Data

The presence of abundant spatio-temporal information based on the location of mobile objects in publicly accessible GPS mobile devices makes it crucial to collect, analyze, and mine such information. Therefore, it is necessary to index a large volume of trajectory data to facilitate efficient trajectory retrieval and access. It is difficult for existing indexing methods that primarily rely on data-driven indexing structures (such as R-Tree) or space-driven indexing structures (such as Quadtree) to support efficient analysis and computation of data based on spatio-temporal range queries as a service basis, especially when applied to massive trajectory data. In this study, we propose a massive GPS data storage and indexing method based on uneven spatial segmentation and trajectory optimization segmentation. Primarily, the method divides GPS trajectories in a large spatio-temporal data space into multiple MBR sequences by greedy algorithm. Then, a hybrid indexing model for segmented trajectories is constructed to form a global spatio-temporal segmentation scheme, called HHBITS index, to achieve hierarchical organization of trajectory data. Eventually, a spatio-temporal range query processing method is proposed based on this index. This paper implements and evaluates the index in MongoDB and compares it with two other spatio-temporal composite indexes for performing spatio-temporal range queries efficiently. The experimental results show that the method in this paper has high performance in responding to spatio-temporal queries on large-scale trajectory data.

Unleashing the potential of urban jogging: Exploring the synergistic relationship of high-density environments and exercise on Residents' Health

This study investigates the influence of high-density urban environments on residents' health, focusing on jogging as a mediating activity. Based on the 5D framework (Density, Diversity, Design, Destination accessibility and Distance to Transit), this study examines GPS trajectory data from 1621 residents in Shanghai, analyses the relationship between high-density urban environments, jogging exercise, and residents' health, using jogging frequency, duration, and distance as mediators, and BMI as a health metric, alongside socio-environmental factors. Key findings include the design of the path environment has a direct positive impact on residents' health. Jogging frequency plays a negative mediating role between Density and residents' health, whereas Jogging distance has a positive masking effect between Density and health, but a negative masking effect between Distance of the path environment and health. Socio-environmental factors indirectly affect health indicators by adjusting the impact of built environment elements on jogging exercise. The Socio-environment has a regulatory effect on the mediating effect of jogging exercise. When the Socio-environment is better, the effects of the Design of the path environment on residents' health through Jogging frequency are more positive, but the effects on residents' health through Jogging distance are more negative. The duration and distance of jogging exercise have a nonlinear effect on health. These research results reveal how high-density urban environments affect residents' health by influencing jogging exercise, providing useful information for future formulation of healthy urban construction strategies.

Decision Model of Taxi Drivers at High-Speed Railway Stations Based on Profit Priority

ABSTRACT After dropping off passengers at the high-speed railway (HSR) station, taxi drivers face a crucial decision: whether to stay at the station or return empty. To solve the decision problem, this paper establishes a weighting model utilizing the Analytic Hierarchy Process (AHP) combined with the entropy weight method. This model determines the significance of various factors influencing drivers’ decisions. Subsequently, a decision model for taxi drivers at the HSR station is established by comparing the profitability of staying at the station versus leaving. Furthermore, the sensitivity analysis method is employed to quantitatively assess the model's dependency on influencing factors to confirm the model's effectiveness. Utilizing GPS trajectory data from Lanzhou taxis, this paper validates the decision-making model. The findings suggest that in the afternoon and evening, drivers are more likely to wait in the parking lot for passengers, whereas in the morning, they are more inclined to return empty.

Investigation of private and public bikes usage patterns considering GPS trajectory based cycling features

Understanding bike usage patterns is essential for increasing bike demand, evaluating the effectiveness of bike facilities, and improving the quality of the bike system. This study aims to identify factors that influence bike usage for zone-based departure and arrival trips. It also seeks to address knowledge gaps by comparing private and dock-based public bikes, as well as use on weekdays and weekends. This study employed GPS trajectory data from 12,106 private bikes and 29,776 dock-based public bikes over one week in Seoul, South Korea. The dataset comprises zone-based physical characteristics and cycling patterns for 19,153 census zones. To tackle the problem of excessive zero values for bike demand in many zones, zero-inflated negative binomial models were employed for analysis. The findings suggested that realized detours, arising from natural barriers and a lack of infrastructure, are associated with a reduction in arrival trips on weekends but an increase in departure trips on weekdays. Behavioral detours, linked to route preferences, tended to decrease usage of public bikes on weekdays. Departure and arrival trips increased in districts with a higher level of mixed land use and larger commercial areas on weekdays. Moreover, bicyclists tended to prefer districts with longer separated bikeways, while bike usage was likely to decrease in districts with a high floating population and vehicle lanes. These findings provide valuable insights into cycling usage patterns and associated factors, encouraging more extensive use of bikes.

Data-Driven Geofencing Design for Point-of-Interest Notifiers Utilizing Genetic Algorithm

This study proposes a method for generating geofences driven by GPS trajectory data to realize scalable point-of-interest (POI) notifiers, encouraging walking tourists to discover new local spots. The case study revealed that manual geofence settings degrade the location relevance and user coverage—key objectives of POI notifiers—and hinder the scalability and reliability of services. The formalization presented computationally equips geofence designers with practical solutions through two implementations based on prior GPS trajectory logs: (1) a multiobjective genetic algorithm that suggests cost-effective geofences by providing trade-off visualizations and (2) a user coverage-penalized genetic algorithm that determines an optimal geofence based on the designers’ expectations. The feasibility and stability of the proposed implementations were tested in areas with varying tourist flow patterns. A comparative survey among manual settings, settings incorporating a reliability simulation, and data-driven settings demonstrates significant performance improvements for geofence services.

Self-paced Gaussian-based graph convolutional network: predicting travel flow and unravelling spatial interactions through GPS trajectory data

ABSTRACT Spatial interaction research is particularly important for geographical analyses, as it plays a crucial role in extracting travel patterns. However, previous studies on spatial interactions have not adequately considered regional population variations over time, resulting in insufficiently precise travel predictions. Moreover, the threshold of spatial correlations is difficult to determine. Existing studies have assumed fully connected spatial correlation matrices, which is not realistic. To address these limitations, we proposed the Self-paced Gaussian-Based Graph Convolutional Network (SG-GCN) to automatically estimate the threshold of spatial correlations for travel flow predictions. It incorporates a temporal dimension into spatial relationship matrices to enhance the accuracy of vehicle flow predictions. In particular, Gaussian-based GCN identifies patterns in a time series of regional flows, enabling more precise capturing of spatial relationships while fusing node and edge features. Building on this model, self-paced contrastive learning automatically sets thresholds to determine the presence or absence of spatial relationships. The model's performance was verified through two empirical case studies conducted in New York City, USA, and Ningbo, China, using 2.8 million bicycle-sharing records and 1.25 million taxi trip records, respectively. The proposed model helps delineate mobility patterns in cities of varying scales and with different modes of transportation.

Measuring exposure and contribution of different types of activity travels to traffic congestion using GPS trajectory data

This study proposes a data-driven framework for understanding the space-time patterns of exposure and contribution of different activities to traffic congestion in urban road networks by using GPS trajectory and Point-of-Interest (POI) big datasets. Taking taxi trips related to traffic congestion in Wuhan, China as a case study, we first infer the types of individual activities from GPS trajectories and POIs and identify traffic congestion on each road. Then we develop two indicators to measure the congestion exposure of different activity types. Further, we reveal the space-time patterns of activity-related congestion through spatiotemporal analysis of the indicators of traffic congestion associated with different activities. The findings of this study shed light on how different types of activities contribute to the space-time heterogeneity of traffic congestion, and highlight the significance of considering the space-time patterns of congestion related with different activity types in urban transportation management.

Study on Spatiotemporal Behavior of Self-driving Tourists in Inner Mongolia

Against the backdrop of normalized tourism and upgraded consumption, self-driving tours have become one of the main choices for tourists. Faced with the booming self-driving tour market, it is of great significance to explore the spatial behavior characteristics and temporal constraint factors of self-driving tourists within a geographical range using GPS trajectory data. This exploration aids in the planning and design of self-driving tour routes, tourism products, and promotes the healthy and sustainable development of the self-driving tour industry. This paper selects Inner Mongolia Autonomous Region as the self-driving tour hotspot area, collects GPS data of self-driving tourists from 2017 to 2022 on the online platform "Foooooot" using Python, and conducts in-depth exploration and comparative analysis of the spatiotemporal behavior trajectories of self-driving tourists, aiming to explore the spatiotemporal behavior characteristics of self-driving tourists and their constraining factors.

Streamlining trajectory map-matching: a framework leveraging spark and GPU-based stream processing

Real-time online trajectory map-matching has emerged as a critical component in the era of location-based services (LBS) and intelligent transportation systems (ITS). It refers to the process of aligning a user’s GPS trajectory data with the corresponding road network in real-time. This technology has significant implications for various industries and applications. As our reliance on LBS and ITS continues to grow, the demand for faster, more accurate, and more reliable trajectory map-matching methods becomes increasingly important. Contemporary online map-matching predominantly employs stream processing techniques. Based on stream processing frameworks, we propose a heterogeneous hybrid architecture for map-matching. The architecture integrates Spark Streaming and graphics processing unit (GPU) heterogeneous computing for the first time. The hidden Markov model is employed as the map-matching algorithm, and Spark Streaming serves as the distributed processing platform. We conduct map-matching experiments using a GPS taxi trajectory dataset in Beijing’s Haidian District. The results demonstrate that in comparison to other analogous research, our framework’s performance has increased by over ten times, possessing a superior data processing capability and lower latency. This research provides a novel approach of stream-based heterogeneous computation for processing large-scale geographic data.