Distributed MobilityDB [2] (DistMobilityDB) extends MobilityDB [7], the PostgreSQL/PostGIS-based moving-object database, into a clustered, SQL-native system for large spatiotemporal trajectory datasets. It introduces a distributed architecture that (1) partitions data with spatiotemporal awareness, (2) plans and executes a wide range of distributed spatiotemporal queries (range, distance/knn, intersection, self-joins) across worker nodes, and (3) is delivered as a PostgreSQL extension so it can be deployed on existing servers without recompilation or downtime. Experiments on real (AIS ship) and synthetic (BerlinMOD) datasets across on-prem and cloud clusters show that DistMobilityDB achieves scalable performance while preserving MobilityDB's rich SQL semantics.

The propagation of Internet-of-Things (IoT) sensors aboard moving objects has led to continuous spatiotemporal data streams that demand on-device, low-latency analysis. However, the underlying systems for processing this streaming data are ill-prepared. On the one hand, common stream processing engines lack support for spatiotemporal operations. On the other hand, existing libraries for spatiotemporal data are optimized for historical data rather than real-time processing. To bridge this gap, we present MobilityNebula, an integration between MEOS (Mobility Engine Open Source), a lightweight C library for spatiotemporal data management, and NebulaStream, a stream processing system designed for the edge-fog-cloud continuum. We evaluated MobilityNebula by deploying the system on edge devices, ingesting data from the Belgian railway operator (SNCB) trains, and performing real-time geospatial processing for point-based queries.

Geospatial data analysis plays a crucial role in tackling intricate societal challenges such as urban planning and climate modeling [17]. However, employing tools like GeoPandas, a prominent Python library for geospatial data manipulation, necessitates expertise in complex domain-specific syntax and workflows. GeoPandas-AI addresses this gap by integrating LLMs directly into the GeoPandas workflow, transforming the GeoDataFrame class into an intelligent, stateful class for both data analysis and geospatial code development. This paper formalizes the design of such a smart class and provides an open-source implementation of GeoPandas-AI in PyPI package manager. Through its innovative combination of conversational interfaces and stateful exploitation of LLMs for code generation and data analysis, GeoPandas-AI introduces a new paradigm for code-copilots and instantiates it for geospatial development.

Foundation models are transformative in artificial intelligence, but building them from scratch, especially for mobility trajectories, is not yet clear or documented. This tutorial bridges this gap by demonstrating the steps and code of a minimal implementation of a trajectory-focused foundation model starting from GPT-2. Through a concise, step-by-step, code-driven process, we demonstrate adapting GPT-2 for spatiotemporal data. We then review and compare representative trajectory foundation models, such as TrajFM and TrajGPT, highlighting their architectural innovations and differences. Additionally, we introduce complementary techniques from related domains, like TimesFM's patching approach. Targeted at researchers and practitioners, this tutorial aims to explain the concepts and terminology of foundation models, at the implementation level. We find it timely and indispensable to create this educational material in order to support the SIGSPATIAL community in building and evaluating mobility foundation models, enhancing both research clarity and peer-review effectiveness in mobility AI.

Accurate estimation of urban traffic conditions is essential for effective transportation planning and smart city management. Conventional traffic monitoring typically relies on infrastructure-based sensors. Despite their precision, these sensors require costly maintenance, and provide limited spatial coverage. To overcome these challenges, probe vehicle data offer an appealing alternative due to their broader spatial reach.

Today it is simple to collect and transmit GNSS data from vehicles with a high frequency. However, there is a storage and processing cost related to handling the data. Further, some data has limited value, e.g., redundant GNSS data from a vehicle stopped at an intersection. In this paper, sampling methods for GNSS data focusing on time, distance, speed, and heading changes are systematically analyzed. The goal is to retain only valuable data. A set of metrics is proposed to quantify the value of the data, e.g., no redundancy and retention of the spatial and temporal distributions. An existing commercial approach to GNSS-based travel time computation in road networks is used to measure if the sampled GNSS is accurate for this important purpose. The results show that sampling methods using individual properties, such as time, space, or speed, have their own strengths and weaknesses. However, with hybrid methods, it is possible to retain the strengths and eliminate most weaknesses. Using a large, real-world GNSS dataset, we show that a hybrid method that retains only 20 % of the original data can achieve travel time estimation with an error of just 1.0 – 1.3%.

The ship trajectories collected by the Automatic Identification System (AIS) are widely used in maritime applications. However, a significant issue with AIS data is that large AIS gaps occur. Existing trajectory imputation methods for AIS data have three main limitations: (1) the temporal aspect is ignored; (2) the methods fall short when dealing with complex ship movements; (3) the common-route assumption does not always hold. To overcome these limitations, we propose TrajImpMC, a tracking-based framework that uses polygon-based ship location estimates from multiple cameras to impute large AIS gaps. TrajImpMC combines speed constraints and Kalman filters, and can return imputed trajectories that contain both spatial and temporal information. Extensive experiments are conducted on real datasets. In terms of the quality of the imputed trajectories, TrajImpMC improves the RMSE errors by at least one order of magnitude over two existing state-of-the-art AIS imputation methods. In addition, a visual comparison shows that the imputed trajectories of TrajImpMC align very well with the real ship trajectories during AIS gaps. The code for this paper is available at: https://github.com/songwu0001/TrajImpMC.

Monitoring roundabouts and signalized intersections in a road network is important, e.g., to reduce travel time and greenhouse gas emissions. The monitoring of such junctions is a challenging problem, and current approaches mainly use high-cost solutions for a selected few. In this work, we present a framework for the automated identification and monitoring of all junctions in a road network. The framework utilizes detailed trajectory data or high-level segment-based data to compute travel time and energy consumption for all turn directions. These metrics are then aggregated per junction to enable a fair comparison between roundabouts and intersections. The aggregated metric is used to provide an overview of all junctions and to pinpoint those performing poorly. An analysis of 1,394 junctions using 334,081 trajectories quantifies the different benefits of roundabouts and intersections, e.g., the travel time in roundabouts varies little, and turns are 21% to 155% more energy-consuming than going straight in intersections. Further, the aggregated junction metric makes it simple to monitor all analyzed junctions and detect the worst-performing. The analysis also clearly shows the benefits of trajectory data over segment-based data for junction monitoring.

The idea of generalized indices is one of the success stories of database systems research. It has found its way to implementation in common database systems. GiST (Generalized Search Tree) and SP-GiST (Space-Partitioned Generalized Search Tree) are two widely-used generalized indices that are typically used for multidimensional data, e.g., to index spatial or spatio-temporal data. Currently, the generalized indices GiST and SP-GiST represent one database object using one index entry, e.g., a scalar value or abounding box for each spatial or spatio-temporal object. However, when dealing with complex objects, e.g., moving object trajectories, a single entry per object is inadequate for creating efficient indices. Previous research has highlighted that splitting trajectories into multiple sub-trajectories or bounding boxes prior to indexing can enhance query performance as it leads to a higher index filter. In this paper, we introduce MGiST and MSP-GiST, the multi-entry generalized search tree counterparts of GiST and SP-GiST, respectively, that are designed to enable the partitioning of objects into multiple entries during insertion. The methods for decomposing a complex object into multiple sub-objects differ from one data type to another, and may depend on some domain-specific parameters. Thus, MGiST and MSP-GiST are designed to allow for pluggable modules that aid in optimizing the split of an object into multiple sub-objects. We demonstrate the usefulness of MGiST and MSP-GiST using a trajectory indexing scenario, where we realize several trajectory indexes using MGiST and MSP-GiST and instantiate these search trees with trajectory-specific splitting algorithms. We create and test the performance of several multi-entry versions of widely-used spatial index structures, e.g., R-Tree, Quad-Tree, and KD-Tree. We conduct evaluations using both synthetic and real-world data, and observe up to an order of magnitude enhancement in performance of point, range, and nearest neighbor queries.

Recent advances, especially in deep learning, allow to effectively detect ship targets in surveillance videos. However, the translation of these detections to the real-world locations of ships has not been sufficiently explored. The common approach in the literature is using a transformation matrix to convert a pixel to a real-world coordinate. However, this approach has three shortcomings: first, a set of reference point pairs has to be manually prepared to establish the matrix; second, the matrix always maps a pixel to the same real-world coordinate, ignoring that there is no one-to-one correspondence between discrete pixel coordinates and continuous real-world coordinates; third, this approach can only work with one camera. In light of this, we propose a technique PixelToRegion that explicitly takes into account the uncertainty in coordinate conversion by mapping each pixel to a spatial polygon. Next, we propose a new algorithm MCbSLE that can estimate ship locations using pixel sets from multiple cameras. The precision of location estimation by MCbSLE is enhanced through spatial intersection between polygons from different cameras. Experiments are conducted under 16 carefully designed multi-camera settings to evaluate MCbSLE w.r.t. four factors: different ports, the number of cameras, the distance between cameras, and camera headings. Results on one-day ship trajectory data show that (1) an 79.8% accuracy in the number of coordinates can be achieved by MCbSLE when there are no more than 10 ships in camera views; (2) using multiple cameras can improve the precision of location estimation by one order of magnitude compared with using one camera.

In this paper, we present an architecture for mobility data spaces enabling trustworthy and reliable data operations along with its main constituent parts. The architecture makes use of a data lake for scalable storage of diverse mobility data sets, on top of which separate computing and storage layers are implemented to allow independent scaling with a data operations toolbox providing all data operations. Furthermore, to cater for mobility analytics, machine learning and artificial intelligence support, an edge analytics suite is provided that encompasses distributed algorithms for mobility analytics and federated learning, thereby exploiting edge computing technologies. In turn, this is supported by a resource allocator that monitors the energy consumption of data-intensive operations and provides this information to the platform for intelligent task placement in edge devices, aiming at energy-efficient operations. As a result, an end-to-end platform is proposed that combines data services and infrastructure services towards supporting mobility application domains, such as urban and maritime.

Urban mobility evolution poses challenges that necessitate the integration of quality-assured, diverse data. Although open data initiatives provide some resources, the published data often lacks utility and quality. This challenge is further magnified by the lack of specialized skills needed to process mobility data. This paper introduces the Brussels Mobility Observatory (BMT), a platform for publishing mobility data in Brussels. The platform collects, archives, enriches and republishes open mobility data, serving as a centralized access point for mobility-related information. It presents the opportunity for bringing research results in data cleaning and fusion to end users in authorities and businesses, transforming challenges into opportunities for innovation in urban mobility.

Accurate estimation of shipping CO2 emissions is important for developing regulations to combat the greenhouse effect. Many shipping CO2 emissions models have been proposed in the past decades. However, most of them are only validated for a few specific ships, and there is a lack of data-driven validation and comparison of these models on a large scale. To fill this gap, this study proposes a general evaluation framework to quantitatively validate and compare different emission models. This framework is based on data integration of three types of data sources: ship technical details, AIS trajectory, and weather. Along with emission models, these data are fed into three carefully-designed modules that perform analysis at both grid and trajectory level as well as use annually aggregated fuel consumption ground truth. Extensive experiments are conducted on one-month data from 1,571 ships passing Danish waters to demonstrate the utility of the framework and insights into the accuracy of five popular CO2 emission models are presented.

Detection of fishing activities in trajectory data is important for authorities to develop fishery management policies and combat illegal, unreported, and unregulated (IUU) fishing at sea. However, the complex movement patterns of fishing activities challenge existing trajectory segmentation approaches, which may not identify complete fishing activities. In light of this, we propose a window-based trajectory segmentation algorithm which aims to detect fishing activities as completely as possible. Firstly, we introduce a visualization-based technique TPoSTE to help design features characterizing different movement patterns. Secondly, a window-based segmentation algorithm WBS-RLE is proposed to split a trajectory into fishing and non-fishing segments. WBS-RLE first utilizes a pre-trained classifier to label windows in a trajectory as fishing or non-fishing, then it uses the run-length encoding technique to merge those labeled windows into complete fishing activities. The effectiveness of our approach and its advantages over existing approaches are evaluated on a real-world trajectory dataset.

Various applications process geospatial trajectories of moving objects, such as cars, ships and robots. There is thus a need for a common conceptual framework to model and manage these objects, as well as to enable data interoperability across tools. The International Organization for Standardization ISO® has responded to this need and created the standard ISO 19141- Schema for moving features. Among its types, it defines a schema for rigid temporal geometries, which represent the movement of spatial objects translating and rotating over time, while preserving a fixed shape. Despite the abundance of these objects in real-world, there exists no reference implementation of this type of data in a common system, which causes them to usually be represented as temporal points without taking into account their spatial extents and shapes. In this paper, we aim to provide an implementation of rigid temporal geometries into MobilityDB, an open-source moving object database, that extends PostgreSQL and PostGIS. We provide a data model for rigid temporal geometries and propose efficient algorithms for the operations defined in ISO 19141. A use case on real AIS ship trajectories is illustrated to validate the proposed implementation. A synthetic data generator for temporal geometries is also proposed. Finally, we review the standard from an implementation point of view and provide insights on possible improvements.

Nowadays, the collection of moving object data is significantly increasing due to the ubiquity of GPS-enabled devices. Managing and analyzing this kind of data is crucial in many application domains, including social mobility, pandemics, and transportation. In previous work, we have proposed the MobilityDB moving object database system. It is a production-ready system, that is built on top of PostgreSQL and PostGIS. It accepts SQL queries and offers most of the common spatiotemporal types and operations. In this paper, to address the scalability requirement of big data, we provide an architecture and an implementation of a distributed moving object database system based on MobilityDB. More specifically, we define: (1) an architecture for deploying a distributed MobilityDB database on a cluster using readily available tools, (2) two alternative trajectory data partitioning and index partitioning methods, and (3) a query optimizer that is capable of distributing spatiotemporal SQL queries over multiple MobilityDB instances. The overall outcome is that the cluster is managed in SQL at the run-time and that the user queries are transparently distributed and executed. This is validated with experiments using a real dataset, which also compares MobilityDB with other relevant systems.

Mobility applications involve large amounts of data that must be managed and queried in a scalable way. MobilityDB1 is an SQL moving object database system. It is an extension of PostgreSQL and PostGIS that supports storing and querying mobility data. This paper demonstrates the distributed query management capabilities in MobilityDB using a cluster that contains 2 billion real AIS ship trajectory points obtained from the Danish Maritime Authority.

Despite two decades of research in moving object databases and a few research prototypes that have been proposed, there is not yet a mainstream system targeted for industrial use. In this article, we present MobilityDB, a moving object database that extends the type system of PostgreSQL and PostGIS with abstract data types for representing moving object data. The types are fully integrated into the platform to reuse its powerful data management features. Furthermore, MobilityDB builds on existing operations, indexing, aggregation, and optimization framework. This is all made accessible via the SQL query interface.

This paper demonstrates the MobilityDB moving object database system. It is an extensive implementation on top of PostgreSQL and PostGIS with multiple novel aspects. MobilityDB defines multiple spatiotemporal types for moving geometry and geography points, as well as for temporal integers, reals, Booleans, and strings. It also defines a rich set of operations on these types. The types are supported with spatiotemporal index access methods by extending GiST (Generalized Search Tree) and SP-GiST (Space Partitioning GiST). The query interface is SQL. MobilityDB thus extends the PostgreSQL optimizer with statistics collectors and selectivity estimation functions. It is available as open source. The demonstration includes a scenario with multiple queries, and a publicly accessible query interface on the Web.

A data marketplace is a platform that facilitates online data trading. It gives sellers the capabilities of publishing data, defining the price function, the authorization rules and the data contract. It gives customers the capabilities to search for data services, to query for their price, to invoke them and to pay for the usage. As data are more and more emphasized as a payable asset, the need for data marketplaces rises. There is a lack of marketplace proposals for spatial data. This paper contributes to fill-in this gap. We propose a data model for representing the different types of data involved. The logical representation of data is a quadtree, and the physical representation fits into a key-value store. The model is thus logically simple and extensible, and physically scalable and highly available. We discuss price queries in detail and propose evaluation algorithms.

In previous work, we have proposed a tool for Spatiotemporal Pattern Query. It matches individual moving object trajectories against a given movement pattern. For example, it can be used to find the situations of Missed Approach in ATC data (Air Traffic Control systems, used for tracking the movement of aircrafts), where the landing of the aircraft was interrupted for some reason. This tool expresses the pattern as a set of predicates that must be fulfilled in a certain temporal order. It is implemented as a Plugin to the Secondo DBMS system. Although the tool is generic and flexible, domain expertise is required to formulate and tune queries. The user has to decide the set of predicates, their arguments, and the temporal constraints that best describe the pattern. This paper demonstrates a novel solution where a Visual Analytics system, V-Analytics, is used in integration with this query tool to help a human analyst explore such patterns. The demonstration is based on a real ATC data set.

In this paper, we demonstrate our results of designing and implementing spatiotemporal pattern queries. That is, one can specify temporal order constraints on the fulfillment of predicates on moving objects. Compared with other approaches, our approach is not restricted in terms of the types of moving object or the sets of predicates. Our work covers the language integration in SQL, the evaluation of the queries, and the integration with the query optimizer. We provide a complete implementation in C++ and Prolog in the context of the SECONDO platform. The implementation is made publicly available online as a SECONDO Plug in We have also made available automatic scripts for executing the examples in this demonstration.